RR280 - Real time evaluation of health and safety management in the National Health Service
Transcript of RR280 - Real time evaluation of health and safety management in the National Health Service
HSEHealth & Safety
Executive
Real time evaluation of health and safetymanagement in the National Health Service
Prepared by the Occupational Health and Safety AdvisoryService (OHSAS) and the University of Aberdeen
for the Health and Safety Executive 2004
RESEARCH REPORT 280
HSEHealth & Safety
Executive
Real time evaluation of health and safetymanagement in the National Health Service
Karen J M Niven PhD MSc FFOH FIOSH RSPOccupational Health and Safety Advisory Service
(OHSAS) and University of AberdeenNavy HouseStuart Road
Rosyth Industry ParkRosyth
Fife KY11 2BJ
The design of an impact evaluation of a health and safety management system in the National HealthService (NHS) was informed by a review of relevant literature, which identified a design comprising sixmain aspects. These were a longitudinal design; inclusion of comparison groups; an intervention that wasof interest to the NHS; a participative style; multiple measurement methods and multiple indicators ofeffectiveness.
Field study data was generated using a prospective longitudinal before-and-after design with a multiplebaseline. Seven NHS Trusts participated; two of which were used as comparison groups. The interventionwas a safety management workbook, introduced only to the test group. Evaluation of the workbook impacton safety performance involved two identical phases, approximately 12 months apart. Each phasecomprised of a staff opinion questionnaire survey, based on previously validated work; and a new HSEmethodology involving analysis of accident data to derive costs, which could be linked to management rootcauses.
The most frequently encountered system failure was that of risk assessment, with planning, implementing,measuring and reviewing as the main root causes. Estimated extrapolated costs were between 0.06% and1.44% of the running costs of the NHS. Responses from the questionnaires showed significant differencesbetween the Trusts and a significant improvement in staff opinion in some safety climate dimensions.
This report and the work it describes were funded by the Health and Safety Executive (HSE) Withadditional funding/support from: Scottish Executive Health Department, Directorate of HumanResources,University of Aberdeen, Tayside Primary Care NHS Trust and Fife Primary Care NHS Trust. Itscontents, including any opinions and/or conclusions expressed, are those of the authors alone and do notnecessarily reflect HSE policy.
HSE BOOKS
ii
© Crown copyright 2004
First published 2004
ISBN 0 7176 2911 2
All rights reserved. No part of this publication may bereproduced, stored in a retrieval system, or transmitted inany form or by any means (electronic, mechanical,photocopying, recording or otherwise) without the priorwritten permission of the copyright owner.
Applications for reproduction should be made in writing to:Licensing Division, Her Majesty's Stationery Office, St Clements House, 2-16 Colegate, Norwich NR3 1BQ or by e-mail to [email protected]
iii
ACKNOWLEDGEMENTS
The research described in this report has a single author. This is potentially misleading and does little justice to the individuals and organisations without whose participation, commitment and enthusiasm this work would have been impossible. My grateful thanks and unbounded admiration for their tenacity are extended to them.
The funding bodies The Health and Safety Executive The Scottish Executive, Department of Health Tayside Primary Care NHS Trust Fife Primary Care NHS Trust
The research team Elaine Ferguson – Research Assistant Louise Webster – Research Administrative Assistant
The staff and long-suffering principal contacts within the participating Trusts Dumfries and Galloway Acute and Maternity Hospitals NHS Trust Fife Primary Care NHS Trust Highland Primary Care NHS Trust Lomond and Argyll Primary Care NHS Trust Lothian University Hospitals NHS Trust Tayside Primary Care NHS Trust Tees and North East Yorkshire NHS Trust
Other sources of advice and encouragement John Cairns – University of Aberdeen, Health Economics Research Unit John Cherrie – University of Aberdeen, Department of Environmental and Occupational Medicine Liz Archibald – Cormack Consulting Amanda Ridings - Originate Alistair Cheyne - Loughborough University, Centre for Hazard and Risk Management Andrew Turner – Focused Marketing
iv
v
CONTENTS
ACKNOWLEDGEMENTS ................................................................................................iii
EXECUTIVE SUMMARY .............................................................................................. ix
INTRODUCTION ............................................................................................................xiii
1 LITERATURE REVIEW .........................................................................................1
1.1 Literature review methodology ……………………………………….......11.2 Findings ....................................................................................................2
1.2.1 Literature review objective 1 ...........................................................2 1.2.2 Literature review objective 2 ...........................................................4 1.2.3 Literature review objective 3 ...........................................................5
2 METHODS ...............................................................................................................7
2.1 Practical factors relevant to the research design ......................................72.1.1 Variability between NHS Trusts.......................................................7
2.1.2 Recruitment of research participants ...............................................7 2.1.3 Relationship between researchers and Trusts....................................82.2 Evaluation methodology design (Research objective 1)............................9
2.2.1 Objective measurement component ..................................................9 2.2.2 Subjective measurement component............................................... 11
2.3 Intervention design (Research objective 2) ............................................. 14 2.3.1 Method of introduction ................................................................ 15
2.4 Research design to measure change in health and safety performance (Research objective 3) ........................................................ 16
2.4.1 Method of monitoring workbook use.............................................. 17 2.4.2 Study composition ......................................................................... 18
2.5 Data analysis strategy .............................................................................. 21 2.5.1 Incident data ................................................................................. 21 2.5.2 Questionnaire data ........................................................................ 22 2.5.3 Workbook monitoring data ............................................................ 23
3 RESULTS ............................................................................................................... 24
3.1 Trust descriptive/observational results.................................................... 243.2 General incident findings......................................................................... 26
3.2.1 “Special” category........................................................................ 31 3.3 Risk control system failures and root causes of incidents....................... 37
3.3.1 Risk control system failures (HSE level 2)...................................... 38 3.3.2 Management root causes (HSE level 1).......................................... 41
3.4 Cost of incidents....................................................................................... 44 3.4.1 Staff/patient incidents .................................................................... 46
3.4.2 Special category ............................................................................ 48
vi
3.4.3 Management root causes of incidents ............................................ 49 3.4.4 Categories, factors and types of incidents ...................................... 53
3.5 Questionnaire data................................................................................... 59 3.5.1 Response rates .............................................................................. 60 3.5.2 Questionnaire confirmatory factor analysis ................................... 62 3.5.3 Dimension score responses............................................................ 63
3.5.4 Analysis of responses by Trust ....................................................... 64 3.5.5 Analysis of responses by job category ............................................ 65 3.5.6 Analysis of responses by individual members of staff ..................... 66
3.5.7 Questionnaire respondents’ additional comments .......................... 67 3.6 Results from workbook usage monitoring (Project phase 3).................. 67
3.6.1 Telephone survey........................................................................... 67 3.6.2 Visual inspections.......................................................................... 71
3.7 Triangulation of results ........................................................................... 72 3.8 Summary of findings................................................................................ 76
3.8.1 General findings............................................................................ 76 3.8.2 Special category ............................................................................ 77 3.8.3 Identification of risk control system failures and root causes of incidents.................................................................................... 77
3.8.4 Costs of incidents .......................................................................... 77 3.8.5 Staff opinion survey ....................................................................... 78 3.8.6 Monitoring and audit of the use of the management system workbook ...................................................................................... 79
3.8.7 Triangulation of results ................................................................. 79
4 DISCUSSION ......................................................................................................... 80
4.1 Health and safety management performance evaluation methodology.. 80 4.1.1 Performance evaluation based on the use of reported incidents ..... 80
4.1.2 Use of the HSE root cause analysis methodology........................... 82 4.1.3 Investigator bias and its implications for the training and development of safety advisers ........................................................................... 85
4.1.4 Costs and economic aspects .......................................................... 86 4.1.5 Health and safety culture............................................................... 89
4.1.6 Triangulation of health and safety measurement methods .............. 90 4.2 Health and safety management intervention .......................................... 91
4.2.1 The degree of use of the workbook................................................. 92 4.2.2 The time interval between the benchmarking phases ...................... 92
4.3 Health and safety management performance change measurement methodology ............................................................................................. 94
4.3.1 Intervention objectives................................................................... 94 4.3.2 Conceptual basis ........................................................................... 94 4.3.3 Research format and design........................................................... 94
4.3.4 Threats to external validity ............................................................ 95 4.3.5 Threats to internal validity ............................................................ 95
4.3.6 Outcome measurement .................................................................. 97 4.3.7 Other limitations ........................................................................... 97
4.4 Assessment of the degree to which the research aim and objectives have been achieved ........................................................................................... 97
vii
4.5 Summary of conclusions and recommendations ..................................... 98 4.5.1 Health and safety management performance evaluation methodology.................................................................................. 98 4.5.2 Health and safety management intervention................................... 99 4.5.3 Health and safety management performance change measurement
methodology................................................................................ 100
APPENDICES
APPENDIX 1 HSE ROOT CAUSE ANALYSIS METHODOLOGY ......................... 101 APPENDIX 2 COST INCLUSIONS AND EXCLUSIONS.......................................... 116 APPENDIX 3 COVERING LETTERS, QUESTIONNAIRE AND
QUALITY CONTROL SYSTEM FOR INPUTTING QUESTIONNAIRE DATA.................................................................... 118 APPENDIX 4 MANAGEMENT SYSTEM WORKBOOK.......................................... 125 APPENDIX 5 PROTOCOL FOR BRIEFING OF WORKBOOK HOLDERS ............................................................................................. 158 APPENDIX 6 PROTOCOL FOR FOLLOW-UP OF WORKBOOK HOLDERS ............................................................................................. 163 APPENDIX 7 INCIDENT DESCRIPTIVE DATA ...................................................... 166 APPENDIX 8 SPECIAL CATEGORY INCIDENT DESCRIPTIVE DATA ............. 186 APPENDIX 9 INDIVIDUAL TRUSTS BREAKDOWN OF TELEPHONE INTERVIEW RESPONSES ................................................................. 208
REFERENCES ................................................................................................... 213
BIBLIOGRAPHY ................................................................................................... 218
viii
ix
EXECUTIVE SUMMARY
A high priority for the NHS is ensuring patient safety and minimising staff absence due to incidents and work-related illness. The research aim was to evaluate whether it was possible to reliably measure change in health and safety performance when a formal health and safety intervention was introduced to the UK healthcare sector. The aim was supported by three research objectives, all of which were informed by a review of relevant literature.
The first objective was to design a health and safety performance measurement methodology. This included a new method, developed by the HSE, for assessing costs, and health and safety management system root causes of reported incidents that met a predetermined set of inclusion criteria. This quantitative data was supported by a qualitative set via a staff opinion questionnaire survey, based on previously validated work, administered to a stratified 10% sample of staff from participating Trusts.
The second objective was to evaluate the use of a health and safety management system workbook, developed by the author for the research. Use of this was intended to improve health and safety management performance. It was issued to selected staff from participating Trusts, who were asked to use it for approximately a year. Workbook users were monitored during this period by telephone interviews and visual inspection of workbooks. Some Trusts already had a workbook-based health and safety management system in place. They were used as a comparison group and did not receive the workbook or the follow-up monitoring.
The third objective was to design and implement a field study to measure change in health and safety performance. Seven NHS Trusts participated in this research; two being the comparison group. The method developed to measure health and safety performance, was used to establish a baseline of health and safety performance in all Trusts (phase 1), before the workbook was introduced to the test group. At the end of the research, health and safety performance was again measured (phase 2). The results, including the monitoring data on level of use of the workbook, were examined to establish whether change in health and safety performance could be detected. Field study data was therefore generated using a design which included: use of a before-and-after (longitudinal) design; use of comparison groups; use of an intervention that was of interest to participating NHS Trusts; use of a participative style which involved the participating Trusts; use of multiple measurement methods and multiple indicators of effectiveness.
The results from the performance measurement methods (to support the first objective) showed that the risk control system of risk assessment was the major source of management root causes in both phases. These root causes were associated with planning, implementing, measuring and reviewing of the management system. Therefore improvements in the planning and implementation of risk management should be a priority for action within the NHS. Where management root causes could be assigned, incidents involving staff were nearly three times more frequent and represented approximately four times the cost than for patients. The total cost assigned to accidental injury was an order of magnitude higher than any other category, while staff absence and replacement labour represented approximately 75% of the total incident factor costs. Also person to person assault, slips and trips, and patient lifting/handling represented 75% of the incident type costs.
x
For the questionnaire, data pooling of data from individual Trusts was avoided because their results were statistically significantly different from each other (P<0.05). There were significant correlations (P<0.05) between the rank order of some staff opinion questionnaire dimensions, captured incidents and workbook use. Higher reported use of the workbook was significantly associated with both higher opinion on the questionnaire dimension of working environment and lower captured incident rate. There was no correlation found between reported incident rate and the other performance measurement methods.
Use of the management workbook was low. For example, use of workbooks decreased significantly over time (P<0.01), with approximately one third of workbook holders having dropped out by the end of the monitoring period and overall perceived workbook usefulness decreased significantly over time (P<0.001). Nevertheless all Trusts reported a significant increase in the use of particular workbook sections (P<0.001) but low implementation was confirmed by a visual inspection of a 10% sample of workbooks.
There was little evidence of change in health and safety performance between the two benchmarking phases. The exception to this was in the responses from the questionnaires, which showed significant differences between all Trusts (P<0.05) and a significant improvement (P<0.05) in staff opinion in some safety climate dimensions for two Trusts. A significantly lower (P<0.05) opinion on health and safety was noted in the medical and dental staff in one Trust. This effect had disappeared by the second project phase and was thought to be linked to the annual intake of NHS junior doctors, which coincided with the first data collection phase in the Trust. The mean number of reported incidents rose by 24% in the second phase, which was marginally significant (P=0.06). There was no significant difference between reported incidents the test and comparison Trusts (P=0.49 for phase 1 and P=0.58 for phase 2).
In the discussion, the wide variation in incident reporting rates was thought to be more likely to be related to reporting culture rather than health and safety performance. However, the smaller group of captured incidents was thought to offer a potentially useful alternative performance indicator and raised questions about the cost effectiveness of collecting data on every reported incident.
A potential limiting factor was inter-investigator bias in benchmarking root causes of incidents, so far as learning lessons from incidents was concerned. It was concluded that there exists a need for improvements in consistency and in incident investigation skills for safety practitioners and others with health and safety management duties as well as a need for a common standard for incident reporting and costing. It was concluded that incident reporting in the NHS has largely been used as a performance indicator and that the process of learning from adverse events has been largely underdeveloped.
Management root causes could not be identified for approximately half of the total incidents reported (defined as a “special” category of incident). Most (67%) were not thought to have been preventable. The potential for many of this type of incident to occur in the NHS should be investigated and debated further to establish whether this phenomenon is largely confined to healthcare and how to interpret their existence in the context of health and safety management performance.
A number of issues were not captured by either the HSE incident analysis method or the Trust incident reporting systems. For future use these should be refined to improve their ability to
xi
effectively capture and manage the risks from cases of occupational ill health; hospital acquired infections and other patient-related incidents; and the potential for impact by civil litigation.
The costing strategy used in the method was conservative. Estimated extrapolated national costs were between 0.06% and 1.44% of the running costs of the NHS, which were low compared to previously published data. The potential future use of such a conservative costing strategy is discussed. So far as the evaluation of the workbook was concerned, it failed to produce a significant effect on health and safety performance. It was not possible to conclude that it produced an effect that was too weak to be detected because, had it been more fully used, there might have been a detectable effect. Therefore, with further refinement it could be a useful tool within a health and safety management system and it was recommended that, following refinement, it be made freely available. It was apparent that a simple approach to health and safety management, which did not ensure the workbook was used and implemented, would not produce significant change in performance. It was recommended that the issue of how to improve implementation and use should be addressed, possibly via a multi-factorial strategy.
In the performance change measurement field trial it was possible to conclude that, despite the fact that minimal change in performance was detected, the measurement methodology had the ability to detect change in health and safety performance. For example the questionnaire could detect significant change in staff opinion and therefore represented a potentially useful measure of culture change. The absence of significant change in questionnaire responses from five of the seven Trusts may represent evidence that the underlying health and safety culture in those Trusts was relatively stable. The key message of the research was that there exist potentially effective methods to evaluate health and safety performance but that to make a lasting effect on change in performance represents a greater challenge, possibly requiring a less simple, multi-factorial strategy, which takes account of both systems and behavioural approach. In addition to the areas identified above, the main area for further work was to seek further cause and effect evidence for the major finding of the research, that is the significant correlation between the use of the workbook and lower captured incident rates and higher staff opinion for some questionnaire dimensions.
xii
xiii
INTRODUCTION
The United Kingdom (UK) National Health Service (NHS) employs more than 1 million people at a cost of around £17.7 billion for salaries and £1.2 billion for agency staff to cover for vacancies and staff absences(1). There are chronic staff shortages caused by vacancies (approximately 3.5% across the whole sector) and staff absences (in 2001/2002 this was running at an average of 4.9% across all NHS Trusts)(1). Staff incidents and work-related illness are known to be major factors in these staff absences(1).
So far as patients are concerned, the Department of Health (DoH) has concluded that in the past the healthcare sector has underestimated the scale of unintended harm or injury to patients and that the issue of patient safety has become a high priority for improvement(2).
There is therefore a need to examine more closely factors that might lead to improvements in these areas and how these improvements might be measured and evaluated. This introductory Chapter therefore comprises of three parts:
Part 1: Explores the background to health and safety management in the context of the strategic policy approaches of both the European Union (EU) and UK. The UK tactical approach to health and safety management and how this applies to the healthcare sector is then described; Part 2: Considers measures currently used to describe current UK health and safety performance, explores how this performance might be improved and how change might be evaluated; Part 3: Sets out the research aim and objectives.
BACKGROUND TO POLICIES ON HEALTH AND SAFETY MANAGEMENT
EU and UK Current health and safety policy in the UK has been largely influenced since 1974 by the Health and Safety at Work etc. Act (HASAWA)(3). Since the late 1980’s the EU “Framework Directive” (89/391/EEC) brought about the enabling of the UK Management of Health and Safety at Work Regulations 1992(4) and 1999(5). These regulations made explicit the implied requirement in HASAWA to manage health and safety risks. It required risks to be assessed and controlled using an approach that was integrated with other business management activities. UK policy on the management of health and safety has therefore tended to embrace the use of workplace risk assessment as an instrument through which the technical aspects of risk are evaluated(6).
The healthcare sector The DoH and the Scottish Executive (SE) are together responsible for overseeing occupational health and safety strategy in the NHS in England, Wales and Scotland. They have a multi-factorial strategy, which involves targets aimed at patients, employees, the general public and employers(7, 8). So far as employees are concerned, these targets center on reductions in incidents and sickness absence.
xiv
In the NHS an “incident” had a broad meaning covering a number of different types of event. The most widely used definition of an incident in the NHS at the time of the research was1:
�� Accidents: defined as any incident, no matter how small, which did or could have adversely affected any person, not caused deliberately (e.g. some acts of violence or fire);
�� Violence/abuse/harassment: defined as any incident involving verbal abuse, unsociable behaviour, racial or sexual harassment or physical assault, whether or not injury was a result;
�� Clinical incidents: defined as any incident related to patient treatment or care which did, or could have, resulted in an adverse outcome (e.g. treatment error, medical equipment failure);
�� Ill health: defined as any case of known or suspected work or environment related ill health (e.g. infection, dermatitis, asthma);
�� Fire and security: defined as any incident, no matter how small, involving fire or fire warning systems (including false alarms);
�� Vehicle incidents: defined as any incident involving a vehicle, excluding vandalism or theft (regarded as a security incident);
�� Complaints: defined as any adverse formal or written communication from a person or organisation which required a response;
�� Other incidents: defined as near misses (an incident which had the potential for serious consequences); accidental property damage or loss; environmental incidents (e.g. accidental discharge to drains or the atmosphere); or food safety incidents.
In 1996 the National Audit Office (NAO) published the results of an audit of thirty acute Trusts in England(9). They found that health and safety management systems were poorly developed and characterised by a generalised lack of standardisation in the recording, costing and definitions of accidents.
Since then the healthcare sector has been encouraged to improve its health and safety management. For example, the Health and Safety Commission (HSC) Strategic Plan 2001-2004 for the UK was published in October 2001(10). As one of the eight priority programmes within the strategic plan, the health service has received, during 2001/02, targeted Health and Safety Executive (HSE) inspections on the management of: violence; manual handling; and slips/trips(11). These inspections included arrangements for both staff and patients. Enforcement action in England and Wales increased by 225% (from 8 to 26 improvement notices) and was thought by HSC to be a causal factor in a slow but steady decline in the rate of incidents reported to HSE(11). Conversely the NAO reported that in 2001/02 the overall number of recorded incidents increased by 24%(1). Reasons for these differences are unclear but may represent an improving culture of incident reporting in the NHS with fewer incidents meeting the HSE reporting criteria(12). Nevertheless these conflicting data should be interpreted with caution until reasons can be established with more certainty.
Two DoH publications, “An organisation with a memory”(13), and “Building a safer NHS for patients”(2) set out to develop a system in England and Wales for adverse clinical incident reporting and establishing a system for reducing the risk of unintended harm to patients. This was intended to establish a mechanism for learning from adverse events. The establishment of the National Patient Safety Association (NPSA) in July 2001 activated the means for implementation of a third 1 Based on that incorporated in the “SAFECODE-Plus” risk and quality management suite, widely used within the NHS (http//www.safecode.co.uk). Throughout this report the above definition of an “incident” will be used unless otherwise stated.
xv
document “Doing less harm”(14). This document set out key implementation requirements for healthcare providers, such as incident investigation and root cause analysis as well as implementation and monitoring of improvement strategies.
So far as the staff of the NHS are concerned, a set of standards for the effective management of occupational health and safety services has been published for both England and Wales(15) and Scotland(8). This has been combined with financial incentives via the insurance schemes covering the NHS in England and Wales (Clinical Negligence Scheme for Trusts(16) (CNST)) and Scotland (Clinical Negligence and Other Risks Indemnity Scheme(17) (CNORIS)). These schemes include basic standards for health and safety management, broadly based on HSE guidance(18).
The above initiatives have therefore laid a foundation for the development of management systems for occupational health and safety risks for staff and patients in the NHS.
HEALTH AND SAFETY PERFORMANCE IN THE UK
The health and safety system in the UK is well established, integrated across industry sectors, based on tripartism and social involvement(19). A major tactic of the HSE to achieve progress with its market forces policy has been to strongly advocate that positive health and safety management represents good business sense.
Their “Good health is good business” (GHGB) campaign ran for five years between 1996 and 2001 with the aim of increasing awareness of occupational health and safety in the workplace and improving employers’ competence in managing health risks in the workplace. The campaign was found to have been associated with significant improvements in employer attitude and the quality of their approaches to risk management(20). The evaluation had an acknowledged selection bias, in that organisations with a positive predisposition towards health and safety were more likely to sign up for the campaign. Indeed some organisations were not convinced. For example Monnery(21) questioned the concept that incident management was always good for business and suggested that other factors, such as the moral and legal aspects of health and safety and potential loss of reputation were better motivational factors for line managers.
However, proposals for reform of the arrangements for employers’ liability insurance(22), may help to redress the balance in the distribution of the costs of health and safety failures. This proposal is one of ten strategic aims of the “Revitalising health and safety” strategy (RHS), which contains Government and HSC targets for improving health and safety performance over the ten years to 2010(22). The targets seek to:
�� Reduce the number of working days lost per 100,000 workers from work-related injuries and ill-health by 30% by 2010;
�� Reduce the incidence rate of cases of work-related ill-health by 20% by 2010; �� Reduce the incidence rate of fatalities and major injuries by 10% by 2010.
This approach has been supplemented by the long-term occupational health strategy for England, Scotland and Wales “Securing health together” (SH2) (23). This seeks to establish evidence-based best practice in five key areas (compliance, continuous improvement, knowledge, skills and support).
xvi
An assessment of the likelihood of success of reforms to employers liability insurance(24), based on the experiences of other countries, has concluded that employers could be motivated to improve occupational health and safety via insurance premium incentives. The financial penalty for non-compliance however would require to be high, estimated at approximately 3% of payroll costs(24).
STRATEGIES FOR IMPROVING HEALTH AND SAFETY PERFORMANCE
Within the UK there is an abundance of readily available guidance on the practicalities of achieving robust health and safety management systems(25). The cornerstone is “Successful health and safety management” (HSG65), published by HSE in 1991(25) with the second edition in 1997(18). It has come to be a “best seller” for HSE, with its lay language and readily remembered mnemonic POPIMAR to signpost the seven main steps of a closed-loop management system (policy, organising, planning, implementing, measuring, audit & review).
HSE have also issued guidance on behavioural based safety(26), recommended as a method that can affect the number of incidents and thereby positively influence health and safety management performance.
A British Standard (BS8800:1996)(27) on occupational health and safety management systems has been developed. Based on HSG65, it currently has the status of voluntary guidance and cannot therefore be used for certification purposes. However, the subsequent standard (OHSAS: 18001 detailing the specification(28) and OHSAS: 18002 giving guidelines for implementation(29)) has the potential to form the basis of an assessment and certification scheme for a UK health and safety management system. BS8800 and OHSAS 18001/18002 are fundamentally the same in terms of structure and requirements. Presently, neither is recognised as either an UK or International standard despite the fact that all major certification bodies are certifying to OHSAS 18001 using OHSAS 18002 to provide generic guidance. Whether a formal certification scheme will result or not remains unclear, given the current UK national policy to allow market forces to bring about improvement in workplace health and safety(30).
The Scottish Executive Department of Health (SEDoH) were interested in the improvement of health and safety management within the healthcare sector and funded the author to develop a health and safety management tool to assist with this process.
This workbook was designed by the author to assist line managers with their health and safety management responsibilities. Its' use is described in Chapter 2.
SEDoH were also interested in whether use of this tool could help to improve health and safety management performance in the healthcare sector and commissioned the author to investigate this. This required a method for the measurement and evaluation of change that might occur as a result.
STRATEGIES FOR EVALUATING CHANGE IN HEALTH AND SAFETY PERFORMANCE
The need for more research to help inform policy and strategy on improving workplace health and safety has been consistently identified (e.g.(23, 31-33)).
In the UK much effort has been directed at evaluating costs of accidents and ill health both to employers and the economy as a whole (9, 34-38). The purpose of this effort has generally been to
xvii
encourage performance improvement, (i.e. used as an incentive rather than measurement tool). Cost estimates have ranged from between 1% and 5% of annual running costs for a hospital, incurred by that hospital(34, 36), to between 1% and 2% of UK total Gross Domestic Product (GDP) for the economy(37, 38). The range of estimates was due to differences in approaches and difficulties in measuring cost, but nevertheless suggested that the costs of accidents and ill health are significant to both employers and the economy as a whole. Auditing has also been extensively used in health and safety performance evaluations(39-41). Its main uses have been to:
�� Identify strengths and weaknesses of occupational health and safety management systems; �� Measure occupational health and safety performance; �� Establish the extent of legal compliance; �� Identify and define areas for improvement; �� Compare organisational performance with an established standard.
Gay and New(42) described the view of HSE on the use of auditing as a means of formally assessing health and safety management systems. The authors regarded auditing as an essential regular part of a management system and identified that it had further potential to be linked to costs and causes of accidents. The basis for this was the underlying paradigm that where audit identified weaknesses in health and safety management systems, there were more accidents and higher costs. Conversely, where a component of a management system was managed effectively there could be expected to be fewer accidents and lower costs. Therefore, if causes of accidents were tracked back to their root cause in the health and safety management system, these should correlate with health and safety management audit findings.
HSE subsequently commissioned the development of a new health and safety management performance measurement tool, based around the potential linkages between audit, costing and root cause analysis of incidents(34, 43).
To do this HSE identified a three-level model to define health and safety management (described below):
�� Workplace precautions (“Level 3 arrangements”) to protect workers health and safety (e.g. ensuring appropriate machinery guarding or the use of safety helmets on construction sites). These were regarded as part of;
�� Risk control systems (“Level 2 arrangements”) (e.g. systems for management of hazardous substances or fire precautions. These were, in turn, part of;
�� Management arrangements (“Level 1 arrangements”) for health and safety (e.g. policy or planning)(18).
In addition, HSE commissioned a literature review(43) of incident investigative tools which could be used for root cause analysis (defined as the level 1 management system causes of incidents). The resulting report concluded that none of the techniques available at the time adequately addressed every stage in the investigation process and that those models that attempted to capture all the relevant points did so at the expense of comprehensibility. The tool with most potential was identified as Management Oversight Risk Tree (MORT). It took the form of a generic fault tree analysis of an incident, which required the identification of factors that must have been present for it
xviii
to occur. However, the full fault tree was very large and it was judged to be impractical for all but high hazard industries, such as those in the nuclear and petrochemical sectors.
As a result of this review, HSE then commissioned the development of a new root cause analysis method for industry in general and linked it to a costing methodology, which was a simplified version from that originally published in 1993(35) and 1997(36). Finally, a pilot study in a hospital setting was conducted by HSE and the author(34). The purpose of this was to evaluate the ability of the method to be used as an investigative tool, to consistently identify costs and rates of incidents that could be mapped onto safety management failures that were consistent with the elements of the HSE model(18). The pilot study also included a health and safety management systems audit, conducted by HSE inspectors. The results of the audit and the root cause analysis were broadly found to compare well. For example, where the HSE audit identified aspects of the health and safety management system to have less than legal compliance (such as in the planning and implementing aspects of the health and safety management system), the root cause analyses of incidents tended to identify similar weaknesses (by virtue of more incidents and higher costs). Conversely, areas identified in the audit as being well managed (such as policy and communication) had fewer incidents and lower costs.
It was concluded that, in practice, an audit component was an unnecessary component because the pilot study had provided evidence that the outcome of the root cause analysis was similar to that obtained by audit. Additionally, the approach was thought to offer the potential for greater sensitivity and objectivity than auditing alone, as well as sharing with audit the potential advantages of use in benchmarking, baseline or before-and-after measurements, thus offering the potential for an effective novel method for measuring change in performance.
The HSE method had not been published because of the need for further validation to establish more objective field data on the performance of the method. HSE commissioned the author to commence the validation process by including the method in the research described in this report.
RESEARCH AIM AND OBJECTIVES
This Chapter has highlighted questions on methods for defining, evaluating and improving health and safety performance.
The aim of this research is therefore to evaluate whether it is possible to reliably measure change in health and safety performance when a formal health and safety intervention is introduced to the healthcare sector. This aim will be achieved by setting the following three research objectives:
Objective 1: The design of a methodology for evaluation of health and safety performance. This design will include the HSE methodology described above but will also be supported by an in-depth critical literature review to assess whether the HSE methodology would benefit from additional components; Objective 2: The evaluation of the strengths and weaknesses of the use of the health and safety management workbook-based intervention tool for the improvement of health and safety performance. This will also be supported by the literature review; Objective 3: The design of a research methodology to measure change in health and safety performance. This will also be supported by the literature review and will also include a field study within the healthcare sector to test the methodology.
1
1 LITERATURE REVIEW
The definition of a safety intervention that was identified as most closely suited to the research aim and objectives was:
“An attempt to change how things are done in order to improve safety”(44)
The main objective of the literature review was to identify suitable measures for evaluating health and safety performance; to evaluate the design of interventions to improve health and safety management performance; and to identify a suitable design for a field study to measure change in health and safety performance.
This Chapter first summarises the review methodology, then describes the findings. The last section forms conclusions about the implications for research designs most suited to intervention studies in the field of occupational health and safety management.
1.1 LITERATURE REVIEW METHODOLOGY
As observed by others(45-50) an issue for researchers in occupational health and safety management is the breadth and depth of the literature, which is spread across several disciplines. Therefore, the search was carried out across professional and scientific disciplines, and included several different data sources and literature collections (listed below).
The review was conducted during parts of 2002 and 2003. Relevant studies were retrieved, using a systematic approach. Searches were restricted to articles in English, and conducted using combinations of the following search words:
Accident, assessment, change, evaluation, impact, incident, intervention, management, measurement, meta-analysis, prevention, review, root-cause and safety.
The following sources of information were used:
�� Biomedical - BioMedNet; Cambridge Scientific Abstracts (CSA); Cochrane Library; Cumulative Index to Nursing and Allied Health (CINAHL) Database; Internet Database Service; MEDLINE; NeLH; OVID.
�� General - Emerald Fulltext; Emerald Management Reviews; Ingenta Services; NESLI (National Electronic Site Licence Initiative); ZETOC –BL.
�� Government - Canada – EU Cooperation on Workplace Safety & Health (www.eu-ccohs.org/); Department of Health (http://www.doh.gov.uk); European Agency for Safety & Health at Work (http://europe.osha.eu.int/); Health & Safety Executive (http://www.hse.gov.uk/); International Labour Organisation (ILO) (http://www.ilo.org/); National Institute for Occupational Safety & Health (NIOSH) (http://www.cdc.gov/niosh/homepage.html); Scottish Executive Publications online.
�� Health & safety - Directory of Occupational Health and Environmental Hygiene Links (http://www.agius.com/hew/links); Embase (covering occupational health, environmental health and ergonomics databases); International Occupational Safety and Health Information Centre (CIS); The Institution of Occupational Safety & Health (http://www.iosh.co.uk).
2
�� Life sciences - Edina CAB Abstracts; ISI Web of Science Citation databases. �� Science & engineering - EDINA Compendex (indexes to engineering articles and conference
proceedings); European Network for Process Safety (SAFETYNET)(http://www.safetynet.de/); Science Direct.
�� Social science - Applied Social Sciences Index and Abstracts (ASSIA); BIDS International Bibliography of the Social Sciences (IBSS); PsychINFO.
Additional information was obtained by checking author names and key publications via citation indices.
1.2 FINDINGS
A noticeable general feature was that, despite regular overlapping of the publication years of the articles examined, few of the published reviews evaluated the same studies. For example Livingston et al(43) did not cite the same work on accident causation theories published earlier by a study group commissioned by the Advisory Committee on the Safety of Nuclear Installations (ACSNI)(48) despite having clear objectives to review occupational health and safety-related intervention studies. Reasons for this are unclear, but one possible cause could have been the multidisciplinary nature of the subject area and the breadth and depth of the literature.
1.2.1 Literature review objective 1 (To identify suitable measures for evaluating health and safety performance)
Four main approaches to health and safety performance evaluation were identified. They were: (a) Methods that could be used in an objective manner, such as those based on audit or incident
data; (b) Subjective methods such as safety culture/climate evaluations; (c) Economic evaluations; (d) Methods that used more than one type of method.
(a) Performance evaluation methods that could be used in an objective manner. It was intended to use the method developed by HSE, which was based on the analysis of incident data. There was a large literature base identified on the use of incident data to evaluate health and safety performance. There was also a large literature on methods that were based on the use of audit.
So far as limitations to using incident investigation as a performance indicator was concerned, a major potential source of bias was the degree of intra- or inter- investigator consistency (47) with an extensive literature base from the 1980s and 1990s, supporting the theory that both experts and novices exhibited systematic biases in their health and safety assessments, with strong evidence that experts may have become over-confident. It was clear that evaluation methods that used techniques of incident investigation or workplace inspection should attempt to control for sources of inter- and intra-investigator bias.
Another significant limitation of the use of incident data as a performance measurement method was the potential for differences in incident reporting conventions between and within organisations. This could be further confounded by the possibility that, as staff became more aware of health and safety, they might tend to report more incidents(51). This effect could have a
3
profound confounding effect on performance evaluation. The main method to control for this risk would be to use more than one evaluation method, preferably linked via triangulation. None of the identified methods attempted to link analysis of incidents to the health and safety management system of an organisation, a finding also noted by Hale et al(52). Within the NHS this was thought to be a negative feature since there were strong drivers to ensure robust health and safety management systems were developed(1, 8, 15, 53, 54). An evaluation method that was linked to the health and safety management system would also be more likely to be understood by staff working in the NHS. Therefore the HSE method was the only method, which could be used for this purpose.
The use of audit has been well established since the 1970s as a means to evaluate health and safety management performance(51, 55). Using audit to evaluate health and safety management performance was identified as a potentially powerful method. This was also acknowledged by HSE who ensured that the pilot work for their method was validated against an HSE inspection audit(34). This study found good agreement with the findings of the HSE root cause analysis method and the audit. This meant that resources could be devoted to the root cause investigation of incidents. Had this finding not been present it would have been necessary to have considered the inclusion of an audit component as part of the evaluation.
(b) Subjective performance evaluation methods. Guldenmund(56) and Hale and Hovden(57) have reviewed the theory and research base on the nature of safety culture and concluded that the scientific approach was immature given that there was no single accepted model or definition of safety climate or safety culture.
Given there was no prerequisite by the sponsoring bodies for the research project for a subjective performance evaluation method, the literature search therefore sought methods that were:
�� Fully developed and ready for use. This was because the research objectives were not intended to include the development and validation of a qualitative method;
�� Applicable to a range of occupational groups across a diverse industry such as the NHS; �� Applicable to general health and safety management rather than a specific hazard or group of
hazards.
The method that offered most potential was that of Cox and Cheyne(58) who published a nine-dimension methodology for assessing general health and safety management culture in all staff groups in the offshore industry (the Loughborough Safety Climate Assessment Toolkit). The method was based on a systems approach to organisational culture, which had been refined using focus groups, factor analysis, and field trials. The questionnaire, along with a “toolkit” for use in data analysis were freely available via the Internet(59). This was assessed as potentially adaptable for use as an evaluation method in the field of health and safety performance evaluation in the NHS. Although developed for the offshore industry its question set was sufficiently generic to offer potential for use in the NHS, subject to confirmatory factor analysis, such as that conducted by Brown and Holmes(60) and Dedobbeleer and Beland(61).
(c) Economic Evaluations. The author has reviewed economic evaluations in the field of health and safety in healthcare(62). It was concluded that including a costing aspect in the research design, with other performance evaluation indicators was regarded as potentially useful. Justification for this would be to
4
investigate whether cost could be linked to other performance evaluation methods. This was a potential use of costing data that has not been widely explored, except by HSE and the author(34).
(d) The use of more than one method. Many authors have advocated that more than one method should be used to measure health and safety performance (e.g. (56, 58, 63-73)). Guldenmund(56) identified that few studies have attempted to establish correlation (“triangulation”) between safety performance measures (such as accident/incident data or safety audits) and safety culture or climate assessments, and called for these relationships to be explored in more detail.
HSE linked their root cause analysis approach with previously published guidance on health and safety management(74) and the costs of incidents(35). Their new methodology therefore linked incident investigation, root cause analysis and costing(34). This method, however, did not include any aspect of measurement of safety climate. If this were included the potential for triangulation of data would be greatly strengthened.
1.2.2 Literature review objective 2 (To evaluate the design of an intervention to improve health and safety management performance)
To fulfill the project research aim required a formal health and safety intervention to be introduced into the NHS. For this to be realistically achievable across the whole sector it was regarded as needing to possess the following features:
�� Acceptable to NHS Trusts, otherwise willingness to participate in the research would be unlikely to be secured;
�� Able to be implemented within current resources (i.e. cost-effective). This was because cost and spending targets and priorities tended to be set at national level and Trusts had comparatively restricted freedom to act outwith these targets;
�� Generic, so as to be applicable across the wide variety of types of services and staff groups within the NHS;
�� In keeping with other health and safety drivers within the NHS (such as from HSC(10, 11), DoH(2, 75) and SE(8)).
The intervention workbook2 used for the research was developed to ensure that it met the above criteria. In practice this meant taking a general systems approach, which had a long history in the NHS.
Nevertheless, it was acknowledged that there were limitations to a systems-based approach of this type. These risks were:
�� The intervention might be overly generic to make an impact on health and safety performance at local departmental level. Although an iterative approach may have been desirable for the research in this report it was regarded as insufficiently generic for application within the NHS as a whole. The risk was addressed by ensuring that examples, relevant to the NHS were included in the workbook and local conventions and terminology were used.
2 The workbook is described in Chapter 3.
5
�� The intervention might not be sufficiently simple to be used intuitively by line managers as a relatively “stand alone” resource. It was therefore regarded as important to use a combined approach, using a written guidance/workbook approach supplemented by education and training as a cost-effective delivery mechanism;
�� The intervention might not produce an effect strong enough to be detected. To address this issue the workbook was aimed at line managers who had acknowledged legal responsibilities for health and safety management. Additionally it was written in a non-technical language and Trusts were encouraged to ensure that managers appreciate the need to use it regularly;
�� The effectiveness of the method might be reduced because it did not make use of health and safety management programmes based on psychological or behavioural models. Its relatively under-developed nature meant that this approach was, at the time, insufficiently generic for use as the sole approach in a general intervention.
Therefore, on balance, a systems-based approach(57) based on currently accepted guidance on health and safety management(18) was most likely to harmonise with other NHS drivers and therefore represent a potentially effective tool for the NHS
1.2.3 Literature review objective 3 (To identify a suitable design for a research study to measure change in health and safety performance)
The purpose of a research design has been defined as ensuring that there are measures in place to organise research activity, including the collection of data, in ways that are most likely to achieve the research aims(68). Therefore the overarching strategy for the literature review was to search for research designs that had the potential to do this.
Meta-analysis was known to be a powerful method for assessing the validity of research by statistically combining results of comparable studies(76-78). Given their potential for identifying strong research designs, meta-analyses and systematic reviews in occupational health and safety evaluation and intervention studies were specifically sought. Only four meta-analyses of performance change evaluation (45, 79-81) and four systematic reviews of intervention studies (46, 50, 73, 82) were identified. This was thought to be a low incidence of these types of publication, particularly considering the wide-ranging multidisciplinary professional and scientific basis of the search strategy.
A possible explanation was that the primary research base on which the reviews were based was insufficiently robust to support this type of analysis. For example, Shannon et al(73) were unable to make quantitative comparisons between ten studies. This was supported by Rivara and Thompson(83) who attempted to conduct a meta-analysis to assess the effectiveness of different strategies to prevent falls from heights in the construction industry. Their review methodology was based on Cochrane collaboration guidance(76, 83) and identified only three studies suitable for review. However, the methodological quality was of such poor quality (no controls or appropriate multivariate analyses to control for potential confounding factors) that no attempt was made to undertake the meta-analysis and few conclusions about the effectiveness of the intervention strategies were made.
Oliver et al (80) conducted a meta-analysis of twenty-one papers, which met strict pre-determined inclusion criteria for hospital fall prevention programmes. The authors concluded that there was a tendency for the most rigorously controlled trials to produce the smallest effects. For example, Lingard and Rowlinson conducted a behaviour based approach to safety management in the
6
construction industry in Hong Kong(84). In common with Harper et al(85), the authors used a “multiple baseline across groups” design but also included a reversal aspect, cited as being particularly suited to measurement of safety behaviour(44). The results were mixed in terms of detected improvements in performance.
Conversely, it was noticeable that the studies most likely to quote impressive effects on health and safety performance following interventions were usually of the non-experimental type (e.g. (48, 86)). Non-experimental before-and-after designs were not regarded as suitable for evaluating change in health and safety performance because of methodological limitations of the research design resulting in uncontrolled sources of bias.
The main finding of the review has been that there exist in the literature base severe limitations in scientific rigor relating to impact evaluations of health and safety interventions. A possible reason for these limitations was thought to be that the subjective benefits of health and safety intervention programmes have been so impressive and the likelihood of impressive effects so great that there have been few demands for more reliable data(62). Other factors that may have also contributed to the lack of a robust literature base:
�� A general lack of scientific training for safety practitioners(87, 88); �� A general lack of funding for evaluation research(45, 89); �� Ethical issues relating to research designs that might withhold risk control measures from
control groups(90); �� The pragmatic nature of health and safety(91); �� Rule-driven basis of prevention strategies(92, 93).
Hillage et al(94) suggested a comprehensive strategy for improving the evidence base, including ensuring evaluation was included in the planning of interventions; developing a stronger evidence base by conducting more systematic reviews and meta-analyses; and adopting minimum standards. Taking into account the findings of the literature review, it has been concluded that a suitable research design should posses the following features:
�� A base of currently accepted UK national guidance on safety management; �� An economic evaluation component; �� A linkage between the causes of incidents and accidents to the health and safety management
system; �� An intervention in the form of written guidance or work book format, supplemented by an
educational component; �� A quasi-experimental or experimental design with baseline measurements; �� A comparison group; �� Random sampling where possible; �� Validated methods; �� A control strategy for potential inter and intra-observer bias; �� More than one performance evaluation method including both objective and subjective
methods; �� An attempt to correlate the results of the performance evaluation methods (triangulation); �� Sample sizes of sufficient size to allow statistical inferences to be made, where possible; �� Validity checks to ensure that pooling of data from different sites was appropriate before
proceeding with analysis.
7
2 METHODS
This Chapter describes the research methodology and identifies how each aspect combined and contributed to the research design. The Chapter has five parts:
Part 2.1: Practical factors relevant to the research design; Part 2.2: The evaluation methodology design (research objective 1); Part 2.3: The intervention design (research objective 2); Part 2.4: The research design for measuring change in health and safety performance
(research objective 3); Part 2.5: The data analysis strategy.
These will be considered in turn.
2.1 PRACTICAL FACTORS RELEVANT TO THE RESEARCH DESIGN
2.1.1 Variability between NHS Trusts
Within the NHS there was a high degree of variability between individual operational units (i.e. NHS Trusts), as a result of population, demographic, regional, and cultural factors. There was also local autonomy within broad, national performance targets and so individual units within the various Health Authorities or Boards tended to have different priorities and approaches to the development of their arrangements for the management of health and safety.
It was recognised that these potential differences should be taken into account, particularly during the analysis of results by testing the appropriateness of combining results from different trusts. This was done by multiple comparison testing (see Section 3.5 on data analysis).
2.1.2 Recruitment of research participants
Random selection of Trusts was not practicable because participation in the fieldwork required their agreement and for them to be willing to be involved. This involvement included:
�� Commitment to all phases of the project over its 30 month duration; �� Co-operation with the requirements of the methodology; �� Allocation of resources (e.g. those with operational responsibilities for health and safety
working within the participating Trusts were expected to be fully involved, especially in day to day contact with the research team).
Therefore a convenience sample approach was adopted (after Harper et al(85)). A call for volunteers was made via the NHS in Scotland Directors of Human Resources (HR) forum. This forum was chosen because the project had the formal support of the HR Directorate within the SEDoH. Additionally, the majority of NHS Trusts in Scotland had assigned executive responsibility for health and safety management to their HR Directors, which was a positive feature of the recruitment strategy. In England the HSE had contacts with NHS Trusts who had previously expressed willingness to participate in project work.
8
Using this approach it was intended to recruit between six and nine Trusts representing a mixture of urban and rural locations and acute and primary care sectors. This number was selected for pragmatic reasons in that it was estimated at the beginning that the project team could devote sufficient time to support and guide the Trusts with this number of participants. Also, this number was thought likely to generate sufficient data for useful inferences to be made. At the end of this process seven NHS Trusts participated in the research.
Variability was expected in the degree of development of health and safety management systems between the Trusts. In the event it was possible to identify two distinct groups from the volunteers.
One group comprised two Trusts with an existing formal health and safety management system, typified by the presence of a workbook in the possession of line managers. This group was particularly interested in health and safety performance. For these Trusts, the performance measurement exercise was the main interest, so feedback on findings was provided, via regular meetings and interim reports. They were assigned to the test group.
The other group of five Trusts did not have a formal, systematic health and safety management system based on a workbook approach. The Trusts comprising this group were interested in introducing one to their organisation. The project offered the opportunity to implement a system and to evaluate its impact. They were assigned to the control group.
The participating trust were therefore as follows:
Trust A: A test primary care Trust (PCT); Trust B: A test PCT; Trust C: A test PCT; Trust D: A test PCT; Trust E: A test acute Trust; Trust F: A control acute Trust; Trust G: A control PCT.
2.1.3 Relationship between researchers and Trusts
Recognising that non-random selection of the participating Trusts would limit the ability to make statistically significant inferences about the findings, the field study relied on the commitment and participation of the Trusts who agreed to take part.
Ultimately the health and safety management system was introduced for the benefit of the participating Trusts, and it was regarded as important that they had ownership from the beginning. This was because the researchers would eventually withdraw leaving the Trusts able to take forward the day-to-day management of the system. It was therefore in the best interests of all concerned to be involved from the beginning.
For data collection on the scale proposed it was vital that positive working relationships were built with the participating organisations to ensure co-operation. This was achieved by the following combination of measures:
�� Regular, informal dialogue with the main Trust contacts to ensure a positive relationship between the participants and the researchers;
9
�� Ensuring regular verbal and formal written feedback at each milestone of the project; �� Adopting an approach that was as flexible as possible, while recognising the need to minimise
sources of bias within the research design. For example, some Trusts wished more assistance with data processing than others, while some requested that some of the investigation of incidents were undertaken by members of the research team while others were content to undertake this activity themselves, under supervision (these differences were not thought likely to have introduced bias but were documented);
�� Regular project meetings. The project team met at least weekly to discuss day-to-day issues, while formal meetings with the Trusts were usually held at least quarterly.
2.2 EVALUATION METHODOLOGY DESIGN (RESEARCH OBJECTIVE 1)
The research design ensured that data was collected on both objective and subjective aspects of health and safety performance.
The method comprised tracking reported incidents and investigating them to establish the objective components. It had been developed on the paradigm that weaknesses in the health and management system underpinned the occurrence of incidents. Linking causes of incidents to the underpinning health and safety management system of an organisation would therefore provide a measure of its’ performance. This was a novel approach, replacing the need for health and safety management systems audits to be conducted in parallel with incident investigation. The pilot work conducted by the author(34) and described in Chapter 1 provided evidence that this was a valid strategy.
There were three objective components to the research design. These were:
1. Identification of incident management system failures; 2. Root causes of incidents; AND 3. Costs of incidents.
There were also two subjective parts. These were:
4. A staff opinion survey of a stratified random sample of NHS staff; 5. Measurement of staff views on the usefulness of the intervention used as the vehicle to
introduce the health and safety management system to the participating Trusts.
Each of these aspects will be described in turn.
2.2.1 Objective measurement component
The HSE method identified in Chapters 1 and 2 was adopted as the measurement method for this aspect of the evaluation. The method will be described below but the detailed methodology, including flow chart, incident checklist and the recording forms, with an example to illustrate their use, is included in Appendix 1.
Only those incidents that met criteria that had been pre-defined by HSE were included. These were:
�� Whether the incident caused actual harm or had the potential to cause harm; �� Whether the incident was within the power of The Trust to control;
10
�� Whether the cost of the incident exceeded £5 or 15 minutes of costed staff time.
The process that was undertaken by the investigators is summarised below:
�� Initial follow up of the incident with the person who reported it; �� Establishment of whether the incident met the above inclusion criteria. If not the incident was
excluded from the study); �� For each incident included in the study, an investigation pack was allocated, noting the
date/time of incident on all paperwork; �� Detailed information was obtained about the incident. The form used to record this data
contained prompts, which encouraged the investigator to consider three major groupings of control measures (i.e. physical, behavioural and organisational controls) (data capture form Appendix 1 Form 1A);
�� Failed risk systems were then identified (e.g. control of contractors, management of violence and aggression) (also recorded on Appendix 1: Form 1A);
�� The main contributions to costs were then grouped against whether they applied to staff, objects, materials, or equipment (recorded on Appendix 1: Form 1B). A detailed specification for participants as to which staff and non-staff costs to include was available (set out in Appendix 2);
�� An Events and causal factors chart was drawn up to assist with the identification of any further areas for follow-up;
�� Details were recorded on an Evidence table (Appendix 1: Form 2A); �� Management system root causes were then identified (e.g. policy, organising, planning etc.)
(Appendix 1: Form 3) and added to the Evidence table (Appendix 1: Form 2B); �� Costs arising from the circumstances and outcome of each incident were further described by
assigning them to relevant headings within each of three different groups. These groups were:
1. Incident category: accidental injury; accidental property damage; fatality; fire; ill health; near miss; other; physical violence; theft; vandalism; verbal abuse/threatening behaviour.
2. Incident factor: absence; cleaning up; damage/repair/staff treatment; hiring/purchasing; initial response to accident; lost production/wasted time; patient treatment; replacement labour; transport.
3. Incident type: contact with electricity; contact with equipment/machinery; cut with sharp material/object; exposure to fire; exposure to harmful substance; fall from height; hot or cold contact; manual lifting/handling; needlestick/sharps injury; other; patient lifting/handling; person to person assault; slip/trip/fall on same level/stairs; struck against something; struck by an object;
�� Verification was undertaken that no costs or other contributory factors were outstanding before completing the case;
It can be seen from the above that the approach was iterative and that the evidence was recorded in such a manner to render it capable of being audited.
All incidents that were reported via each Trusts incident reporting system were assessed against the above criteria. Those incidents that met one or more criteria were included in the study (i.e. those that were “captured”).
11
The Trusts used reporting schemes that were similar to each other, based on the IR1 form(9, 95), which was produced specifically for use in the NHS, and in this respect could be regarded as comparable.
Nevertheless, these systems were undoubtedly prone to local reporting bias. In addition, the introduction of a health and safety management system could have resulted in an increase in reporting of incidents, due to an increased awareness of the need to report. These possibilities were borne in mind during the analysis and interpretation of the results.
It was also recognised that there was a need to control potential investigator bias during the incident investigations and root cause analyses. The ability to demonstrate consistency of investigation and root cause analysis was recognised as important to the credibility of the research design. A combination of measures was therefore implemented to ensure as high a degree of intra- and inter-observer consistency as possible.
These measures included initial training for investigators, which was undertaken by staff from the HSE, at the start of the first data collection period. Refresher training supervised by HSE, at the start of the second phase, a year later was also undertaken. In addition, all investigation and follow-up decisions in all seven participating Trusts were under the control of the author. In cases where there was ambiguity, a decision was reached following discussion between the author, researcher, investigator and relevant experts (e.g. ergonomists or specialist clinicians such as psychiatrists) as judged appropriate by the research team.
2.2.2 Subjective measurement component
The method selected was the safety climate questionnaire and toolkit developed by Cox and Cheyne(59). This model was chosen because it was based on a systems approach to organisational culture and had been piloted, as a potential benchmarking tool, under the auspices of a joint industry and HSE partnership.
The questionnaire was regarded as validated and reliable in use because of extensive analysis of pilot data, conducted in the offshore industry(58). This included factor analysis, internal-scale consistency and alternate forms reliability tests.
The questionnaire consisted of 43 questions aimed at eliciting individual views on nine dimensions of safety climate, grouped into four categories. These were:
Category (a): Organisational context covering four dimensions: 1. Management commitment 2. Communication 3. Priority of safety 4. Safety rules & procedures
Category (b): Social environment covering two dimensions: 5. Supportive environment 6. Involvement
Category (c): Individual appreciation covering two dimensions: 7. Personal priorities and need for safety
12
8. Personal appreciation of risk
Category (d): Work environment covering a single dimension: 9. Physical work environment
Each question was answered by ticking a box to indicate the response. The whole questionnaire took between 10 and 15 minutes to complete. A free-text area was included for respondents to note any additional information if they wished.
Each item was scored by assigning a value to the question response as follows (the higher the score the more positive the opinion):
1 - strongly disagree 2 - disagree 3 - neither agree nor disagree 4 - agree 5 - strongly agree
Some questions had been negatively worded. The guidance provided with the questionnaire required these to be coded by subtracting the item score from 6 (e.g. a score of 2 on a negatively worded item would be reversed to a score of 4). Scores were then averaged for each question, across each Trust. Since the dimensions in the questionnaire had different numbers of questions they were standardised before being compared, by converting the scores to a 1 to 10 scale, by dividing the actual score by the total possible score and then multiplying by 10.
The resulting standardised dimension scores then had the following meaning:
2 - strongly disagree 4 - disagree 6 - neither agree nor disagree 8 - agree 10 - strongly agree
The lower the score below 6 (representative of the mid-point on the scale) the more negative the opinion(58).
The questionnaire, and detailed guidance for use, had been published for use by managers and safety professionals within the offshore oil extraction industry(59). This detailed guidance was followed closely.
Questionnaire confirmatory factor analysis The questionnaire had been developed for the offshore industry(58). To ensure that the scales in the original questionnaire were appropriate for use in a healthcare setting, confirmatory factor analysis (CFA) was undertaken. The data was subjected to CFA using version 6 of the EQS (structural equation) program(96). Internal-scale reliability (or consistency) was also examined to investigate the degree to which the various questions measured different aspects of the same concept. Finally the data was examined to establish the degree of fit for each Trust.
13
Questionnaire sampling strategy Staff lists were obtained from the Trusts and random 10% sample, stratified by job title/grade was selected. The lists obtained from the Trusts were ordered alphabetically according to job title/grade so that individual staff groups were segregated. Every tenth person on this list was then sampled, to produce the stratified random sample.
Questionnaires were sent to named individuals so that they could be followed up. This was done so that change in opinion, if it were present, could be detected with minimal threats to the validity of the results from confounding history effects. A potential source of bias in the results by using this approach was that individuals might have been concerned about issues of confidentiality. This was not raised as an issue during a pilot exercise (described below). Nevertheless the potential was recognised and the wording of the covering letters was chosen to reassure staff that their responses would be treated in strict confidence. The letters are reproduced in Appendix 3.
Questionnaires were distributed at the start of a twelve week window during which incident data was also investigated, to ensure that the opinions expressed in the survey were matched in time to the incident data. Those individuals who were classified as non-responders at the end of the 12-week period were not followed up any further.
To ensure an optimum response rate, an opt-out system was included. This allowed individuals to inform the project team that they were unlikely to be available for a second questionnaire round. Those who had not responded within 4 weeks were issued with a reminder. Those formally opting out and those who did not respond to one reminder, within the 12 week window, were replaced with an additional individual, randomly selected from the relevant area in the stratification.
Questionnaire pilot and quality control Although its designers had extensively validated the questionnaire, it had never before been used in the healthcare sector. The wording of two of the questions was altered to reflect different terminology used in the NHS.
These changes were not thought to be likely to have an impact on the final analysis. Nevertheless the use of the modified NHS-version was piloted to ensure that the wording was relevant to staff from the healthcare sector. The pilot was carried out in two departments within a Trust not participating in the main study, resulting in 22 completed questionnaires (100% response from those issued).
An additional question was included for the pilot only to elicit any concerns regarding the changed wording in the questionnaire. This question was worded:
“I found the questions in this survey easy to understand and apply to my situation”.
The responses to this question were:
Strongly agree 1 respondent Agree 7 respondents Neither agree nor disagree 10 respondents Disagree 3 respondents Strongly disagree 1 respondent
14
Since most (82%) respondents were either positive or neutral in their responses. This was judged to be reasonable and it was therefore decided to proceed with the questionnaire in its modified form, but without the additional question on ease of use.
A quality control system for questionnaire data entry was developed, using a C-Chart, a form of statistical process control(97, 98). This required that each time a batch (a batch was done in one sitting, so there could be more than one batch in a day) of questionnaires was entered onto the database, five forms were randomly selected, from across the whole batch, and checked for errors. If an error was found the whole batch was rechecked. Once all the data had been entered, the points were added and divided by the total number of data entries to calculate the final percentage error rate.
2.3 INTERVENTION DESIGN (RESEARCH OBJECTIVE 2)
The intervention was a health and safety management system, introduced, via a record keeping based system in a workbook format. It was designed to be a manager’s tool rather than a reference resource. This was because background information and guidance was already available to managers via their professional associations and HSE publications that were freely available to the general public and employees of the Trusts3. Also, users in the pilot had expressed a desire for a pragmatic document, which gave step-by-step guidance on required action. The author therefore designed a new workbook for the research described in this report. It is reproduced in Appendix 6-4. Its purpose was intended to provide a common standard with which the Trusts could implement a formal health and safety management system, based on HSG65 (74, 99).
It was written in a style that followed the general guidance of the Plain English Campaign4. The design of the workbook followed the same structure and terminology of HSG65, which was familiar to managers. Thus the document comprised five main Sections, which are described below:
�� Policy. Including space for a general health and safety policy as well as other policies on specific risks (e.g. policy on the management of violence and aggression), whereby a clear direction for the organisation to follow is described;
�� Organising. Including management structure and arrangements for delivering the policies. In this section are lists of health and safety roles and responsibilities, as well as data on consultation with staff, records of meetings, and training needs analyses;
�� Planning and implementing. Including the process for recognising, assessing and controlling workplace risks. Action planning is encouraged and various examples of risk assessments are given (e.g. manual handling and display screen equipment assessments);
�� Measuring performance. Including pro-active methods (e.g. workplace inspections) and reactive methods (e.g. incident data analysis) to monitor against policy standards. A checklist for workplace inspections is given as well as ideas for other activities, such as safety tours;
�� Audit and review. Including arrangements for a systematic review of performance based on data from monitoring and audits. A managers’ self-audit is included and the reader is guided to undertake this activity at the start of the process and to use the findings to form a yearly plan, with quarterly reporting milestones.
3 e.g. http://www.hse.gov.uk/sources/ or http://www.tuc.org.uk/h_and_s/index.cfm 4 http://www.plainenglish.co.uk/
15
An economic tool, based on the technique of option appraisal had also been developed by the author in response to a stated need for managers to be able to use economic assessment of their control measures as they undertook risk assessment(100). This method was also included in the workbook, although managers were encouraged to seek specialist help from either the author or their local safety adviser if they identified a complex assessment that might benefit from the inclusion of an economic option appraisal.
Early drafts of the workbook were piloted by initially being shown to colleagues and to employees within a NHS Trust that was not part of the fieldwork. After further modification the author formed a small working group, comprising representatives from management, staff and the health and safety department of the Trust. This group discussed and agreed how the workbook should look and how it should be introduced and used. The consensus was that it was most suited to line managers with responsibility for more than one tier of staff and that training in use was required. The third draft was used by a group of 12 managers, within the same Trust, who were asked to use the workbook for approximately three months and to offer suggestions for final refinement. The author interpreted the feedback translated this into modifications of the workbook and produced the version that was used for the research.
In its final form the workbook represented a useful, user-friendly health and safety management tool, which had proved to be popular with users. It was therefore judged to be suitable for use in the research project.
2.3.1 Method of introduction
Although use of the workbook was planned as a standardised system, a degree of flexibility with the different Trusts was necessary. This was because once the workbooks were issued it was important for the users to have ownership of the system, since at the end of the research the workbooks were intended to remain in situ.
This generic workbook was shown to the participating Trust contacts who were asked to specify minor amendments so that it could be tailored to the terminology, management arrangements, recording conventions and culture of each Trust. Examples of typical changes included:
�� Inclusion of an organisational diagram of the management structure of the Trust; �� Inclusion of contact details for relevant individuals and departments (e.g. the safety Trust safety
adviser or infection control department); �� Inclusion of risk assessment forms used by the Trust (these tended to be different in each Trust); �� Use of font, point size and layout for individual Trust documents.
The changes were not regarded as likely to introduce significant differences between the Trusts since the modified workbooks contained the same sections and guidance. The research team incorporated the detailed changes into the workbook and arranged for sufficient copies of the tailored version to be made available. Each set of workbooks therefore had the corporate identity of the individual Trusts. The decision as to who should be designated as a workbook holder, and therefore how many workbooks were necessary, was also agreed with the Trusts. As described above, the workbook was designed for middle managers, such as a charge nurse with responsibility for several wards. The numbers issued to each Trust are shown in Table 1. From this it can be seen that there were large proportional differences between Trusts.
16
Table 1 Numbers of workbooks issued per Trust
Test Trusts Number of employees Number of workbooks issued Trust A 824 36 Trust B 3,378 23 Trust C 1,922 29 Trust D 5,636 40 Trust E 1,853 40 TOTAL 13,613 177
These differences proved to be necessary because each Trust differed in their management structure (for example, some were much less devolved than others). The number issued to each Trust was based on a balance of factors to ensure all areas within the Trusts were covered by a workbook-holder. If each person within a Trust who was responsible for managing staff had a workbook there would have been multiple repetitions of risk assessments. Conversely, if only the most senior managers were issued with a personal copy, insufficient detail of local arrangements would result. Therefore because the grade and job description of workbook holders was broadly similar between Trusts these differences were not regarded as significant to the performance of the workbook.
Issuing of workbooks to named recipients was done via a two-hour briefing, during which time the background to the project was explained and each section of the workbook was explored in detail. A detailed framework for the briefings was prepared and the author always carried out the training (the training protocol is reproduced in Appendix 5). A representative from HSE sat in on a number of the briefings and approved the delivery against the protocol to confirm consistent delivery.
The briefings took place over a 22 week period, between 10th November 2000 and 17th April 2001. This staggered introduction represented a multiple baseline across groups design(85) (44).
2.4 RESEARCH DESIGN TO MEASURE CHANGE IN HEALTH AND SAFETY PERFORMANCE (RESEARCH OBJECTIVE 3)
The aim of the research was to evaluate whether it was possible to reliably measure change in health and safety performance in a NHS setting. A study design was therefore needed to control, as far as possible, for inter-trust differences, which also allowed for reliable performance measurement both before and after an intervention.
The method adopted was a prospective empirical field epidemiology study using a longitudinal cohort observational study design of the before-and-after type, with a multiple baseline. The literature review had identified this as a suitable design that had been validated for use in health and safety impact evaluation research(44).
The Trusts were assigned to one of two cohorts: a test group; and a comparison (control) group5 (64). Those without a formal system were assigned to the test group, which received the
5 Rakel et al defined the difference between a control group and a comparison group as whether the evaluation programme used a randomised or non-randomised sampling strategy. The method adopted uses non-random sampling of the participating Trusts, therefore the control group is, strictly speaking, a comparison group.
17
intervention, while those with an existing workbook-based system already in use were assigned to the control group, which did not receive the intervention. This was a valid strategy because any effect that might be noted in the results from the test group and not the control group could have been as a result of the intervention. Given that the research aim was to assess the effect of the introduction of a formal health and safety management system the Trusts with an existing system did not require another. Another factor was that the Trusts assigned to the control group were not disadvantaged by not receiving an intervention that could have benefited their health and safety performance.
2.4.1 Method of monitoring workbook use
The workbooks were issued via a closed-loop quality management system, so that monitoring of usage could be carried out. Therefore each workbook was given a unique identification code, which specified both the Trust and the identity of the holder. Additionally, each page recorded the section of the workbook, the title of the section, the issue number and date and cumulative page number. This allowed for the possibility of changes being made and tracked. The Trusts senior management and the author both held copies of these codes.
A protocol for workbooks monitoring was developed for the test Trusts only. All workbook holders were contacted by telephone for feedback regarding their use of the workbook in accordance with a telephone interview protocol (Appendix 6). Each person was contacted on three separate occasions and asked the same questions. This occurred within three designated windows, each lasting for four months (between 1st April 2001 and 31st March 2002). Approximately 10% of those individuals who were recorded as using their workbooks were selected at random for a visual examination of their workbooks.
Any confounding effect produced by the presence of the researchers could have been controlled for in the research design if the control trusts had also been included in the monitoring scheme. However, in practice the two control trusts had had their safety management systems in place for several years and personal contact by the researchers were thought to be liable to produce a different reaction from the test Trusts. This could have introduced complex methodological confounding of the results. This was because the workbook holders in the test Trusts all had been advised, as part of their briefing, when and how they would be contacted. There was therefore little surprise when this contact occurred. The workbook holders in the control trusts had not received a briefing, because their system was already in place and training had previously been given.
Therefore, although regular contact was made with the workbook holders to gauge usage of the system via the monitoring protocol, other forms of contact were deliberately avoided, to minimise confounding effects from the research team. The protocol itself was designed to keep contact to the least necessary to gauge usage (for example, making contact three times in a year rather than four, making contact via the telephone and only visually inspecting 10% of workbooks). This effect is discussed further in Chapter 4.
The Trusts were intended to be the principal catalysts for development and integration on a day-to-day basis of the Trust management system. They had been encouraged to adopt the workbook
However, parts of the strategy used random sampling so, for simplicity, this group is referred to as the control group throughout this report.
18
system and to manage its use as they would any other management initiative. The project team took no part in these activities, other than to be aware of what had been implemented.
2.4.2 Study composition
Over 23,500 employees were encompassed by the seven Trusts. They comprised five test Trusts and two control Trusts. There was a mixture of acute and primary care Trusts in both the test and control groups, described in Table 2:
Table 2 Description of Participating Trusts
Test Trusts Control Trusts Type of Trust No. of employees Trust A Primary Care 824 Trust B Primary Care 3,378 Trust C Primary Care 1,922 Trust D Primary Care 5,636 Trust E Acute 1,853 Total number of participants in test group 13,613 Trust G Primary Care 3,978 Trust F Acute 5,944 Total number of participants in control group 9,922
To ensure that the distribution of the questionnaire was even across staff groups, the sample was stratified by selecting every 10th name from a list of staff, which had been split into eleven occupational groups, which were common to the payroll departments of all the participating Trusts. These were (alphabetical order):
�� Administration and clerical (A&C) �� Domestic and catering �� Estates (e.g. joiners, plumbers, engineers, maintenance workers) �� Medical and dental �� Nurse manager �� Nursing and midwifery (qualified) �� Nursing and midwifery (unqualified) �� Other (e.g. Chaplain) �� Professional, technical and Professions Allied to Medicine (PAMs) (unqualified) �� PAMs (qualified) �� Senior manager
The study design then followed the sequence in Figure 1:
19
TEST group CONTROL group
Benchmarking 1
Intervention “Workbook”
Benchmarking 2
Benchmarking 1
No intervention
Benchmarking 2
Compare benchmarking 2 with benchmarking 1
Compare benchmarking 2 with benchmarking 1
Compare TEST/CONTROL
Job Category & Trust
Figure 1 Relationship between test and control groups
Each measurement method had its own indicator(s) of effectiveness. For the objective methods the measures were: change in number of reported incidents; pattern of root causes of incidents; change in pattern of system failures as causal factors within incidents; and change in costs of incidents. The subjective method indicator of effectiveness was change in dimension score for safety climate opinion. The field study plan therefore involved five consecutive phases over 30 months (summarised in Table 3.
20
Table 3 Summary of project phases
Phase 1 (Section 2.2)
Recruitment of participating Trusts Training Publicity campaign
Phase 2 (Sections 2.3 & 2.4)
1st Benchmarking Phase lasting 12 weeks, consisting of: -Incident screening -Incident costing -Incident investigation -Management root cause analysis -Issue/collation of questionnaires to 10% of staff
Phase 3 (Section 2.5)
Introduction of workbook system to five participating “test” Trusts: -Introduced via 2-hour briefing sessions -Used over a 12 month period -Monitored via telephone interviews of workbook holders -10% visually inspected to assess completion of documentation Re-training and publicity reminders also undertaken
Phase 4 (Sections 2.3.& 2.4)
2nd Benchmarking Phase lasting 12 weeks, consisting of: -Incident screening -Incident costing -Incident investigation -Management root cause analysis -Issue/collation of questionnaires to 10% of staff (same individuals as for 1st benchmarking phase)
Phase 5 (Section 2.6)
Data analysis
Phases 1 and 5 involved recruitment of Trusts and data analysis respectively. The test Trusts were encouraged to work on their developing health and safety management systems during the middle phase, Phase 3.
Phases 2 and 4 were identical measurement benchmarking phases where health and safety performance is evaluated. The literature review and guidance from HSE project supervisors helped to inform the final decision on the features and optimum design of the measurement method. It was designed to have an objective component (based on costs and management root cause analysis of reported incidents) and a subjective component (based on staff opinion of the safety climate within their workplace).
All investigators attended training in the implementation of the methodology. Prior to the first data collection benchmarking phase (phase 2), HSE personnel familiar with the methodology, provided the training for all Trusts. This training lasted a full day. Prior to the second phase (phase 4) the author, under the supervision of HSE, carried out the re-training.
A detailed week-by-week project plan was drawn up and used throughout to assess progress. The timescale is shown in Figure 2.
21
Figure 2 Project timescale
2.5 DATA ANALYSIS STRATEGY
All data was entered onto a spreadsheet computer software package (Microsoft Excel) and analysed using this or the Statistical Package for the Social Sciences (SPSS) (version 10)(101).
Much of the data was of the ordinal type. The preferred analysis strategy with this data was to use parametric tests, where possible. Researchers in the behavioural sciences have differed as to whether the more powerful parametric significance tests are appropriate for use with ordinal data(102). The social sciences have been divided between those whose position is that the use of parametric tests on ordinal data is incorrect on both theoretical and practical grounds(103). At the other extreme are those who advocate that this approach is usually acceptable(104). The approach adopted in this work took a more neutral approach, advocated by a majority of authors(102, 105-108), by accepting the use of parametric testing where the data appeared to approach a normal distribution.
Therefore as a general principle, descriptive summary statistics were explored, by examining distributions of the data, their means and medians, prior to estimations of confidence intervals or hyporeport testing. This was done, to test the data for normality before making a decision as to whether to use parametric testing methods in preference to non-parametric methods.
The 5% significance level was used unless otherwise indicated in the results.
2.5.1 Incident data
A computerised database was provided by the HSE for the initial analysis. The outputs of this package were:
�� Summary management system failures and root causes of incidents; �� Frequency of occurrence; �� Costs of incidents; �� Magnitude of cost associated with system failures and root causes.
22
Subsequently incident data was analysed by profiling and other descriptive methods, followed by statistical analysis. The following non-parametric tests were used:
�� Wilcoxon rank sum W test (also known as the Mann-Whitney U test) was chosen because it is suitable for small sample sizes and where normality of the data cannot be assumed. This test was used to compare the following general types of pairs of ordinal variables:
�� Staff numbers and rates between the two project benchmarking phases; �� Numbers of reported incidents between phases; �� Numbers of incidents captured by the project inclusion criteria, across both Phases; �� Incident root causes data and risk control system failures. �� Spearman’s rank-order correlation was used for examination of the rank order of Trusts, by way
of their incident reporting and capture rates as well as costs of incidents and ranks of frequency of incident root causes;
�� Kruskal-Wallis one-way analysis of variance (ANOVA) was used to examine relationships within the ordinal root cause data.
Parametric tests used were:
�� Pearson’s product moment correlation matrices to investigate the relationship between the ratio-scaled variables of incident reporting rates and Trust size. This test was also used to attempt to triangulate results from all data;
�� Independent and paired sample T-testing was used to compare incident data between test and control Trusts and to examine the proportions of incidents reported against those captured by the project inclusion criteria.
2.5.2 Questionnaire data
Evidence for normality of the questionnaire response data was obtained by visual inspection of their distributions. Kruskal-Wallis one-way ANOVA was used to examine the mean dimension scores for each Trust. One-sample Kolmogorov-Smirnov testing was carried out to test the data for normality and to provide supportive evidence that the data could be treated as normal, by using parametric methods. Analysis of Variance (ANOVA) on dimension score responses between Trusts was undertaken. Comparisons within each data collection phase were made using multiple comparison tests. Differences in responses between different job categories were explored using Scheffe multiple comparison testing. The Scheffe test was used since it is suitable for use with unequal sample sizes. The test is of particular value to identify where differences lie following an ANOVA, which indicates differences in the means under testing6 (107).
Changes in responses from individuals between the two data collection phases were analysed using paired sample T-tests.
�� Power: Based on the pilot data, with a standard deviation of 0.83 units in dimension score measurements, a sample size of 1200 would have 99% power in detecting a 0.1 shift in dimension score (say, from 6.2 to 6.3). This premise was dependent on the data from the
6 http://www.richland.cc.il.us/james/lecture/m113/post_anova.html
23
individual Trusts being pooled to create a single sample. A sample size of 200 (if the data was not pooled) would have 96% power in detecting a 0.2 shift in dimension score.
2.5.3 Workbook monitoring data
This data was nominal and was firstly analysed using descriptive methods. Associations between telephone interview responses were subsequently analysed using Cramer’s V.
24
3 RESULTS
As described in the introduction research design comprised both objective and subjective aspects. The three objective components were:
1. Identification of risk control system failures leading to the establishment of; 2. Root causes of incidents; 3. Costs of incidents.
The two subjective methods were:
4. Staff opinion survey; 5. Monitoring and audit of the use of the management system workbook, used as the vehicle to introduce the health and safety management system to five of the seven participating Trusts.
Following a general description of each of the participating Trusts, the results from each aspect of the research design will be described in turn.
3.1 TRUST DESCRIPTIVES/OBSERVATIONAL RESULTS
The seven participating Trusts were described, in general terms, in Chapter 2. Further background and more detailed descriptive information are included below.
Trust A 824 employees Trust A provided community and hospital based services over 10,000 square miles, in the north of Scotland. It had a budget of approximately £118 million, 18 hospitals and nine Local Health Care Co-operatives (LHCCs). The project was restricted to two main areas, representing some 60% of the total staff complement. These were Acute Mental Health and Learning Disabilities and a LHCC.
The Trust operated the IRIS incident reporting system within the “Safecode” suite of risk management software programmes7. Incident reports were produced quarterly. Local managers reported incidents captured by the requirements of the RIDDOR Regulations(109) to HSE
Trust B 3,378 employees The Trust provided a range of specialist services to a population of 800,000 in the north-east of England. It employed approximately 3,500 staff, with a budget of approximately £90 million for patient care. Its specialist services included mental health and learning disabilities services; community mental health services.
The entire Trust was included in the research project. Incidents (both clinical and non-clinical) were reported using a single page form. All incidents were given a risk rating by the risk management department with trend analysis produced quarterly. All RIDDOR reporting was undertaken centrally.
7 http://www.show.nhs.uk/sehd/mels/1999_18.doc
25
Trust C 1,922 employees The Trust provided primary health and mental health care services to communities in the west of Scotland. The Trust was configured into two LHCC’s and a Mental Health Directorate:
Locality Managers/Department Heads were responsible for incident reporting. Copies of all incidents were forwarded to the Occupational Health and Safety Department where a central database was maintained. Analysis was provided quarterly to localities and General Managers. The same form was used for both clinical and non-clinical incidents. Regular reports were provided to the Trust Management Team, including an annual report. RIDDOR reporting was devolved to locality managers/Risk Co-ordinators.
Trust D 5,636 employees The Trust provided primary care, community and hospital based services and supporting services to a population of approximately 400,000 in the east of Scotland. A number of the Trust's services were delivered in partnership with other statutory and voluntary organisations and the Trust had close links with and provided teaching facilities for two local Universities. The Trust employed approximately 5,500 staff and had a budget of around £234 million made up of £125 million for hospital and community health service budget and £109 million for primary care services.
Incident reporting was largely centralised although incident investigation and RIDDOR reporting was delegated via the line management function.
Trust E 1,853 employees The Trust came into being on 1st April 1994. It was the main provider of district general care and maternity services to a population of approximately 147,000 covering an area of approximately 2,500 square miles in south-west Scotland. In 1999/2000 the Trust reported that it employed 1841 staff (1417 whole time equivalents) with a projected income of approximately £50 million and a capital programme of approximately £0.8 million.
Incident reports were followed up by the in-house occupational health service.
Trust F (Control) 5,944 employees The Trust provided a comprehensive range of acute adult and paediatric care to a large city in central Scotland. 12,500 staff provided a wide range of specialist services for approximately 142,000 adults and 19,000 children each year.
The area of the Trust participating in the project was a single large teaching hospital site as it had an established health and safety management system already in place. All incidents were reported on a single initial report form regardless of type or consequence. A Trust-wide database allowed the analysis of trends. RIDDOR reporting was centralised in the Health and Safety office.
26
Trust G (Control) 3,978 employees This was a Primary Care Trust responsible for primary services, community and inpatient mental health services, child health and a range of other NHS services throughout a semi rural area in the east of Scotland. The Trust was moving through Area Redevelopment Teams to reshape mental health services towards a more community-based approach.
Incident and RIDDOR reporting was delegated to four localities.
3.2 GENERAL INCIDENT FINDINGS
The data collection periods were at different times to ensure that there was a multiple baseline(44, 85) but all periods were the same length (12 weeks or 84 days). The Trusts were allocated to a time period that was most convenient to them. The periods were as follows:
Phase 1: 1st July 2000 – 22nd September 2000 Trusts C, E 1st August 2000 – 23rd October 2000 Trusts A, B, F 1st September 2000 – 23rd November 2000 Trusts D, G
Phase 2: 1st January 2002 – 25th March 2002 Trusts A, B, F, G 1st February 2002 – 25th April 2002 Trusts C, D, E
Summary descriptive data is shown in Table 4. The first two columns show the staff head-count and the whole-time equivalent (WTE) staff numbers for both phases of the project. Not all of the Trusts were able to provide staff head count data for the second phase but all were able to provide WTE data. Therefore only WTE data was used later in the results to calculate rates.
The third column shows the WTE data, standardised for the length of the data collection period (84 days), with the total incidents reported in the fourth column. The average reported incident rate (expressed in per person working years) is in the fifth column.
The last two columns show the number of incidents included in the study (referred to as “captured” by the selection criteria) and the percentage of incidents captured.
27
Tabl
e 4
Gen
eral
inci
dent
resu
lts fr
om b
oth
phas
es
Trus
t St
aff h
ead
coun
t W
hole
-tim
e st
aff
equi
vale
nt
(WTE
)
Pers
on w
orki
ng
days
(WTE
*84
days
)
Tota
l in
cide
nts r
epor
ted
Mea
n re
porte
d in
cide
nt ra
te
(per
per
son
wor
king
ye
ar8 )
Tota
l in
cide
nts
incl
uded
in st
udy
(“ca
ptur
ed”)
Perc
enta
ge in
cide
nts
capt
ured
Trus
t A 1
st p
hase
65
4 54
,936
46
51.
91
40
8.6%
Tr
ust A
2nd
pha
se
824
662
55,6
08
488
(+5%
) 1.
98
47
9.6%
Tr
ust B
1st p
hase
2,
977
250,
068
2,03
4 1.
82
20
1.0%
Tr
ust B
2nd
pha
se
3,37
8 2,
969
249,
396
2,92
7(+4
4%)
2.63
16
0.
6%
Trus
t C 1
st p
hase
1,
334
112,
056
583
1.17
33
5.
7%
Trus
t C 2
nd p
hase
1,
922
1,40
6 11
8,10
4 80
0 (+
37%
) 1.
53
42
5.3%
Tr
ust D
1st p
hase
4,
011
336,
924
1,99
2 1.
33
59
3.0%
Tr
ust D
2nd
pha
se
5,63
6 3,
927
329,
868
2,09
3(+5
%)
1.42
12
5 6.
0%
Trus
t E 1
st p
hase
1,
430
120,
120
290
0.54
31
10
.7%
Tr
ust E
2nd
pha
se
1,85
8 1,
5699
131,
796
220
(-24
%)
0.38
20
9.
1%
Trus
t F 1
st p
hase
4,
121
346,
164
547
0.36
32
5.
9%
Trus
t F 2
nd p
hase
5,
944
Con
trol
4,15
0 34
8,60
0 57
7 (+
5%)
0.38
30
5.
2%
Trus
t G 1
st p
hase
2,
787
234,
108
635
0.61
30
4.
7%
Trus
t G 2
nd p
hase
4,
290
Con
trol
2,92
2 24
5,44
8 10
02 (+
58%
) 0.
92
26
2.6%
To
tal 1
st p
hase
17
,314
1,
454,
376
6,54
3 1.
01
245
3.7%
To
tal 2
nd p
hase
23
,852
17
,605
1,
478,
820
8,10
7 (+
24%
) 1.
24
306
3.8%
8 Ass
umes
225
wor
king
day
s per
yea
r (45
wee
ks x
5 d
ays p
er w
eek)
9 T
his i
ncre
ase
was
due
to ta
king
ove
r hot
el se
rvic
es st
aff
29
The numbers of staff covered by the research ranged from 824 (representing 654 WTE at the first baseline) within Trust A, to 5,944 (4,150 WTE) within Trust F (control). The difference in WTE between the two phases did not vary by more than 5%, with the exception of Trust E which rose by 10% as a consequence of taking over hotel services staff from another unit during the study. Nevertheless the differences between the mean WTE numbers between the two phases were not significant (Wilcoxon rank sum W test P=0.20).
The total number of incidents reported was 6,543 in Phase 1 and 8,107 in Phase 2. The number of incidents reported varied between Trusts, with an approximate order of magnitude between the lowest (255 for Trust E (mean of both phases)), and the highest (2,481 for Trust B (mean of both phases)). The mean number of reported incidents rose by 24% in the second phase. However, in Trust E fewer incidents were reported in the second phase (mean fell by 24%). The overall reduction was marginally significant (Wilcoxon rank sum W test P=0.06). However, given the small sample size it would be prudent to assume a lack of significance.
The number of incidents reported from the control Trusts was not found to be different from the test Trusts (Independent samples t-test P=0.49 for phase 1 and P=0.58 for phase 2, although because of the small sample size the confidence intervals were wide (between –1728 and +2760)).
The total number of incidents that met all three incident inclusion criteria, (i.e. the total number of incidents captured) was 245 for phase 1 and 306 in phase 2, an increase that was not significant at the 0.05 level (Wilcoxon rank sum W test P=0.61).
The captured incidents represented an average of 3.7% of the total number reported (range 0.6% to 10.7%). This is a very small proportion, brought about by large number of reported incidents, which did not meet the inclusion criteria largely because of their trivial nature. The mean number of incidents captured per Trust was 40, with more captured in phase 2 (35 in phase 1 against 46 in phase 2), again an increase that was not significant (Wilcoxon rank sum W test P=0.73).
Although the total numbers of reported incidents were broadly within the same order of magnitude for each Trust across both data collection phases, intra-Trust variation between the two phases was noticeable (between +5% for Trust A and Trust D and +58% for Trust G (control). Therefore the relationship between the reported incidents were standardised by person working years. However, the resultant differences between rates within each Trust across the two phases (column 5 in Table 4) were not significant between the two phases (Wilcoxon rank sum W test P=0.61).
The mean incident-reporting rate increased from 1.01 incidents reported per person per year to 1.24 incidents reported per person per year. This was marginally significant (Wilcoxon rank sum W test P=0.09). Due to the small sample size it is regarded as prudent to regard the result as non-significant. The rank order of reporting rates (expressed as per person working year) is shown in Table 5.
30
Table 5 Rank order for incident reporting rates between both phases
Phase 1 Average reported incident rate (per person working year)
Phase 2 Average reported incident rate (per person working year)
Trust F (Control) 0.36 Trust F (Control) 0.38 Trust E 0.54 Trust E 0.38 Trust G (Control) 0.61 Trust G (Control) 0.92 Trust C 1.17 Trust D 1.42 Trust D 1.33 Trust C 1.53 Trust B 1.82 Trust A 1.98 Trust A 1.91 Trust B 2.63 MEAN 1.01 MEAN 1.24
None of the Trusts changed their rank by more than one place between the two phases of data collection. Three Trusts maintained their rank between phases while the ranking interchanged between the remaining four Trusts (Trusts C and D, and A and B). The acute Trusts (Trust F and Trust E) and the control Trusts (Trusts F and G) had, between them, the lowest rates in both data collection phases. The four test PCT Trusts had the highest reporting rates. The Spearman rank order correlation coefficient between the rates for the two phases was 0.919 (P<0.01), confirming a significant relationship.
This suggests that the rates of reported incidents were independent of the size of the Trust. This was confirmed when the Pearson correlation between WTE and incident reporting rate (per person per year) were examined and found to be unassociated (P=0.48)).
Intuitively this seems likely, given the different client groups within the two types of Trust and the likelihood of recording more incidents related to the management of patients with mental health problems and learning difficulties in the PCTs. However, the variation between the five PCTs was approximately four-fold (between 0.61 and 2.63 incidents per person per year). This points to the variation in reporting rates being more related to the reporting culture and conventions within the individual Trusts, rather than differences between the type of Trust.
Despite the large variation in numbers of incidents reported, the numbers actually captured by the study criteria were relatively low (between 0.6% and 10.7% of the total number of incidents reported), because of the large numbers of incidents that were reported but not captured. These captured incidents can be regarded as the most serious or the reported incidents. The ratio of serious to minor (i.e. captured versus non-captured) reported incidents is sometimes regarded as a measure of safety culture – the lower the proportion of serious incidents the better the (reporting) culture (e.g. (66)). The low percentage of captured incidents adds weight to this hyporeport.
The rank order of the rates of captured incidents, across the two data collection phases, is shown in Table 6.
31
Table 6 Rank order for incident capture rates between both phases
Phase 1 Average captured incident rate (per person working year)
Phase 2 Average captured incident rate (per person working year)
Trust B 0.018 Trust B 0.014 Trust F (control) 0.021 Trust F (control) 0.020 Trust G (control) 0.029 Trust G (control) 0.025 Trust D 0.039 Trust E 0.037 Trust E 0.058 Trust D 0.083 Trust C 0.066 Trust C 0.084 Trust A 0.164 Trust A 0.193 MEAN 0.056 MEAN 0.065
The difference between phases of the mean rates of number of incidents captured per person per year was not significant (Wilcoxon rank sum W test P=0.61). The ranking of individual Trusts was maintained across the two data collection phases with the exception of Trusts D and E. The lowest capture rate was Trust B and the highest was Trust A.
Although Trust B had the lowest capture rate it also had one of the highest reported incident rates (see Table 4). When this relationship was explored further with all of the Trusts there was no correlation between rates of reported and captured incidents (P=0.54 for phase 1 and P=0.70 for phase 2). This indicates that the rates of captured reported incidents were not linked to rates for all reported incidents.
Unlike rates of reported incidents, there was an association between the percentage of captured incidents and the size of Trust. This was significant in the first phase (P=0.01) but only marginally significant in phase two (P=0.09). Given the small sample size it is prudent to interpret this finding with caution but it may be that the more serious incidents are linked to some other source than reporting culture, possibly safety performance. This possibility is explored further in Chapter 4.
The test and control Trusts were not significantly different from each other in either phase for either reported or captured incidents (P=0.50 for both phases).
3.2.1 “Special” category
At the start of the field work it was anticipated that most incidents that were identified would have readily identifiable management root causes associated with them(43). This was because of general acceptance of the paradigm that accidents are largely a consequence of deficiencies in the health and safety management system(99).
However investigation of approximately half of the incidents captured did not reveal readily identifiable risk control system failures or management root causes. These incidents fell into a “grey” area between obvious system failures and root causes and meeting the incident inclusion criteria. For example:
�� By the time a junior doctor reported a needlestick incident and it was subsequently identified for further investigation, the doctor had moved to another placement elsewhere in the UK and new contact details were unavailable;
�� In the care of elderly, frail patients, individual clinicians routinely had to make professional judgments as to when it was appropriate to encourage mobilisation so that independence,
32
and ultimately quality of life, could improve. This benefit had to be balanced against the risk that, in so doing, the patients’ risk of falling increased;
�� Some mental health patients were being encouraged to take a more active part in society by being given periods of parole, during which time their freedom was increased. The purpose of this care approach was to build trust with the aim of the patient being able to re-integrate into society. This aim was in the context of a subjective balance with the risk that the patient might abscond or self-harm.
It was judged inappropriate to simply exclude these cases from the study since, on balance they were judged to fall within the inclusion criteria already described. They were therefore assigned to a third “special” category of incident types10.
After the completion of the first data collection exercise special category incidents were identified as belonging to one or more of four different types11. These are listed below but also discussed further in Chapter 4:
�� Health and safety systems were in place within the bounds of what was considered (by the research team) to have been reasonably practicable. An example of this type of incident was where a patient fell, who had been assessed as independently mobile, with no history of falls and already under general observation. It was judged that the same assessment would have been made in similar circumstances;
�� Insufficient information emerged from investigations to come to reliable conclusions. An example of this type of incident was the fall of a patient where the fall was unwitnessed and the patient was unable to offer any explanation for what happened;
�� With hindsight the incident may have been preventable but, based on the facts that were known at the time, the risk was unlikely to have been foreseen. An example of this type of incident was where a gust of wind smashed an open window. The window was normally opened for ventilation and had never before caused problems;
�� The key to assessing how well the risk was managed was associated with individual practitioners’ clinical judgement. An example of this type of incident was where a patient on a soft-food diet was being re-introduced to textured foods. The patient choked and had to be taken to the casualty department for treatment.
For some incidents of this type it was subsequently possible to identify management root causes. Where an incident was identified as having the potential to be included within the special category its various merits were discussed in a case-conference by the project team and with the involvement of various specialists, independent of the project, where necessary. Extensive measures were applied to ensure that every identified possibility was examined in detail before a consensus was reached as to whether the incident was to be assigned to the special category group and, if so, to which type.
It can therefore be assumed that the assignment to the categories was internally consistent. However, a potential limitation was that the need for a special category was only identified during the first data collection phase, in response to the increasing numbers of incidents that had
10 These incidents are listed and described in Appendix 8
11 The distinction between the types was not exact and there were occasions where more than one special category type was assigned to a single incident. For example, if a patient who was not assessed as being at risk from falling, and had never fallen before (risk possibly preventable but not foreseeable) fell but the exact circumstances were not witnessed and the patient was not able to offer any explanation as to what happened (insufficient information to come to a reliable conclusion), the incident would be assigned under both types.
33
defied efforts to identify system failures and root causes. It is therefore possible that the consistency was less reliable with incidents that occurred at the start of phase 1. These incidents were reviewed again during the second phase, to minimise this effect, but this could only achieve partial assurance, given that so much time had elapsed since they had occurred.
Table 7 shows the total numbers of incidents reported; those that were captured by the inclusion criteria; and those that were assigned to the special category. The table also shows the rates (per person per year) for incidents for which root causes could be identified (column 4) and those assigned to the special category (column 6).
35
Tabl
e 7
Inci
dent
s sp
lit b
y m
anag
emen
t roo
t cau
se a
nd s
peci
al c
ateg
ory
from
bot
h ph
ases
Tr
ust
Tota
l in
cide
nts
repo
rted
Tota
l in
cide
nts
capt
ured
Num
ber
of
inci
dent
s (r
oot
caus
es a
ssig
ned)
Aver
age
inci
dent
rat
e (r
oot
caus
es
assig
ned)
(p
er
pers
on w
orki
ng y
ear)
Num
ber
of in
cide
nts
(roo
t ca
uses
not
ass
igne
d) (
i.e.
spec
ial c
ateg
ory)
Aver
age
inci
dent
rat
e (r
oot
caus
es
not a
ssig
ned)
(i.e
. spe
cial
cat
egor
y)
(per
per
son
work
ing
year
) Tr
ust A
1st
pha
se
465
40
20
0.08
20
0.
08
Trus
t A 2
nd p
hase
48
8
47
27
0.11
20
0.
08
Trus
t B 1
st p
hase
2,
034
20
15
0.01
5
0.01
Tr
ust B
2nd
pha
se
2,92
7 16
11
0.
01
5 0.
01
Trus
t C 1
st p
hase
58
3 33
16
0.
03
17
0.03
Tr
ust C
2nd
pha
se
800
42
13
0.02
29
0.
06
Trus
t D 1
st p
hase
1,
992
59
28
0.02
31
0.
02
Trus
t D 2
nd p
hase
2,
093
125
34
0.02
91
0.
06
Trus
t E 1
st p
hase
29
0 31
22
0.
04
9 0.
02
Trus
t E 2
nd p
hase
22
0 20
16
0.
03
4 0.
01
Trus
t F 1
st p
hase
54
7 32
16
0.
01
16
0.01
Tr
ust F
2nd
pha
se
577
30
20
0.01
10
0.
01
Trus
t G 1
st p
hase
63
5 30
10
0.
01
20
0.02
Tr
ust G
2nd
pha
se
1002
26
10
0.
01
16
0.01
To
tal 1
st p
hase
6,
543
245
127
0.03
11
8 0.
03
Tota
l 2nd
pha
se
8,10
7 30
6 13
1 0.
03
175
0.03
36
The number of incidents captured was particularly influenced by an increase in those from the special category in Trust C (increased from 17 to 29) and Trust D (increased from 31 to 91). Both increases were for incidents involving violence and aggression. These resulted in an increase in the total number of special category incidents of 48% between the two phases. However, the differences between the Trusts were not significant (Wilcoxon rank sum W test P=0.69).
Some of the cases included repeated incidents involving a single patient. It could be argued that, once an incident had occurred its re-occurrence could be predicted and therefore should have a root cause. This applied to some of the incidents and, in these cases, the repeated incidents were root caused. However, some were still classed as special category. These were where, for example, the decision to maintain a care plan was a clinical judgment such as a decision to continue to give parole to a patient with a history of absconding, so that they could continue to attempt rehabilitation.
The total number of incidents that were found to have management root causes associated with them (i.e. those minus the special category incidents) were 127 in phase 1 and 131 in phase 2. This represented an increase of 3% between the two phases, which was not found to be significant (Wilcoxon rank sum W test P=0.87)12.
There were also more special category incidents between phase 2 and phase 1 (175 special category in phase 2 against 118 in phase 1) but again this increase was not significant (Wilcoxon rank sum W test P=0.69). This is further discussed in Section 4.4.3.
The percentage of special category incidents did not differ significantly between Trusts (Spearman’s rank order correlation P=0.7). The percentage of special category incidents that were assigned to each of the four categories identified above is shown in Figure 3. It can be seen that most incidents were assigned to the category where the project team judged that avoidance of the incident was out with what could be thought of as reasonable (50% of number of incidents). One quarter of the incidents was based on a clinical judgment that was under the responsibility of the clinician in charge of the patient (26%). The risk was not foreseeable in 17% of the special category incidents and the remaining group was characterised by insufficient information on which to base a judgment (7%).
The percentage of special category incidents that were assigned to each of the four categories identified in 3.2.1 is shown in Figure 3.
12 These incidents are listed (for both phases) and described in Appendix 7
37
Figure 3 Comparison between number and cost of special category incidents
The traditional assumption has been that most, if not all, incidents can be prevented by the implementation of robust risk management policies and procedures(18, 25, 110). The special category incidents are in potential conflict with this paradigm and may represent a consequence of the culture and inherent risks within the NHS whereby a percentage of this type of incident may always exist. It may also be an outcome of the wide definition of “incidents” used in the NHS, which was described in the introductory chapter. The issue is discussed further in Chapter 4.
3.3 RISK CONTROL SYSTEM FAILURES AND ROOT CAUSES OF INCIDENTS
The HSE root cause analysis methodology used in the research was based on their three-level model, described in Chapter 1. This provided a mechanism for identifying management root causes (level 1 failures) from existing workplace precautions (level 3 failures) using identified risk control system failures (level 2) as a means of linking the two(99). Therefore the identification of risk control system failures for each incident was an important pre-requisite to identifying the underlying management root causes.
50
7
17
25
51
16 16 16
0
10
20
30
40
50
60
Outwith reasonablepracticability
Insufficientinformation
Risk was notforeseeable
Risk managementdecision based onclinical judgement
Numbers of incidents (%)Cost of incidents (%)
38
The process identified in Chapter 2 for assessing incidents against the inclusion criteria and then identifying sufficient details about each incident to allow risk control systems and root causes to be identified consistently proved to be challenging for both the project team and the Trust representatives. The process undoubtedly raised confidence in the research results but it was time consuming to undertake.
In practice the investigators in the individual Trusts tended to be resistant to attempts by the research team to modify their normal approach to investigation. Most issues centered on evidence for conclusions reached regarding root causes of incidents. These normally involved requests from the research team for more information on which to base judgments. These requests were generally liable to be interpreted, particularly by safety practitioners with several years’ experience, as potentially calling their competency into question. Incident investigation seemed to have taken on an almost artistic form with the Trust investigators generally taking the view that they could usually establish a root cause with the most meager of basic information, making assumptions (“gut instinct”) based on and justified by their detailed knowledge of the organisation. This issue will be further discussed in Chapter 4.
Nevertheless, the measures taken, to ensure consistency of investigation between Trusts (described in Chapter 3), were thought to have been effective and that the results can therefore be regarded as robust.
The risk control system failures identified will be described in Section 3.3.1. The management root causes will be described in Section 3.3.2.
3.3.1 Risk control system failures (HSE Level 2(99))
A range of 17 different risk control system failures was identified in phase 1. The number of times a root cause was found in each system is shown in Table 8. The risk control systems ranged (alphabetically) from control of contractors to work organisation. The level 1 management root causes, based on HSE guidance(18), ranged from policy to reviewing. Failures within the risk control system for risk assessment predominated with 48% of the identified root causes assigned to this system. With the exception of the risk control systems for procedures (10%) and training (11%) all of the other systems represented less than 10% of the total percentage of root causes identified.
39
Table 8 number of management root causes associated with risk control system failures in seven NHS trusts (Phase 1)
Root cause Risk control system Policy Implem-
enting Control Plan-
ning Compet-ence
Commun-ication
Coop-eration
Meas-uring
Review-ing
Control of contractors
0 1 1 0 0 1 0 2 2
Disposal 0 6 2 2 0 2 2 5 5 Drug administration
0 1 0 1 0 0 0 0 0
Emergency procedures
0 7 1 4 4 6 4 7 6
Handling sharps 1 0 0 0 0 0 0 0 0 Inspections 0 2 0 2 0 0 0 2 1 Maintenance 1 3 0 2 0 0 0 1 1 Management of violence and aggression
0 9 3 5 3 2 2 7 6
Procedures 4 15 1 4 2 5 2 11 11 Procurement 0 2 0 0 0 0 0 0 0 Responsibilities 0 7 0 0 0 0 0 0 0 Risk assessment 0 60 16 36 9 17 21 58 62 Security 6 0 0 0 0 0 0 0 0 Supervision 1 4 0 1 0 0 0 3 3 Training 1 12 3 10 5 6 6 11 8 Work environment
9 0 0 0 0 0 0 0 0
Work organisation
0 5 2 2 0 0 0 1 1
TOTAL 23 134 29 69 23 39 37 113 109
The most noticeable feature in Table 8 was the extent to which the risk control system of risk assessment dominated the others. For example, the number of times that management root causes were assigned in any of the other systems did not exceed 15, whereas for the system of risk assessment, planning was assigned over 30 times, measuring over 50 times and implementing and reviewing in excess of 60 times each. This indicates that the main management root causes of incidents in the first phase were related to these four root causes in the risk control system of risk assessment.
In the second data collection phase 18 different systems were identified, one more than in the first phase. The pattern is shown in Table 9. Fifteen of the systems were shared between the two phases. Two were unique to the first phase (supervision, and maintenance) while three were only identified in the second phase (control of infection, stress management, and hazard reporting).
40
Table 9 number of management root causes associated with risk control system failures in seven NHS trusts (Phase 2)
Root cause Risk control system Policy Implem-
enting Control Plan-
ning Compet-ence
Commun-ication
Coop-eration
Measur-ing
Review-ing
Control of contractors
0 3 1 3 1 1 1 2 2
Control of infection
0 2 0 1 0 1 0 0 2
Disposal 0 4 0 3 0 2 0 4 4 Drug administration
0 1 0 1 0 0 0 1 1
Emergency procedures
2 7 1 6 1 0 0 6 7
Handling sharps
0 17 7 12 7 9 0 9 10
Hazardreporting
0 1 0 0 0 1 0 1 1
Inspections 0 1 1 1 0 0 0 1 1 Management of violence and aggression
0 21 9 20 18 7 13 22 21
Procedures 8 7 0 3 1 2 1 4 3 Procurement 0 1 0 0 0 1 0 1 1 Responsibilities 0 10 0 0 0 0 0 0 0 Risk assessment
0 51 10 31 12 14 2 39 47
Security 2 1 0 1 0 0 0 1 1 Stress management
2 0 0 0 0 0 0 0 0
Training 1 9 6 8 1 4 1 6 6 Work environment
1 8 5 8 5 0 5 0 0
Work organisation
1 0 0 0 0 0 0 0 0
TOTAL 17 144 40 98 46 42 23 97 107
The total numbers of root causes assigned to risk control systems was not significantly different between phase 1 and phase 2 (Wilcoxon Signed Ranks Test P=0.515). The system of risk assessment was again the dominant source of management root causes, as noted in phase 1 (Table 8), although the percentage had significantly reduced from 48% in phase 1 to 34% in phase 2 (P=0.02). Four other risk control systems had marginally significant reductions in numbers of assigned root causes. These were:
Disposal (P=0.07); Procedures (P=0.05); Training (P=0.06); Work organisation (P=0.07).
However, root causes assigned to the system of management of violence and aggression rose significantly from 6% in the first phase to 21% in the second (P=0.01). The system of handling sharps also increased significantly from a negligible percentage (0.2%) in the first phase to 12%
41
in the second (P=0.02). The system of control of contractors also rose but this increase was only marginally significant (P=0.07).
There were undoubtedly more incidents involving violence and aggression reported in the second phase, as compared to the first. It is unclear whether this was a real increase in incidence of this type of incident or as a consequence of better reporting possibly due to heightened awareness.
In the first phase the only management root cause identified for the system of handling sharps was policy. In the second phase seven additional root causes were evident in the system for handling sharps. These were in the areas of planning, implementing, control competence, communication, measuring and reviewing. The NHS has been engaged in major campaigns to raise awareness regarding the risk of needlestick injuries or potential occupational exposure to blood-borne viruses. It is therefore possible that this has improved reporting.
However, it is also possible that the effect could be caused by a systematic error brought about by increased familiarisation by the research team with the root causation methodology in the second phase when compared with the first. Also, in the root cause methodology, if an incident was identified as not having a policy there was no further assessment against the other criteria. This was because HSE had reasoned that the absence of a policy was indicative of the absence of any other aspects of a management system. This approach was not as strict as might be assumed since, during investigations, the presence of an informal local arrangement that was known to staff was taken to be a policy for the purposes of project data capture. These points are discussed further in Chapter 4.
3.3.2 Management root causes (HSE Level 1)
The number of times a management root cause(99) was identified within the incidents captured for all seven Trusts is shown, for both Phases, in Figure 4. Given there was no significant difference between the test and control data for risk control system failures and management root causes, the data was combined. This was done so that the overall visual impression of differences between the individual management root causes could be revealed.
42
Figure 4 Management root causes in seven NHS Trusts
The range for phase 1 was between 23 times for policy and 134 times for implementing. The four main root causes most frequently identified were implementing (134 times), planning (69 times), measuring (113 times), and reviewing (109 times).
The range of citations for phase 2 was between 10 times for policy and 116 times for implementing. In Phase 2 the number of times management root causes were cited was, on average, approximately 14% less than in the first phase. Control, planning and competence increased by 31%, 13% and 57% respectively. The same four root causes as in phase 1 were the most frequently identified (implementing (116 times), planning (78 times), measuring (81 times), and reviewing (92 times)).
The differences in the numbers of cited management root causes between the two phases were not significant (Wilcoxon rank sum W test P=0.92).
The rank order was broadly similar between the two phases. Table 10 shows the ranks split into two arbitrary groups. These are the major group (cited more than 50 times) and the minor root causes group (cited less than 50 times).
0
20
40
60
80
100
120
140
160
Polic
y
Impl
emen
ting
Con
trol
Plan
ning
Com
pete
nce
Com
mun
icat
ion
Co-
oper
atio
n
Mea
surin
g
Rev
iew
ing
Num
ber o
f man
agem
ent
root
cau
ses
iden
tifie
d
Phase 1 Phase 2
43
Table 10 Rank order of frequency of citation of management root causes
Phase 1 Phase 2 Implementing (134 times) Implementing (116 times) Measuring (113 times) Reviewing (92 times)Reviewing (109 times) Measuring (81 times)
Major management root causes
Planning (69 times) Planning (78 times)Communication (39 times) Control (38 times) Co-operation (37 times) Competence (36 times)Control (29 times) Communication (27 times)Competence (23 times) Co-operation (19 times)
Minor management root causes
Policy (23 times) Policy (10 times)
Implementing, planning, and policy held the same rank between phases (1st, 4th and 9th). Reviewing and measuring shared 2nd and 3rd ranking between phases, and communication, co-operation, control and competency shared the rankings between 5th and 8th. The Spearman order correlation coefficient was 0.695 (P<0.05), indicating a significant correlation between the ranks of the root causes between the two phases.
This ranking was broadly similar to the findings of the pilot work(34), where the main root causes were found within the areas of planning and implementing. Measuring and reviewing were not identified as significant root causes within the pilot, although control and competence were highlighted. These two root causes were among the least frequently cited root causes in the main project. A possible explanation for this could be related to history effects between the pilot work, which was undertaken in 1998, and the main study conducted in 2000 and 2001.
It may also be that the root causes identified in the pilot site were characteristic of that hospital and that each Trust might have a distinctive root cause “profile” related to the relative strengths and weaknesses of its health and safety management system. Evidence for this hypo-report was inconclusive. For example, Trusts A, D, E, F, and G were significantly associated between the first and second phases (P<0.05). However, Trusts B and C were not. The pattern in the first phase for Trust C was not significantly correlated with any other Trust except that of Trust G in the second phase (Spearman’s rank order correlation P=0.03). Conversely Trusts E and G were significantly associated with all of the other Trusts (with the exception of Trust C) (P<0.05). The remaining four Trusts had varying degrees of associations with the others.
Another interpretation might be that more risk assessments were carried out but the main cause of incidents was that these had not been implemented sufficiently robustly. This is consistent with subjective observations within the Trusts whereby, in most cases, measures had been put in place to enhance aspects of the health and safety management system during the period following the first data collection phase. However, unless these were adopted and implemented by line managers and staff, incidents with this root cause were predominant.
This results from the risk control systems failure and root cause analyses was potentially useful information for the participating Trusts, in that it gave an indication of where practical effort might be best directed for maximum impact in reducing incident rates. In other words, if effort was directed at the planning, implementing, measuring and reviewing aspects of the system of risk assessment, then some incidents might be prevented or their severity reduced.
This information was communicated to all participating Trusts at feedback sessions conducted at the end of the second data collection phase.
44
3.4 COSTS OF INCIDENTS
As described in Chapter 2 and Appendices 1 and 2, costs of incidents were collected for both phases.
Those incidents that had management root causes associated with them were assigned costs in both phases, while the costs of the special category incidents were only included in phase 2. This was because, as described in 3.2.1, the need for a special category was only identified during the first data collection phase. Costs were not collected for these incidents during the initial investigation and once the special category groupings were established it was judged that too much time had elapsed to obtain reliable information on which to base cost estimations.
With the exception of some staff opportunity costs, all costings were carried out to the same protocol as in the pilot work(34). This exception was to include costs for staff taken away from their normal duties by the effects of the incident. This was because the bulk of costs associated with incidents were incurred by staff absence and replacement labour. The breakdown across each Trust is shown in Table 11. The costs for each phase incurred by each Trust, is shown along with the costs incurred as a result of the incidents that were allocated to the special category.
Table 11 Costs incurred by Trusts in both phases
13 Note that the costs were at 2000 prices for both data collection phases 14 Figures in (brackets) represent the percentage increase/decrease in cost between phase 1 and phase 2
13 Trust A Trust B Trust C Trust D Trust E Trust F (Control)
Trust G (Control)
Total
Phase 1 127incidents
£3,494 20incidents
£22,285 15incidents
£1,960 16incidents
£3,867 28incidents
£8,470 22incidents
£4,198 16incidents
£6,822 10incidents
£51,096
Phase 2 131incidents
£14,779 (+323%)14 27incidents
£11,272 (-49%) 11incidents
£3,592 (+83%) 13incidents
£7,711 (+99%) 34incidents
£1,207 (-86%) 16incidents
£12,734 (+203%) 20incidents
£1,914 (-72%) 10incidents
£53,209 (+4%)
Phase 2 special category 159incidents
£15,986
21incidents
£256
5incidents
£4,563
13incidents
£13,948
91incidents
£3,852
4incidents
£6,602
9incidents
£10,387
16incidents
£55,593
Phase 1 rate (cost per person working day (WTE*84)
£0.06 £0.09 £0.02 £0.01 £0.07 £0.01 £0.03 £0.04 (mean)
Phase 2 rate (cost per person working day (WTE*84)
£0.27 £0.05 £0.03 £0.02 £0.01 £0.04 £0.01 £0.04 (mean)
46
In the first data collection phase, the total cost of incidents that met the inclusion criteria for all seven Trusts was £51,096 over twelve weeks (127 incidents). The costs incurred by each Trust varied between £1,960 (Trust C) and £22,285 (Trust B). In the second phase, the total cost was £53,209, an increase of 4% from the first phase (131 incidents)15. The range was between £1,914 (Trust G) and £12,734 (Trust F). These were both control Trusts.
Costs varied considerably both between Trusts and within phases. Between the first and second phase costs increased in four Trusts and decreased in three. The largest change occurred in Trust A where an increase of +323% in total costs occurred between the two phases. Trust F (control) increased by just over 200% (+203%). Costs fell for three Trusts (-49% for Trust B, –72% for Trust G (control) and –86% for Trust E). However, the overall total average cost only changed by +4% between phases.
This may represent the inherent variability in costs of incidents since the differences between the test and control Trusts was not significant in either phase (Mann-Whitney U Test: P= 0.86 (phase 1); P=1.0 (phase 2)). Between the first and second phase the costs across all seven Trusts were not found to be significantly different (Wilcoxon rank sum W Test P=0.87).
So far as rates were concerned (cost per person working day (i.e. WTE*84)), these ranged between £0.01 per person working day and £0.27 per person working day. These rates were not significantly different between phases (Wilcoxon rank sum W Test P=0.20)
3.4.1 Staff/patient incidents
Incidents involving staff were nearly three times more frequent and represented approximately four times the cost than those for patients. The distribution of incidents incurred by staff, patients and others (e.g. members of the public or property damage) is shown in Figure 5. It can be seen that the two phases were broadly similar and that most incidents involved staff (70% in phase 1 and 69% in phase 2). Incidents involving patients accounted for 20% and 29% for Phases 1 and 2 respectively. In the “other” category there were 10% in phase 1 and 2% in phase 2.
15 Costs were assigned within both phases at 2000 prices
47
Figure 5 Number of incidents classified as staff, patient or other (not including special category incidents)
Figure 6 Cost of incidents classified as staff, patient or other (not including special category incidents)
26
89
12
38
90
3
0
10
20
30
40
50
60
70
80
90
100
No. of patient incidents No. of staff incidents O ther
No.
of i
ncid
ents
Phase 1 Phase 2
£9,534
£39,826
£1,730
£9,640
£43,508
£60£0
£5,000
£10,000
£15,000
£20,000
£25,000
£30,000
£35,000
£40,000
£45,000
£50,000
No. of patient incidents No. of staff incidents Other
Cos
t
Phase 1 Phase 2
48
When cost was taken into account (Figure 6) this distribution was maintained with incidents involving staff incurring the highest cost (78% in phase 1 and 81% in phase 2). Incidents involving patients accounted for 19% and 18% of the costs incurred in phases 1 and 2 respectively. In the “other” category the costs represented 3% in phase 1 and 1% in phase 2.
3.4.2 Special category
So far as special category incidents were concerned, costs were only obtained for the second data collection phase, as explained earlier. There were 159 special category incidents in the second phase accounting for a total loss of £55,593. This is a significant finding in that the total costs for these incidents was more than either of the two data collection phases for the incidents where root causes could be assigned.
With the exception of Trust B the highest costs were incurred by the primary care trusts. In addition, most special category costs were attributed to incidents involving patients, which is contrary to the findings for those incidents for which root cases could be established, as shown in Figure 7.
Figure 7 Cost of incidents classified as staff, patient or other (phase 2 and special category incidents)
Figure 7 shows that the cost of patient incidents was approximately three and a half times (3.64) that of the captured incidents in phase 2. Special category staff incidents were approximately two and a half times less than patient incidents from phase 2. When the phase 2 and special category costs were combined, costs for incidents involving staff (£59,896) were still in excess of that for patients (£44,736).
£9,640
£43,508
£60
£35,096
£4,146
£16,388
£0
£5,000
£10,000
£15,000
£20,000
£25,000
£30,000
£35,000
£40,000
£45,000
£50,000
Cost of patient Incidents Cost of Staff Incidents Cost of Other
49
Therefore costs for staff were consistently higher than those incurred for patients. Possible reasons for this may be that the management of health and safety risks for patients is more controlled than that for staff. This is discussed further in Chapter 4.
3.4.3 Management root causes of incidents
Each incident usually had multiple management root causes. In the HSE methodology the software had the facility to allow more than one management root cause to be assigned per incident. The software then divided the total cost for the incident by the allocated number of management root causes to obtain a cost per management root cause for that incident. The method did not include an evaluation of the relative importance of each management root cause for each incident so the costs were assigned equally. This may have reduced the sensitivity of the method but ensured reproducibility between incidents and phases.
The total cost assigned to any particular management root cause was then obtained from all of the incidents. The numbers of incidents and their costs allocated to identified management root causes (i.e. not including the special category incidents) for both phases are shown in Table 12.
The pattern and distribution of costs was broadly similar to the root causes associated with failed systems (Table 12), in that the main areas of weakness were related to the categories of planning, implementing, measuring, and reviewing. For example, in Trust E costs decreased for every management root cause in the second Phase while in Trust A only communication cost decreased. The pattern of costs for each Trust was similar for both data collection Phases.
50
Tabl
e 12
Num
ber o
f inc
iden
ts a
nd c
osts
per
man
agem
ent r
oot c
ause
16 P
hase
2 d
ata
in (b
rack
ets)
Trus
t A
Trus
t B
Trus
t C
Trus
t D
Trus
t E
Trus
t F
(Con
trol)
Trus
t G
Cont
rol)
Ove
rall
Tota
ls Fo
r 7
Trus
ts N
umbe
r of i
ncid
ents
20
(2
7)16
15 (1
1)16 (1
3)28 (3
4)22 (1
6)16 (2
0)10 (1
0)12
7(1
31)
Ave
rage
cos
t per
man
agem
ent r
oot c
ause
Po
licy
£7
(£56
2)
£31
(£23
9)
£1,3
20
(£0)
£1
58
(£5)
£3
2 (£
3)
£190
(£
5,48
7)
£263
(£
132)
£2
,001
(£
6,42
8)
Impl
emen
ting
£792
(£
4,69
7)
£5,7
26
(£3,
963)
£3
01
(£1,
239)
£9
61
(£3,
070)
£1
,420
(£
260)
£8
03
(£3,
387)
£2
,334
(£
323)
£1
2,33
7 (£
16,9
39)
Con
trol
£179
(£
321)
£2
8 (£
389)
£2
6 (£
779)
£1
38
(£44
) £1
,255
(£
17)
£477
(£
0)
£0
(£11
) £2
,103
(£
1,56
1)
Plan
ning
£3
49
(£3,
652)
£2
,785
(£
385)
£2
5 (£
939)
£3
01
(£1,
103)
£2
37
(£23
0)
£519
(£
1,17
8)
£785
(£
303)
£5
,001
(£
7,79
1)
Com
pete
nce
£0
(£1,
398)
£2
8 (£
10)
£0
(£52
) £1
51
(£42
) £1
70
(£12
) £1
82
(£11
0)
£245
(£
270)
£7
76
(£1,
895)
C
omm
unic
atio
n £1
86
(£36
) £0
(£
7)
£6
(£0)
£2
21
(£56
2)
£1,3
18
(£21
) £4
94
(£1,
235)
£2
(£
270)
£2
,227
(£
2,13
0)
Co-
oper
atio
n £3
57
(£1,
094)
£3
,565
(£
0)
£0
(£2)
£1
71
(£44
9)
£225
(£
193)
£3
0 (£
885)
£2
(£
0)
£4,3
50
(£2,
622)
M
easu
ring
£812
(£
1,49
0)
£5,3
42
(£3,
332)
£0
(£
26)
£883
(£
1,17
3)
£2,4
78
(£23
7)
£794
(£
279)
£2
,327
(£
303)
£1
2,63
6 (£
6,84
1)
Rev
iew
ing
£812
(£
1,52
7)
£4,7
80
(£2,
947)
£2
82
(£3,
592)
£8
83
(£1,
262)
£1
,335
(£
235)
£7
09
(£17
4)
£864
(£
303)
£9
,665
(£
7,00
2)
TRU
ST T
OTA
L £3
,494
(£
14,7
79)
£22,
285
(£11
,272
) £1
,960
(£
3,59
2)
£3,8
67
(£7,
711)
£8
,470
(£
1,20
7)
£4,1
98
(£12
,734
) £6
,822
(£
1,91
4)
£51,
096
(£53
,209
)
52
Total cost increased in four Trusts (Trusts A, C, D and F (control)) and decreased in the other three (Trusts B, E and G (control)). The total costs incurred increased by approximately 4% between the two phases.
Statistical testing revealed that the management root cause costs assigned to the test and control trusts, were not significantly different from one another in either phase (Mann-Whitney U-Test phase 1 P=0.57, phase 2 P=0.38). The costs for management root causes in each Trust were not significantly different between phases (Wilcoxon rank sum W Test P=0.87). Spearman’s rank-order correlation indicated a correlation coefficient between both phases of 0.334 (P<0.01) indicating a significant association (i.e. no significant change between the two phases).
The cumulative distribution of individual incident costs (i.e. from highest to lowest cost, not cumulative over time) across all seven Trusts, for both phases, is shown in Figure 8. The majority of incidents were relatively low cost (for example, in Trust A during the first data collection phase, a small group of 6 incidents accounted for approximately 85% of the total cost incurred).
Figure 8 Cumulative % accident costs
This is in keeping with the HSE concept of accident ratio pyramids, whereby there was an inverse relationship between the severity of the incident outcome and the number of incidents that exhibited that outcome(36). Also, the universal Pareto 80:20 law appears to broadly apply,
0
20
40
60
80
100
0 20 40 60 80 100
Cumulative % numbers of incidents
Cum
ulat
ive
% c
ost
Trust A PCT Trust B Trust C PCTTrust D PCT Trust E Trust F UH (Control)Trust G PCT (Control) Trust A PCT (2) Trust B (2)Trust C PCT (2) Trust D PCT (2) Trust E (2)Trust F UH (2) (Control) Trust G (2) PCT (Control)
53
whereby approximately 80% of the costs were attributed to approximately 20% (actually 10% - 40%) of the observed incidents(111).
This is potentially useful to NHS Trusts if they could focus on prioritising attention on the 20% highest cost incidents to influence safety performance. To further evaluate whether it is possible to predict for the Trusts which incidents on which to concentrate, further examination of the categories, factors and types of incidents is included in the following section (Section 3.4.4.)
3.4.4 Categories, factors and types of incidents
As described in Chapter 2, the costs arising from the circumstances and outcome of each incident was further described by assigning them to relevant headings within each of three different groups of incident category, factor and type. Appendix 7 contains a short description of each incident and their cost, category, and type.
Incident category Categories of incidents are shown in Figure 9.
Figure 9 Total cost per incident category (both phases)
The total costs incurred as a result of accidental injury were an order of magnitude higher than any other category in both phases. Costs from physical violence were nearly four times that of the first phase.
£0
£5 ,000
£10 ,000
£15 ,000
£20 ,000
£25 ,000
£30 ,000
£35 ,000
£40 ,000
£45 ,000
£50 ,000
Acci
dent
al in
jury
Phys
ical
vio
lenc
e
Fata
lity
Thef
t
Nea
r mis
s
Acci
dent
al p
rope
rty d
amag
e
Vand
alis
m
Oth
er
Verb
al a
buse
/thre
aten
ing
beha
viou
r
Fire
Ill h
ealth
A c cid e n t ca teg o ry
Cos
t
P ha se 1 P ha se 2
54
Table 13 shows the average cost per incident category. It can be seen that most incidents were assigned to the category of accidental injury (65% in phase 1 and 56% in phase 2). None of the other categories exceeded 5% of the total cost, with the exception of physical violence (10% in phase 1 and 15% in phase 2) and near miss in the first phase (12%).
Table 13 Average costs per incident category (all Trusts)
No costs were incurred for the category of ill health in phase 1. This was likely to be due to the relatively long timescales involved, but may also be related to relatively poor reporting mechanisms for ill health, as compared to other incidents, and therefore less likely to be picked up by the research design. Three cases of ill health were identified in the second phase. These were:
�� A member of staff absent from work due to prolonged work related stress; �� An enrolled nurse contracted an elbow infection from a splash of urine; �� Pressure sores were identified in an elderly patient.
This general lack of incidents from the category of ill health is important to consider when interpreting the costing data, since it is likely that this represents an underestimate of the true cost to the NHS of inadequate health and safety management. For example, it seems inconceivable that there could have only been one case of pressure sores in the whole sample from both data collection phases. It seems more likely that the study design or the in-house Trust reporting system or culture did not capture them. This is discussed further in Chapter 4.
Incident factor Total cost by incident factor for both phases is shown in Table 14.
Incident category (phase 1 total cost)
No. of incidents
Average cost/category of incident (phase 1)
Incident category (phase 2 total cost)
No. of incidents
Average cost/category of incident (phase 2)
Accidental injury (£44,529)
83 £537 Accidental injury (£32,013)
74 £433
Physical violence (£2,409)
13 £185 Physical violence (£11,075)
20 £554
Fatality (£2,091) 1 £2,091 Fatality (£1,718) 3 £573 Theft (£1,171) 2 £586 Theft (£0) 0 0 Near miss (£211) 15 £14 Near miss (£426) 5 £85 Accidental property damage (£308)
3 £103 Accidental property damage (£0)
0 0
Vandalism (£148) 4 £37 Vandalism (£281) 1 £281 Other (£131) 5 £26 Other (£2,122) 16 £133 Threatening behaviour (£72)
1 £72 Threatening behaviour (£160)
6 £27
Fire (£0) 0 0 Fire (£16) 1 £16 Ill health (£0) 0 0 Ill health (£5,397) 3 £1,799
55
Table 14 Total cost per incident factor
Incident factor Cost (phase 1)
Incident factor Cost (phase 2)
Absence £21,350 Absence £25,798 Replacement labour £18,122 Replacement labour £18,130 Patient treatment £9,184 Patient treatment £7,846 Damage repair/staff treatment
£2,344 Damage repair/staff treatment
£739
Initial response to accident £73 Hiring/purchasing £400 Transport £67 Transport £120 Hiring/purchasing £20 Lost production/ wasted time £119 Cleaning up £4 Initial response to accident £58 Lost production/ wasted time
£0 Cleaning up £0
Staff absence incurred the greatest total cost in both phases, followed closely by replacement labour. These two Factors accounted for approximately 75% of the total costs. Patient and staff treatment accounted for approximately 20% of costs with the remaining 5% spread between the remaining factors.
The costs per incident factor were not significantly different when the test Trusts were compared with the control Trusts (Kruskal-Wallis one-way analysis of variance (P=0.85 (Phase 1); P=0.77 (Phase 2)). For all Trusts the differences between the two phases were also not significant (Wilcoxon rank sum W test P=0.33). Spearman’s rank order correlation coefficient was 0.76 (P<0.01) indicating a significant correlation between the factors for all Trusts across both phases (i.e. no significant change between the two phases).
Incident type The total cost by incident type is shown in Table 15. This shows some variation between the two phases, although the same three types incurred most of the cost in both phases. These were person to person assault, slip trip or fall on the same level/stairs and patient lifting/handling. These three types accounted for 71% of the total cost in the first phase and 78% in the second.
For all Trusts the differences in total cost in incident types between the two phases were not significant (Wilcoxon rank sum W test P=0.69). Spearman’s rank order correlation coefficient was 0.76 (P<0.01) indicating a significant correlation between the types for all Trusts across both phases (i.e. no significant change between the two phases).
56
Table 15 Total cost per incident type (all Trusts)
Incident type Phase 1 Incident type Phase 2 Patient lifting/handling £22,565 Slip, trip or fall on same level/stairs £20,231 Slip, trip or fall on same level/stairs £10,874 Patient lifting/handling £12,223 Person to person assault £2,974 Person to person assault £9,273 Struck by an object £2,909 Other £6,369 Manual lifting/handling £2,583 Manual lifting/handling £1,848 Other £2,500 Fall from height £1,231 Fall from height £2,091 Struck by an object £481 Contact with equipment/machinery £1,824 Needlestick/sharps injury £431 Struck against something £1,797 Contact with electricity £400 Cut with sharp material/object £676 Exposure to fire £297 Needlestick/sharps injury £268 Hot or cold contact £243 Hot or cold contact £28 Cut with sharp material/object £101 Exposure to harmful substance £7 Struck against something £53 Exposure to fire £0 Contact with equipment/machinery £24 Contact with electricity £0 Exposure to harmful substance £4 TRUST TOTAL £51,096 TRUST TOTAL £53,209
Three groups of incident types were apparent:
�� Those that tended to exceed £10,000 total cost (slip trip or fall on the same level/stairs and patient lifting/handling);
�� Those that tended to exceed £5,000 total cost (other and person to person assault); �� Those that tended to fall below £5,000 total cost (the remaining types).
This means that the first two groups of type of incidents were the most costly for the participating Trusts.
The other group (64 incidents (32 in each phase)) was those that did not fit within an incident type. Examples of these were:
�� A fire alarm was activated by burnt toast. A patient who was unsupervised was making the toast, despite a Trust policy that all toast making should be supervised17;
�� A puppy in a patient’s home bit a community psychiatric nurse; �� A visitor, driving a high-topped van, struck an overhead archway while driving under an
archway in the hospital.
The classification in the HSE methodology did not include groupings for incident types involving animals or transport. It is not known why this was done, given that these two types are included in the Reporting of Injuries, Diseases, and Dangerous Occurrences Regulations 1995 (RIDDOR)(12). These may be useful additional Types to include in any future revisions of the methodology.
The average incident type cost is shown in Table 16. 17 This incident was not assigned to the exposure to fire type because a fire had not occurred and there was evidence that toaster was not Trust property
57
Tabl
e 16
Ave
rage
cos
t per
inci
dent
type
(all
Trus
ts)
Inci
dent
type
(pha
se 1
tota
l cos
t) N
o. o
f in
cide
nts
Aver
age
cost/
type
of i
ncid
ent
(pha
se 1
) In
cide
nt ty
pe (p
hase
2 to
tal c
ost)
No.
of
inci
dent
s Av
erag
e co
st/ty
pe o
f inc
iden
t (p
hase
2)
Fall
from
hei
ght (
£2,0
91)
1 £2
,091
Sl
ip/tr
ip/fa
ll on
sam
e le
vel/s
tairs
(£
20,2
31)
14
£1,4
45
Patie
nt li
fting
/han
dlin
g (£
22,5
65)
15
£1,5
04
Patie
nt li
fting
/han
dlin
g (£
12,2
23)
14
£873
St
ruck
by
an o
bjec
t (£2
,909
) 3
£970
Pe
rson
to p
erso
n as
saul
t (£9
,273
) 15
£6
18
Slip
/trip
/fall
on sa
me
leve
l/sta
irs
(£10
,874
) 15
£7
25
Man
ual l
iftin
g/ha
ndlin
g (£
1,84
8)
4 £4
62
Man
ual l
iftin
g/ha
ndlin
g (£
2,58
3)
8 £3
23
Con
tact
with
ele
ctric
ity (£
400)
1
£400
C
onta
ct w
ith e
quip
men
t/mac
hine
ry
(£1,
824)
6
£304
Fa
ll fr
om h
eigh
t (£1
,231
) 6
£205
Pers
on to
per
son
assa
ult (
£2,9
74)
13
£229
O
ther
(£6,
369)
32
£1
99
Stru
ck a
gain
st so
met
hing
(£1,
797)
9
£200
St
ruck
by
an o
bjec
t (£4
81)
3 £1
60
Cut
with
shar
p m
ater
ial/o
bjec
t (£6
76)
4 £1
69
Expo
sure
to fi
re (£
297)
2
£149
O
ther
(£2,
500)
32
£7
8 H
ot o
r col
d co
ntac
t (£2
43)
2 £1
22
Nee
dles
tick/
shar
ps in
jury
(£26
8)
16
£17
Cut
with
shar
p m
ater
ial/o
bjec
t (£
101)
4
£25
Hot
or c
old
cont
act (
£28)
2
£14
Nee
dles
tick/
shar
ps in
jury
(£43
1)
26
£17
Expo
sure
to h
arm
ful s
ubst
ance
(£7)
1
£7
Stru
ck a
gain
st so
met
hing
(£53
) 4
£13
Expo
sure
to fi
re (£
0)
0 £0
C
onta
ct w
ith e
quip
men
t/mac
hine
ry
(£24
) 3
£8
Con
tact
with
ele
ctric
ity (£
0)
0 £0
Ex
posu
re to
har
mfu
l sub
stan
ce (£
4)
1 £4
58
As with total costs, the incident types average cost could be grouped into three:
�� Those that tended to exceed £1,000 average cost (fall from height (phase 1 only), patient lifting/handling (phase 1 only), and slip, trip or fall on the same level/stairs (phase 2 only));
�� Those that tended to exceed £500 average cost (struck by an object (phase 1 only), slip, trip or fall on the same level/stairs (phase 1 only), patient lifting/handling (phase 2 only) and person to person assault (phase 2 only));
�� Those that tended to fall below £500 average cost (the remaining types).
As with total cost, the same three types of incident had the highest average cost (i.e. person to person assault, slip trip or fall on the same level/stairs and patient lifting/handling).
However, struck by an object and fall from height also fall within the high cost group. this was due to a small number of high cost incidents in phase 1.
When this finding was examined further, the differences in average cost in incident types between the two phases were not significant (Wilcoxon rank sum W test P=0.68). However, Spearman’s rank order correlation coefficient was 0.45 (P<0.1) indicating a marginal rejection of the null hypo-report of no correlation (i.e. no significant difference between the two phases.
Table 17 shows the range of costs for each incident type.
Table 17 Range of costs per incident type (all Trusts)
As with total and average cost, incident type cost ranges could be grouped into three:
�� Those where the cost range tended to exceed £1,000. Approximately half of the incident Types (47%) fell within this grouping (patient lifting/handling, slip, trip or fall on the same level/stairs, struck by an object, manual lifting/handling, struck against something (phase 1 only), person to person assault (phase 2 only) and other);
18 NA = not applicable as there was either none or only a single incident in this type.
Incident type Cost range(phase 1)
Cost minimum/ maximum (phase 1)
Cost range (phase 2)
Cost minimum/ maximum (phase 2)
Patient lifting/handling £9,847 £2 -£9,849 £2,682 £6 -£2,688 Slip/trip/fall on same level/stairs £5,706 £7 -£5,712 £7,022 £57 -£7,078 Struck by an object £2,244 £66 -£2,310 £7,260 £2 -£7,262 Struck against something £1,334 £5 -£1,339 £35 £3 -£37 Manual lifting/handling £1,204 £5 -£1,209 £1,525 £14 -£1,540 Other £1,064 £4 -£1,069 £3,383 £2 -£3,385 Contact with equipment/machinery £704 £6 -£710 £17 £2 -£19 Person to person assault £667 £22 -£689 £7,260 £2 -£7,263 Cut with sharp material/object £227 £10 -£237 £64 £4 -£68 Needlestick/sharps injury £27 £4 -£31 £96 £2 -£98 Hot or cold contact £26 £14 -£40 £26 £109 -£134 Fall from height NA18 NA £650 £6 -£650 Exposure to fire NA NA £265 £16 -£280 Contact with electricity NA NA NA NA Exposure to harmful substance NA NA NA NA
59
�� Those that tended to exceed £100 cost range, which represented approximately one third (27%) of the types (contact with equipment/machinery (phase 1 only), cut with sharp material/object (phase 1 only), fall from height (phase 2 only) and exposure to fire (phase 2 only));
�� Those that tended to fall below £100 average cost, which represented approximately one tenth (13%) of the incident types (needlestick/sharps, and hot or cold contact).
Two of the types were not included (contact with electricity and exposure to harmful substance). This is because a range could not be established for either phase because either zero or one incident occurred in either type.
For all Trusts the differences in cost range for incident types between the two phases was not significant (Wilcoxon rank sum W test P=0.35). Spearman’s rank order correlation coefficient was 0.59 (P<0.05) indicating a significant correlation with the range of types in both phases (i.e. no significant change between the two phases).
It should be noted that the incident types in the first grouping (where the range tended to exceed £1,000) all had minimum costs that were less than £10 with the exception of one type (struck by an object), which was less than £100. This represents weak evidence that many of the incident types could incur high costs. However, there is much stronger evidence that some types consistently tended to incur higher costs. Using the cost range data as the measurement criteria these were (alphabetical list):
�� Manual lifting/handling; �� Other; �� Patient lifting/handling; �� Person to person assault; �� Slip/trip/fall on same level/stairs; �� Struck against something; �� Struck by an object.
This indicates a potential cost-effective prioritisation strategy for Trusts, based on placing effort to control risks from incident types that incur highest cost.
Whether they should be prioritised in this way is, according to Hale(112), linked to whether they had the same pattern of underlying causes. As has been shown earlier (section 4.3.3.), the majority of incidents had the risk control system of risk assessment and, specifically, the management root causes of planning and implementing. This is evidence that the underlying causes were the same and that the above strategy could be potentially effective.
3.5 QUESTIONNAIRE DATA
As described in Chapter 2, the safety climate questionnaire used was that developed by Cox and Cheyne(58). Each questionnaire consisted of 43 questions aimed at eliciting individual views on nine dimensions, grouped into four categories. These were:
Category (a): Organisational context covering four dimensions: 1. Management commitment 2. Communication 3. Priority of safety 4. Safety rules & procedures
60
Category (b): Social environment covering two dimensions: 5. Supportive environment 6. Involvement
Category (c): Individual appreciation covering two dimensions: 7. Personal priorities and need for safety 8. Personal appreciation of risk
Category (d): Work environment covering a single dimension: 9. Physical work environment
Within this section of the Chapter the response rates and job category distribution will be described followed by general analysis of the dimension score results and the responses grouped by Trust, job category, and individual, across both data collection phases. The section ends with analysis of questionnaire respondent’s additional comments.
3.5.1 Response rates
The numbers of questionnaires issued, received and derived response rates are shown in Table 18. The data for phase 2 are shown in brackets. For example in Trust A, 82 questionnaires were issued at the start of phase 1. By the end of the second phase 43 valid questionnaires had been returned. The overall project response rate was therefore 52%. Since the questionnaire sample was 10% of Trust staff this represented 5.2% of the total Trust staff (final column).
Table 18 Response rates to the questionnaires issued (1st & 2nd phases19)
Trust Number of questionnaires issued
Number of completed questionnaires received
%Response rate
Change in response rate between phases
Overall project response rate (%)
% of Trust staff
Trust A 82 (56) 56 (43) 68% (77%)
+ 13% 52% 5.2%
Trust B 338 (171) 171 (114) 51% (67%)
+31% 34% 3.4%
Trust C 194 (81) 81 (74) 42% (91%)
+117% 38% 3.8%
Trust D 561 (337) 337 (274) 60% (81%)
+35% 49% 4.9%
Trust E 185 (135) 135 (94) 74% (70%)
-5% 51% 5.1%
Trust F (Control)
592 (220) 220 (145) 37% (66%)
+78% 25% 2.5%
Trust G (Control)
397 (207) 207 (167) 52% (81%)
+56% 42% 4.2%
TOTAL 2349 (1207) 1207 (911) MEAN 55%
(76%) +46% 42% 4.2%
SD 13% (9%) +41% 10% 1.0%
Therefore, it can be seen that in the first data collection phase 2439 questionnaires were issued with 1207 returned satisfactorily completed (55% response rate). In the second phase 1207 questionnaires were issued to the same individuals as before. This resulted in 911 valid 19 2nd benchmarking data in (brackets)
61
questionnaires being returned (76% response rate). This group of respondents represents the individuals who completed two valid questionnaires resulting in an overall project response rate of 42%, representing 4.1% of the total headcount available to the project across the seven Trusts.
The highest response rate in the first data collection phase was 74% (Trust E) and the lowest 37% (Trust F (control)). The rates in the second phase varied between 91% (Trust C) and 66% (Trust F). The differences across Trusts were not significantly different (Kolmogorov-Smirnov Test P=0.94).
The response rates were all higher than in the first phase (between +13% for Trust A and +117% for Trust C), with the exception of Trust E, which fell by 5%. This was not anticipated in that the proportion of those who responded during the first phase represented the entire pool for the second. Reasons why Trust E differed from the other Trusts are unclear.
As described in Chapter 3, efforts were made to optimise the response rate throughout the 12 week sampling phases by sending a reminder letter. However, it was regarded as important that the questionnaires were completed during the same time period as the set of reported incidents. Therefore individuals who had not returned a valid questionnaire after the end of these time periods were not followed up any further.
This data represent an overall project response rate for all seven Trusts of 42% (between 24% (Trust F (control)) and 52% (Trust A)). This is an average of 4.1% of all staff employed during the data collection phases. This is regarded as a good response rate, given the challenge of ensuring that the same individual completed and returned two questionnaires with a year between the events.
Since the responses from individuals in the second phase was dependent on those returning completed questionnaires during the first phase it was necessary to check that each occupational group was adequately represented in both phases. The occupational group data was compared across the two phases and the results represented in Figure 10. This confirms that, despite fewer questionnaires being received in the second phase, a broad balance of respondents across occupational groups was maintained between the two phases and that no group was under-represented.
62
Figure 10 Questionnaire job category distribution
3.5.2 Questionnaire confirmatory factor analysis
The questionnaire had been developed for the offshore industry(58). To ensure that the scales in the original questionnaire were appropriate for use in a healthcare setting, confirmatory factor analysis (CFA) was undertaken.
The data was subjected to CFA using version 6 of the EQS (structural equation) program(96). The original nine-factor model(58) had a relatively good fit for the data (CFI=0.85)20. The CFA for the healthcare data revealed a relatively good fit for the data with no new factors revealed (CFI=0.84).
Internal-scale reliability (or consistency) was examined to investigate the degree to which the various questions measured different aspects of the same concept. The accepted level for this statistic (Cronbach’s alpha) is approximately 0.7(58). All the alphas reported were in the range 0.62 to 0.89 (original range 0.53 to 0.84), which was regarded as acceptable reliability.
It was therefore concluded that the original scales were appropriate for use in a healthcare setting, without modification. However, the instrument could not be regarded as able to distinguish whether a lower opinion score was indicative of a more critical attitude as a result of climate improvement.
20 Cox and Cheyne identified that a value of approximately CFI=0.9 was accepted as indicating good model fit.
0 1 2 3 4 5 6 7 8 9 10
Estates
Senior manager
Professional, Technical & PAMs (qualified)
Medical & Dental
Other
Domestic & Catering
Nursing & Midwifery (qualified)
Professional, Technical & PAMs (unqualified)
Nursing & Midwifery (unqualified)
A&C
Nurse manager
Unknown
Staf
f gro
up
Questionnaires received as a % of total trust staff
Phase 1 Phase 2
63
3.5.3 Dimension score responses
The dimension scores were calculated from the responses for each of the nine dimensions(59). The resulting standardised dimension scores then had the following meaning (with a score below six (scale mid-point) indicating increasingly low opinion(58)):
2 - strongly disagree 4 – disagree 6 – neither agree nor disagree 8 – agree 10 – strongly agree
Summary dimension scores obtained for both data collection phases are shown in Table 19.
Table 19 Questionnaire mean response dimension scores (1st & 2nd phases)21)22
Dimension�
Trust �
Manage-ment commit-ment
Commun-ication
Priority of safety
Safetyrules
Supp-ortive environ-ment
Involve-ment
Pers-onal prio-rities
Apprec-iation ofrisk
Workenviron-ment
MeanTrust score
Trust A 6.4 (6.6)
6.5(6.5)
6.3(6.7)
6.2(6.0)
7.0(7.0)
6.4(6.7)
7.8(7.6)
6.7(6.7)
5.9(6.0)
6.6(6.7)
Trust B 7.2 (7.0)
6.8(6.8)
7.0(7.0)
7.0(7.0)
7.3(7.4)
6.9(7.1)
7.9(8.1)
7.0(7.0)
6.3(6.2)
7.0(7.1)
Trust C 6.6 (7.0)
6.6(6.7)
6.8(7.1)
6.3(6.5)
7.2(7.4)
6.8(6.9)
8.0(8.1)
6.9(6.9)
5.7(5.8)
6.8(6.9)
Trust D 7.2 (7.2)
6.8(6.9)
7.1(7.2)
6.8(6.9)
7.3(7.5)
6.9(7.0)
7.9(8.1)
7.1(7.2)
6.1(6.2)
7.0(7.1)
Trust E 6.8 (6.7)
6.4(6.4)
6.8(6.7)
6.4(6.3)
7.1(7.1)
6.5(6.8)
7.6(7.8)
7.1(7.1)
6.0(5.8)
6.7(6.8)
Trust F (Control)
6.7(6.7)
6.5(6.4)
6.5(6.5)
6.5(6.4)
7.0(7.2)
6.5(6.9)
7.6(7.7)
6.8(6.9)
5.7(5.6)
6.6(6.7)
Trust G (Control)
7.0(6.9)
6.7(6.6)
7.0(6.9)
6.9(6.8)
7.3(7.3)
6.9(6.9)
7.8(7.9)
7.2(7.1)
6.1(6.2)
7.0(7.0)
Mean 6.9 (6.9)
6.6(6.6)
6.9(6.9)
6.7(6.7)
7.2(7.3)
6.7(6.9)
7.8(7.9)
7.0(7.1)
6.0(6.0)
6.9(6.9)
Median 7.1 (7.1)
6.8(6.8)
7.0(7.0)
6.7(6.7)
7.3(7.3)
6.7(7.3)
8.0(8.0)
7.0(7.5)
6.0(6.0)
6.9(7.0)
SD 1.4 (1.4)
1.3(1.3)
1.5(1.5)
1.4(1.5)
0.9(0.9)
1.6(1.6)
1.0(1.1)
1.3(1.2)
1.4(1.4)
0.9(0.9)
Gross change for most scores was either an increase (53%) or no change (24%). The scores for the remaining 23% decreased. There was also a visual impression that three of the nine dimension scores (supportive environment, involvement and personal priorities) were consistently higher between the two phases, across all the Trusts. The mean Trust score was also higher for all Trusts with the exception of Trust G, which stayed the same.
Statistical testing was used to examine further these trends. The null hypo-report used for interpretation of the results was that an increase in dimension score represented an improvement
21 2nd benchmarking data in (brackets) 22 The mean dimension scores are out of a possible score of 10 and, the higher the score, the more positive the opinion. A score of 6.0 equals a neutral opinion so any scores above this level can be regarded as marginally positive.
64
in opinion of the aspect of safety culture. However, a decrease in score could also represent greater awareness of safety culture whereby individuals had become more critical. This could represent an early stage in improving safety culture and is discussed further in Chapter 4.
3.5.4 Analysis of responses by Trust
The dimension scores represented ordinal data. Non-parametric hypo-report testing was therefore used initially to compare responses across all seven Trusts during both data collection phases. Kruskal-Wallis One-way Analysis of Variance for independent samples was chosen as a validated, reliable test for data that are amenable to ranking(103, 106, 113). The results are shown in Table 20. All were highly significant (P<0.01) indicating that the null hypo-report (that the responses were the same from each Trust) should be rejected and the responses treated as if they were from seven distinct sources.
Table 20 Kruskal-Wallis test for mean dimension scores across both phases
Dimension (grouped by Trust) Chi-square Degrees of freedom Significance Mean response per person 75.010 13 0.000 Management commitment 68.748 13 0.000 Communication 43.138 13 0.000 Priority of safety 56.341 13 0.000 Safety rules 57.315 13 0.000 Supportive environment 49.581 13 0.000 Involvement 31.886 13 0.002 Personal priorities 42.518 13 0.000 Personal appreciation of risk 31.677 13 0.003 Work environment 50.331 13 0.000
As discussed in Chapter 2, where possible, the use of parametric testing was the preferred approach for this data provided it was not characterised by gross inequality of intervals(108, 113, 114). To do this the distribution of the data was examined to check whether it approximated to normality. The distribution of the data was examined and found to be close to normality. The means and medians of the questionnaire responses were found to overlap within one standard deviation of the mean in every dimension for both data collection phases for each Trust (P<0.01). Since the mean and median are the same in a normal distribution this was strong evidence of approximate normality within the data. It was decided that this evidence, combined with the near-normal distribution shape, was sufficiently persuasive to proceed with ANOVA.
When ANOVA was carried out the mean responses to the various dimensions for each Trust were significantly different from each other for both phases of the project (P<0.001). This confirmed the result from the Kruskal-Wallis test that the null hypo-report (that the questionnaire responses were not different between individual Trusts) should be rejected. This finding held true when mean dimension scores were analysed within each Trust (Tables 21 and 22). This showed that, for the first data collection phase, all dimension scores differed between Trusts (P<0.05). For the second data collection phase all scores differed between Trusts (P<0.01) with the exception of the dimensions of involvement and appreciation of risk.Subsequent analysis of pooled questionnaire data was therefore avoided with responses from individual Trusts analysed individually.
65
Table 21 One-way ANOVA for mean dimension scores (1st data collection phase)
Sum of squares df Mean square F Sig. Management commitment Between Trusts 78.78 6 13.13 7.276 .000 Communication Between Trusts 31.86 6 5.311 3.300 .003 Priority of safety Between Trusts 63.47 6 10.57 4.859 .000 Safety rules Between Trusts 61.40 6 10.23 5.297 .000 Supportive environment Between Trusts 23.10 6 3.851 4.385 .000 Involvement Between Trusts 40.20 6 6.700 2.717 .013 Personal priorities Between Trusts 21.19 6 3.533 3.235 .004 Appreciation of risks Between Trusts 33.98 6 5.665 3.529 .002 Work environment Between Trusts 52.32 6 8.721 4.787 .000 Mean response/person Between Trusts 35.66 6 5.944 7.339 .000
Table 22 One-way ANOVA for mean dimension scores (2nd data collection phase)
Sum of squares
df Mean square F Sig.
Management commitment Between Trusts 40.965 6 6.827 3.388 .003 Communication Between Trusts 33.515 6 5.586 3.180 .004 Priority of safety Between Trusts 54.079 6 9.013 4.141 .000 Safety Between Trusts 71.267 6 11.878 5.343 .000 Supportive environment Between Trusts 16.185 6 2.697 2.993 .007 Involvement Between Trusts 9.535 6 1.589 .647 .693 Personal priorities Between Trusts 25.697 6 4.283 3.909 .001 Appreciation of risks Between Trusts 16.383 6 2.730 1.764 .103 Work environment Between Trusts 44.606 6 7.434 3.905 .001 Mean response/person Between Trusts 27.937 6 4.656 5.571 .000
3.5.5 Analysis of responses by job category
The responses for each Trust were categorised according to the twelve job categories identified in Chapter 2. Multiple comparison testing was carried out, using the Scheffe test. The null hypothesis was that there was no difference in mean response between staff groups within each Trust.
Trust F (a control Trust) was the only Trust to have significant differences between staff groups. Table 23 shows these differences. The mean questionnaire responses for the A&C staff differed significantly from the nursing and midwifery (unqualified) group. The medical and dental group differed significantly from the nursing and midwifery group (both qualified and unqualified), the professional, technical and PAMs (both qualified and unqualified) group, and the senior manager group. This finding was mostly confined to the first data collection phase (with the exception of the nursing and midwifery (unqualified) group, and the professional, technical and PAMs (unqualified) group. Possible explanations for this finding will be discussed further in Chapter 4.
66
Table 23 Trust F Multiple Comparisons by Job Category (Showing Only Significant Associations) †††
Job Category23 �� Medical & Dental A&C
Nursing and Midwifery (qualified) <0.01 (phase 1) Nursing and Midwifery (unqualified) <0.001 (phase 1) <0.05 (phase 2) <0.05 (phase 1) Professional, Technical & PAMs (qualified) <0.01 (Phase 1) Professional, Technical & PAMs (unqualified) <0.05 (both phases) Senior Manager <0.05 (phase 1)
3.5.6 Analysis of responses by individual members of staff
The research design meant that two questionnaires were available for each individual, with an approximate year in between each questionnaire. Since it was already known that the responses from within each Trust were different individual the data was analysed to investigate whether any change could be detected in individual responses between the two phases by grouping the responses, across all job categories, by individual Trust (Table 24).
Table 24 Paired samples T-tests (individual responses grouped by Trust)
Dimension Asymp. Sig. (2-tailed) Trust
ATrust B
Trust C Trust D Trust E
Trust F
Trust G
Management commitment +0.453 -0.200 +0.127 +0.714 -0.786 +0.805 -0.226 Communication +0.869 +0.705 +0.341 +0.529 +0.774 -0.153 -0.246 Priority of safety +0.160 +0.486 +0.255 +0.025* -0.178 -0.787 -0.668 Safety rules & procedures -0.403 -0.425 +0.036* +0.005* -0.807 -0.090 -0.578 Supportive environment +0.617 +0.728 +0.117 +0.021* +0.982 +0.236 -0.557 Involvement +0.405 +0.608 +0.718 +0.218 +0.065 +0.060 -0.416 Personal priorities/ need for safety
-0.368 +0.720 +0.618 +0.004* +0.067 -0.361 +0.145
Personal appreciation of risk +0.688 +0.776 +0.634 +0.011* +0.690 +0.262 -0.584 Physical working environment +0.566 -0.586 +0.216 +0.135 -0.227 -0.117 +0.788 Mean response per person +0.499 -0.970 +0.079 +0.002* +0.839 -0.747 -0.387
* = Differences significant at the 0.05 level
The only significant change in opinion was found in Trust D, with positive significant changes in 6 of the 9 dimensions. Trust D was also the only Trust to show a significant overall positive change in the mean response per person (P<0.05). Trust C also showed significant improvements in staff opinion in the dimension of safety rules & procedures.
The results show that the questionnaire methodology could detect significant change in staff opinion and was therefore a good measure of culture change. The absence of significant change in five of the seven Trusts may be evidence that the underlying health and safety culture within these Trusts is relatively stable.
23 A distinction was made between unqualified and qualified staff in the categories of Nursing & Midwifery and Professional, Technical and PAMs. This was because the staff databases enabled this distinction to be made.
67
3.5.7 Questionnaire respondents additional comments The pattern of additional comments was similar between Trusts. Approximately two-thirds of those who returned questionnaires did not make additional comments (68% for phase 1 and 69% for phase 2).
Of the remaining approximate third that did comment, approximately one tenth (9%) were either positive or neutral for both phases. Approximately one tenth of comments related to specific issues within the employing Trust (9% for phase 1 and 11% for phase 2). The remaining approximate tenth focused on health and safety management issues within the Trust (11% for phase 1 and 9% for phase 2). Less than 4% of respondents reported that they found the questionnaire difficult to interpret (3% for phase 1 and 2% for phase 2).
3.6 RESULTS FROM WORKBOOK USAGE MONITORING (PROJECT PHASE 3)
As described in Chapter 2, the workbooks were issued to each Trust via a series of briefings to a pre-defined schedule. The number issued to each Trust was agreed in partnership with the Trusts. This took account of the Trust line management structure so that workbooks were issued to managers with responsibilities for a number of staff (e.g. nurses with responsibility for several wards rather than a single ward were specifically encouraged to take part). The final selection of individuals to be designated as workbook holders was however the responsibility of the Trusts.
The participating Trusts were encouraged to introduce measures to ensure the workbooks were used but, other than this, the research team took no further part in ensuring their usage was maximised. The research team took on a monitoring role to estimate and compare the levels of workbook usage in each of the five Trusts that had been issued with workbooks (the two control Trusts were not issued with workbooks and were not included in the monitoring).
3.6.1 Telephone survey
Attempts to contact each workbook holder by telephone were made during three time-periods (the method used is detailed in Appendix 6). Contact was made with the workbook holder and answers to nine questions were collated.
The time periods and percentages of successful contacts made are summarised below:
�� Period 1: o 1st April 2001 - 31st July 2001 o 86% successful contacts (153 workbook holders out of 177)
�� Period 2: o 1st August 2001 – 30th Nov. 2001 o 84% successful contacts (149 workbook holders out of 177)
�� Period 3: o 1st Dec. 2001 – 31st March 2002 o 64% successful contacts (114 workbook holders out of 177)
There was a drop from over 80% successful contacts in the first two periods to 64% in the third period. This was as a result of difficulties in making contact with the workbook holders. There were two main reasons for this, approximately equal in occurrence:
�� The workbook holder had been moved to another area or job and no one had contacted the research team or resumed ownership of the workbook;
68
�� Contact could not be made with the workbook holder despite repeated attempts (a typical reason was that the individual was too busy to participate or otherwise declined).
Therefore approximately one third of workbook holders had dropped out by the end of the monitoring period. The percentage responses for all questions asked during the three periods for all five test Trusts are shown in Figure 11.
Figure 11 Workbook holder telephone interviews feedback from three monitoring periods
Significant associations were tested using Cramer’s V, which is a suitable test for investigating associations between nominal variables. The results showed that the percentage of workbook holders did not change significantly (P<0.1) although use of the workbook reduced significantly
Note: Error bars show standard error
of the mean
0
10
20
30
40
50
60
70
80
90
100
Are
you
still
the
hold
er o
f the
wor
kboo
k?
Hav
e yo
u us
ed th
ew
orkb
ook?
Hav
e yo
u fo
und
the
wor
kboo
k he
lpfu
l?
Hav
e yo
u us
ed th
em
anag
ers'
aud
it?
Hav
e yo
u us
ed th
e w
orkb
ook
gene
rally
?
Hav
e yo
u fo
cuss
ed in
on
parti
cula
r sec
tions
?
Hav
e yo
u im
plem
ente
d an
yco
ntro
l mea
sure
s as
a re
sult
of a
risk
ass
essm
ent?
Hav
e yo
u us
ed th
e op
tion
appr
aisa
l?
Hav
e yo
u an
y su
gges
tions
for i
mpr
ovem
ents
?
Question posed
% o
f w
ork
bo
ok h
old
that
rep
lied
yes
1 Apr - 31 Jul 2001 1 Aug - 30 Nov 2001 1st December 2001 to 31st March 2002
69
over the course of the 12 months of use (P<0.01). This is regarded as low implementation of the workbook.
If the respondent indicated that they were no longer the workbook holder or had not used the workbook the interview was terminated. If the workbook had been used the next three questions were asked. These were intended to gain information on whether the workbook had been used in a general sense or whether particular sections had been used. The percentage that had used the workbook in a general sense had reduced significantly (P<0.05) although the percentage that focused on particular sections had nearly doubled between the first monitoring period and the last (P<0.001). Over the three monitoring periods the use of the manager’s audit did not change significantly (P<0.1).
The percentage that found the workbook helpful decreased by almost half in the third monitoring window (P<0.001).
The percentage who had implemented control measures as a result of risk assessment increased significantly in the middle window (P<0.01) but fell back to just below that of the first window by the time of the third window, resulting in no overall significant change (P<0.1).
The percentage of workbook holders that used the option appraisal was less than 5% across all three windows and did not vary significantly (P<1.0). Verbal feedback from users indicated that the method might benefit from development to make it easier to use. There may also be implications for redesigning training in use of the method. However, the option appraisal was always intended for use in circumstances where the solution to a risk control problem was not clear, and where there were financial implications. The circumstances where the option appraisal could be useful were therefore likely to be limited. This means that the low usage could also reflect the low number of circumstances where it was needed, rather than an avoidance of use.
There was no significant change in the percentage of users who made suggestions for improvement (P<0.1)
The results of the significance testing for individual Trusts, across the time intervals, (Cramer’s V) are shown in Table 25.
Table 25 Workbook holder telephone interview feedback associations
Cramer’s V Question Trust A Trust B Trust C Trust D Trust E Are you still the holder of the workbook? -0.346 -0.002* -0.001* -0.700 +0.096 Have you used the workbook? +0.040* -0.777 -0.083 -0.001* -0.001* Have you used the workbook generally? +0.377 -0.753 -0.024* -<0.001* -0.069 Have you focused on particular sections? +<0.001* +<0.001* +<0.001* +0.645 +0.027* Have you used the manager’s audit? -0.367 -0.856 -0.003* +0.228 +0.700 Have you found the workbook helpful or not?
+0.066 -0.001* -0.003* -<0.001* -<0.001*
Have you implemented any control measures as a result of risk assessment?
+<0.001* -0.635 -0.094 -0.016* -0.020*
Have you used the option appraisal? +0.081 +0.268 ZERO -0.004* -0.004* Have you any suggestions for improvement?
-0.661 +0.002* +0.083 -0.788 -0.471
* = Differences significant at the 0.05 level
70
The workbook holder telephone interview results for individual Trusts are shown in Appendix 9. General conclusions can be drawn from visual inspection of these five Figures. These are:
�� Numbers of holders of the workbooks tended to decrease (significantly in Trusts B and C); �� Usage of workbooks tended to decrease (significantly in Trusts D and E), although in Trust
A usage increased significantly; �� Perceived workbook usefulness decreased significantly in four Trusts (Trusts B, C, D and
E); �� Usage of the manager’s audit did not change significantly except in Trust C, where its use
decreased; �� General use of the workbook tended to decrease (significantly in Trusts C and D); �� All Trusts reported increased use of particular sections (significantly in four Trusts (Trusts
A, B, C and E)); �� Implementation of control measures as a result of risk assessment tended to decrease
(significantly in Trusts D and E). In Trust A implementation increase significantly; �� Use of the option appraisal decreased significantly in two Trusts (Trusts D and E); �� Suggestions for improvement tended to decrease but increased significantly in Trust B.
Although it can be seen from the above descriptive data that individual Trusts varied in their responses, statistical testing was undertaken to establish the degree of association between the Trusts in terms of the percentages of positive responses to the telephone interview questions. No statistically significant differences were found between any of the Trusts (k-sample chi-square between P=0.80 and P=0.99). Table 26 shows the results of non-parametric correlation testing (Spearman’s Rank Order Correlation) for the percentages recorded during the third interview period (the other two periods produced similar results). The null hypothesis was that there was no relationship between Trusts. The table shows significant observed correlation between all Trusts, indicating that the null hypothesis should be rejected thus confirming highly significant correlation.
Table 26 Between trust non-parametric correlations of workbook holder telephone interview feedback (third telephone interview period)
Spearman’s rank order correlation Trust A Trust B Trust C Trust D Trust E Correlation coefficient .975(**) .904(**) .819(**) .775(*) Sig. (2-tailed) <0.001 <0.001 <0.01 <0.05
Trust A
N 9 9 9 9 Correlation coefficient .975(**) .936(**) .836(**) .865(**) Sig. (2-tailed) <0.001 <0.001 0.005 <0.005
Trust B
N 9 9 9 9 Correlation coefficient .904(**) .936(**) .856(**) .846(**) Sig. (2-tailed) <0.001 <0.001 <.005 <0.005
Trust C
N 9 9 9 9 Correlation coefficient .819(**) .836(**) .856(**) .785(*) Sig. (2-tailed) <0.01 <0.005 <0.005 <0.05
Trust D
N 9 9 9 9Correlation coefficient .775(*) .865(**) .846(**) .785(*) Sig. (2-tailed) <0.05 <0.005 <0.005 <0.05
Trust E
N 9 9 9 9 ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).
71
If the percentage of positive answers to the telephone interview questions were assumed to represent an indicator of workbook implementation, it can be concluded that the degree of implementation was similar for each Trust. The implications of this finding in terms of the incident reporting and climate measurement changes are discussed in Chapter 4.
3.6.2 Visual inspections
The visual inspection of a 10% random sample of workbooks in each Trust was carried out during a 12 week period between 27th September 2001 and 10th December 2001, towards the end of the second telephone survey period and the beginning of the third. This means that the workbooks had been in place at least six months. Summary results are shown in Figure 12.
Figure 12 Use of documentation within a 10% random sample of workbook holders
Approximately 80% of holders who took part in the visual inspection had used their workbook, with very few having had pages removed and, encouragingly, approximately 75% had added items, such as policies or written risk assessments. The workbook sections used most were
0 10 20 30 40 50 60 70 80 90
Is the workbook holder accurately recorded?
Are cooperation measures entered or a marker included?
Is there evidence of quarterly reports having been completed?
Is there evidence of safety inspections having been carried outor a marker included?
Are national/trust policies listed with locations or a markerincluded?
Are items from departmental safety meetings documented or amarker included?
Are there training records in the workbook or a marker included?
Are there operational procedures listed or a marker included?
Are there objectives entered following completion of the audit?
Was the model business case used?
Insp
ectio
n qu
estio
n
Percentage of workbooks inspected exhibiting some element of documentation or a marker
72
policies, roles and responsibilities, measures to ensure co-operation, evidence of safety inspections, and quarterly reports.
However, the finding of low implementation following the telephone interviews was generally confirmed by the visual inspection in that, of the 19 questions asked within the inspection, there were only 5 questions where more than 50% of workbooks inspected exhibited some element of implementation.
Workbook sections used least were operational procedures, and incident reporting (possibly because all Trusts had a separate system). The model business case was not used in any of the workbooks inspected. Proactive management activities were among the least used aspects of the workbook. For example, less than one third of workbooks inspected had completed the managers’ audit (27%) with even fewer of this group using the outcome of the audit to set objectives (11%). Evidence of action or forward planning was found in only 37% of workbooks.
The impact of this low implementation on health and safety management performance is discussed further in Chapter 4.
3.7 TRIANGULATION OF RESULTS
As discussed in Chapter 1, Guldenmund recommended attempting to correlate safety performance measures with climate assessments(56). Although the data from the incident analysis and the workbook usage monitoring were not directly comparable with the climate survey results because their units and sampling strategy were different, broad categorisation using the ranks of the Trusts within each category was undertaken.
Tables 27 to 29 show the results of correlation testing (Spearman’s rank order correlation) for both phases. Tables 27 and 28 include:
�� Trust rank for the nine safety climate dimensions plus mean dimension score per person (from rank 1, representing the lowest mean dimension score to 5, representing the highest);
�� Trust rank for both the total reported incidents and those captured by the inclusion criteria and for which root causes could be assigned (i.e. minus the special category incidents) (from rank 1, representing the highest incident rate to 5, representing the lowest).
Table 29 also includes:
�� Trust rank of workbook usage (this was taken from the mean percentage of respondents to the telephone interviews that reported they had used the workbook (data in Appendix 8) (from rank 1, representing the lowest mean percentage reported use to 5, representing the highest).
73
Tabl
e 27
Ran
k or
der c
orre
latio
n m
atrix
bet
wee
n fiv
e te
st tr
usts
rank
ing
for h
ealth
and
saf
ety
clim
ate
mea
sure
s, a
nd re
porte
d an
d ca
ptur
ed
inci
dent
s (p
hase
1) (
Spea
rman
’s ra
nk o
rder
cor
rela
tion)
(* =
Sig
nific
ant c
orre
latio
n (2
-taile
d)).
DIM
ENSI
ON
Ca
ptur
ed
inci
dent
s Cl
imat
e m
ean
Com
mun
-ic
atio
n In
volv
e-m
ent
Man
age-
men
t co
mm
i-tm
ent
Pers
onal
ap
prec
iatio
n of
risk
Pers
onal
pr
iorit
ies
Prio
rity
of safe
ty
Repo
rted
inci
dent
sSa
fety
ru
les/
pr
oced
ures
Supp
ortiv
e en
viro
nmen
tW
orki
ng
envi
ronm
ent
Cap
ture
d in
cide
nts
<0
.01*
0.
10
<0.0
1*
0.10
0.
62
0.28
<0
.05*
1.
00
<0.0
5*
<0.0
1*
0.19
Clim
ate
mea
n <0
.01*
0.10
<0
.01*
0.
10
0.62
0.
28
<0.0
5*
1.00
<0
.05*
<0
.01*
0.
19
Com
mun
icat
ion
0.10
0.
10
0.
10
0.19
1.
00
0.28
<0
.05*
0.
62
0.28
0.
10
0.39
In
volv
emen
t <0
.01*
<0
.01*
0.
10
0.
10
0.62
0.
28
<0.0
5*
1.00
<0
.05*
<0
.01*
0.
19
Man
agem
ent
com
mitm
ent
0.10
0.
10
0.19
0.
10
0.
19
0.87
<0
.05*
0.
75
<0.0
5*
0.10
0.
10
Pers
onal
ap
prec
iatio
n of
ris
k
0.62
0.
62
1.00
0.
62
0.19
0.50
0.
50
0.19
0.
28
0.62
0.
39
Pers
onal
pr
iorit
ies
0.28
0.
28
0.28
0.
28
0.87
0.
50
0.
39
0.87
0.
75
0.28
0.
87
Prio
rity
of
safe
ty
<0.0
5*
<0.0
5*
<0.0
5*
<0.0
5*
<0.0
5*
0.50
0.
39
0.
87
0.10
<0
.05*
0.
28
Rep
orte
d in
cide
nts
1.00
1.
00
0.62
1.
00
0.75
0.
19
0.87
0.
87
0.
87
1.00
0.
75
Safe
ty ru
les &
pr
oced
ures
<0
.05*
<0
.05*
0.
28
<0.0
5*
<0.0
5*
0.28
0.
75
0.10
0.
87
<0
.05
<0.0
5*
Supp
ortiv
e en
viro
nmen
t <0
.01*
<0
.01*
0.
10
<0.0
1*
0.10
0.
62
0.28
<0
.05*
1.
00
<0.0
5*
0.
19
Wor
king
en
viro
nmen
t 0.
19
0.19
0.
39
0.19
0.
10
0.39
0.
87
0.28
0.
75
<0.0
5*
0.19
74
Tabl
e 28
Ran
k or
der c
orre
latio
n m
atrix
bet
wee
n se
ven
trust
s ra
nkin
g fo
r hea
lth a
nd s
afet
y cl
imat
e m
easu
res,
and
repo
rted
and
capt
ured
in
cide
nts
(pha
se 1
) (Sp
earm
an’s
rank
ord
er c
orre
latio
n) (*
= S
igni
fican
t cor
rela
tion
(2-ta
iled)
).
DIM
ENSI
ON
Ca
ptur
ed
inci
dent
s Cl
imat
e m
ean
Com
mun
-ic
atio
n In
volv
e-m
ent
Man
agem
ent
com
mitm
ent
Pers
onal
ap
prec
iatio
n of
risk
Pers
onal
pr
iorit
ies
Prio
rity
of safe
ty
Repo
rted
inci
dent
sSa
fety
ru
les &
pr
oced
ures
Supp
ortiv
e en
viro
nmen
tW
orki
ng
envi
ronm
ent
Cap
ture
d in
cide
nts
0.
19
0.36
0.
08
0.19
0.
52
0.63
0.
27
0.41
<0
.05*
0.
41
0.57
Clim
ate
mea
n 0.
19
<0
.05*
<0
.01*
<0
.05*
0.
18
0.12
<0
.01*
0.
76
<0.0
5*
<0.0
1*
<0.0
5*
Com
mun
icat
ion
0.36
<0
.05*
<0.0
5*
0.09
0.
59
0.05
<0
.05*
0.
34
0.16
<0
.05*
0.
07
Invo
lvem
ent
0.08
<0
.01*
<0
.05*
<0.0
5*
0.34
0.
09
<0.0
1*
0.82
<0
.05*
<0
.01*
0.
07
Man
agem
ent
com
mitm
ent
0.19
<0
.05*
0.
09
<0.0
5*
0.
09
0.64
<0
.01
0.70
<0
.05*
<0
.05*
<0
.05*
Pers
onal
ap
prec
iatio
n of
ris
k
0.52
0.
18
0.59
0.
34
0.09
0.94
0.
15
0.38
0.
18
0.21
0.
18
Pers
onal
pr
iorit
ies
0.63
0.
12
0.05
0.
09
0.64
0.
94
0.
15
0.18
0.
59
0.09
0.
48
Prio
rity
of
safe
ty
0.27
<0
.01*
<0
.05*
<0
.01*
<0
.01*
0.
15
0.15
0.88
0.
05
<0.0
5*
<0.0
5*
Rep
orte
d in
cide
nts
0.41
0.
76
0.34
0.
82
0.70
0.
38
0.18
0.
88
0.
70
0.43
0.
48
Safe
ty ru
les &
pr
oced
ures
<0
.05*
<0
.05*
0.
15
<0.0
5*
<0.0
5*
0.18
0.
59
0.05
0.
70
0.
07
0.07
Supp
ortiv
e en
viro
nmen
t 0.
41
<0.0
1*
<0.0
5*
<0.0
1*
<0.0
5*
0.21
0.
09
<0.0
1*
0.43
0.
07
<0
.01*
Wor
king
en
viro
nmen
t 0.
57
<0.0
5*
0.07
0.
07
<0.0
5*
0.18
0.
48
<0.0
5*
0.48
0.
07
<0.0
1*
75
Tabl
e 29
Ran
k or
der c
orre
latio
n m
atrix
bet
wee
n fiv
e te
st tr
usts
rank
ing
for h
ealth
and
saf
ety
clim
ate
mea
sure
s, re
porte
d an
d ca
ptur
ed
inci
dent
s an
d w
orkb
ook
use
(pha
se 2
) (Sp
earm
an’s
Ran
k O
rder
Cor
rela
tion)
(* =
Sig
nific
ant c
orre
latio
n (2
-taile
d)).
DIM
ENSI
ON
Capt
ured
in
cide
nts
Clim
ate
mea
n Co
mm
un-
icat
ion
Invo
lve-
men
t M
anag
e-m
ent
com
mi-
tmen
t
Pers
onal
ap
prec
-ia
tion
of
risk
Pers
onal
pr
iorit
ies
Prio
rity
of safe
ty
Repo
rted
inci
dent
sSa
fety
ru
les &
pr
oced
ures
Supp
ortiv
e en
viro
nmen
tW
orkb
ook
use
Wor
king
en
viro
nmen
t
Cap
ture
d in
cide
nts
<0
.05*
0.
17
<0.0
1*
0.09
0.
55
0.09
0.
32
0.55
<0
.01*
0.
22
0.61
0.
43
Clim
ate
mea
n <0
.05*
0.10
<0
.01*
. <0
.05*
0.
39
<0.0
5*
0.28
0.
62
<0.0
1*
.019
0.
39
0.28
C
omm
uni-
catio
n 0.
17
0.10
0.10
<0
.05*
0.
50
<0.0
5*
<0.0
5*
0.62
0.
10
0.10
0.
19
<0.0
5*
Invo
lvem
ent
<0.0
5*
<0.0
1*
0.10
<0.0
5*
0.39
<0
.05
0.28
0.
62
<0.0
1*
0.19
0.
39
0.28
M
anag
emen
t co
mm
itmen
t 0.
09
<0.0
5*
<0.0
5*
<0.0
5*
0.
19
<0.0
1*
0.10
1.
00
<0.0
5*
<0.0
5*
0.28
0.
19
Pers
onal
ap
prec
iatio
n of
risk
0.55
0.
39
0.50
0.
39
0.19
0.19
0.
62
0.28
0.
39
0.28
0.
39
0.62
Pers
onal
pr
iorit
ies
0.09
<0
.05*
<0
.05*
<0
.05*
<0
.01*
0.
19
0.
10
1.00
<0
.05*
<0
.05*
0.
28
0.19
Prio
rity
of
safe
ty
0.32
0.
28
<0.0
5*
0.28
0.
10
0.62
0.
10
0.
87
0.28
<0
.05*
0.
50
0.19
Rep
orte
d in
cide
nts
0.55
0.
62
0.62
0.
62
1.00
0.
28
1.00
0.
87
0.
62
0.75
0.
75
0.50
Safe
ty ru
les &
pr
oced
ures
<0
.05*
<0
.01*
. 0.
10
<0.0
1*
<0.0
5*
0.39
<0
.05*
0.
28
0.62
0.19
0.
39
0.28
Supp
ortiv
e en
viro
nmen
t 0.
22
0.19
0.
10
0.19
<0
.05*
0.
28
<0.0
5*
<0.0
5*
0.75
0.
19
0.
62
0.39
Wor
kboo
k us
e0.
61
0.39
0.
19
0.39
0.
28
0.39
0.
28
0.50
0.
75
0.39
0.
62
<0
.05*
Wor
king
en
viro
nmen
t 0.
43
0.28
<0
.05*
0.
28
0.19
0.
62
0.19
0.
19
0.50
0.
28
0.39
<0
.05*
76
The results indicated a number of significant correlations. In the first phase (Table 27) the Trust rank of captured incidents was significantly associated with the Trust ranks of four of the nine questionnaire dimensions (involvement, priority of safety, safety rules & procedures and supportive environment) and the overall climate questionnaire mean score.
Significant associations with overall climate questionnaire mean score were present in both phases. In addition, there was a lack of correlation between climate dimensions and reported incident rates. This indicates that the Trust rank of incidents for which root causes could be assigned was potentially a better indicator of health and safety performance than total incident reporting rate.
In other words, higher overall mean questionnaire dimension score was significantly correlated with lower incident capture rate.
When the control Trusts were included in the first phase rankings the findings were less clear with the only significant correlation between captured incidents and questionnaire response rank was with the dimension of safety rules & procedures (Table 28). Trust rank of reported incident rate was again not significantly correlated with any other group.
For the five test Trusts in the second phase (Table 29) Trust rank of percentage reported workbook use was also included. The Trust rank for use of the workbook was significantly correlated with Trust rank for the climate dimension of working environment. As well as the overall climate questionnaire mean score the questionnaire dimensions of involvement and safety rules & procedures were significantly correlated with Trust rank for captured incident rate.
Therefore higher reported use of the workbook was significantly correlated with both higher overall mean questionnaire dimension score for working environment and lower captured incident rate.
This data should be interpreted with caution, given the small sample size (n=5) and the absence of data on workbook usage in the control Trusts. However, implications of these significant correlations are discussed in Chapter 4.
3.8 SUMMARY OF RESULTS
3.8.1 General findings
�� Between phase 1 and phase 2 the mean number of incidents reported increased by 24% although the deference was only marginally significant at the 0.05 level (P=0.06);
�� Between phase 1 and phase 2 there was highly correlated rankings of reported and captured incident rates for each Trust (P<0.01);
�� Reporting rates were not associated with the size of the Trust (P=0.48); �� Capture rates were significantly associated with the size of the Trust in phase one (P=0.01)
but only marginally associated in phase 2 (P=0.09); �� The numbers of incidents captured represented an average of 6% of the total that were
reported; �� Between phase 1 and phase 2 there was an increase in the number of incidents captured,
although it was not significant (P=0.41); �� Between phase 1 and phase 2 there was no correlation between Trust reporting and capture
rates (P=0.54);
77
�� The mean reporting and capture rates for the control Trusts were not significantly different (P=0.50 for both phases) from the test Trusts.
3.8.2 Special category
�� A challenge exists for those conducting incident investigation within the NHS to identify clearly cases where management root causes cannot reasonably be established;
�� There was an increase in special category incidents between phase 1 and 2, although not significant (P=0.69);
�� Most special category incidents were either not foreseen or outwith reasonable practicability (67% of special category incidents);
�� The remaining group involved clinical judgements that attempted to balance potentially conflicting health and safety and patient care factors (25%). There was a small proportion where there was insufficient information on which to base a decision (7%);
�� The traditional paradigm that most, if not all, incidents can be prevented by implementation of robust risk management arrangements has been challenged by the special category incidents. In the NHS it is likely that a percentage of incidents will always fall into the special category.
3.8.3 Identification of risk control system failures and root causes of incidents
So far as results that were relevant to research objective 1 (The design of a methodology for evaluation of health and safety performance) were concerned:
�� The risk control system of risk assessment was the major source of management root causes in both phases;
�� The four major management system root causes were those of planning, implementing, measuring and reviewing;
�� There were no significant differences between the test and control trusts in either phase of the project.
So far as results that were relevant to research objective 3 (The design of a methodology to measure change in health and safety performance) were concerned:
�� The numbers of root causes assigned to risk control systems was not significantly different between phases (P=0.51);
�� The number of times risk assessment was identified as a risk control system failure decreased significantly between phases (p=0.02), while violence and aggression (p=0.01) and handling sharps (p=0.02) significantly increased. This could indicate better reporting of incidents involving violence and aggression and sharps;
�� Effort on the Implementation of control measures as a result of risk assessment was thought likely to produce most impact on health and safety performance;
�� The differences in the numbers of cited management root causes between the two phases were not significant (P=0.92);
�� There was significant correlation between the ranks of frequency of citation of root causes between the two phases (P<0.05).
3.8.4 Costs of incidents
So far as results that were relevant to research objective 1 (The design of a methodology for evaluation of health and safety performance) were concerned:
78
�� The total cost for special category incidents (£55,593) was more than either of the two data collection phases (Phase 1: £51,096; Phase 2: £53,209) for the incidents where root causes could be assigned;
�� Most special category costs were attributed to accidents involving patients; �� Where root causes could be assigned, incidents involving staff were nearly three times more
frequent and represented approximately four times the cost than those for patients; �� The Pareto 80:20 rule broadly applied to the incident costing data in that approximately
80% of costs were attributed to approximately 20% of the observed incidents; �� The total cost assigned to the incident category of accidental injury was an order of
magnitude higher than any other category in both phases. Costs from physical violence increased by nearly four times between phases;
�� The category of ill health may underestimate the true cost to the NHS; �� The incident factors staff absence and replacement labour costs were approximately 75% of
the total costs; �� For incident types, person to person assault, slip trip or fall on the same level/stairs and
patient lifting/handling were approximately 75% of the total costs.
So far as results that were relevant to research objective 3 (The design of a methodology to measure change in health and safety performance) were concerned:
�� There was no significant difference between the two phases for total costs per incident factor or type (P=0.33 (incident factor), P=0.69 (incident type);
�� Total incident factor costs for the test Trusts were not significantly different from the control Trusts (P=0.85 (phase 1), P=0.77 (phase 2));
�� The average costs for incident types were not significant between phases (P=0.68). However, the rank order of average cost of incident type between the two phases was marginally correlated (P<0.1);
�� The difference in cost range for incident types was not significant between phases (P=0.35).
3.8.5 Staff opinion survey
So far as results that were relevant to research objective 1 (The design of a methodology for evaluation of health and safety performance) were concerned:
�� There was a 42% response rate from individuals who completed two valid questionnaires; �� The distribution of questionnaire response data was approximately normal allowing
parametric statistical testing to be applied; �� Pooling of data from individual Trusts was avoided because their results were statistically
significant from each other (P<0.05); �� A significant difference (P<0.05) in the opinion of medical and dental staff was found in
one Trust in phase 1.
So far as results that were relevant to research objective 3 (The design of a methodology to measure change in health and safety performance) were concerned:
�� Significant change in opinion was found in two Trusts (P<0.05). In one Trust a single dimension showed improvement while in the other six of the dimensions none showed improvement;
�� The questionnaire methodology could detect significant change in staff opinion and therefore represents a potentially useful measure of culture change;
79
�� The absence of significant change in five of the seven Trusts may be evidence that the underlying health and safety culture within these Trusts is relatively stable.
3.8.6 Monitoring and audit of the use of the management system workbook
There was generally low implementation of the workbook across all Trusts. Evidence for this conclusion included:
�� Over time the percentages of holders of workbooks tended to decrease, although not significantly (P<0.1);
�� Within Trusts use of workbooks decreased significantly over time (P<0.01) with approximately one third having dropped out by the end of the monitoring period;
�� General workbook use decreased significantly (P<0.05); �� All Trusts reported a significant increase in the use of particular sections (P<0.001); �� Use of the manager’s audit did not change significantly (P<0.1); �� Overall perceived workbook usefulness decreased significantly over time (P<0.001); �� Implementation of control measures as a result of risk assessment did not change overall
(P<0.1); �� Use of the option appraisal and suggestions for improvement did not vary significantly
(P<0.1). �� There were no statistically significant differences between individual Trusts in terms of
their responses to telephone interview questions (P<1.0). This was confirmed by the presence of significant rank order correlations between Trusts (P<0.05);
�� Visual inspection of a 10% sample of workbooks also confirmed low implementation of the workbooks.
3.8.7 Triangulation of results
�� Test Trust rank order of captured incident rates (i.e. incidents for which health and safety management system root causes could be established) was significantly correlated with climate questionnaire mean score in both phases (P<0.05);
�� Test Trust rank order of captured incident rates was also significantly correlated with the questionnaire dimensions of involvement, priority of safety, safety rules & procedures and supportive environment in the first phase and involvement and safety rules & procedures in the second phase (P<0.05);
�� When the control Trusts were included in the first phase rankings only significant correlation between captured incidents and questionnaire response rank was with the dimension of safety rules & procedures (P<0.05);
�� Trust rank of reported incident rate was not significantly correlated with any other group; �� The Trust rank for use of the workbook was significantly correlated with Trust rank for
captured incidents and the climate dimension of working environment (P<0.05).
80
4 DISCUSSION
There were two main findings in the research. These were, firstly, that there were significant correlations between the performance evaluation components. The second major finding was that there was little change in performance between the health and safety evaluation phases.
The discussion, conclusions and recommendations arising from these findings comprises of four parts:
�� Part 1: Consists of a discussion and evaluation of the individual components of the health and safety management performance evaluation method and how they might be refined to optimise their potential to reveal correlations;
�� Part 2: Comprises a discussion and evaluation of the strengths and weaknesses of the health and safety management workbook so as to form conclusions as to its impact on the lack of change in health and safety performance;
�� Part 3: Also comprises a discussion and evaluation of the strengths and weaknesses of the health and safety management performance change measurement methodology as to its impact on the lack of change in health and safety performance;
�� Part 4: Concludes with an assessment of the degree to which the research aim and objectives have been achieved.
4.1 HEALTH AND SAFETY MANAGEMENT PERFORMANCE EVALUATION METHODOLOGY
There are six sections to this part of the discussion. These are:
�� Performance evaluation based on the use of reported incidents; �� The use of the HSE root cause analysis methodology; �� Investigator bias and its implications for the training and development of safety advisers; �� Costs and economic implications; �� The Health and safety climate assessment; �� Triangulation of health and safety measurement methods.
Each will be considered in turn.
4.1.1 Performance evaluation based on the use of reported incidents
The Health services sector is currently a priority area for the HSC. One reason for this is because it is a major employer. The NHS employs approximately 1.1 million people, with more than 0.5 million in the private healthcare sector. Although it is not regarded as a high hazard industry, high rates of incidents are routinely reported to HSE(109).
This presents an interesting paradox because the activity of reporting incidents for the purpose of deriving health and safety performance data has been extensively encouraged in the NHS, in order to facilitate comparisons of performance between and within Trusts (1, 9).
In addition, Government and HSCs national targets(22) for improving health and safety performance over the ten years to 2010 are ambitious and partly reliant on incident reporting
81
data24. The Health Services Advisory Committee (HSAC) of the Health and Safety Commission (HSC) set a ten-year target:
“By 2004, health and safety performance across the health care sector will present a model for other sectors to emulate.”(54)
This aim was supported by a number of objectives, including one that set targets for the reduction of the incidence and costs of work-related incidents and ill health. In NHS Scotland incident reporting data (defined in the introductory Chapter) have been routinely collected since 200125 and are set against the context of a national strategy for health and safety performance targets(115).
It has been planned to use these minimum dataset indicators to monitor ongoing performance in occupational health and safety within the NHS(116). The numbers and rates of reported incidents were to be two of the indicators used.
Therefore it can be concluded that there exists in the NHS a strong culture of reporting incidents, no matter how trivial. The “if in doubt – report it” message, has been strongly advocated and implemented at local level by senior management and safety departments alike. The outcome of this strategy could be observed in the research results whereby some Trusts were found to deal with as many as 250 reported incidents per week.
However, in Chapter 3 it was shown that the number of reported incidents (between 255 and 2,481) and the reported incident rate (between 0.36 and 2.63 per person working years) differed between Trusts by an approximate order of magnitude. This is regarded as wide variability.
The cause of this variability could be related to differences in reporting rates between the Trusts but it could also represent an indicator of health and safety performance in that, Trust staff with more awareness of health and safety might report more incidents, a possibility also highlighted by the NAO(1). This is likely given that the majority of reported incidents were found to have little or no potential for harm. However, the total incident reporting rates did not significantly correlate with any of the other performance measures in the triangulation conducted in Chapter 3. It is suggested that his makes the variation more likely to be related to reporting culture, rather than health and safety performance, a limitation of using incident rates that has already been identified (e.g.(44)).
The captured incident rates also differed by an approximate order of magnitude (between 0.02 and 0.19 per person working years) and were a much smaller percentage of the total incidents reported (between 0.6% and 10.7% of the total numbers of incidents reported), however they were found to correlate significantly with workbook use and one of the questionnaire dimensions. Therefore this might be a more useful performance indicator for future use.
Nevertheless there remain a number of unresolved limitations with the method as it currently stands. These are: �� As discussed above, the main purpose of collecting incident data in the NHS has been to
benchmark health and safety performance between Trusts. However, there is frequent
24 The targets were to reduce the number of working days lost from work-related injuries and ill-health by 30%, to reduce the incidence rate of cases of work-related ill-health by 20% and to reduce the incidence rate of fatalities and major injuries by 10% (achieving half the improvement by 2004) 25 at the time of writing the first data collection for the year ending 31 March 2001 was still ongoing had had not yet been published. It is therefore not yet possible to compare the research data with this routine data.
82
reference to learning lessons from incident data in health and safety publications referred to by safety practitioners working in the NHS (e.g. (1, 9, 18, 25, 35, 36, 117)). It is questionable as to whether there is sufficient clarity between Trusts as to the distinction between these two purposes. There is work to do to ensure that policy makers, occupational health and safety professionals, line managers and local staff understand why they are collecting and reporting incident data;
�� Although the Trusts participating in the research tended to use a similar approach to incident reporting, there is an undoubted variation in conventions for reporting clinical versus non-clinical incidents and also accidents and ill-health between Trusts. This is because the reporting mechanisms tend to be different for these types of incident. For example, clinical incidents (such as infections, clinical negligence etc.) tend to be reported via the clinical network (such as director of medicine or nursing) where as non-clinical incidents tend to be reported through the health and safety department. Cases of ill health tend to be reported or recorded by the HR or occupational health departments, where as accidents tend to be reported via the health and safety department. Communication and collaboration between these various departments, who tend to use their own reporting systems, has been highlighted as lacking by the NAO(1). The recent moves towards more multidisciplinary occupational health and safety services within the NHS may offer potential for improvements in this area(8, 23).
Therefore it is concluded and recommended that national strategy based on the use of incident rates to benchmark health and safety performance should use a tighter definition that the current one (described in the introductory Chapter), possibly similar to that used in the research (i.e. the presence of actual or potential for damage; that the incident was under the control of the Trust; and incurring more than £5 cost or 15 minutes time). In any case it is recommended that efforts be made within the NHS to work towards a common standard for incident reporting, supported by guidelines, which clarify the purpose for which the incident reports are used, in order to better control the risk of reporting inconsistencies.
4.1.2 Use of the HSE root cause analysis methodology
The previous section (4.1.1.) explored the use of reported incident data as performance indicators. The other use of reported incident data is to learn lessons so as to control health and safety risks by identifying underlying causes (e.g. (18, 118)).
In this regard, it has been suggested that there are six key features for a suitable incident analysis method (118). These are that the method should:
�� Distinguish between events and underlying causes; �� Allow for the grouping and pooling of accident data; �� Identify remedial action to prevent future accidents; �� Offer reliability between different investigators; �� Be pragmatic and capable of being used by non-specialists; �� Allow conclusions to be reached and prioritised action plans to be produced.
The HSE methodology used in the research26 met all of these criteria. The circumstances of the incident could be linked to risk control system failures and thereby to the root causes within the health and safety management system. (99). This information could then be used to identify remedial action, in other words, to learn lessons. Therefore, the purpose of the root cause analysis aspect of the HSE methodology was different from that of the incident frequency data.
26 Described in Chapter 3 and reproduced in Appendix 1.
83
The methodology provided information to allow conclusions to be formed that, not only was risk assessment the most significant underlying system failure, but that it was associated with the root causes of planning and implementing, and that they were the most significant management system root causes. This has implications for action planning for the employer. This finding was communicated to the participating Trusts at the end of the second phase of the research project. It is therefore not known if any action they took had an affect on health and safety performance.
In some respects the finding of a lack of risk assessment at the root of most of the incidents is unsurprising. For example, in the personal experience of the author, many safety practitioners working in the NHS would subjectively concur, as this is what they find on a day-to-day basis. On the other hand, the requirement for risk management systems has been very much part of the rhetoric of the HSE since the early 1990’s (when Crown immunity was lifted from the NHS). HSE has issued a plethora of advice and guidance on the systematic management of health and safety with risk assessment at its core. It is therefore disappointing that, despite the considerable efforts of nearly 15 years, the message still has not apparently translated into effective control of risk.
Therefore, it is recommended that priority should be given to ensuring that the system of risk assessment was operating effectively and, in particular, the planning and implementation of the system, it is possible that improvement in health and safety performance might result. This approach would certainly direct effort where it was most needed in the health and safety management system.
Another strength of the method was that its use revealed the existence of the group of incidents, which were thought to be unpreventable, those grouped under the title “special category”. Others have noted this phenomenon. For example, Neale et al (119) estimated that approximately half of a group of adverse incidents that occurred in two hospitals were not preventable while Oliver et al(80) also noticed that some health and safety risks in a healthcare setting were difficult to manage. For example, for frail, elderly patients recovering from acute illness, a delicate balance of risk factors is needed because the risk of falling can be challenging to prevent without adversely affecting rehabilitation, a finding also present in the current research, where the special category incidents represented approximately half of those incidents that met the research inclusion criteria and also about half the total cost.
Therefore, although unexpected at the start of the research, they must be regarded as an important group of incidents. It would undoubtedly be possible to argue that some of the incidents could have been categorised differently, but the overall picture remains that, within healthcare, there exist incidents, which do not fit within the adopted health and safety management model.
Although the special category comprised incidents that occurred to both staff and patients, the legal obligations are different. So far as staff is concerned, legislation in the UK centers, within the Health and Safety at Work etc. Act 1974, on the concept of risk control “so far as is reasonably practicable” (SFARP) which is, in effect, a requirement to undertake a cost-benefit analysis whereby risk is weighed against the sacrifice (in money, time or trouble) involved in averting the risk. The decision as to whether risk was controlled in any particular set of circumstances is generally left to the legal system to decide, with its large body of legal precedent on which to base its judgment.
So far as incidents involving staff and patients are concerned the legal test is one of a common law “duty of care” against which claims of negligence or clinical negligence (broadly defined as
84
“careless conduct which injures another”) can be submitted, via the civil courts, again drawing on the body of legal precedent to establish on the “balance of probability”, whether the duty of care was breached(120). So far as patients are concerned the definition of reasonable practicability was one applied by the research team, as a convenient label for this group of incidents. However, it must be remembered that this does not constitute a legal definition.
It is concluded that the special category results have an impact on the definition of reasonable practicability within the NHS as an industry. It is therefore recommended that this issue must be debated further.
Certainly the concept that nearly all incidents have underlying management system root causes(74) and that they are, therefore, preventable seems to be challenged by this group of incidents. It may be that healthcare is a “special case” in that its client group is, by definition, vulnerable and that some incidents, particularly those occurring to patients, may be inevitable, taking account of the needs of the patient and available resources.
However, if this is the case for some incidents, it should not be used as an excuse to avoid the implementation of robust risk management procedures. Nor should the apparent special category of an incident be used as justification for curtailing detailed investigation of serious incidents.
This is not judged to be likely, so far as patient care is concerned. This is because, in many ways clinical management in the NHS is more developed than health and safety management. For example, there exist organisations such as the National Patient Safety Association (NPSA)27 and the National Institute for Clinical Excellence (NICE)28, set up as a Special Health Authority for England and Wales on 1 April 1999 as part of the NHS to provide patients, health professionals and the public with authoritative, robust and reliable guidance on current “best practice” on individual health technologies (including medicines, medical devices, diagnostic techniques, and procedures) and the clinical management of specific conditions. Managers can and should continue to apply these robust management procedures to all aspects of their work, managing risk in the broadest sense.
The author has classified these incidents as a special category. This may be misleading in that the category was not intended to be definitive but rather a convenient label under which to group the incidents. A more appropriate description for the group may be necessary. They all share a common link in that they are incidents occurring as an outcome of the healthcare industry. It is therefore recommended that, in future, investigations of incidents should bear in mind the possibility of a classification of “intrinsic incidents”.
It is therefore concluded that the HSE root cause analysis method was a powerful tool, which could identify consistently health and safety management system weaknesses and, importantly, identify whether there exist incidents in the data set, which cannot be prevented within the bounds of accepted best practice.
Limitations of the method were that it was useful as a research tool but if it were used for more routine applications, further development is recommended. In particular its performance against a range of industries should be established and how it might be applied to workplace ill health. The method would also benefit from conversion to a software tool, possibly based on an expert system, to ensure it is intuitive and easy to use. This may assist with some of the more laborious tasks associated with processing investigation data. 27 http://www.npsa.nhs.uk/ 28 http://www.nice.org.uk/cat.asp?c=20
85
4.1.3 Investigator bias and its implications for the training and development of safety advisers
Wagenaar has suggested that approaches to incident analysis should be reliable in that various independent analysts should reach the same conclusions(118). The literature review (Chapter 1) identified evidence that the current UK standard of incident investigation was poor in that wide variations in approach, and therefore unreliability, were commonplace.
This was confirmed during the incident investigations and root cause analyses described in Chapters 2 and 3, which were typified by a challenge to maintain consistency against the pre-set standard. The consensus approach that was adopted and described in Chapters 2 and 3 has also been discussed by Culvenor and others, who found that consistency of judgments of teams were better than the average of the individuals that comprised the team(121, 122). This adds weight to the conclusion that the results of the incident investigations were, as a result of the control measures, internally consistent.
The main limitation with adopting this consensus approach was not with the research project where the thorough investigation was resources and planned for. The challenge lies in achieving incident investigation consistency within and between safety professionals working on a day-to-day basis, usually with limited time and resources to devote to thorough data gathering prior to incident analysis. This is particularly relevant in the NHS where, the results in Chapter 3 have shown, some Trusts were processing up to 250 incident reports each week. This begs the question as to whether the purpose of collecting the incident data was sufficiently clear within the healthcare organisations.
For example, HSE have suggested that collecting information on injuries and ill health should not present major problems for most organisations but that learning lessons from these incidents can prove more challenging(18). The author has found this to be the case both before and during the research, where the collection of incident data and the preparation of trend analyses and reports seemed the paramount purpose of some health and safety departments. As described in Chapter 2, thorough investigation of incidents seemed to be a secondary task. An illustration of this was when HSE issued a consultative document(123) on proposed changes to the Reporting of Diseases and Dangerous Occurrences Regulations (RIDDOR) 1995(109). Informal discussions with the health and safety practitioners participating in the research project indicated that their perception of the impact of a proposed new duty to investigate all RIDDOR incidents was that it would not be an onerous task. This was thought to be because the task was already routinely undertaken.
Therefore it can be concluded that the purpose of incident reporting in the NHS has been largely as a performance indicator (already discussed above in Section 4.1.1.) and that the process of learning from adverse events has been comparatively underdeveloped and largely based on analysis of frequency, something Hale has cautioned against(112). Hale has also suggested that there may be differences between the causes and underlying conditions between minor and major accidents(112). The use of minor incidents and near misses as a performance measure was thought by Hale to be erroneous and particularly inappropriate if extrapolated to apply to performance in major hazard control. In the experience of the author it was common for underlying causes of minor incidents and near misses to be assumed to be the same as more serious incidents. This illustrates further the underdeveloped culture of learning from adverse events in health and safety in healthcare.
In the search for reasons to explain this underdeveloped situation it is possible to look at training for safety professionals, which has concentrated on techniques of incident investigation
86
and root cause analysis for industries with the highest hazards29. For “softer” industries, such as healthcare, without engineering data on which to base failure probability calculations, the professional safety adviser has to rely on a combination of subjective judgment based on their professional background, experience and interviews or discussions with those involved.
However, this position may be starting to change. For example, training for safety advisers has begun to shift towards producing agents of change rather than the more traditional adviser/officer role. For example, Swuste and Arnoldy have reported on how they have developed their postgraduate masters’ course in health and safety at the University of Delft to meet this need(88). This approach is in response to a perceived lack of progress with management of health and safety using traditional methods.
4.1.4 Costs and economic aspects
If it is assumed that the two twelve week data collection phases were typical, the costs represent an extrapolated equivalent annual total cost for the seven Trusts of £221,416 for phase 1 and £230,572 for phase 2. Taking the speculation to the entire NHS in Scotland this would equate to a cost for captured incidents of approximately £1.3 million per annum (£1,284,996 for phase 1 and £1,338,133 for phase 2). This assumes 23,852 staff in the seven participating Trusts and 138,426 total staff employed in the NHS in Scotland, as of September 2001(124). For the NHS in England the equivalent figure (based on a total staff compliment of just under 1 million (948,275)) is approximately £9 million per annum30 (£8.8 million in Phase 1 and £9.2 million in Phase 2).
However, a cost range may be more appropriate given the wide cost variation observed between the Trusts. If the extrapolated annualised costs for the NHS were based on the lowest rate observed (£2.25/person/annum (based on £0.01 per person working day * 225) the overall losses were of the order of £300,000 (£311,458) in Scotland and approximately £2million (£2,133,619) for England. If the highest observed rate were used on which to base the estimate (£60.75/person/annum), the potential losses increase to just under £8.5million (£8,409,379) in NHS Scotland and just over £57.5million (£57,607,706) in England.
This is a wide range of estimated costs. If the 2000/2001 running costs for the Scottish NHS are taken into account (running cost estimated at approximately £2.5 billion31) incidents accounted for between 0.01% and 0.31%. The equivalent annual estimate for the NHS in England and Wales (which receives proportionately less funding) is between 0.06% and 1.44% (running cost estimated at approximately £40billion (£39,883million)).
These costs are relatively low and represent approximately 50% of those identified in the pilot work reported in Chapter 1 and approximately 20% of that estimated by HSE in 1992(35). They estimated, using a total loss approach, that direct financial and opportunity costs amounted to 5% of a hospital’s running costs (incurred by accidents alone). HSE’s estimation of the costs to UK employers of work injuries and non-injury accidents for health and social work in 1995/96 was approximately £240million(125), approximately two and a half times the highest estimate for the current data (i.e. 40%). More recently, the HSC launched a web-based “ready reckoner” to assist businesses to work out costs of work-related accidents and ill health(126). Although the method takes a similar approach to the research methodology (i.e. calculating costs as a product of time spent and hourly rate), the range of categories is wider and therefore
29 Information on course contents for UK training courses for health and safety advisers available from NEBOSH (http://www.nebosh.org.uk/) and IOSH (http://www.iosh.co.uk/). 30 Source http://www.doh.gov.uk 31 Source http://www.show.scot.nhs.uk/isd
87
costs likely to be higher for a given incident. For example, costs for sanctions and penalties such as fines and legal expenses are included in the ready reckoner but were not included in the current research unless they occurred within the 12 week data collection windows.
The only study to publish costing data that were significantly lower than the current research was a National Audit Office (NAO) study into NHS acute hospitals in England(9). Data wascollected in 30 hospitals over an eight week period in 1995. The immediate costs of accidents were estimated for England at approximately £12million. However, the costs were taken over a shorter time period (eight weeks against twelve weeks) and the data was nearly six years old. Cost ranges are not given and primary care Trusts were not included. It is therefore unlikely that the data is directly comparable. The 2003 review of this work by the NAO(1) concluded that the direct cost of work-related accidents was at least £173 million, but estimated that this was a conservative estimate. The results in this report were just over one third (38%) of the NAO estimate.
However, despite being conservative the research costing protocol was rigorous and applied, via a standardised costing frame, consistently in each Trust. The resultant costs can therefore be regarded as reliable, thus ensuring the validity of making comparisons between Trusts. However, they included only direct costs to the participating Trusts. An illustrative example was a case where a patient was found hanging using a rope suspended from a window catch. The ligature point had been identified as a high risk, but remedial measures had not been actioned by the time of the suicide. The incident was allocated a cost of £19.67, to cover the direct cost of staff time dealing with the immediate actions necessary. The costing protocol did not include the undoubtedly higher cost to the Trust of longer-term issues such as legal action and any official enquiry. The cost to the individual and their family are also not included, nor was any account taken of emotional issues.
In addition, the method did not lend itself to inclusion of cases of ill health, unless they were diagnosed within the data collection windows. For example, most needlestick incidents were costed at between £20 and £25. Obviously the long-term consequences of some of these incidents could result in ill health and the cost obviously does not account in any way for anxiety or distress suffered by the individual.
It is therefore apparent that a possible reason for the difference in cost estimates could be the rigorous inclusion criteria for the current study. It has been suggested that one reason for using extrapolated national costs has been to persuade senior business managers to invest in prevention(112). It has also been suggested that the larger these sums are, the more persuasive they are likely to be. However, it has already been suggested that UK business may not be fully persuaded of this argument(127). Koopmanschap and Rutten identified a body of opinion supporting the exclusion of indirect costs in healthcare-based economic evaluations unless compelling evidence for their inclusion was present(128). An alternative paradigm is therefore proposed, which concentrates more on attempting to change boardroom attitudes to costs of health and safety. This is regarded as a potential key to success.
In the experience of the author it can be more effective to take a conservative incident cost estimate to a board of Directors. Their natural, and correct, response when presented with financial data is to question their validity. When the estimates are high senior managers’ tend to look for reasons why and then to revise the estimates downward. When they are conservative they are more likely to interpret that the true loss to the business may be worse than estimated and are therefore, ironically, more likely to invest in prevention. It is therefore suggested that the results, far from being non-persuasive because of their lower magnitude relative to other estimates, could be more influential since they are obviously pragmatic, understandable by senior managers, and still represent significant losses to employers. It is recommended that
88
further behavioural based research should be conducted to test the potential usefulness of this phenomenon.
In any case the current data undoubtedly represent a significant underestimate of the true cost of occupational health and safety to the healthcare sector. Estimation of this cost represents a considerable challenge. In the participating Trusts sickness absence was not studied, but it is known that they represented approximately 6% of salary costs. If it were assumed that half of this was related to workplace occupational health and safety32 then this represents approximately £1.2billion for the NHS in England and approximately £2.5million in Scotland.
However, these estimates are, at best, speculative, given there is, as yet, no agreed standard for cost estimation or incident reporting within the NHS. This will continue to limit progress in that the debate is currently focused on measuring the magnitude of cost, rather than on means to control them.
It is not suggested that Trusts routinely cost incidents, since the cost-benefit of this action is not likely to be favourable. In any case, there is also sufficient data in the literature to convince that costs, especially of occupational ill health, are considerable. It is however recommended that costing exercises are carried out on an occasional basis in order to check progress and to reaffirm the presence of sources of high cost.
As already noted by others(128), there was no doubt that some of the costs recorded in the data collection were borne by a Trust other than the one in which the incident occurred. This was particularly noticeable with incidents occurring in primary care Trusts. If treatment was needed this was usually done via the local casualty department of an adjacent acute Trust, which would then incur the cost of treatment and potential extended stay of the patient. None of these transferred costs were estimated but they represented a significant group. This transfer of burden of cost from the primary to the secondary care establishments could represent, particularly for the primary care sector, a disincentive to accident prevention on the basis of business management. Although the cost of treatment would usually ultimately be borne by the same health authority/board, the accounting centers would deduct the cost of treatment from the budget of the relevant clinical group in the acute Trust. Additionally, the acute Trust would not recognise cases of this sort as relevant to its accident management strategy because it would regard them as externalities and not in their interests to control.
That the NHS can be seen to routinely transfer costs within its own organisation is a minor issue once consideration is given to the massive financial burden on acute Trusts from every other industry who also transfer costs via their accident and ill-health victims. This has been estimated at hundreds of millions of pounds each year(129). Proposals to recoup these costs have been included in the debate for changes to employers liability insurance. Proposals exist for a “no fault” scheme, which would ensure that the proportion of costs borne by the employer rose to a level that acted as a driver for improved health and safety management(129). Proposals such as this have the potential to assist with the difficulties already discussed with regard to enforcement versus business planning and may be more effective at changing attitudes than the current arguments.
32 This is based on the Revitalising Health and Safety Strategy, described in Chapter 1, which seeks to reduce the number of working days lost from work-related injuries and occupational ill health by 30% by 2010. If it is thought possible to reduce the rate by such a figure it must be assumed that it is currently higher than this.
89
4.1.5 Health and safety culture
The decision to use a previously extensively validated climate questionnaire(58, 59) has been vindicated by the robust nature of the data obtained, as shown by the confirmatory factor analysis (Chapter 3). This data could then be used to gain insight into the safety opinion of the participants, and ultimately an assessment of the safety climate(66). The use of a longitudinal design yielded matched pairs of responses from each individual and reduced the effect of confounders on the data while the sample size also meant that statistically significant results were obtained. The issue of triangulation of the questionnaire responses with the other performance measurement methods is dealt with in Section 3.8.7.
Although the overall questionnaire response rate at 52% was regarded as good, especially given each respondent had to provide two valid questionnaires with a year in between, consideration should be given to additional methods that could have optimised this rate further. For example, the Trusts could have encouraged staff to complete the questionnaire; perhaps by giving them protected time in which to do so. Also, the period in which the respondents were given to respond could have been extended and the non-responders followed up again. A major disadvantage to this approach was that the time period during which the questionnaire was completed might therefore not coincide with the incident data collection phase. This was something that was regarded as important to achieve if the triangulation was to be valid. Therefore, neither of these options was judged to be a cost-effective option because of the additional resources required and the statistical power of the results within the existing response rate.
The job title profile of questionnaire responders was shown in Chapter 3 to be representative of all staff groups. It can therefore be concluded that the non-responders were also representative of all staff groups. However, it is not possible with the current data set to conclude more about the non-responders. It is likely that those who responded had more positive opinion about health and safety but this risk was controlled for in the research design methodology with the same individuals providing a valid questionnaire in the two benchmarking phases to establish whether there was any change in opinion.
The most significant difference found in the questionnaire responses was at Trust level with the between Trust variation larger than the within Trust variation (p<0.05). It had been assumed at the start of the research that grouping of the data from individual units, such as NHS Trusts, would be a valid strategy. However, the apparent similarities between NHS organisations which can be experienced, at a macro level, by those who work in it appear to be over-ridden by the local culture which, presumably, has developed as a result of devolved management to the Trusts and health authorities. Whether this is something that has changed over the lifetime of the NHS is not known. However, the finding was consistent between both data collection phases and therefore represents a snapshot of the situation between the years 2000 and 2002, when the NHS was less devolved than it was during, say, the early 1990’s.
Whether there exists a demographic aspect is also unclear since none of the participating Trusts were from the same geographical area. There was a mixture of secondary and tertiary type of Trusts but the Trusts were all so significantly different from each other that it was not valid to combine them so that they could be analysed with any grouping. In retrospect it may have been useful to have included Trusts from the same health authority area. However, personal experience of tertiary, acute hospitals (in this research, Trusts E and F) is that they care for different client groups and provide different services from those in the secondary care setting, such as primary care Trusts (i.e. Trusts A, B, C, D, & G). In addition their cultures are different. It therefore seems unlikely that if this grouping had been available that the findings would have changed.
90
The other factor that yielded significant differences in questionnaire response, and therefore safety climate, were the staff categories. With the exception of Trust F, none of the other Trusts showed any significant differences between the staff groups, again indicating that the Trust unit was the grouping that exerted the greatest influence on staff opinion.
When the difference in Trust F was examined further it was only present in the first data collection phase. This effect was so strong that when the initial data, pooled from all Trusts, were examined it appeared that the medical & dental grouping represented a group with opinion on health and safety that was significantly different (lower) from the others.
The most obvious unique difference between Trust F and the other Trusts, which might explain the finding was that it formed part of a large medical school and, as such, would have a large number of student doctors and also doctors whose role included a large teaching component. This may have influenced the results, given the knowledge that health and safety was not part of the curriculum of this medical school33. The timing of Phase 1 (1st August 200 to 23rd October 2000) coincided with the annual intake of junior doctors to the Trust. Phase 2 (1st January 2002 to 25th March 2002) was at a more stable time of the year for this Trust. It seems likely that doctors could have been under greater workload pressure during the first Phase as they balanced the needs of their job with the need to ensure adequate induction and supervision of a large number of inexperienced students.
Nevertheless, it is reassuring that the method and sample size was sensitive enough to detect statistically significant differences. This creates a strong message for other researchers and highlights the danger of pooling data without first checking whether it is valid to do so. It also gives a clear addition to the body of research knowledge that it is likely that this phenomenon applies to all NHS establishments and research designs can be tailored accordingly.
4.1.6 Triangulation of health and safety measurement methods
Triangulation was a feature of the research design and is regarded as strengthening the design. Its use has been described and discussed in Chapter 1 and identified as having many advocates, although no identified proof of its effectiveness. The research design made possible effective triangulation across paradigms, leading to the identification of a number of significant correlations.
The first significant correlation was that those tests Trusts whose staff had a higher overall mean questionnaire dimension score (i.e. more positive opinion of health and safety management) were significantly correlated with Trusts with lower incident capture rates. When the control Trusts were included the only questionnaire dimension to be significantly associated with lower incident capture rate was that of safety rules & procedures. Intuitively this seems logical, since having clear safety procedures is indicative of a strong health and safety management system, which is supported by having fewer serious incidents (if the captured incidents are regarded as the most serious). This conclusion is reinforced by the consistent finding that rank of reported incident rate (i.e. all incidents), was not significantly correlated with any questionnaire dimension. In the second phase, these significant findings were repeated, thus increasing confidence in the results.
Also in the second phase was available the workbook monitoring data for the test trusts. Higher reported use of the workbook was significantly correlated with both lower captured incident rate 33 Personal communication regarding an ongoing project between HSE Health Services Unit and the Council of Heads of Medical Schools (CHMS)
91
and higher overall mean questionnaire dimension of working environment. This is an important finding but needs more work to establish the mechanism behind the correlation.
For example, it would be tempting to suggest that more use of the workbook had led to lower numbers of serious incidents and better opinion on the working environment by staff. However, there is no proof of cause and effect with the current data set, especially given that no change in performance was detected following the use of the intervention (discussed further in Section 4.3). It may be that Trusts with better commitment to health and safety might encourage more the use of the workbook and other factors were responsible for the lower serious incident rate and higher opinion. The inclusion of a behavioural aspect may have helped interpretation of these results, already identified as a possible limitation of the intervention.
Nevertheless, the results of the triangulation add to the knowledge base by providing supportive evidence for the paradigm that having a positive health and safety management system is associated with better health and safety performance(18, 25, 130-137).
The main limitation to these significant findings is that, because the control Trusts were not included in the monitoring scheme, there is no data on which to confirm further, or otherwise, the findings. This is an acknowledged limitation of the research design and future work using research designs of this type should ensure that the control received the same monitoring as the test Trusts. The reason that this was not done was that, although the control Trusts had established health and safety management systems supported by workbooks, there was incomplete data on workbook holders, on which to base telephone interviews. This was because the workbooks had been in place for some time and gradually the original list of holders had changed.
4.2 HEALTH AND SAFETY MANAGEMENT INTERVENTION
The workbook was based on the main principles underpinning HSE’s guidance to health and safety management(74, 99). Rather than a detailed information resource, it was intended to be a pragmatic tool that could be used by line managers to ensure that they had addressed all of the aspects of the management system.
It has been suggested that a possible limitation of impact evaluations is that interventions may produce a small effect that is too weak to be detected(138). In the case of this intervention, the field data can provide evidence as to whether the effect was small.
Firstly, the maximum effect that would not be detected can be defined from the power of each aspect of the performance evaluation:
�� For the incident reporting data, based on a standard deviation of 0.014 units in average reported incident rate per person working days, a sample size of seven Trusts would have 98% power in detecting a 2% shift in reported incident rate;
�� For the questionnaire responses (standard deviation of 0.83) units in dimension score measurements sample size of 200 would have 96% power in detecting a 0.2 shift in dimension score;
�� For the workbook usage data (based on an estimated standard deviation of 0.49 units in answers to telephone interview questions) a sample size of 100 would have 98% power in detecting a 15% shift in answer response.
92
The power of the study was therefore regarded as high. It can be assumed that, if there was any change in performance brought about by the use of the workbook, its effect was below these thresholds.
There are two factors that may have impacted on effectiveness of the intervention in this study:
�� The degree of use of the workbook; �� The time interval between the benchmarking phases.
These will be considered in turn.
4.2.1 The degree of use of the workbook
The five Trusts that participated in this phase of the research were free to manage deployment and use, as they saw fit, within the general guidance given by the project team. This meant that there would be a natural variability of attitude and tactics of use within Trusts.
However, as has been shown in Chapter 3, approximately one third of workbook holders had ceased using the workbook by the end of the 12 months implementation phase. It was also shown that, of those who remained, their use of the workbook decreased significantly over the period.
This data on workbook use was obtained from the telephone interviews and inspections of workbooks. As has been described in Chapter 3, many of those who dropped out of the telephone interviews were either unavailable for comments or declined to take further part in the research. It is likely that this group were not using the workbook. There was therefore likely to have been a self-selection bias within the group who provided telephone interview data, probably those who had the most positive views. Although it is not possible to conclude this with certainty it should be borne in mind when interpreting the results. If anything, it indicates that workbook usage could have been worse than measured.
This serves to reinforce the conclusion that there was low implementation of the workbook. It cannot therefore be concluded that the workbook produced an effect that was too weak to be detected. Had the workbook been more fully used as intended, there might have been an effect on health and safety performance.
The decision to allow Trusts relative freedom to manage implementation of use of the workbook was deliberate in that, with the limited resources available to the project, it was necessary to have the co-operation of the participating Trusts. This would have been difficult to secure if participation meant the imposition of a rigorous common standard across all Trusts, since Trusts worked in a culture of establishing procedures, which took account of local factors to broad national standards.
However, this limitation of the research design should be controlled for in future research. Attempting to secure greater commitment from Trusts could achieve this, possibly by securing advance agreement to predetermined measures designed to encourage use.
4.2.2 The time interval between the benchmarking phases
The second factor is the length of time between the first and second benchmarking, which, at approximately 12 months, may have been too short for change to become established. In other
93
words, the process of change is complex and, in a complex organisation like the NHS, likely to take time.
For example the European Foundation for Quality Management (EFQM)34 suggest that the process of implementation of their Business Excellence Model (BEM) takes typically between three and five years. Wright et al(139) adopted the model as part of a Safety culture improvement matrix and advocated that improvement was multi-factorial, therefore likely to take a similar time as the BEM to establish. Ogden(140) described the stages of change model as a five-stage dynamic process which takes place over time. Actively engaging in a new behavior is the fourth stage, preceded by pre-contemplation contemplation and preparation. The author contended that individuals might move between preparation and contemplation several times before taking any action. At group level Houghton et al(141) introduced the concept of a “group mind” in their five-stage cognitive model of team behaviour sustainability. This took account of the group dynamic that undoubtedly exists in the NHS.
The initial benchmarking data was designed to provide, not only a measure of health and safety performance, but also, via the HSE root cause analysis methodology, a baseline measure of the health and safety management system. This was because it was used to track causes of incidents to their root in the health and safety management system of the organisation and had been validated against audit, the more commonly used method to assess health and safety management systems(34).
Further benchmarking exercises with the same Trusts and individuals would be a way to test whether a longer time interval would result in a greater effect. However, the use of the workbook has already been shown to be low, decreasing significantly over time. Unless the Trusts had put greater effort into encouraging use the greater time period would be unlikely to show a difference. Nevertheless, if detailed information about the measures taken by Trusts to encourage use could be obtained and controlled for it might be a worthwhile exercise.
An alternative possibility may be via the use of certification standards, such as BS8800:1996 or OHSAS18001:1999(30), to support the use of workbooks of this type, used as a form of quality manual. This type of practice is already well established and, the benefit of independent inspections of the workbooks, tied into retention of certification, is likely to be a major motivating factor in their use. Unless and until such a scheme is introduced, consistency of use within individual Trusts will continue to be dependent on the motivation and enthusiasm of the senior and line management of the organisation. Consistency will be much more likely to be attributed to chance in this scenario. The continuing UK policy of self-regulation is more likely to continue with the use of standards of this sort on a voluntary basis, for the foreseeable future.
Nevertheless in its relatively crude format, the workbook did meet an apparent need, illustrated by the degree of usefulness reported by users in the telephone interviews conducted to assess feedback on its use. With further refinement, possibly involving some of the individuals who used it during the project, the workbook and its briefing schedule could be made freely available via the Internet. This could be done via the HSE or Scottish Executive sites.
However, it is suggested that the workbook be regarded as only a part of any strategy to improve health and safety performance and that for significant change to occur it should be used as part of a multi-factorial model, such as the ones described above. This would offer a potentially fruitful avenue for further research.
34 http://www.efqm.org
94
4.3 HEALTH AND SAFETY MANAGEMENT PERFORMANCE CHANGE MEASUREMENT METHODOLOGY
A critical evaluation of the research design used in this report was carried out to the strategy identified in Chapter 1. Seven main headings were used so that a relative analysis of the strengths and weaknesses of the research can be made. These are:
�� Intervention objectives; �� Conceptual basis; �� Research format and design; �� External validity; �� Threats to internal validity; �� Outcome measurement; �� Other limitations.
Each will be considered in turn.
4.3.1 Intervention objectives
The aim of the intervention was stated and supported by three relevant objectives (reviewed in Section 4.4). The aim was, broadly, to measure change in health and safety performance in the healthcare sector. The maximum effect that could be detected by the method was established.
4.3.2 Conceptual basis
Although the literature base was diverse in terms of subject base, it mainly comprised non-experimental or quasi-experimental approaches, from a positivist perspective. Although currently popular, behavioural research and that based on management principles, such as qualitative research from a phenomenological standpoint (for example (78, 105)) were relatively less well established in the health and safety field. More lateral thinking is needed as to the foundations of health and safety research, perhaps using other models more normally associated with business(91) or organisational development(142).
The research design included both positivist and phenomenological aspects(68). The objective incident analysis and properties sought data to explain findings. Descriptive data was obtained from the questionnaire and were used to try to understand why individuals, and the Trusts for which they worked, had different experiences within a similar environment of provision of healthcare. An assessment of the baseline health and safety system, as well as performance, was established with the first benchmarking phase. The detection of change was made possible by the second benchmarking phase.
This combined approach was largely successful in that there was more than one indicator of effectiveness successfully measured by triangulation.
4.3.3 Research format and design
So far as the positivist paradigm is concerned, the definitive source of guidance on research design within occupational health and safety was that of NIOSH(44). However, other disciplines, such as medical research and the work of the Cochrane Collaboration(76), bears close scrutiny by occupational health and safety researchers. Whether there is sufficient momentum for an independent health and safety collaboration, which could seek to establish best practice and guidance for research and to form a focus for prioritising research strategies is
95
unclear. The Cochrane system provides a validated pre-existing methodology for research reviews and meta-analyses but the primary research on which it is based must also be of high quality, possibly based on the features of randomised controlled trials.
To attempt to mirror, so far as possible, this best-practice guidance, the research format was that of primary research of a health and safety impact evaluation, the design of which was informed by a review of relevant literature. It was a quasi-experimental, longitudinal design, which included a comparison group. It included introduction of an intervention that was staggered over a three-month period and multiple measures of outcome were used. It was not feasible within existing project budgetary and time constraints to consider the use of a time series design by including a third data collection period (say a further twelve months after the second), or for reversal of the intervention.
However, as has already been discussed, if the intervention produced an effect, it was below the limit of detection of the method. This may have been related to low implementation or the time period that may have been too short for change to become established. These are regarded as two limitations of the research design.
4.3.4 Threats to external validity
The experimental design included a comparison group in that two of the seven participating Trusts did not receive the workbook intervention because they already had a health and safety management system supported by the use of a workbook. This meant that there were no ethical issues associated with not issuing a workbook to these two Trusts. This is a positive feature and strengthens the research design in that if, the intervention produced an effect it should have been already present in the comparison Trusts and that change would therefore occur only in the test Trusts. The main limitation to this was that the degree of use of the comparison Trusts workbooks was not established.
Another limitation was that the selection of the participating Trusts was non-random. This was because the intervention had not only to be of interest to the participating organisations but also for them to be committed to working within the research resources and framework. This limitation was a compromise that was necessary to ensure the continuing participation of the Trusts.
The initial selection of the individuals to receive a safety climate questionnaire was a stratified random selection. The same individuals completed two questionnaires, approximately twelve months apart. This is a strong feature of the research design.
4.3.5 Threats to internal validity
There was potential for the research results to be sensitive to a history threat. The research took place over a twenty-two month window. The NHS is characterised by constant changes in both management and organisational practices and, although formal major reorganisation did not occur during the research period, there were still extraneous influences. For example, two of the Trusts were involved in planning and implementing major construction projects. Also, major governmental initiatives, such as the Revitalising Health and Safety Strategy were launched and publicised. At the outset of the research these possibilities were recognised. The choice of length of the intervention phase was largely influenced by the need to minimise this risk, while still giving sufficient time for any effect to become established. The choice therefore represented a balance of these risks. The use of control Trusts also strengthened the strategy.
96
By the time of the second data collection period the participants had become familiar with the root causation methodology. This represented a potential instrumentation threat (i.e. the measurement method altered between the two project phases). An example to illustrate this is that the number of times a root cause was cited reduced in the second phase. This could be interpreted as an improvement in performance but was identified at an early stage as a consequence of increased familiarity with the methodology. This measure was therefore not used as an indicator of performance (the identification of the root cause itself was used instead). It is therefore unlikely that an instrumentation bias existed in the research results.
Extreme values were detected in the results. In particular, the medical and dental staff of Trust F was identified as having a significantly different opinion within the questionnaire responses. This represents a regression-to-the-mean threat. This was overcome by the analysis strategy, which avoided pooling data once it became clear that results from within individual Trusts were significantly different from each other. This meant that the extreme values were identified and classified as being from within a single Trust and present only in the first data collection phase. Had this strategy not been adopted it is likely that a wrong conclusion would have been reached that medical and dental staff across the whole of the NHS had significantly different views from other staff.
The benchmarking events themselves will have had an effect on the outcome. The inclusion of a comparison group should have ensured that testing and placebo threats were controlled. So far as the potential for an effect to occur because of the involvement of the research team (i.e. Hawthorne effect(143)) is concerned, the comparison group was treated as close as possible to the same as the intervention group, with the exception of the intervention and monitoring of workbook holders. This was also the reason why contact with the intervention group was kept to an absolute minimum during the intervention period. For example the telephone monitoring of workbook holders within the test Trusts was restricted to three times in twelve months (rather than the more traditional three-monthly grouping) and also only a single visual inspection of workbooks was carried out. Apart from these times contact was avoided with the workbook holders. It is therefore thought unlikely that these threats have exerted undue influence on the research outcomes.
The presence of a maturation threat could have influenced the results, particularly given the increasing emphasis within the NHS on risk management and their related insurance schemes (CNST(144)/CNORIS(17)). These schemes were being publicised at the time of the research and Trusts were being encouraged to develop measures to ensure effective management of risks. However, the presence of the control group should have minimised this risk.
Dropout threats were similarly controlled in that none of the participating Trusts left the study once the first benchmarking data was obtained. Within the random sample of questionnaire respondents all staff groups were represented equally in both data collection phases. However, although the initial selection of participants was random, and efforts were made to encourage completion of the questionnaire, replacing non-responders with others selected at random from within the relevant staff group, it is likely that a selection bias still exists within the data. This is because the questionnaires were likely to have been returned by individuals who were more motivated to do so. This is demonstrated by the increase in response rate between the first and second data collection phases. However, the mean response rate from the first phase was satisfactory (55%) and not regarded as a major threat to the results.
It is unlikely that a contamination (or diffusion) threat existed. This is because NHS Trusts tend to work autonomously from each other and, so far as the author is aware, were not in communication with or influenced by each other. For the same reason, it is unlikely that a
97
resentment threat existed between those Trusts in the control group and those implementing the intervention.
4.3.6 Outcome measurement
Outcome measurements were carried out to a rigorous protocol. The incident data comprised of a combination of measurement scales, which were analysed using descriptive methods, including bivariate analyses and cross-tabulations of rates. Ordinal data, such as system failure and management root causes, were analysed using non-parametric tests. Parametric tests such as correlation and T-testing were used for interval and ratio data such as types and costs of incidents.
However, a limitation of the incident data analysis was the small sample size of five test and two control Trusts. This limitation did not exist with the questionnaire response data. Parametric testing (such as ANOVA) was only used once it had been established that it was appropriate to do this with the ordinal data from the questionnaire responses. The analysis strategy was robust in that it avoided pooling of data from individual trusts once it was identified that it was inappropriate to do so. Multiple comparison testing (Scheffe Test) was used to identify where the differences lay. This strategy was robust and gave reliable results.
The workbook usage telephone interview monitoring data was useful in that it clearly established the low degree of implementation of the workbook. The rank order data was also useful for use in the triangulation exercise and were shown to correlate significantly with captured incident rate and the questionnaire dimension of Working Environment.
4.3.7 Other limitations
The potential for selection threats has been identified. This is because any effect from the intervention could be due to differences between the Trusts rather than the intervention itself. The inclusion of the two comparison Trusts was intended to control for these threats. However, in practice the influence of individual Trust units proved to be the overriding finding and differences between the comparison and test group could not be detected. It is suspected that this may have been due to the low use of the intervention workbook. An intervention that sought to control this effect would be useful although, because this would require more intervention from external researchers, this may lead to an increased risk of Hawthorne effects.
A number of issues were not captured by the method. These included cases of occupational ill health; hospital acquired infections; the considerable potential for impact by civil litigation.
4.4 ASSESSMENT OF THE DEGREE TO WHICH THE RESEARCH AIM AND OBJECTIVES HAVE BEEN ACHIEVED
In Chapter 1 it was identified that there was a need to establish best practice in terms of research methodology to aim to evaluate whether it is possible to reliably measure change in health and safety performance when a formal health and safety intervention is introduced to the healthcare sector. This was done by a combination of a review of the literature and primary research. Three research objectives were set, which are reviewed against the outcomes of the research below.
Objective 1: The design of a methodology for evaluation of health and safety performance. This was informed by the literature review described in Chapter 1, which allowed the selection of an appropriate research design. The objective was achieved with the selection of the incident analysis methodology that allowed numbers, costs and management system root causes (which
98
also was used as an assessment of the health and safety management system) to be measured plus the safety climate questionnaire that allowed the measurement of NHS staff opinion on health and safety; Objective 2: The design of a health and safety management instrument, which could be used as an intervention tool for the improvement of health and safety performance. This was informed by the literature review and achieved with the development, by the author, of the managers workbook; Objective 3: The design of a research methodology to measure change in health and safety performance. This objective was achieved with the implementation of an appropriately designed field study, which allowed the evaluation methodology to be used in a quasi-experimental before-and-after longitudinal design, which included a comparison group.
The research has added to the knowledge base by identifying a number of important concepts and issues that can help inform future research.
Despite the fact that minimal change in performance was detected, the research design was able to show significant correlations between three of the performance evaluation aspects (i.e. degree of workbook use, rate of captured incidents and staff opinion on the questionnaire dimension of working environment). The method was robust and possessed an ability to detect change in health and safety performance. It is highly likely that, if a change in health and safety performance had occurred, the method would have detected it, although the cause of the change would not be identified. It seems likely that the small improvement in climate that was detected was at the limit of detection of the method.
This is contrary to claims of dramatic changes in performance cited by others(86, 145) (and reviewed in Chapter 1) but supportive of claims by others who have also failed to detect impressive changes(80, 84, 146). This provides evidence that claims for improvements in health and safety performance should be treated with caution, unless the research design is robust.
Future research strategies for health and safety research programmes now have evidence that priorities for action must include measures to ensure the improvement in research quality. The research in this report concentrated on the healthcare sector but the findings are likely to be of relevance of findings to other industry sectors.
It seems likely that, given the sensitivity of the research method, a simple approach to health and safety management is unlikely to improve performance in the long term. Concerns have been raised earlier in this Chapter that, despite the effort of HSE over two decades, effective management of risk remains elusive. Their systems based approach would benefit from a more in-depth critical evaluation, such as how to positively affect safety behaviour – a challenging task(147).
It is nevertheless concluded that the three project research objectives, set at the start of the project have all been achieved, and that the research aim has been fulfilled.
4.5 SUMMARY OF CONCLUSIONS AND RECOMMENDATIONS
4.5.1 Health and Safety Management Performance Evaluation Methodology
Conclusions �� The conceptual basis for the research was robust in that it included both positivist and
phenomenological aspects. Triangulation across these aspects was achievable and showed statistically significant correlations;
99
�� Although captured incident rates offered potential, as a performance evaluation method, the definition of an incident should be made clearer across the NHS. This would facilitate the production of better benchmarking data;
�� Incident reporting in the NHS has been largely used as a performance indicator. The process of learning from adverse events is comparatively under-developed, especially for non-clinical incidents. The HSE root cause analysis method could reliably identify health and safety management system weaknesses and therefore offered strong potential to be used by organisation to learn lessons from their incident data;
�� Estimated costs ranged between 0.06% and 1.44% of the running costs of the NHS. These are low costs compared to other published data (between 20% and 50%). The costing protocol is therefore regarded as very conservative (cost for ill health, infection and litigation not reliably captured). However, it was rigorous and reproducible;
�� The use of the questionnaire yielded useful, robust data; �� The results of the triangulation added supportive evidence that efforts in health and safety
management are associated with better health and safety performance.
Recommendations �� A common standard for incident reporting and costing should be developed for the NHS,
supported by guidelines. This would ensure a clearer purpose for incident reporting and facilitate improved communication and collaboration between departments currently responsible for individual aspects of incident reporting;
�� Priority should be given to ensuring the effective management of the risk control system of risk assessment, in particular the areas of planning and implementing. If this were done it is likely that health and safety performance would improve;
�� The potential for many incidents in the NHS to be unpreventable (the special category incidents in this research) should be debated and investigated further. This should be done to establish if a clearer definition of reasonably practicable could be established for the NHS;
�� The HSE root cause analysis method should be further developed before it is suitable for use as a routine method;
�� Training providers and accreditation bodies should look at their course content for training of safety advisers to ensure that the skills of behavioural safety and change management are given a higher priority than at present;
�� Further behavioural-based research should be conducted to investigate whether a conservative costing estimate protocol is likely to convince senior managers and budget controllers to take action to control health and safety risks
4.5.2 Health and safety management intervention
Conclusions �� The workbook intervention failed to produce a detectable effect on health and safety
performance; �� The maximum effect detectable by the evaluation method was a 2% shift in reported
incident rate, 0.2 point shift in questionnaire dimension score, and 15% shift in users answers to telephone monitoring questions;
�� Use of the workbook decreased significantly over the 12 months implementation period,indicating low implementation.
�� It cannot be concluded that the workbook produced an effect that was too weak to be detected. Had the workbook been more fully used, there might have been a detectable effect;
�� The time between the two phases might have been too short for change to occur.
100
Recommendations�� Future research should ensure a standardised approach in individual Trust units to
encourage use of the intervention; �� The use of certification standards as a motivating factor in workbook usage should be
further explored; �� Further research into the impact of workbook-type interventions should include other
aspects of multi-factorial change models; �� Since it is likely that other organisations could benefit from the workbook it should be
refined and adapted for wider use, possibly via the Internet.
4.5.3 Health and safety management performance change measurement methodology
Conclusions The main limitations of the method were: �� The intervention did not produce an effect that could be detected (this may have been
caused by low implementation or too short an intervention period). The comparison Trusts did not receive follow-up monitoring, which might have assisted with interpretation of this finding;
�� Non random selection of participating Trusts; �� History threats that may have occurred over the project duration; �� Small sample size in the incident reporting and root cause analysis aspects of the method; �� Cases of ill health, infection, or litigation were not easily included.
The main strengths of the method were: �� Intervention objectives that were stated and defined; �� The use of more than one indicator of effectiveness; �� The use of a comparison group; �� Questionnaire sampling strategy which produced a robust sample, sensitive to statistical
testing and identified the need to avoid pooling data from individual Trusts; �� Effective use of triangulation.
Recommendations:
Further debate suggested exploring whether an independent health and safety collaboration could be established to identify examples of best practice and form a focus for prioritising research strategies.
101
APPENDIX 1 HSE ROOT CAUSE ANALYSIS METHODOLOGY
Figure 1 Process for screening and investigating incidentsIncident Occurs
Incident Report Form isCompleted and Submitted
Harm orPotential for
Harm?
Outwith Power ofTrust to Control? Discuss with RA
Greater than £5 or 15minutes of costed staff
time?
Exclude fromStudy
Allocate Investigation Pack
Events & Causal FactorsChart
Conduct Interviews
Complete Forms 1A & 1B No ApparentSystem Failures
Discuss with RA
Place In SpecialCategory File
Transfer System FailuresFrom Form 1A to Form 2A
Refer to Root CausingTables & Complete Form 3
Transfer Root Causes & Codes to Form2B & Enter Evidence
Complete Project Checklist
END POINT
END POINT
END POINT
NO
YES
NO
NO
YES
NO
YES
NONE FOUND
Process for Screening & Investigating Incidents
102
Incident investigation example
Agitated patient referred to medical admissions ward as a potential suicide risk
Two incident forms were submitted for this patient. These were regarded as the same incident. In both cases a “red alert” was activated. This meant staff attended from other areas of the hospital to assist in a restraint manoeuvre, therefore there was more than 15 minutes of costed staff time involved. Incident was screened – INCLUDED IN STUDY.
Incident form submitted at 18:15 �� 26 year old male. Was supposed to be lying on bed eating �� Located in side room of medical admissions unit �� Description of incident: “Threw meal tray onto floor. Staff nurse in room at time tried
to restrain patient. Became physically aggressive towards social worker. Trying to get out of room. Nursing staff assisted to restrain patient. Red alert put out. Self poisoning – suicidal intent. Major social problems. Patient nursed in single room.”
Incident form submitted at 19:30 �� Patient trying to leave side room �� Description of Incident: “Staff were waiting for psychiatrist to come from Psychiatric
Facility, as patient was to be sectioned. Patient burst from room with social worker trying to prevent him from leaving. Patient became physically aggressive – red alert put out. Self-poisoning – suicidal intent. Major social problems. Police had arrested girlfriend on ward. Patient had been and continued to be nursed in single room.”
What were the crucial questions?
Was there potential for harm?Although no one was actually hurt, there was potential for harm to both patient and staff.
Was there a cost to the Trust? Staff were called out from other areas of the hospital to attend the incident therefore this incident should be included due to the opportunity cost of staff who did not work in the ward where the incident took place.
Was the incident within the power of the Trust to control?It was not reasonable to expect an acute NHS Trust to have measures in place to avoid disturbed/aggressive patients. However elements of management failure may emerge in the way in which such incidents are dealt with. There should be systems in place to reduce the risk to staff/patients so that the risk is as low as is reasonably practicable.
An assessment of the control measures supported by appropriate risk control systems must be made. The following would be needed to do this:
1. What events led to the occurrence of the incidents? 2. What are the Trust-wide and local policies and procedures for dealing with violence
and aggression and detaining patients, regarded as a suicide risk? 3. What actions are necessary to legally detain a patient against their will?
103
Interviews conducted
�� A member of staff on the ward who was not involved in the incident regarding the procedures for emergency assistance
�� The staff nurse who reported the incident/restrained the patient prior to arrival of the “fast action response team”
�� The senior staff nurse on the ward �� A nurse manager who is involved in training of the response team �� The health & safety co-ordinator �� The consultant in charge of Accident & Emergency Dept �� The clinical director of medicine �� A violence & aggression adviser
Questions to Consider in Identifying Management System Root Causes
Policy Reference was made to local or Trust wide policies. The Health & Safety Policy may have been a good starting point, but many other systems may have had their own policies, e.g. risk assessment, manual handling, sharps handling etc.
Implementing Were there means to operate the system? (e.g. within the system of risk assessment, there may have been trained risk assessors, forms disseminated, specialist advisers etc.)
Had the system been implemented partially or not at all? What was the position regarding risk assessment (the risk may have been documented but appropriate control measures may have not been in place, or not utilised).
Control Was there a standard that defined responsibilities? This may have been laid out in the relevant policy, or within individual employment contracts. In most cases it will have been assigned through the line management chain.
The main thing that should have been considered was whether responsibility was adequately assigned not whether the individual or group concerned met it. Not meeting responsibilities should have been regarded as equivalent to not implementing the system. Awareness of responsibility could have been a factor.
Planning Was there a plan to implement the system in the organisation (resources, training, procedures etc.). Also was there local planning, which might have involved task management, supervision, communication etc.
CompetenceWhat standard of competency was required by the organisation? What was the skill knowledge and experience of the individual or group involved? Were they trained appropriately to implement this system? Was their training up to date?
104
Communication What was the standard for communication? This may have been written information in policy files, on Trust intranets, on staff notice-boards, verbal information via departmental meetings, induction training, visits by specialist advisers’ etc. When interviewing individuals, who might have breached the policy, try to ascertain whether they were aware of this? Was this a one-off slip up – people might have said, “I don’t usually work this way but…” or was the person adopting a method that was used by everyone in their workplace? This may have helped to assess the effectiveness of the communication methods employed.
Co-operationWhat were the consultation mechanisms that were in place? What ways were staff involved in the development and planning of the system? Were they given the opportunity to work together to develop systems that can work in their particular environment? This root cause should have been associated with team issues, not whether or not someone implemented a system.
Measuring How was the system measured? This may have been written into a policy or individuals employment contract. Consider how the effectiveness of the system was determined. Were measurements made at a regular frequency?
Reviewing This involved review of the system, not of individual risk assessment etc. Was there a formal process for review of the system? If so had the system been reviewed when it should have been and was the outcome from the review been acted upon.
Now we know WHAT HAPPENED we need to establish the ROOT CAUSES
(i.e. Identification of system failures & management root causes)
What systems were involved in handling an incident such as this? In this case the investigator chose three relevant systems to explore.
1. Training – Management of violence and aggression 2. Procedures – Detention for mental health assessment 3. Emergency procedures – Dealing with aggressive patients.
1. Training
What were the “active failures”?�� Improper restraint of the patient prior to the arrival of the red alert team. �� Inadequate training provided, particularly with regard to de-escalation techniques and
breakaway techniques.
What was the evidence? �� Only individuals who were on the response team rota were trained.�� Training took place over one day, with a half-day refresher on a yearly basis. This is
not enough time to learn both de-escalation and control and restraint techniques.�� Internal staff who had attended a course delivered training themselves, not by
experienced trainers.�� The training programme was targeted on the basis of strength, i.e. mostly male staff
trained, rather than areas where violent incidents occurred most frequently.
105
What were the management root causes? �� Policy was inadequate
2. Procedures
What were the “active failures”?�� The patient was detained (illegally) for four hours against his will.
What was the evidence? �� Three red alerts were called over an extended period. No record of detention of the
patient under the Mental Health Act during that period.
What were the management root causes? �� There was no Trust-wide or departmental policy in place regarding the procedure to
adopt for detention of patients requiring psychiatric assessment (based on the Mental Health Act).
3. Emergency procedures
What were the “active failures”? �� Improper restraint of the patient prior to the arrival of the red alert team
What was the evidence? �� Staff nurse involved stated in interview that himself, the social worker and the doctor
were wrestling with the patient on the ground attempting to detain him on the floor. Only the staff nurse had received any training on control and restraint techniques. Therefore this group manoeuvre put the patient/staff at risk.
What were the management root causes? �� Implementation - Lack of implementation of management of violence & aggression
policy – Two individuals restraining the patient were untrained. �� Planning - Inadequate resources allocated to the development, implementation and
maintenance of emergency procedures. �� Competency - Standard of competency was inadequate for two reasons
- Did not specify evaluation of training content - Standard did not specify the provision of competent advice e.g. from a
psychiatrically trained nurse – staff had very few psychiatric skills to draw from. �� Communication – Not all relevant staff groups were informed of the system. �� Co-operation – Not all relevant staff groups were consulted and few were involved in
the planning, measuring and reviewing of the system.
The information obtained from the investigation has been inserted into the project forms below. The codes obtained in Form 3, the “System failure form” should be entered into the project database for subsequent analysis.
106
Accident Ref: Leave blank Date: 28/09/00 Time 1815
INITIAL FOLLOW UP
1. Follow up incident form with person who reported it �
2. Does the incident cost more than £5 or take up more than 15 minutes of costed staff time?
No end Yes ��
3. Was there harm or potential for harm? �
______________________________________________________________
INVESTIGATION
4. Allocate investigation pack and mark date/time of incident on forms �
5. Obtain details about costs and incident. Identify failed systems. Forms 1A and 1B
�
6. Draw up Events and Causal factors chart to identify further issues �
7. Transfer details to Evidence table Form 2A �
8. Examine root causes Form 3 �
9. Complete Evidence Table Form 2B �
10. Verify that no costs are outstanding �
COMPLETED �
Date Actions complete / Outstanding / Notes Initials 27/11/00 All incident investigation/root cause analysis complete. Logging of
incident outstanding.
Figure 2 HSE Project Checklist
107
Figure 3 Event and causal factors chart
108
Department where incident occurred: Medical assessment Main Department* where management failure arose: Medical assessment
ACDC Dep’t Code*: 64
Accident Ref: Leave blank Date: 28/09/00 Time: 1815 Describe the sequence of events and consequences (What, When, Where, Who, Which, How, Why?)
Interviewed
26 year old male patient admitted as ?overdose to medical assessment ward, accompanied by social worker. Psychiatrist called from local primary care hospital to assess patient but takes four hours to arrive. During this time the patient attempts to leave the ward and becomes aggressive upon being held there against his will. Three V&A alerts were called and the patient was restrained.
Incident Reporter�Injured Party Line Manager �Senior Nurse�Nursing Manager�Director (Med) �Doctor �Witness �Others (please state)Health & Safety Co-ordinator A & E Consultant V & A Adviser
Identify the physical controls: Were they used? Were they effective? Elimination Substitution Enclosure Time 72 hr detention SpaceContainment design Insulation Guards Interlocks Other: (Please state)
LEV Safety valves Two handed control PPE Access/Egress Workstation layout Man/machine interface Work environment Equipment design
No, a registered medical practitioner could have used this to legally detain patient
Not applicable
Identify organisational controls: Were they used? Were they effective? Responsibilities Risk assessment Training �(1)Supervision procedures�(2)DetentionWork organisation Emergency procedure�(3)Work environment Maintenance Disposal Management of violence & aggression Control of contractors Procurement
Security Handling sharps Recruitment Installation Commission Specification Inspections Decommission Change management Design (Humanfactor) Other (please state)
(1) YES (2) NO (3) YES (V&A alert was activated)
(1)NO(2)Not applicable (3)Partially, due to increased numbers of staff present to deal with the situation
Identify who was present (1) social worker (2) nursing staff (3) doctor (4) police
Were they supposed to be there? (1) No (2,3,4) Yes
Identify who was absent Psychiatrist
Where were they? Not established
Figure 4 Data capture form – Form 1A
109
Figure 5 Accident/incident costs – Form 1B
110
Acc
iden
t Ref
: Le
ave
blan
k Su
mm
ary:
Vio
lent
inci
dent
invo
lvin
g su
icid
al
patie
nt a
wai
ting
psyc
hiat
rist
Dat
e:
28/0
9/00
Tim
e: 1
815
FOR
M 2
A –
leve
l 2
FOR
M 2
B –
leve
l 1
SYST
EM
Act
ive
failu
re
Evid
ence
M
anag
emen
t sys
tem
fa
ilure
(roo
t cau
se)
Roo
tca
use
code
Evid
ence
TRA
ININ
G
1. P
atie
nt w
as im
prop
erly
rest
rain
ed
befo
re th
e re
spon
se te
am a
rriv
ed.
2. In
adeq
uate
trai
ning
pro
vide
d,
parti
cula
rly w
ith re
gard
to d
e-es
cala
tion
tech
niqu
es a
nd b
reak
away
, but
als
o fo
r co
ntro
l & re
stra
int t
echn
ique
s.
1. T
he p
atie
nt w
as
wre
stle
d to
the
floor
by
a st
aff n
urse
, do
ctor
and
soci
al
wor
ker.
Onl
y th
e st
aff n
urse
had
any
tra
inin
g in
safe
re
stra
int t
echn
ique
. 2.
Mos
t sta
ff
mem
bers
on
the
war
d ha
ve h
ad n
o de
-esc
alat
ion
train
ing.
Tho
se th
at
are
train
ed to
re
stra
in h
ave
rece
ived
min
imal
train
ing.
POLI
CY
1.
2 1.
Tra
inin
g on
ly fo
r 1 d
ay in
itial
ly
+ ha
lf da
y re
fres
her y
early
. 2.
Tra
inin
g is
del
iver
ed b
y in
tern
al
staf
f who
hav
e at
tend
ed a
cou
rse,
no
t by
expe
rienc
ed tr
aine
rs.
3. T
rain
ing
prog
ram
me
was
not
ta
rget
ed to
thos
e ar
eas w
here
vi
olen
t inc
iden
ts o
ccur
mos
t fr
eque
ntly
.
111
Acc
iden
t Ref
: Le
ave
blan
k Su
mm
ary:
Vio
lent
inci
dent
invo
lvin
g su
icid
al
patie
nt a
wai
ting
psyc
hiat
rist
Dat
e:
28/0
9/00
Tim
e: 1
815
FOR
M 2
A –
leve
l 2
FOR
M 2
B –
leve
l 1
SYST
EM
Act
ive
failu
re
Evid
ence
M
anag
emen
t sys
tem
fa
ilure
(roo
t cau
se)
Roo
tca
use
code
Evid
ence
PRO
CE
DU
RE
S (T
O L
EG
ALL
Y
DET
AIN
PA
TIEN
TS
AG
AIN
ST T
HEI
R
WIL
L)
Patie
nt w
as d
etai
ned
for 4
hou
rs u
ntil
a ps
ychi
atris
t cou
ld a
rriv
e an
d m
ake
an
asse
ssm
ent.
As i
t was
the
patie
nt’s
will
to
leav
e th
e ho
spita
l the
det
entio
n w
as
illeg
al a
nd a
dded
to th
e pa
tient
’s st
ate
of
agita
tion.
Had
he
been
lega
lly d
etai
ned
chem
ical
rest
rain
t wou
ld h
ave
been
an
optio
n th
at m
ay h
ave
redu
ced
the
risk
to
staf
f. A
llow
ing
the
patie
nt to
dis
char
ge
him
self
wou
ld a
lso
have
redu
ced
the
risk
to st
aff.
1. O
n in
terv
iew
the
staf
f nur
se p
rese
nt
was
una
war
e of
the
cond
ition
s of
dete
ntio
n.
2. N
o re
cord
exi
sts
of d
eten
tion
of th
e pa
tient
und
er th
e m
enta
l hea
lth a
ct.
3. 3
V&
A a
lerts
w
ere
calle
d ov
er a
n ex
tend
ed p
erio
d.
POLI
CY
1.
1 N
o Tr
ust-w
ide
or d
epar
tmen
tal
polic
y in
pla
ce (c
omm
unic
ated
in
writ
ing
or v
erba
lly) r
egar
ding
the
proc
edur
es to
ado
pt fo
r det
entio
n of
pat
ient
s req
uirin
g ps
ychi
atric
as
sess
men
t (ba
sed
on th
e m
enta
l he
alth
act
). C
onfir
med
by
the
clin
ical
dire
ctor
of m
edic
ine.
EMER
GE
NC
Y
PRO
CE
DU
RE
Im
prop
er re
stra
int o
f the
pat
ient
prio
r to
the
arriv
al o
f the
resp
onse
team
. Th
e pa
tient
was
w
rest
led
to th
e flo
or
by a
staf
f nur
se,
doct
or a
nd so
cial
w
orke
r. O
nly
the
staf
f nur
se h
ad a
ny
train
ing
in sa
fe
rest
rain
t tec
hniq
ue.
IMPL
EMEN
TIN
G
7.4
Res
train
t of t
he p
atie
nt sh
ould
be
carr
ied
out,
only
by
mem
bers
of
the
resp
onse
team
trai
ned
to
rest
rain
safe
ly.
PL
AN
NIN
G
6.2
The
reso
urce
s allo
cate
d to
the
deve
lopm
ent,
impl
emen
tatio
n an
d m
aint
enan
ce o
f the
em
erge
ncy
proc
edur
e fo
r vio
lent
inci
dent
s w
ere
inad
equa
te.
112
Acc
iden
t Ref
: Le
ave
blan
k Su
mm
ary:
Vio
lent
inci
dent
invo
lvin
g su
icid
al
patie
nt a
wai
ting
psyc
hiat
rist
Dat
e:
28/0
9/00
Tim
e: 1
815
FOR
M 2
A –
leve
l 2
FOR
M 2
B –
leve
l 1
SYST
EM
Act
ive
failu
re
Evid
ence
M
anag
emen
t sys
tem
fa
ilure
(roo
t cau
se)
Roo
tca
use
code
Evid
ence
C
OM
PETE
NC
E 3.
2 1.
Sta
ndar
ds o
f com
pete
ncy
did
not s
peci
fy e
valu
atio
n of
trai
ning
co
nten
t. 2.
Sta
ndar
ds o
f com
pete
ncy
did
not s
peci
fy th
e pr
ovis
ion
of
com
pete
nt a
dvic
e –
staf
f hav
e ve
ry
few
psy
chia
tric
skill
s to
draw
fr
om.
C
OM
MU
NIC
AT
ION
4.
4 N
ot a
ll re
leva
nt st
aff g
roup
s are
in
form
ed o
f the
syst
em.
C
O-O
PER
AT
ION
5.
2 N
ot a
ll re
leva
nt st
aff g
roup
s wer
e co
nsul
ted
and
few
wer
e in
volv
ed
in th
e pl
anni
ng, m
easu
ring
and
revi
ewin
g of
the
syst
em.
Figu
re 6
Evi
denc
e –
Form
2A
and
2B
114
Instructions: (1) Enter a cross for NO (2) Where there is no policy in place STOP root causing completely.
(3) Once a cross is entered within a series of four move onto the next set. (4) Upon completion transfer all root causes & codes into evidence form
(Form 2B) and complete the evidence column.
Accident Ref: LEAVE BLANK Date: 28/09/00 Time: 1815 SYSTEM (from Form 2a)
Training Proced-ures
Emer. Proc.
1.1 Was a policy in place regarding the system?
X
1.2 Was the policy adequate? X IMPLEMENTING 6.3 Was a standard for the system
specified?
6.4 Was the standard for the system adequate?
7.3 Was the system implemented? 7.4 Was the system implemented to
the required standard? X
CONTROL 2.1 Was a standard specified defining
responsibilities?
2.2 Was the definition of responsibilities adequate?
2.3 Was responsibility for the system assigned?
2.4 Were responsibilities assigned according to the standard?
PLANNING 6.1 Was there a plan to implement the
system?
6.2 Was the plan adequate? X 7.1 Was the plan put into operation? 7.2 Was the plan put into operation
adequately?
COMPETENCE 3.1 Were standards of competency
specified?
3.2 Were the standards for competency adequate?
X
3.3 Was the person/persons competent?
3.4 Was competency assessed against the standard?
COMMUNICATION 4.1 Was a standard for communication
of the system specified?
4.2 Was the standard for communication adequate?
4.3 Was the system communicated? 4.4 Was the communication in line
with the standard? X
115
CO-OPERATION 5.1 Was a standard for co-operation
specified?
5.2 Was the standard for ensuring co-operation adequate?
X
5.3 Was there co-operation? 5.4 Was the co-operation in line with
the standard?
MEASURING 2.5 Was a standard for measuring the
effectiveness specified?
2.6 Was the standard for measuring effectiveness adequate?
8.1 Was the system measured for effectiveness?
8.2 Were these measures in line with the standard?
REVIEWING 2.7 Was a standard for reviewing the
system specified?
2.8 Was the standard for reviewing the system adequate?
9.1 Was the system reviewed? 9.2 Were these reviews in line with the
standard?
Figure 7 System failure - Form 3
116
APPENDIX 2 COST INCLUSIONS AND EXCLUSIONS
Non staff costs
Include
1. Cost of disposable materials used to treat staff or patient injuries that occur as a result of an incident.
1.1 NOTE: In the case of needlestick injuries this will depend on whether the Trust provides its own services or has a service level agreement (SLA) with an Occupational Health Service (OHS) run by another Trust. It may also depend on the location of the member of staff and the time that the injury occurs, as treatment may be delivered at the A&E department instead of the OHS Department.
2. Cost of disposable materials used to make the area safe again.
2.1 EXAMPLE: To mop up a spillage.
3. Cost of an extended patient stay that is either the direct or indirect result of an incident.
4. Cost of replacing damaged equipment.
5. Cost of drugs that are administered as a result of the incident.
6. Cost of raw materials used to repair damage to equipment or property.
6.1 EXAMPLE: Glass and other materials used to repair a broken window.
Exclude
7. Cost of equipment that is in-situ and is re-useable.
8. Cost of equipment that has been damaged but will not be replaced.
9. Costs that are associated with improvement/upgrading during replacement.
10. Patient facilities that are unavailable due to damage i.e. short-term reduction in the patient capacity of a ward.
11. Vaccination and disposable materials costs when they are covered by a service level agreement.
12. Cost of materials that are purchased to prevent further harm (as part of risk control measure) following the incident.
12.1. EXAMPLE: Hip protector pads purchased following patient fall.
Staff costs
Include
13. Time spent receiving or administering first aid to a member of staff, visitor, contractor involved in an incident.
14. Time spent during a shift visiting A&E or OHS.
15. Sick leave.
117
16. Time spent performing duties, not normally part of job, as a result of an incident.
16.1 EXAMPLE: A needle goes missing during eye surgery, theatre staff & porters etc. spend time searching for the needle.
16.2 EXAMPLE: A needlestick injury is sustained from a needle sticking out of a general waste bag. The bag is passed onto a member of nursing staff & they have to carefully dispose of the needle & search through the contents of the bag to ensure there are no more needles.
17. Time spent performing regular duties outwith normal hours.
17.1. EXAMPLE: Engineer called out to reset a fire alarm system.
18. Time spent outwith designated ward areas managing violence & aggression.
19. Time spent outwith designated ward areas searching for absconded patients.
20. Time spent outwith designated ward areas escorting a patient involved in an incident.
21. Replacement labour costs if additional staff are brought in to cover absence due to an incident.
22. Contractor’s fees for services required following an incident such as repairs to equipment etc.
Exclude
23. Time spent by doctors & nurses administering first aid to a patient on their ward. The situation becomes slightly more complex if the patient is injured outwith their ward area.
23.1. EXAMPLE: If a radiographer provides first aid to a patient their time should be costed.
24. Staff time to care for patients, where the provision of care forms part of their everyday role.
24.1. EXAMPLE: If a patient requires an x-ray, do not cost for the radiographers’ time.
25. Time spent being interviewed & completing forms that are part of normal reporting & investigation procedures following an incident. Thus, incident reporting, police interviews & internal investigations involving staff associated with the incident or staff with an investigative role in the Trust are excluded from the costing process.
26. Additional staff required following the incident as a result of a review of the risk assessment following the incident.
26.1. EXAMPLE: Increased staff numbers to allow for constant observations of a patient who is considered high risk following an incident.
118
APPENDIX 3 COVERING LETTERS, QUESTIONNAIRE AND QUALITY CONTROL SYSTEM FOR INPUTTING
QUESTIONNAIRE DATA
Dear Colleague,
Safety Questionnaire
Tayside Occupational Health and Safety Service is undertaking a research project in collaboration with the NHS Management Executive and the Health and Safety Executive, aimed at improving the management of health and safety in the NHS and reducing the costs of accidents at work. As part of this endeavour it is our intention to obtain some indication of employee perception of health and safety by conducting a confidential survey of 10% of staff working at all levels in the NHS.
You have been randomly selected to participate in the survey and we would be very grateful for your co-operation. This entails completion of the enclosed questionnaire and a further identical questionnaire which will be issued to you in 12 to 14 months. It is important that you complete and return the tear-off slip at the bottom of this letter if you do not wish to participate or are unlikely to be available to complete the second questionnaire. All responses will be treated in strict confidence, however you are required to put your name on the questionnaire so that I can monitor changes in your perception of health and safety.
The questionnaire is straightforward to complete and asks about your attitudes to safety issues as well as any suggestions you might have to improve things. Please try to answer all of the questions. The findings will be fed back to you on completion of the study.
Should you have any queries relating to the nature of the study or the questionnaire please do not hesitate to contact me. I appreciate that it is often difficult to find time to respond to the requests of researchers but I would urge you to take this rare opportunity to express a personal opinion on an important issue which affects everyone in the NHS. I will contact you again if I do not hear from you in 10 days.
Many thanks for your assistance.
Yours sincerely,
----------------------------------------------------------------------------------------------------------
I do not wish to participate in the study.
Name ____________________________________________________________ Trust ____________________________________________________________ Department____________________________________________________________ Job Function____________________________________________________________
Figure 1 Covering letter (Phase 2)
119
Dear Colleague,
Safety Questionnaire
You may recall being contacted just over a year ago by Tayside Occupational Health & Safety Service to ask for your help in an externally funded study aimed at improving the management of health and safety in the NHS and reducing the costs of accidents at work. Since the occupational health services that cover Tayside and Fife have now merged, the name of our organisation has changed to the Occupational Health & Safety Advisory Service (OHSAS). However, the remit of the research team is unaltered and I am writing to thank you for returning a completed questionnaire in the first round and to ask for your cooperation again so that we can determine whether the overall opinion of staff regarding health and safety issues has changed over the last year.
As before, the survey is strictly confidential and will be processed by a member of the research team in OHSAS. We are issuing the questionnaire to 1547 people across 7 NHS Trusts and in some cases the reply envelope may bear an internal address. This is purely to reduce postage costs, the collated envelopes will be collected unopened by a member of the research team and only the overall opinion of the sampled group will be fed back to the Trust.
10% of staff working at all levels of the Trust were randomly selected to take part in this survey which is aimed at assessing the effectiveness of a new health and safety management system that has been introduced in 5 of the 7 NHS Trusts participating. Only those that responded in the first round of the survey have been contacted again as we feel it is important to compare the opinions, over time, of the same group of people. This means that your cooperation is even more valuable to us at this crucial stage of the project and we would be very grateful if you would complete and return the enclosed questionnaire.
The questionnaire is identical to the last one you completed. It asks about your attitudes to health and safety issues and the importance assigned to them by the Trust. Please try to answer all of the questions. The findings of the survey will be fed back to you once we have collated the responses from this second round and have established whether staff opinion of the management of health and safety has changed for the better, worse or stayed the same over the period of the study.
The results from the whole programme of research, of which this is the only opinion based aspect, will be fed back to line managers, health and safety committee members and Trust executives. I would therefore urge you to take this rare opportunity to express a personal opinion on an important issue which affects everyone in the NHS.
Should you have any queries relating to the nature of the study or the questionnaire please do not hesitate to contact me. I will be in touch again if I do not hear from you in 14 days.
Many thanks for your assistance.
Yours sincerely,
Figure 2 Covering letter (Phase 4)
120
SAFETY SURVEY
We would like to find out how you feel about safety practices and principles in the NHS and in order to do this we would like you to complete this questionnaire.
It is important for you to be completely honest about your feelings. All responses will be treated in strict confidence and will be processed by the Tayside Occupational Health & Safety Service.
It should take 15 to 20 minutes to complete this questionnaire.
We would like you to enter your name (so that we can link two questionnaires administered at different time points in the study), NHS Trust, Department and Job function to assist us with the interpretation of the results.
Thank you for your co-operation.
Name ___________________________________________________
Trust ___________________________________________________
Department ___________________________________________________
Job Function ___________________________________________________
You will be presented with a series of statements on the following pages about health and safety. You should indicate your response by ticking the appropriate box.
For example, if you agreed with the following statement you would tick under the ‘I agree’ category, thus:
Strongly agree
Agree Neither agree nor disagree
Disagree Strongly disagree
1. Health & safety issues are very important �
121
Please tick the appropriate box to indicate your level of agreement
Strongly agree
Agree Neither agree nor disagree
Disagree Strongly disagree
1. Management operates an open door policy on safety issues
2. Safety is the number one priority in my mind when completing a job
3. Co-workers often give tips to each other on how to work safely
4. Safety rules and procedures are carefully followed
5. Management clearly considers the safety of employees of great importance
6. I am sure it is only a matter of time before I am involved in an accident
7. Sometimes I am not given enough time to get the job done safely
8. I am involved in informing management of important safety issues
9. Management acts decisively when a safety concern is raised
10. There is good communication here about safety issues which affect me
11. I understand the safety rules for my job
12. It is important to me that there is a continuing emphasis on safety
13. I am involved with safety issues at work
14. This is a safer place to work than other organisations* I have worked for
(*“organisations” changed from “companies”)
15. I am strongly encouraged to report unsafe conditions
16. In my workplace management turn a blind eye to safety issues
17. Some safety rules and procedures do not need to be followed to get the job done safely
123
Please tick the appropriate box to indicate your level of agreement
Strongly agree
Agree Neither agree nor disagree
Disagree Strongly disagree
35. Sometimes it is necessary to depart from safety requirements in order to meet the demands of my job*
(* “to meet the demands of my job” changed from “for productions sake”.)
36. A safe place to work has a lot of personal meaning to me
37. There are always enough people available to get the job done safely
38. In my workplace managers/ supervisors show interest in my safety
39. I am never involved in the ongoing review of safety
40. Management considers safety to be equally as important as getting the job done
41. A no-blame approach is used to persuade people acting unsafely that their behaviour is inappropriate
42. Managers and supervisors express concern if safety procedures are not adhered to
43. I cannot always get the equipment I need to do the job safely
Do you have any other comments about health and safety in your workplace? ________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
Thank you for completing this questionnaire.
Figure 3 Safety survey
124
Quality control system for inputting questionnaire data
1. Input a “batch” of questionnaires into the database – a batch should be done in one sitting so there may be a number of batches input in one day.
2. Make a note of the day, date, batch/sample number, number of questionnaires entered, the time data inputting commenced and the time data inputting finished in the data inputting table.
3. Randomly sample five questionnaires from the pile of questionnaires entered – ensure there are some questionnaires sampled from the latterly input half of the pile.
4. Check inputting on the database, count and correct any errors found for each questionnaire (only check errors for multiple choice statements).
5. Plot the total number of errors found in the five questionnaires sampled on the control chart.
6. Monitor the control chart plots checking for:
�� Non-random patterns �� Eight consecutive plots above the mean line �� Two/three plots between the upper warning limit and the upper control limit �� Any plot above the upper control limit
7. In the event of the above, return to the batches concerned and re-check the entire batch, correct the entries and then conduct the quality control exercise again, re-plotting the “corrected” error rate.
125
APPENDIX 4 MANAGEMENT SYSTEM WORKBOOK
Introduction
The Trust aims to achieve a safe and healthy environment for staff, patients and their relatives and visitors.
Clinical leaders/managers have a vital role to play in achieving this because health and safety is NOT an “add-on” to your job. It is a fact that in departments and companies where health and safety is positively managed performance, efficiency, morale, and cost-effectiveness all improve.
A safe and healthy workplace can only be achieved if health and safety is INTEGRATED with the work that we do. Therefore health and safety must be MANAGED. This is where you come in.
Clinical leaders/managers are responsible for managing the health and safety of their staff and the premises in which they work. This is a legal requirement of the Health and Safety at Work Act 1974 and the Management of Health and Safety at Work Regulations 1999.
This workbook has been designed to help you manage health and safety. You are encouraged to use it fully.
On the next page is a FORWARD PLANNER, which you can use to record summary information of your actions to ensure health and safety is managed adequately in your area of responsibility. Use it to plan your year but REMEMBER it is a summary sheet and there are other actions that you will need to take which are contained in each section of this workbook.
A WORKED EXAMPLE of the planner is also included to help you.
REMEMBER: HEALTH AND SAFETY MUST BE MANAGED. YOU WILL FIND THIS WORKBOOK FORMS A CONTINUOUS LOOP AND THAT YOU WILL END BACK AT THE START AGAIN, EACH TIME REFINING AND IMPROVING YOUR SYSTEM. IF YOU ARE NOT SURE ASK FOR HELP (see “Where to find more help & information” at the end of the Workbook)
YOU DO NOT HAVE TO USE EVERY SECTION IN THE WORKBOOK IF YOU ALREADY HAVE SOMETHING THAT PERFORMS THE SAME FUNCTION. IF YOU DO - MAKE A NOTE IN THE RELEVANT SECTION AS TO WHERE THE INFORMATION CAN BE OBTAINED
122
Please tick the appropriate box to indicate your level of agreement
Strongly agree
Agree Neither agree nor disagree
Disagree Strongly disagree
18. I am rarely worried about being injured on the job
19. Management acts only after accidents have occurred
20. I believe that safety issues are not assigned a high priority
21. Some health and safety rules and procedures are not really practical
22. Employees are not encouraged to raise safety concerns
23. Personally I feel that safety issues are not the most important aspect of my job
24. In my workplace the chances of being involved in an accident are quite large
25. I do not receive praise for working safely
26. Corrective action is always taken when management is told about unsafe practices
27. Operational targets often conflict with safety measures
28. My line manager/supervisor does not always inform me of current concerns and issues
29. I can influence health and safety performance here
30. Sometimes conditions here hinder my ability to work safely
31. Safety information is always brought to my attention by my line manager/supervisor
32. When people ignore safety procedures here, I feel it is none of my business
33. In my workplace management acts quickly to correct safety problems.
34. I am clear about what my responsibilities are for health and safety
126
Departmental Health & Safety Management System: FORWARD PLANNER (WORKED EXAMPLE)
Financial year1999-2000
Policy Indiv-idual roles & respons-ibilities
Incident report forms
Training records
Safety meet-ings
Risk assess-ments
Action plan
Safety inspec-tions
Audit Review
April Review �
Produce plan �
General �
�
May Assign �
Manual handling �
�
June � COSHH �
�
July � DSE � � Aug Other � � Sept Review
�
� Produce �
Submit to TEG
�
Oct � � � Nov � Dec � � Jan � � �
Feb � �
Mar Review� � �
WHERE TO FIND MORE HELP & INFORMATION The first person to approach if you have any queries is your clinical leader/manager. However, the following individuals are also responsible for health and safety within the Trust:
Name Position Area of responsibility �or Extension
127
POLICY
Introduction
Policy documents should have three main sections:
STATEMENT OF INTENT - which outlines briefly an organisation or departments position on a certain subject (like health & safety). It should also give a broad description of the strategy to be adopted which will put the statement into practice (e.g. a series of objectives). ORGANISATION - which sets out the roles and responsibilities of the policy. ARRANGEMENTS - which sets out how the policy aims and objectives will be achieved.
(N.B. “PROCEDURES” OR “RISK CONTROL MEASURES”, on the other hand, are detailed descriptions of standardised or understood ways of doing tasks. You probably have many in use in your department).
This section is split into 2 parts:
National/Area/Trust policies (e.g. Trust A Area Infection Control Policy, Trust Health and Safety Policy etc.)
Departmental policies
You should now:
Insert a sheet indicating where all relevant documents are located if they are stored elsewhere
OR
file copies of the documents in these parts.
128
There is a blank index sheet at the start of each part for you to record the document titles (if you don’t already have one).
National/Area/Trust/Policy index
Title of policy Date of policy Comments
Departmental policy index
Title of policy Date of policy Comments
ORGANISING
Introduction
To make your health and safety management system really effective you need to get organised
This is often referred to as a “HEALTH AND SAFETY CULTURE”.
There are 4 aspects to achieving this:
1. Control (this is often maintained by ensuring that all staff know and understand their roles and responsibilities) 2. Co-operation (this comes from consulting staff and involving them in planning, reviewing performance, writing procedures & problem solving) 3. Communication (this involves discussing health and safety regularly and providing information about hazards, risks and preventative measures) 4. Competence (this means ensuring adequate information, instruction, training and supervision are available)
Each of the 4 aspects of organising listed above contributes to a positive health and safety culture.
You should record the roles and responsibilities of individuals within the department in this part of the workbook.
129
Control
To help you understand how the Trust assigns roles & responsibilities, a diagram of the TRUST HEALTH AND SAFETY MANAGEMENT SYSTEM is included overleaf. This identifies how health and safety is managed and supported at Trust level. A description of the roles and responsibilities of the individuals identified in the diagram is given below:
Chief Executive: has overall responsibility for health & safety within the Trust.
Clinical Director/General Manager: is responsible for the health & safety of all staff working within their “patch”.
Managers: are responsible for the health & safety of their staff. Should draw up and implement an action plan, if necessary, following risk assessment. Responsible to the Clinical Director / General Manager. Staff: all staff have a responsibility for their own health & safety. Director of Human Resources: the executive director responsible for co-ordination of health & safety management. Responsible to the Trust Board.
Health & Safety Co-ordinators: act on behalf of their Clinical Director/General Manager to ensure that managers fulfil their legal requirements and to act as a communication channel and local source of advice.
Risk Assessors: responsible for conducting risk assessments as required by their manager (N.B. Not for implementing the recommendations of assessments).
Safety “Reps”: represent a group of staff on health & safety issues. Specialist Advisers: act as a source of advice and guidance on health & safety (N.B. Not for managing health & safety). A blank sheet is provided in this section to allow you to list your specialist advisers.
To help you,
There is a sheet for recording the roles and responsibilities of yourself and your staff.
You should make sure that this sheet is kept up to date and that any gaps are entered into the Record of training.
130
KEY: Line Management
Health & Safety
Support for line management
Specialist advisors
131
List of Trust specialist advisers
Position Name Contact number Safety Officer Occupational Hygienist Infection Control Radiation Protection Adviser Manual Handling Adviser Occupational Health Dept Fire Adviser
List of individual roles & responsibilities
Role Name(s) Date Trained
Date re-training required
General Risk Assessor COSHH Assessor Manual Handling Risk Assessor Display Screen Equipment Assessor Other Risk Assessor Infection Control Link Person Manual Handling Key Worker Safety Inspectors Staff Safety Rep First Aider
Remember:
This list is not exhaustive. You may have specialised roles & responsibilities that are not mentioned. Include them in the spaces provided
You may not need all the types of role described. For example, you might have a general risk assessor but do not need a COSHH assessor. Just use the headings you need
132
Co-operation
Participation, commitment and involvement in health and safety activities at all levels in your department is essential because: You have a LEGAL DUTY to consult your staff Pooling knowledge and experience is a KEY ASPECT OF RISK CONTROL This can be done in a number of different ways such as: Holding regular departmental meetings to:
�� Plan �� Establish departmental policies �� Produce a departmental action plan �� Measure, audit and review performance �� Routinely revise policies, procedures and the action plan �� Introduce improvement initiatives �� Promote health & safety awareness �� Raise awareness of health and safety within the department �� Be the focal point of efforts to prevent accidents and manage risks �� Promote a reduction in environmental emissions and the efficient use of paper,
energy and water �� Ensure action is taken on health and safety matters �� Communicate and manage information �� Co-ordinate the responsibilities of management and staff for health and safety to
improve the health, safety and welfare of staff at work �� Identify training needs
Involving staff in working groups (to develop policies, procedures etc).
Encouraging the nomination of and providing support for staff safety representatives.
Operating suggestion schemes.
Record of consultation with staff.
Measure taken to achieve co-operation Dates of meetings Comments/actions necessary
You should now:
Begin by thinking about how you can improve co-operation within your area of responsibility. Consider the suggestions given above. You DO NOT have to introduce ALL of the suggestions above. They are given as examples. Make a plan for implementation of the measures that you decide upon. Record your actions on the sheet provided.
133
Communication
Effective communication is essential. It may help to think of communication in terms of FLOW both into and out of a department or organisation:
Record of health & safety information
Document title Location
Record of departmental safety meetings
Date of meeting Action taken
You should now record your communication measures. This part of the workbook contains 2 dividers for you to record:
Health and safety information (i.e. your “library” of information)
Departmental safety meetings (i.e. minutes etc.)
134
Competence
Arrangements need to be made to ensure the competence of all employees (including managers) if they are to make the maximum contribution to health and safety.
These should include:
�� Identifying health and safety training needs arising from recruitment, changes in staff, plant, processes or working practices.
�� Maintaining or enhancing competence by refresher training �� Managing contractors �� Ensuring relevant information, instruction, training and supervision are provided �� Making arrangements to ensure cover for staff absences �� Re-organising staff roles in accordance with changes in competence which arise as a
result of external factors such as a recent injury, illness or bereavement
Record of training needs
Name Area of responsibility Training need Date completed
Date re-training due
PLANNING & IMPLEMENTATION
Introduction
Planning, and then ensuring that these plans are implemented, is the key to ensuring that your health & safety efforts really work.
You must make sure you have WRITTEN assessments of risk for all the procedures in your area of responsibility that have SIGNIFICANT health and safety risks associated with them.
A summary sheet for your training records is provided below.
This is a bit like a loop leading to an update of the procedure after a risk assessment.
ProcedureRisk assessment
Planning involves: setting objectives identifying hazards assessing risks implementing standards of performance in your operational procedures.
135
Begin by making a list of all the TASKS your staff undertake. If you have operational procedures (or equivalent) these will be ideal. Do this with your risk assessor(s).
Identify, with your risk assessor(s), which tasks do not have SIGNIFICANT health and safety risks (If you need help with this, a decision matrix can be found at the end of this section).
Remove the tasks that do not need action from your list (if in doubt -include the task), RETAIN these and review them annually or sooner if the task changes in some way.
Now make a new prioritised list of tasks with significant health & safety risks (the decision matrix will also help with this).
Ask your risk assessor(s) to conduct written assessments for the remaining tasks.
Start with GENERAL RISK ASSESSMENTS for compliance with the Management of Health and Safety at Work Regulations, 1999.
If necessary, arrange for a more detailed risk assessment to be carried out (e.g. COSHH, Manual Handling, DSE.
136
Your risk assessments form the foundation of your RISK CONTROL SYSTEM
When preparing your action plan you MUST take account of the hierarchy of risk control principles:
1. Eliminate risks by substituting the dangerous with the less dangerous (e.g. using a less hazardous substance or substituting a better guarded machine)
2. Combat risks at source by engineering controls and giving collective protective measures priority. (e.g. enclose the process, provide guards, suppress or contain air-borne hazards, remote/automatic operation)
3. Minimise risk by designing operational procedures so that they are safer (e.g. reducing exposure processes by using competent or specialist contractors)
4. Use personal protective equipment (remember this is a LAST resort)
Review the assessments once they are complete. Make a written ACTION PLAN based on the conclusions. The plan may consist of a number of different types of action:
If the solution to the problem is obvious, and it is achievable within your range of authority, IMPLEMENT THE SOLUTION; If the solution is obvious but you cannot implement the measure(s) because of constraints, you must IMPLEMENT SHORT-TERM CONTROL MEASURES IMMEDIATELY (to protect staff in the meantime) and then PREPARE A BUSINESS CASE for the relevant member of the Corporate Team (a model business case layout is included) If the solution to the problem is not obvious, identify possible solutions by carrying out an OPTION APPRAISAL (a model for this is included). If the preferred solution is achievable, implement it. If it is not then IMPLEMENT SHORT-TERM CONTROL MEASURES immediately and prepare a BUSINESS CASE for the relevant member of the Corporate Team. Remember to contact your specialist advisers if you need more helpMake sure that you implement your action plan - It’s no good making a lot of good plans and then doing nothing about them.
Remember to record your findings. Blank copies of the forms to use are in each of the labelled sections. You should store OR record where you have stored your completed risk assessments and action plans in these sections too.
Assign a date on your planner when the assessments are due for review. It is your job to instruct your risk assessor to repeat or re-assess risk assessments.
Check that progress is being made with your action plan from time to time. It is easy to let things slip!
137
The process is summarised in the following flow diagram:
138
HAZARD�
RISK�
Low Medium High Very high
Highly likely Action required Urgent action Urgent action Suspend work
Likely Action required Action soon Action soon Urgent action
Possible Action required Action required Action soon Action soon
Not likely No action Action required Action required Action required
Action planning
You should prepare an action plan if any of your risk assessments show action to control risk is necessary. The flow diagram in the previous Section shows that there are 3 levels of action plan. The solution is obvious and achievable. You should implement it. The solution is obvious but you are unable to implement it for some reason (can’t authorise the expenditure, for example). In this case you should ensure that the task can be done safely by some means (which may not be ideal but you must make the task safe). You should then prepare a business case and submit it to your head of department. The solution is not obvious. In this case you should make the task safe in the short-term and carry out an option appraisal before preparing your business case. You may also need to seek help with this from a specialist adviser.
A blank sheet for recording your action plans, model business case and option appraisal are all included in this section.
Record of action plans
Task Action needed Person responsible for action Target date
Use this matrix to help you decide which of your tasks have significant health & safety risks (i.e. those where action is necessary). Remove those where no action is indicated (remember to review them annually or if they change)Make sure detailed, written risk assessments are carried out for the remainder. Prioritise these, tackling the most urgent needs first.
REMEMBER, TACKLE THE MOST SIGNIFICANT RISKS FIRST
139
MODEL BUSINESS CASE
Provide the following information and submit to the Head of Department or equivalent
Signed………………………………………………Date………………… Designation…………………………………………Department…………
1. Description of support requested (attach quotations etc. if appropriate)
2. Evidence for significant health and safety risk (summarise outcome of risk assessment)
3. Are other options available? (If so include an option appraisal with this document
4. Potential consequences should support not be available (attempt a realistic assessment)
140
Operational procedures
Completed risk assessments (see the following Sections) should be used by you to develop changes in your operational procedures.
Remember, it’s like a LOOP:
If you wish you can store your written operational procedures in this workbook. Alternatively, you can store a list of them with their locations using the sheet provided overleaf.
141
List of operational procedures
Title of procedure Date Location
General risk assessment
Written general risk assessments are needed to comply with the legal requirements of the Management of Health and Safety at Work Regulations 1999.
A risk assessment has four main parts:
A blank risk assessment form is included in this part. You should also store your risk assessments here or indicate below where they are stored.
Remember to seek guidance from your specialist advisers (you can refer to the list at the beginning of the workbook) if you need help.
Identify the significant hazards within the task (i.e. aspects of the task with the potential to cause harm such as slipping hazards, electrical hazards, manual handling hazards etc.).
Evaluate the risk (i.e. the probability that things will go wrong. For example, how likely it is that someone will trip in the circumstances of the task? You can use previous experience to help you here. For example has anyone already tripped? Make sure you consider who could be affected and what control measures already exist as part of this process. Also ask yourself before you decide. Is this a foreseeable risk?)
Identify measures needed to control any risks that are unacceptable. (This may need to be a package of measures. It may also include asking for another, more specific risk assessment to be done before you can decide. Examples of this are COSHH, Manual handling, and DSE assessments as well as asking your specialist adviser for an opinion).
Review the risk assessment. (If the risks are controlled to your satisfaction you only need review in a year to check nothing has changed. If there are actions that are needed you should review more frequently to make sure progress is being made.)
142
Control of Substances Hazardous to Health Regulations 1999 (COSHH) Risk Assessments
COSHH assessments concentrate on the use and production of hazardous substances. This means in practice looking for the potential within the task for exposure and harm to occur by:
�� Inhalation; �� Skin contact; �� Skin absorption; �� Injection via a sharps injury; �� Eye contact; �� Ingestion.
Grounds for concluding that exposure is not a significant risk to health are:
Quantities or rate of use/production of the substance(s) are too small to constitute any risk to health under foreseeable circumstances of use, even if control measures break down;
OR
There is no available route of exposure.
If you are not sure if exposure is significant or not, contact your special adviser (refer to the list at the beginning of the workbook), who will be able to give you an opinion or arrange for measurements to be taken to help you decide.
A blank risk assessment form is included in this part. You should store your risk assessments here or indicate below where they are stored
The Manual Handling Operations Regulation 1992 Risk Assessment
The numerical guidelines on the next page provide an initial filter which can help to identify those manual handling operations deserving more detailed examination. The guidelines set out an approximate boundary within which operations are unlikely to create a risk of injury sufficient to warrant more detailed assessment. This should enable assessment work to be concentrated where it is most needed.
There is no threshold below which manual handling operations may be regarded as "safe". Even operations lying within the boundary mapped out by the guidelines should be avoided or made less demanding wherever it is reasonably practicable to do so.
There is a wide range of individual physical capability, even among those fit and healthy enough to be at work. For the working population the higher guideline figures will give reasonable protection to about 95% of men and between one half and two thirds of women. The lower guideline figures will give reasonable protection to about 95% of women.
It is important to understand that the guideline figures are not limits. They may be exceeded where a more detailed assessment shows that it is appropriate to do so, having regard always to the employer's duty to avoid or reduce risk of injury where this is reasonably practicable. However, even for fit, well-trained individuals working under favourable conditions ANY MANUAL HANDLING OPERATION SHOULD NOT EXCEED TWO TIMES THE GUIDELINE FIGURES.
143
Basic guideline figures (kilograms) for manual handling operations involving lifting and lowering are set out overleaf. They assume that the load is readily grasped with both hands and that the operation takes place in reasonable working conditions with the handler in a stable body position.
The guideline figures take into consideration the vertical and horizontal position of the hands as they move the load during the handling operation, as well as the height and reach of the individual handler. It will be apparent that the capability to lift or lower is reduced significantly if, for example, the load is held at arm's length or the hands pass above shoulder height.
CARRYING
Guideline figures for carrying are the same as for lifting and lowering (overleaf), provided: the hands are not below knuckle height the load is held against the body the load is carried no further than 10 metres without resting.
When the load can be carried securely on the shoulder without first having to be lifted (e.g. unloading sacks from a lorry) a more detailed assessment may show that it is acceptable to exceed the guideline figure.
PUSHING AND PULLING
Start / stop: 25 kg force (men); 17 kg force (women)
Steady motion: 10 kg force (men); 7 kg force (women)
These figures assume that the force is applied between knuckle and shoulder height.
HANDLING WHILE SEATED
The figures in the diagram assume: No handling outside the box No twisting
144
Twisting Frequency
0o
<30/hour 0o
<2/min 0o
<12/min 0o
>12/min
Twisting Frequency
45o
<30/hour
45o
<2/min 45o
<12/min 45o
>12/min
Twisting Frequency
90o
<30/hour
90o
<2/min
90<12/min
90o
>12/min
(men)(women
(men)(women)
(men)(wome
145
WEIGHT IN KG A blank risk assessment form is included next. You can store your risk assessments here or indicate below where they are stored
The Health & Safety (Display Screen Equipment) Regulations 1992 Risk Assessment
You are responsible for carrying out workstation assessments for those staff who are display screen equipment (DSE) “users”. A “user” is someone who uses DSE as a significant part of their normal work. Therefore any person who works with DSE more or less continuously on most days should be regarded as a user. If you are not sure whether someone is a “user “ or not the following guide may help.
A DSE “user” will: Normally use DSE for continuous spells of an hour or more at a time; andUse it in this way more or less daily; andHave to transfer information quickly to or from the screen; and also:Need to apply high levels of attention or concentration; orAre highly dependent on DSE or have little choice about using them; orNeed special training or skills to use the equipment.
The DSE assessment has two forms that should be completed for each workstation:
A workstation assessment. This should be used by someone who has attended DSE risk assessor training; A self-assessment checklist. Each person identified as a DSE “user” should complete this. It will probably be best to ensure that any training in setting up a workstation is given to “users” BEFORE they complete this checklist.
A checklist for “users” to follow when setting up their workstation is included on the next page. This can be used as part of departmental training.
For further advice on training contact your specialist Health and Safety Adviser
As a manger of a DSE “user” you are responsible for:
The workstation layout and assessment; The daily work routine; General health & safety training on the use of DSE; Provision of information. This should include entitlement to a free eye test (check with personnel for detailed arrangements for accessing this service).
146
Display Screen Equipment - Checklist for setting up a workstation
Whenever you sit down at a workstation you should go through these 10 steps before you start work. 1. Sit back well in the seat, adjust the angle (and height if appropriate) of the back rest so
your back is well supported. 2. Make sure that the small of your back is well supported. 3. Adjust the height of your chair so that your forearms are approximately horizontal when
you place your hands to use the keyboard (see position 7 in the diagram. 4. Check that you don’t have too much pressure on the underside of your thighs and backs
of your knees or that your feet are not dangling. If you answer YES to any of these carry out 5. If NO go to 6.
5. Support the feet with a foot rest of suitable height. 6. Make sure that you have enough space under the desk with no obstacles which may
restrict your posture. 7. Check again that your forearms are approximately horizontal when using the keyboard
and then, 8. Check that your wrists are in a natural position. 9. Now look at the screen and make sure its height and angle allow you to hold your head
comfortably. You may need a document holder if you do a lot of copy typing. 10. Check the screen for glare. (You may need to close blinds, dim lights, reposition the
screen to do this). 11. Check the characters on the screen are bright enough (but not too bright or they will
become “fuzzy”). 12. Check that you have enough space in front of the keyboard to support your hands/wrists
during times when you are not using the keyboard.
Remember to take breaks away from the screen.
147
MEASURING PERFORMANCE
Introduction
Just like finance or quality of service, you need to measure your health and safety performance to find out if you are being successful.
There are two sorts of ways to measure your performance:
ACTIVE methods i.e. BEFORE things go wrong REACTIVE methods i.e. AFTER things go wrong
It is suggested that you adopt 3 different methods of measuring your performance:
Safety inspections (an “active” method) Safety tours (another “active” method) Accident/incident data (a “reactive” method)
Space has been made in this workbook for you to record relevant information for all three parts.
Safety inspections
These should be carried out regularly (e.g. monthly). A blank safety inspection form is included so you can make a start. A safety inspection is different to a risk assessment and usually concentrates on hazard spotting which produces a list of actions. A safety inspection typically identifies a lot of “maintenance” and “housekeeping” issues which must be managed if accidents are to be avoided. Allocate the duty of carrying out an inspection to a member of staff (it does not HAVE to be a trained risk assessor). When the inspection is carried out make sure you take action on those problems that have been identified.
148
Health & safety inspection checklist
LOCATION: ______________DATE OF INSPECTION:___________ INSPECTED BY:___________________________________________
HAZARD PRESENT �
HAZARD IDENTIFIED BEFORE �
DETAILS / COMMENTS / ACTIONREQUIRED
Relating to both workplaces and activities. Identify only hazards which you could reasonably expect to result in significant harm under the conditions in your workplace, otherwise the list becomes excessive. Use the following examples as a guide. 1. Slipping/tripping hazards 2. Infective agents (e.g. sharps) 3. People falling (e.g. working at height) 4. Falling materials 5. Confined spaces 6. Fire (e.g. flammable materials, exits blocked) 7. Moving parts of machinery (e.g. blades) 8. Ejection of material (e.g. material grindstones) 9. Pressure systems (e.g. boilers, autoclaves) 10.Vehicles (e.g. trolleys, fork lift trucks, wheelchairs) 11. Electricity (e.g. poor wiring) 12. Manual handling 13. Display screen equipment 14. Noise 15. Vibration 16. Poor lighting 17. Radiation (e.g. x-ray, UV, microwave, laser) 18. Inhalation of dusts, fumes, mists, gases, vapours, infective agents 19. Skin contact with chemicals 20. Assault 21. Working with animals 22. Poor housekeeping (e.g. untidy and dirty) 23. Extremes of temperature 24. Inadequate ventilation (e.g. no control, draughty) 25. Overcrowding 26. Poor conditions of floors and traffic routes 27. Inadequate or dirty toilets and rest areas 28.29.30.31.32.
149
Safety tours
These are your opportunity to “walk the job”. As managers we often become isolated from our staff and one way of improving this is to take an informal walk through your “patch” on a regular basis. The benefits are considerable but not always obvious (e.g. improved morale, becoming more “in touch” etc.). Don’t arrive with a notebook and announce your arrival. Be informal and talk to your staff. You will be amazed at what you find. Don’t be put off if staff find it strange at first. Begin by assuring them that you want to listen to their opinion and take it from there. Make sure that you book a regular time in your diary to do a tour and DON’T cancel it when something “more important” comes up - it should have a HIGH PRIORITY.
Accident/incident data
This should be reported to the Trust on the official incident report form. The Trust may issue reports about trends in accidents and incidents. Keep an eye on what is happening in your area and investigate any occurrences, which highlight a problem.
REMEMBER: FOR EVERY ACCIDENT THAT OCCURS ABOUT 30 “NEAR MISSES” WILL ALSO HAPPEN. YOU CAN GET VALUABLE INFORMATION IF YOU ENCOURAGE STAFF TO REPORT THESE TOO.
Record of accident/incident reporting (Incident Report Form & RIDDOR)
Date of incident
Name of reporter
Report form completed
Retained incident record pad reference
Action taken
Reviewing performance
Introduction
This section is about LEARNING FROM YOUR EXPERIENCES and consists of two parts:
Audit Review
You will find a master copy of a departmental audit in this part. Use it every year to evaluate how you are doing and where gaps still exist. You can either conduct the audit yourself or get another member of staff to do it.
Take the findings of the audit and use them to make an ACTION PLAN for the coming year. You can also incorporate this plan into a revision of your DEPARTMENTAL HEALTH AND SAFETY POLICY.
150
This brings us back to the beginning again.
NOW GO TO THE BEGINNING OF THE WORKBOOK AND BEGIN AGAIN BY MAKING (OR REVIEWING) YOUR PLAN!
MANAGERS HEALTH AND SAFETY AUDIT
Note: A= Essential B= Good Practice
Policy
There is a written health and safety policy which has a statement of intent, organisation and arrangement sections. It should be signed and dated by the person in charge. (A)
Yes No (please circle answer)
Comments
Written departmental health and safety policies and procedures exist and are implemented consistent with the Trust health and safety policy. (A)
Yes No (please circle answer)
Comments
Health and safety policies are subject to continuous review. (A)
Yes No (please circle answer) Comments
151
Organisational development
There is an individual at senior management level who has overall responsibility for formulating, developing and implementing health and safety policy. (A)
Yes No (please circle answer)
Comments
Health and safety responsibilities of managers are clearly defined within their job descriptions. (A)
(Note: these managers have the necessary authority and competence to carry out their duties effectively and are held accountable for their actions, health and safety objectives are set and reviewed annually as part of the performance review process)
Yes No (please circle answer)
Comments
Arrangements are in place for obtaining competent safety advice. (A)
Yes No (please circle answer)
Comments
There is a multidisciplinary safety committee (or committees). (A)
(Note: meets regularly includes senior management, staff and staff representation and is consulted on the development, implementation and monitoring of health and safety policy, is actively involved in the setting and monitoring of performance standards for health and safety.)
Yes No (please circle answer)
Comments
The committee reports to the Corporate Team regularly. (B)
Yes No (please circle answer)
152
Comments
A Trust annual health and safety report is produced. (A)
(Note: presented to the executive management team, made available to all staff)
Yes No (please circle answer)
Comments
Local arrangements for consultation with staff are in place and staff are provided with the training necessary to make an informed contribution to health and safety issues. (A)
Yes No (please circle answer)
Comments
First aid arrangements are in place and are in accordance with the Health and Safety (First Aid) Regulations 1997. (A)
Yes No (please circle answer)
Comments
Mechanisms are in place to promote awareness of the Trust and Departmental health and safety policies and health and safety issues (e.g. notice boards, newsletters, etc). (B)
Yes No (please circle answer)
Comments
There is a written departmental safety education programme. (A)
(Note: includes orientation of new employees to safety practices, is reviewed at least annually to determine its effectiveness)
Yes No (please circle answer)
Comments
153
All local orientation and induction programmes include an introduction to the overall health and safety policy and any necessary health and safety instruction. (A)
Yes No (please circle answer)
Comments
Arrangements are in place for identifying and providing on-going health and safety instruction and training (for example, when changes in staff or working practices occur). All instruction and training is recorded. (A)
Yes No (please circle answer)
Comments
Temporary workers on fixed or short-term contracts are provided with information concerning health and safety issues which may be encountered in their work. (A)
Yes No (please circle answer)
Comments
Planning and implementation
There is an up-to-local date plan, which identifies health and safety objectives, targets and timescales and is developed in consultation with staff. (A)
Yes No (please circle answer)
Comments
Departmental hazards are identified and written risk assessments have been carried out in accordance with the Management of Health and Safety at Work Regulations 1999 or other relevant health and safety legislation. (A)
154
Such as: �� Control of Substances Hazardous to Health Regulations 1999 �� Electricity at Work Regulations 1989 �� Health and Safety (Display Screen Equipment) Regulations 1992 �� Manual Handling Operations Regulations 1992 �� Noise at Work Regulations 1989 �� Personal Protective Equipment at Work Regulations 1992 �� Provision and Use of Work Equipment Regulations 1998 �� Lifting Operations and Lifting Equipment Regulations 1998 �� Workplace (Health, Safety and Welfare) Regulations 1992
Yes No (please circle answer)
Comments
Where necessary preventive and protective measures (control measures) are implemented. (A)
Yes No (please circle answer)
Comments
Departmental risk assessments are reviewed and updated on a systematic basis or when circumstances change. (A)
Yes No (please circle answer)
Comments
Measuring performance
Regular departmental safety inspections are carried out in hazardous areas. (A)
Yes No (please circle answer)
Comments
There is a clear reporting procedure in place recording, investigating, reporting and taking action on accidents, incidents, hazards and defects. (A)
Yes No (please circle answer)
155
Comments
The Reporting of Injuries, Diseases and Dangerous Occurrences Regulations 1995 (RIDDOR) are complied with. (A)
Yes No (please circle answer)
Comments
There is a system in place for disseminating safety action notices and hazard notices. (A)
Yes No (please circle answer)
Comments
Is health and safety on the agenda of departmental meetings so that health and safety objectives and effectiveness of arrangements are evaluated annually and modified as required. (B)
Yes No (please circle answer)
Comments
Audit and review
Departmental audit and review systems are established, operated and maintained. (A)
Yes No (please circle answer)
Comments
(Note: These are designed to assess the following elements of the health and safety management system:
�� policy �� organisation �� planning and policy implementation �� measuring systems �� reviewing systems)
156
Review of management system and setting new objectives
Those questions that you answered “NO” in the Managers health and safety audit represent the areas of your departmental health and safety management system that you need to develop. They therefore represent your objectives for the future.
Write your new objectives (with timescales and how you plan to achieve them) below.
Quarterly reports
Each quarter you should prepare a report on the progress you have made developing, implementing and refining your health and safety management system.
The purpose of reports of this kind is not for you to let everyone know how well things are going but rather to state what you planned to do in the relevant period, what you achieved and, in the light of this, what you plan to do next. If you are having problems you should say so. At least this can then be used as an indication to yourself that you need advice and guidance.
It is suggested that you use the “5-steps” format that this workbook follows to structure your report. A blank report is included. (Don’t feel obliged to use it if it doesn’t fit with your style of reporting though).
REMEMBER: NEVER BE AFRAID TO ASK FOR HELP - START WITH YOUR OWN MANAGER, HEALTH & SAFETY CO-ORDINATOR OR SPECIALIST ADVISER.
(Worked Example)
157
QUARTERLY REPORT FOR PERIOD April 2000 TO June 2000
Section of health & safety management system
Policy Organising Planning & implementing
Measuring performance
Audit/ review
Planned activities Distribute new policy
Roles & responsibilities clarified
Undertake general, COSHH & manual handling risk assessments
Review incident forms
None till January
Achievements Done Mostly complete Most done Highlighted trend of staff tripping. Modified floor covering. All OK now
N/A
Outstanding items
None First-aider still to be trained
Need Assistance from Occ.Health with COSHH Assessment for dust exposure
None None
Future plans Arrange seminar with staff to ensure all aware of new policy
Arrange first aid training course with occupational health
Appointment for COSHH survey made for July
Review in July Audit in January
Comments May not get training till August
All assessments will be complete after COSHH survey
Safety inspections done monthly
Any other items
158
APPENDIX 5 PROTOCOL FOR BRIEFING OF WORKBOOK HOLDERS
Aim of the Workbook Briefing: To introduce the workbook to managers, explain its purpose and ensure that manager’s understand how they can use the workbook to maximum effect.
1. Introductions –Project Coordinator, Project Research Assistant (both in attendance at all workbook briefings).
2. Background of the research project – The philosophy behind it, testing the theory, “Is good management of health & safety in the workplace good business?” Who is funding the work and why they are interested in looking at these issues. The various aspects of the project, i.e. data collection, the issue of questionnaires, operating the workbook for a period of a year and trying to detect change. Also why the research fits in so well with the internal health & safety objectives of the participating Trusts and the need for the NHS to improve its health & safety performance and implement appropriate management systems that are user-friendly to the managers who have to use them.
3. Legal requirement to manage health & safety in the workplace – Workbook holders are referred to the Health & Safety at Work Act (1974) and the Management of Health & Safety at Work Regulations (1992/1999) and how these pieces of legislation define the role of managers. Emphasis is placed on the obligation to manage health & safety through the line management chain and the need for acceptance of the responsibility to do this as opposed to the delegation of that responsibility to other individuals in the department who perform a health & safety role. Workbook holders are referred to the HSE guidance on how managers achieve compliance with the law, “Successful Health & Safety Management” (HSG65).
4. Health & Safety Management System – The five steps of the health & safety system are outlined and explained briefly.
�� Policy – A distinction is made between Trust-wide and local policy. �� Organising – Introduction of the “four Cs” of a positive health & safety culture
and a brief explanation of what is meant by each of these. �� Planning & implementing – This is largely tied up with risk assessment, the
importance of conducting risk assessments that feed in directly to the procedures and tasks that are being conducted in the workplace is explained and the concept that these should form a continuous loop. In order for the risk assessments to do any good they must be actioned – therefore an action plan is essential.
�� Measuring performance – Divided into reactive and active measures. Reactive being learning from experience, e.g. trends identified in the incident reports. Also the importance of “near misses” is mentioned and workbook holders are referred to the accident pyramid to demonstrate this.
�� Audit & review – It is made clear that this is not the same as measuring performance. It is an assessment of the management system itself and should feed back into the start of the management system so that it operates as a continuous loop.
5. Going through the workbook page by page – The workbook itself becomes the focus of the rest of the briefing. Although the workbooks have all been tailored to suit each of the participating Trusts the framework remains the same. Any particular sections that relate
159
specifically to the individual Trust e.g. their own business case application rather than the one featured in the general issue are referred to in passing and questions relating to the processes involved forwarded to the Trust representative who is present at the briefing. Copies of the workbook disk are given to the Trust contact who is responsible for setting up a system for providing workbook holders with copies of blank tables etc as required.
(i) Log Page
This is usually completed at the beginning of the briefing. The need for the project team to track the workbooks is explained and the importance of comprehensive coverage of the Trust by the workbook system. Thus workbook holders are asked to inform the Trust contact if their role in the Trust changes or they leave the Trust, so that the workbook can be passed onto someone else who can cover their “patch”. Attention is drawn to the contact details of the Project Research Assistant, for the purposes of answering workbook-related questions.
(ii) Contents, Introduction & Further information
�� Contents Pages – These are bypassed. �� Introduction – Attention is focussed on the bold print at the bottom of the
introduction page which asks workbook holders not to “reinvent” their working methods to adapt to the workbook but to use the workbook so it fits in with what they have in place already. It is anticipated that the adoption of this approach will encourage people to use the workbook and there is nothing to be gained by asking managers to invest more time and energy than is required. The point is made that it is possible to do “too much” in attempting to manage health & safety and managers must prioritise and be sensible in their assessment of the gains to be made from investing effort.
�� Forward planner – It is recommended that managers incorporate their health & safety management system into their forward planning efforts, either by using the table provided, adapting the format if necessary, or by incorporating it into their existing organiser or wall planner etc. However, they are asked to make a note on the page referring to where the information is kept.
�� Where to find more help & information – The last part of this section is highlighted as a source of information that may prove useful to managers. (It usually includes the main individuals responsible for health & safety in the Trust, addresses for organisations like the HSE or Occupational Health and sometimes information on the training courses available in the Trust).
(iii) Policy
�� Workbook holders are given the option to list their main policies in the two tables provided and are asked to make a reference on these pages to where copies of the policies can be found.
(iv) Organising
�� Control – The structure of the line management system in each particular Trust is examined briefly by reference to the list of Trust roles and the flow diagram. The flow diagram is used to reinforce the message of the legal obligation of managers to take responsibility but to use the network of health & safety support that is available. Reference is made to the list of specialist advisers that is provided in this
160
section. Managers are asked to consider the individual roles and responsibilities within their departments and enter them in the table provided, adding other roles to the list if necessary.
�� Co-operation – A number of suggestions to achieve co-operation are listed, attention is drawn to these. Workbook holders are asked to either enter a record of consultation with staff into the table provided or refer to the appropriate documentation e.g. the minutes of a departmental health & safety meeting, stating where these can be found. The need to consult staff on pro-active health & safety issues, like policy development for example, is highlighted rather than limiting consultation to the discussion of solving hazards identified through the occurrence of incidents.
�� Communication – It is recommended that some reference be made to the “stock” of health & safety advisory documents held by the manager, either by listing these in the table and/or stating where they are kept. The importance of documenting actions taken following departmental safety meetings is emphasised, either a note of the main points in the workbook or reference to the minutes of the meetings are required.
�� Competence – The point is made that managers should be identifying the training needs within their department, including their own! Feedback regarding training needs is more effective than sending staff on whatever courses are available. The various factors to consider to ensure staff are competent are listed by bulletpoint and are discussed very briefly. Managers are encouraged to keep some record of staff training needs that they can refer to in order to ensure their staff are competent in their roles and responsibilities (particularly as the centralised recording of training requirements rarely includes the issue of re-training reminders to individuals).
(v) Planning & Implementing
�� Risk assessment/Procedures loop – The need to assess risk in the context of the tasks that are performed within the department is emphasised again. This should be a dynamic process whereby the risk assessments performed should be reviewed on a regular basis, or perhaps following some change in the method, equipment or individual associated with the task.
�� Working through the processes of a “risk control system” (shaded text box) - �� A method for controlling risk is outlined, it is based on the assumption that
managers have done little or nothing to tackle this issue prior to the briefing. It is recommended that workbook holders make a list of the tasks performed in their department. They are then required to assess the tasks (together with their risk assessors and the people who conduct the task if necessary) to determine whether they present a “significant” health & safety risk. At this point reference is made to the decision matrix (in some workbooks replaced by the Trust’s own version) which allows assessors to combine the factors of the severity of the hazard with the likelihood that an incident will occur. It is recommended that tasks that do not represent a significant risk are removed from the list and set aside for annual review. The risk assessor is instructed by the manager to conduct written risk assessments for the remaining tasks and the outcomes of these risk assessments are discussed with the manager on completion. It is stressed that at this point the responsibility is placed back on the manager to decide on suitable action to take to eliminate or reduce the level of risk.
161
�� Three possible outcomes following risk assessment are discussed, as well as the role that a business case and option appraisal (both included in the workbook) can play when managers find themselves constrained or unsure of the best action to take. An important point that is made is the need to prioritise assessment and action plans according to the degree of risk. This will allow managers to make the best use of the time they have available to concentrate on health & safety issues.
�� To assist with these processes tables are provided to record operational procedures and action plans, although workbook holders can refer elsewhere e.g. procedures manual, minutes of health & safety committee, as long as the important facts are documented.
�� Model business case/Option appraisal – The model business case is referred to very briefly as the form has a simplistic design. Workbook holders are encouraged to consider completing this form as a means of taking positive action under circumstances that are beyond their level of authority to action. At least five minutes is spent taking workbook holders through the worked example of the option appraisal (based on a manual handling of patients scenario). The use of this case to demonstrate the method is stressed, as one of the options involves increasing staffing levels and this approach is less up-to-date than the purchase of additional hoists to mechanise lifting. The difference between this “option appraisal” and traditional NHS option appraisal is emphasised and the inclusion of the option appraisal in the workbook in order to field test this new method with what is expected to be a minority of workbook holders. Attention is drawn to the weaknesses of the method – the subjective aspect of the scoring system, and for this reason anyone interested in using it is advised to tackle the analysis as a group. The basis of the scoring method is outlined – low score is good, reversal of the scale for costs and benefits aspects and the need for ranking of the three aspects of the analysis in order to combine the scores in a meaningful way. Finally the importance of making a written statement to support the findings of the analysis and the potential “next stage” of the process, being a full financial appraisal of the favoured options are discussed.
�� Risk assessment regulations/Form inclusion – Workbook holders are referred to the guidance available on the regulations which govern general risk assessment and the more specific risk assessment types such as COSHH and DSE. Copies of the Trust’s own forms are included in these sections and therefore the role of the manager in completion of the forms is again made clear. They should know what the forms look like and understand their content, but they are expected to delegate the task of completing the forms to a trained risk assessor.
(vi) Measuring Performance
�� Active/reactive measurement – These are defined. �� Safety inspections – Many departments have something in place to ensure regular
formal safety inspections. The benefits of conducting relatively informal safety inspections to pick up maintenance of housekeeping issues that may be going unnoticed in the department is highlighted. Workbook holders are referred to the hazard spotting checklist included. The distinction between inspection and task-based risk assessment is made clear.
�� Safety tours – Managers are encouraged to improve communication regarding health & safety issues by conducting informal tours of their department.
�� Accident/incident data – A brief discussion of the systems in place to record incident reports submitted takes place. Depending on the numbers of incidents received and the systems in place, workbook holders are made aware that a table is provided in the
162
workbook for recording the types of incidents that occur within the department so that any local trends might become apparent.
(vii) Reviewing performance
�� Manager’s audit – It is recommended that Managers conduct the audit that is included in the workbook as an entry point to the “loop system” that the workbook operates. This should help them to identify their health & safety objectives. They are asked to document these on a space that is left blank in the workbook and this should lead them to the completion of the forward planner at the beginning of the workbook.
�� Quarterly reporting – Workbook holders’ attention is drawn to the quarterly report format that is provided at the end of the workbook. It is recommended that they complete the reports using brief notes that state clearly their current position and any future plans. They are instructed to channel the reports through their risk management department/the project Trust contact rather than forwarding directly to the project team so that the process can be adopted by the Trusts upon completion of the research.
163
APPENDIX 6 PROTOCOL FOR FOLLOW-UP OF WORKBOOK HOLDERS
Feedback obtained via telephone interviews
1. All workbook holders within each test Trust will be contacted for feedback regarding their use of the workbook in accordance with the telephone interview protocol attached.
2. Each workbook holder will be contacted on three separate occasions and asked the same questions. This should occur within the following designated “windows”.
�� Between 1st April 2001 and 31st July 2001. �� Between 1st August 2001 and 30th November 2001. �� Between 1st December 2001 and 31st March 2002.
3. At first contact the workbook holder will be asked if it is convenient to answer questions regarding the workbook, if not the researcher will make an appointment to call back at an agreed time. The researcher will make it clear that the questions should only take around 10 minutes to answer.
4. Actions to be taken during questioning:
5. Q1. If no, end of questions but try to establish why. 6. Q2. If no end of questions but try to establish why. 7. Q3. List the sections used, try to establish if the workbook holder has completed
the manager’s audit, used the workbook generally or used particular sections of the workbook.
8. Q4. Ask workbook holder to justify their answer. 9. Q5. Record any risk control measures implemented, regardless of cost. Try to
ascertain the cost to the Trust. 10. Q6. If yes obtain a copy of the option appraisal form. 11. Q7. If yes note suggestions. 12. Some of those individuals who are using their workbooks will be selected at
random for examination of their workbooks.
164
1st April 2001 – 31st July 2001
ISSUE NUMBER 1: TRUST A PRIMARY CARE NHS TRUST
Contact details
Workbook code
Number______Name________________________________________
Location_______________________________________________________________Question 1
Are you still the holder of the workbook? YES NO
If not WHY?Question 2
Have you used the workbook? YES NO
If not WHY?
1.1.1 Question 3
If yes WHICH SECTIONS?
Question 3
Have you found the workbook helpful or not? YES NO
REASONS?Question 4
Have you implemented any control measures as a result of any risk assessment? YES NO
If yes, what were MEASURES?Question 5
Question 6
Have you any suggestions for improvements? YES NO
Date of interview______________
166
APPE
ND
IX 7
INC
IDEN
T D
ESC
RIP
TIVE
DAT
A
PHA
SE 1
Ta
ble
1 Tr
ust A
(20
Inci
dent
s)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
peD
escr
iptio
n
1 £3
3.13
O
ther
O
ther
O
ther
Fa
lse
fire
alar
m –
faul
ty sm
oke
dete
ctor
(det
ecto
r req
uire
d cl
eani
ng).
The
re w
as a
m
aint
enan
ce sc
hedu
le b
ut E
stat
es D
ept w
as b
ehin
d w
ith im
plem
enta
tion.
2
£689
.00
Staf
f Ph
ysic
al
viol
ence
Pe
rson
to p
erso
n as
saul
t N
urse
hea
d-bu
tted
by p
atie
nt (a
lthou
gh p
atie
nt w
as o
n cl
ose
obse
rvat
ion
prio
r to
inci
dent
as
it ha
d be
en id
entif
ied
they
wer
e be
com
ing
aggr
essi
ve).
The
pat
ient
was
rest
ricte
d to
the
war
d.3
£130
.27
Staf
f A
ccid
enta
l in
jury
C
ut w
ith sh
arp
mat
eria
l/obj
ect
Two
nurs
es c
ut w
ith p
oten
tially
infe
cted
gla
ss w
hile
tryi
ng to
take
it fr
om a
pat
ient
who
had
al
read
y cu
t her
self
with
it.
The
patie
nt w
as in
tent
on
self-
harm
ing
and
did
not w
ant s
taff
to
rem
ove
the
glas
s, ho
wev
er th
e in
jurie
s to
the
nurs
es w
ere
mor
e ac
cide
ntal
than
inte
ntio
nal.
4 £4
.00
Patie
nt
Nea
r mis
s O
ther
Pa
tient
det
aine
d un
der S
ectio
n 26
MH
A, l
eft w
ard
with
out p
erm
issi
on.
Foun
d by
staf
f in
an
adja
cent
car
par
k. D
ue to
pat
ient
’s h
ostil
ity a
nd a
ggre
ssio
n, p
olic
e as
sist
ance
was
requ
ired
to re
turn
her
to h
ospi
tal.
5 £3
4.44
Pa
tient
N
ear m
iss
Oth
er
A d
etai
ned
patie
nt a
bsco
nded
from
hos
pita
l war
d, ju
mpe
d in
to c
anal
and
refu
sed
to c
ome
out.
Eve
ntua
lly re
mov
ed fr
om th
e ca
nal b
y th
e po
lice.
6
£14.
00
Staf
f A
ccid
enta
l in
jury
H
ot o
r col
d co
ntac
t N
ursi
ng A
ssis
tant
unl
oadi
ng lu
nch
trolle
y tra
nsfe
rrin
g fo
od to
hot
ove
n. S
oup
cont
aine
r slid
to
floo
r as d
oor o
pene
d. F
oot s
cald
ed w
ith so
up.
Off
dut
y fo
llow
ing
first
aid
. 7
£18.
96
Patie
nt
Nea
r mis
s O
ther
Pa
tient
abs
cond
ed fr
om h
ospi
tal.
Pur
sued
by
staf
f bec
ause
of h
isto
ry o
f sui
cide
atte
mpt
s.
Ret
urne
d to
hos
pita
l with
min
imal
rest
rain
t. 8
£493
.31
Staf
f A
ccid
enta
l in
jury
O
ther
Tw
o st
aff n
urse
s pur
sued
an
absc
ondi
ng p
atie
nt w
ho st
rugg
led
whe
n re
stra
ined
. N
urse
was
pu
shed
off
pav
emen
t and
wen
t ove
r on
her a
nkle
. 9
£112
.00
Staf
f A
ccid
enta
l in
jury
St
ruck
aga
inst
so
met
hing
(fur
nitu
re
etc)
Nur
se w
alki
ng th
roug
h an
unl
it co
rrid
or k
nock
ed le
ft ha
nd a
gain
st w
all,
sust
aini
ng in
jury
.
10
£431
.20
Staf
f Ph
ysic
al
viol
ence
Pe
rson
to p
erso
n as
saul
t A
n au
xilia
ry n
urse
was
ass
aulte
d w
hile
tryi
ng to
retri
eve
tabl
ets f
rom
the
floor
bec
ause
a
patie
nt h
ad sp
at o
ut th
e ta
blet
s. 11
£2
80.6
4 St
aff
Phys
ical
vi
olen
ce
Pers
on to
per
son
assa
ult
A st
aff n
urse
ass
aulte
d du
ring
drug
roun
d.
167
12
£88.
00
Staf
f Ph
ysic
al
viol
ence
Pe
rson
to p
erso
n as
saul
t Pa
tient
with
lear
ning
dis
abili
ties a
ssau
lted
fello
w p
atie
nt.
Nur
sing
Ass
ista
nt w
as k
icke
d w
hils
t atte
mpt
ing
to d
efus
e th
e in
cide
nt.
Patie
nt w
as re
turn
ed to
his
room
by
forc
e w
here
he
proc
eede
d to
cau
se d
amag
e to
the
room
and
its c
onte
nts.
13
£413
.13
Staf
f A
ccid
enta
l in
jury
Pa
tient
lifti
ng/h
andl
ing
Nur
se to
iletin
g re
side
nt w
ho d
idn’
t wei
ght-b
ear.
Sus
tain
ed b
ack
inju
ry.
14
£52.
18
Staf
f Ph
ysic
al
viol
ence
Pe
rson
to p
erso
n as
saul
t St
aff n
urse
ass
aulte
d tw
ice
by p
atie
nt a
fter t
hey
beca
me
aggr
essi
ve.
Req
uire
d su
ppor
t of 2
ot
her s
taff
to se
ttle
situ
atio
n.
15
£5.1
0 Pa
tient
N
ear m
iss
Oth
er
Patie
nt w
ith se
nile
dem
entia
left
war
d un
notic
ed.
Staf
f sea
rche
d lo
cal a
rea,
pat
ient
foun
d ab
out 1
½ m
iles a
way
and
retu
rned
to w
ard
unha
rmed
. 16
£6
57.4
5 St
aff
Acc
iden
tal
inju
ry
Patie
nt li
fting
/han
dlin
g St
aff n
urse
and
aux
iliar
y nu
rse
wer
e tra
nsfe
rrin
g pa
tient
from
bed
to w
heel
chai
r. T
he
patie
nt u
nabl
e/un
will
ing
to w
eigh
t-bea
r. A
uxili
ary
nurs
e su
stai
ned
back
inju
ry.
17
£7.7
2 Pa
tient
N
ear m
iss
Oth
er
Patie
nt n
oted
to b
e ab
sent
from
war
d. D
etai
ned
unde
r Sec
tion
26 M
HA
. Fo
llow
ing
a se
arch
, pat
ient
foun
d ou
twith
hos
pita
l gro
unds
aro
und
¾ m
ile a
way
. N
o ap
pare
nt in
jury
. 18
£1
2.88
Pa
tient
N
ear m
iss
Oth
er
Patie
nt a
bsco
nded
, sea
rch
inst
igat
ed.
Patie
nt fo
und
in lo
ng g
rass
in g
roun
ds.
Had
falle
n an
dw
as u
nabl
e to
get
up.
No
appa
rent
inju
ry n
oted
. 19
£1
0.12
Pa
tient
A
ccid
enta
l in
jury
O
ther
Pa
tient
abs
cond
ed fr
om w
ard.
Fou
nd a
bout
½ m
ile a
way
. Pa
tient
had
falle
n, su
stai
ned
smal
l abr
asio
ns.
20
£5.1
0 Pa
tient
N
ear m
iss
Oth
er
Patie
nt n
otifi
ed m
issi
ng.
Spot
ted
by o
ff-d
uty
mem
ber o
f sta
ff w
ho re
turn
ed h
im to
the
war
d.
Tabl
e 2
Trus
t B (1
5 In
cide
nts)
No
Cost
Staf
f/ pa
tient
/ ot
her
Cate
gory
Ty
peD
escr
iptio
n
1 £2
5.16
St
aff
Oth
er
Oth
er
Com
mun
ity p
sych
iatri
c nu
rse
bitte
n by
pup
py in
pat
ient
’s h
ome.
2 £3
0.87
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Nee
dles
tick
inju
ry su
stai
ned
by d
istri
ct n
urse
in p
atie
nt’s
hom
e. D
urin
g ad
min
istra
tion
of
inje
ctio
n, p
atie
nt b
ecam
e an
xiou
s and
mov
ed a
way
from
nur
se, c
ausi
ng a
shar
ps in
jury
to h
er
finge
r. A
t the
tim
e of
the
inci
dent
it w
as n
ot p
ossi
ble
to p
rocu
re re
tract
able
nee
dles
for
inje
ctio
ns c
arrie
d ou
t in
the
com
mun
ity.
This
has
sinc
e be
en c
orre
cted
. 3
£16.
00
Staf
f O
ther
O
ther
A
you
ng p
erso
n’s a
ddic
tion
wor
ker s
usta
ined
a d
og b
ite in
a c
lient
’s h
ouse
. 4
£173
.48
Staf
f A
ccid
enta
l in
jury
O
ther
D
urin
g a
cont
rol a
nd re
stra
int m
anoe
uvre
a n
ursi
ng a
ssis
tant
was
pre
ssed
aga
inst
ano
ther
m
embe
r of s
taff
and
an
alar
m o
n he
r bel
t pus
hed
into
her
ribs
. A
rib
inju
ry su
stai
ned.
5
£4,4
11.8
8 St
aff
Acc
iden
tal
Patie
nt li
fting
/han
dlin
g A
n en
rolle
d nu
rse
was
rolli
ng a
pat
ient
in b
ed in
ord
er to
was
h an
d ch
ange
him
. A
s she
rolle
d
168
inju
ry
the
patie
nt o
ver,
she
felt
a sh
arp
pain
. La
ter d
evel
oped
a p
ainf
ul le
ft hi
p.
6 £2
5.16
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Enro
lled
nurs
e su
stai
ned
shar
ps in
jury
thro
ugh
a cl
inic
al w
aste
bag
into
whi
ch 5
gla
ss
ampo
ules
and
gla
ss to
ps h
ad b
een
plac
ed.
Late
r dis
cove
red
to h
ave
been
dis
pose
d of
by
a do
ctor
dur
ing
the
adm
inis
tratio
n of
an
inje
ctio
n of
Vita
min
K to
a p
atie
nt.
7 £9
,849
.60
Staf
f A
ccid
enta
l in
jury
Pa
tient
lifti
ng/h
andl
ing
A G
rade
E st
aff n
urse
mov
ing
a co
nfus
ed p
atie
nt in
bed
, tw
iste
d he
r nec
k. L
ift h
ad b
een
cond
ucte
d by
the
nurs
e on
her
ow
n.
8 £3
12.1
8 St
aff
Acc
iden
tal
inju
ry
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
N
urse
slip
ped
dow
n 2
stai
rs w
hils
t run
ning
to a
ttend
an
emer
genc
y al
arm
(vio
lenc
e an
d ag
gres
sion
). D
urin
g sl
ip, l
eft a
nkle
turn
ed in
war
ds a
nd a
nkle
inju
ry w
as su
stai
ned.
9
£525
.00
Staf
f A
ccid
enta
l in
jury
Pa
tient
lifti
ng/h
andl
ing
An
auxi
liary
nur
se w
as a
ssis
ting
a pa
tient
to st
and
and
trans
ferr
ing
him
ont
o hi
s whe
elch
air
whe
n sh
e ex
perie
nced
a su
dden
pai
n in
her
mid
bac
k.
10
£1,3
39.2
2 St
aff
Acc
iden
tal
inju
ry
Stru
ck a
gain
st
som
ethi
ng (f
urni
ture
etc
) N
urse
chi
pped
a b
one
in a
rm d
urin
g a
cont
rol a
nd re
stra
int m
anoe
uvre
.
11
£358
.88
Staf
f A
ccid
enta
l in
jury
Pa
tient
lifti
ng/h
andl
ing
Supp
ort w
orke
r atte
mpt
ed to
bre
ak th
e fa
ll of
a p
atie
nt w
ho sl
ippe
d du
ring
a w
alk.
The
y su
stai
ned
a fr
actu
red
thum
b an
d cu
t to
right
han
d.
12
£141
.96
Staf
f A
ccid
enta
l in
jury
Pe
rson
to p
erso
n as
saul
t N
ursi
ng a
ssis
tant
sust
aine
d ha
nd in
jury
bei
ng k
icke
d by
pat
ient
whi
lst a
ttem
ptin
g to
adm
inis
ter
emer
genc
y m
edic
atio
n.
13
£2,3
10.0
0 St
aff
Acc
iden
tal
inju
ry
Stru
ck b
y an
obj
ect
Nur
sing
ass
ista
nt st
ruck
on
foot
by
flip
char
t, w
hich
was
pus
hed
over
by
a pa
tient
. Fl
ip C
hart
shou
ld n
ot h
ave
been
stor
ed in
that
are
a.
14
£2,2
05.0
0 St
aff
Acc
iden
tal
inju
ry
Patie
nt li
fting
/han
dlin
g N
ursi
ng a
uxili
ary
sust
aine
d tw
iste
d ba
ck a
ssis
ting
patie
nt to
stan
d.
15
£560
.81
Staf
f A
ccid
enta
l in
jury
M
anua
l lift
ing/
hand
ling
Tech
nica
l ins
truct
or su
stai
ned
back
inju
ry h
andl
ing
timbe
r whe
n th
ey w
ere
push
ing
timbe
r th
roug
h a
pane
l saw
ben
ch.
Tabl
e 3
Trus
t C (1
6 In
cide
nts)
No
Cost
Staf
f/ pa
tient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £2
2.54
St
aff
Acc
iden
tal i
njur
y Pa
tient
lifti
ng/h
andl
ing
Whi
le a
ssis
ting
in m
ovin
g a
dece
ased
pat
ient
from
the
bed
to a
trol
ley,
the
trolle
y sh
elf
mov
ed a
nd a
n au
xilia
ry n
urse
sust
aine
d an
inju
ry to
her
bac
k. T
he tr
olle
y w
as in
appr
opria
te
for t
his k
ind
of tr
ansf
er b
ut it
was
cho
sen
beca
use
it w
as p
lann
ed to
be
used
to tr
ansf
er th
e bo
dy in
to th
e fr
idge
. 2
£41.
44
Oth
er
Van
dalis
m
Oth
er
Win
dow
s bro
ken
by u
nkno
wn
pers
on(s
). N
o ot
her d
amag
e. R
epai
red
sam
e da
y.
3 £1
2.50
O
ther
V
anda
lism
O
ther
W
indo
w d
isco
vere
d sm
ashe
d. R
epor
ted
to p
olic
e an
d Es
tate
s Dep
t. N
o w
itnes
ses.
4 £4
1.72
O
ther
V
anda
lism
O
ther
Tw
o bo
ys tr
espa
ssin
g th
roug
h a
bath
room
win
dow
, bre
akin
g it.
Pol
ice
wer
e ca
lled
to th
e
169
scen
e. W
indo
w re
paire
d sa
me
even
ing.
5
£52.
11
Oth
er
Van
dalis
m
Oth
er
An
alar
m w
as a
ctiv
ated
and
was
hea
rd b
y st
aff o
n a
near
by w
ard.
Bot
h Po
lice
and
Esta
tes
Dep
t wer
e no
tifie
d. A
win
dow
was
bro
ken
and
the
build
ing
had
been
ent
ered
and
are
as
dist
urbe
d, b
ut n
othi
ng w
as st
olen
. 6
£1,0
68.5
0 O
ther
Th
eft
Oth
er
Bre
ak-in
and
thef
t of c
ompu
ter e
quip
men
t. 7
£102
.94
Oth
er
Thef
t O
ther
Th
e w
indo
w in
the
fron
t off
ice
was
bro
ken
durin
g th
e ni
ght a
nd a
n un
iden
tifie
d in
trude
r st
ole
a co
mpu
ter k
eybo
ard.
No
witn
esse
s. 8
£7.0
6 Pa
tient
A
ccid
enta
l inj
ury
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
Pa
tient
fell
whi
le w
alki
ng to
toile
t. P
atie
nt w
as w
alki
ng in
stoc
king
feet
with
out s
lippe
rs o
r Zi
mm
er.
Patie
nt d
id n
ot c
all f
or a
ssis
tanc
e an
d su
bseq
uent
ly fe
ll ba
ckw
ards
, stri
king
hea
d on
floo
r sus
tain
ing
lace
ratio
n to
hea
d.
9 £2
09.4
1 St
aff
Acc
iden
tal i
njur
y Sl
ip/tr
ip o
r fal
l on
the
sam
e le
vel o
r sta
irs
Enro
lled
nurs
e he
ard
a pa
tient
cho
king
. R
ushe
d to
hel
p, a
nd fa
iled
to n
otic
e a
haza
rd c
one,
an
d sl
ippe
d on
wet
floo
r. 10
£7
0.16
St
aff
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t St
aff n
urse
ass
aulte
d by
pat
ient
with
out w
arni
ng.
11
£8.7
2 Pa
tient
A
ccid
enta
l inj
ury
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
Pa
tient
uno
bser
ved
on c
orrid
or.
Fell,
frac
turin
g hi
p. T
rans
ferr
ed to
Acu
te T
rust
for
treat
men
t. 12
£7
.80
Patie
nt
Acc
iden
tal i
njur
y Sl
ip/tr
ip o
r fal
l on
the
sam
e le
vel o
r sta
irs
Patie
nt b
eing
nur
sed
on fl
oor s
usta
ined
frac
ture
d hi
p. T
rans
ferr
ed to
Acu
te H
ospi
tal f
or
treat
men
t. 13
£2
07.8
0 Pa
tient
A
ccid
enta
l inj
ury
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
Pa
tient
slip
ped
whi
le u
nobs
erve
d, su
stai
ning
an
inju
ry to
the
left
foot
. N
o fr
actu
res
dete
cted
. 14
£3
.55
Staf
f A
ccid
enta
l inj
ury
Nee
dles
tick/
shar
ps
Nur
sing
ass
ista
nt su
stai
ned
lace
ratio
n w
hils
t sha
ving
pat
ient
. 15
£3
0.27
O
ther
A
ccid
enta
l pr
oper
ty
loss
/dam
age
Oth
er
Gut
terin
g bl
ocke
d w
ith le
aves
ove
rflo
wed
, cau
sing
dam
age
to th
e ce
iling
whi
ch fe
ll in
pl
aces
cau
sing
floo
ding
to th
e flo
or.
16
£71.
89
Patie
nt
Ver
bal a
buse
or
thre
aten
ing
beha
viou
r
Oth
er
Patie
nt a
rriv
ed o
n th
e w
ard
acco
mpa
nied
by
2 fr
iend
s but
with
out a
refe
rral
lette
r fro
m th
eir
GP.
The
pat
ient
was
adv
ised
by
staf
f he
coul
d no
t be
asse
ssed
with
out a
lette
r and
su
gges
ted
he g
o to
A&
E or
bac
k to
his
GP.
As t
he p
atie
nt le
ft th
e bu
ildin
g, st
aff h
eard
a
win
dow
bre
akin
g. T
he p
olic
e di
d no
t wan
t to
take
act
ion
until
the
patie
nt w
as a
sses
sed.
Th
e pa
tient
retu
rned
to th
e w
ard
late
r for
ass
essm
ent.
170
Tabl
e 4
Trus
t D (2
8 In
cide
nts)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £2
02.1
0 St
aff
Acc
iden
tal i
njur
y C
ut w
ith sh
arp
mat
eria
l/obj
ect
A c
ater
ing
assi
stan
t sus
tain
ed a
dee
p la
cera
tion
to h
is h
and
from
a b
roke
n m
ilk ju
g w
hils
t was
hing
dis
hes.
Not
wea
ring
glov
es.
2 £3
2.40
O
ther
N
ear m
iss
Oth
er
Elec
tric
grill
in th
e ki
tche
n of
staf
f res
iden
ce w
as le
ft on
. Th
e ki
tche
n do
or w
as
open
and
smok
e tra
velle
d in
to th
e co
rrid
or a
nd a
ctiv
ated
the
smok
e al
arm
. Fi
re
Brig
ade
atte
nded
but
no
actio
n w
as re
quire
d. A
larm
s wer
e re
set b
y th
e on
cal
l el
ectri
cian
. 3
£17.
28
Staf
f A
ccid
enta
l inj
ury
Nee
dles
tick/
shar
ps
A c
omm
unity
staf
f nur
se w
as ta
king
a b
lood
sam
ple
from
a p
atie
nt.
On
rem
oval
of
the
need
le fr
om th
e pa
tient
’s a
rm, t
he n
eedl
e sl
ippe
d an
d pr
icke
d th
e nu
rse’
s lef
t th
umb.
4
£9.4
0 St
aff
Acc
iden
tal i
njur
y C
ut w
ith sh
arp
mat
eria
l/obj
ect
Dom
estic
was
doi
ng d
ishe
s fol
low
ing
a fu
nctio
n in
her
dep
artm
ent.
Not
wea
ring
yello
w k
itche
n du
ty g
love
s pro
vide
d. W
hils
t dry
ing
a fin
e-st
emm
ed g
lass
, it
slip
ped
and
brok
e in
her
han
d. S
ent t
o th
e A
cute
Tru
st fo
r tre
atm
ent.
5 £1
8.80
St
aff
Acc
iden
tal i
njur
y St
ruck
aga
inst
som
ethi
ng
(fur
nitu
re e
tc)
Elec
trici
an st
ruck
hea
d on
pip
e w
ork
whi
le c
heck
ing
batte
ries i
n ge
nera
tor r
oom
. N
o PP
E (b
ump
cap)
was
wor
n, a
lthou
gh th
ese
wer
e av
aila
ble
in th
e w
orks
hop.
In
jure
d pa
rty a
ttend
ed A
cute
Tru
st A
&E
Dep
t for
trea
tmen
t. 6
£66.
27
Staf
f A
ccid
enta
l inj
ury
Stru
ck b
y an
obj
ect
Mai
nten
ance
ass
ista
nt w
as a
djus
ting
the
cloc
ks b
y cl
imbi
ng o
nto
a ch
air.
The
cl
ock
fell
off w
all a
nd st
ruck
mai
nten
ance
ass
ista
nt o
n br
idge
of n
ose.
7
£238
.96
Staf
f A
ccid
enta
l inj
ury
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
El
ectri
cian
wor
king
in a
n at
tic sp
ace
slip
ped
off a
jois
t. T
o av
oid
falli
ng th
roug
h th
e ce
iling
, fel
l ont
o hi
s lef
t sid
e in
jurin
g hi
s lef
t sho
ulde
r and
arm
. 8
£14.
60
Staf
f A
ccid
enta
l inj
ury
Stru
ck a
gain
st so
met
hing
(f
urni
ture
etc
) A
mem
ber o
f the
Est
ates
Dep
t was
esc
ortin
g a
cont
ract
or to
an
unde
r flo
or a
rea
to
cond
uct e
lect
rical
repa
irs.
Nei
ther
man
was
wea
ring
a ha
rd h
at.
The
cont
ract
or
wal
ked
into
a lo
w b
eam
, stri
king
his
fore
head
. Su
stai
ned
a la
cera
tion
and
suff
ered
co
ncus
sion
and
a b
rief l
oss o
f con
scio
usne
ss.
Esco
rted
to A
cute
Tru
st A
&E
Dep
t fo
r tre
atm
ent.
9 £2
38.0
0 O
ther
A
ccid
enta
l pr
oper
ty
loss
/dam
age
Con
tact
with
eq
uipm
ent/m
achi
nery
Po
rter d
amag
ed a
utom
atic
doo
rs b
y st
rikin
g th
em w
ith a
food
trol
ley
whi
ch h
e w
as
push
ing.
10
£806
.40
Staf
f A
ccid
enta
l inj
ury
Patie
nt li
fting
/han
dlin
g W
hils
t giv
ing
a pa
tient
a b
ath,
the
patie
nt d
ecid
ed to
subm
erge
him
self.
A
Hea
lthca
re A
ssis
tant
sust
aine
d ba
ck in
jury
whi
le, t
oget
her w
ith a
seni
or c
harg
e nu
rse,
atte
mpt
ing
to su
ppor
t the
pat
ient
’s h
ead
abov
e w
ater
. Th
e st
aff h
ad d
ecid
ed
that
he
shou
ld b
e gi
ven
a ba
th, d
espi
te h
is c
are
plan
stat
ing
he b
e sh
ower
ed.
171
11
£21.
02
Patie
nt
Nea
r mis
s O
ther
A
lega
lly d
etai
ned
patie
nt w
as se
en le
avin
g th
e w
ard
by st
aff.
At t
he ti
me,
the
absc
onde
d pa
tient
was
und
er g
ener
al o
bser
vatio
n. T
hree
mem
bers
of s
taff
pur
sued
th
e pa
tient
and
retu
rned
him
to th
e w
ard
unha
rmed
. 12
£2
1.02
Pa
tient
N
ear m
iss
Oth
er
Patie
nt a
bsco
nded
from
an
open
war
d w
here
he
was
in g
ener
al o
bser
vatio
n.
Pers
on re
sidi
ng n
earb
y to
the
hosp
ital p
hone
d to
not
ify o
f the
ir w
here
abou
ts a
nd
the
patie
nt w
as c
olle
cted
by
a st
aff m
embe
r and
retu
rned
to th
e w
ard.
13
£4
3.40
St
aff
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t En
rolle
d nu
rse
head
-but
ted
by p
atie
nt w
hile
ass
istin
g w
ith a
vio
lent
out
burs
t. 14
£6
.60
Patie
nt
Nea
r mis
s O
ther
Pa
tient
det
aine
d un
der M
enta
l Hea
lth A
ct u
nder
con
stan
t obs
erva
tion
was
take
n by
se
nior
hou
se o
ffic
er to
be
inte
rvie
wed
in a
n ex
tern
al in
terv
iew
room
. Pa
tient
re
turn
ed to
the
war
d w
ith so
me
forc
e re
quire
d.
15
£67.
51
Staf
f Ph
ysic
al v
iole
nce
Pers
on to
per
son
assa
ult
Nur
sing
ass
ista
nt in
terv
ened
to se
para
te p
atie
nts w
ho w
ere
fight
ing
and
sust
aine
d in
jurie
s to
her n
eck
and
elbo
w.
The
nurs
e se
en b
y th
e A
cute
Tru
st A
&E
Dep
t. 16
£1
0.78
Pa
tient
N
ear m
iss
Oth
er
Abs
cond
ing
patie
nt.
Sear
ch in
itiat
ed b
ut p
atie
nt la
ter r
etur
ned
to th
e w
ard
of h
is
own
free
will
, unh
arm
ed.
17
£5.5
0 Pa
tient
N
ear m
iss
Oth
er
Abs
cond
ed p
atie
nt re
turn
ed to
the
war
d af
ter a
sear
ch.
18
£19.
47
Staf
f A
ccid
enta
l inj
ury
Stru
ck a
gain
st so
met
hing
(f
urni
ture
etc
) G
ener
al a
dmin
ass
ista
nt w
as in
jure
d w
hile
car
ryin
g a
flip
char
t and
mar
ker p
ens
thro
ugh
doub
le d
oors
. Tr
eate
d by
Acu
te T
rust
A&
E D
ept.
19
£95.
20
Staf
f A
ccid
enta
l inj
ury
Patie
nt li
fting
/han
dlin
g A
Hea
lthca
re A
ssis
tant
was
ass
istin
g an
eld
erly
pat
ient
to w
alk
with
the
aid
of a
Zi
mm
er.
Patie
nt le
t go
of th
e Zi
mm
er a
nd fe
ll. T
he H
ealth
care
ass
ista
nt su
ffer
ed
carp
et b
urns
atte
mpt
ing
to c
ontro
l the
fall.
20
£1
2.40
Pa
tient
N
ear m
iss
Oth
er
Dem
entia
pat
ient
abs
cond
ed.
Foun
d 30
min
s lat
er u
nhar
med
, by
nurs
ing
auxi
liary
an
d re
turn
ed to
war
d.
21
£28.
00
Staf
f A
ccid
enta
l inj
ury
Patie
nt li
fting
/han
dlin
g H
ealth
care
Ass
ista
nt tr
ansf
errin
g pa
tient
from
bed
to w
heel
chai
r whe
n pa
tient
st
umbl
ed a
nd tr
appe
d H
CA
’s h
and
betw
een
him
self
and
lock
er.
Nig
htsh
ift st
aff
used
hoi
st to
tran
sfer
pat
ient
due
to h
is d
eter
iora
tion
in h
ealth
, but
this
had
not
be
en c
omm
unic
ated
to d
ay st
aff.
22
£26.
75
Staf
f A
ccid
enta
l inj
ury
Nee
dles
tick/
shar
ps
Staf
f nur
se o
btai
ned
ster
ile u
rine
sam
ple
usin
g a
syrin
ge a
nd n
eedl
e fr
om a
cath
eter
por
t at t
he p
atie
nt’s
bed
side
. Pl
aced
the
used
nee
dle
back
on
the
tray
and
took
to th
e sl
uice
are
a fo
r dis
posa
l. D
istra
cted
whe
n so
meo
ne sp
oke
to h
er, s
he
pick
ed u
p th
e ne
edle
by
the
shar
p en
d an
d w
as in
jure
d.
23
£7.0
7 St
aff
Acc
iden
tal i
njur
y C
onta
ct w
ith e
xpos
ure
to
harm
ful s
ubst
ance
s A
22½
-litre
ble
ach
barr
el b
urst
whi
le b
eing
tran
spor
ted
into
the
laun
dry.
The
bl
each
spill
ed o
nto
the
clot
hing
and
skin
of a
laun
dere
tte su
perin
tend
ent,
caus
ing
her t
o su
stai
n irr
itatio
n to
her
legs
and
che
st.
Mem
ber o
f Est
ates
dec
ante
d th
e bl
each
into
oth
er c
onta
iner
s and
dilu
ted
the
spilt
ble
ach.
24
£3
7.12
O
ther
O
ther
O
ther
R
esid
ents
wer
e sm
okin
g in
the
sitti
ng ro
om, w
hich
act
ivat
ed th
e fir
e al
arm
whi
ch
172w
as se
t to
“sm
oke
dete
ctio
n”.
25
£260
.00
Staf
f Ph
ysic
al v
iole
nce
Pers
on to
per
son
assa
ult
Res
iden
t gra
bbed
an
enro
lled
nurs
e by
her
clo
thin
g as
she
ente
red
a fla
t. T
he
pinp
oint
ala
rm w
as a
ctiv
ated
but
did
n’t w
ork.
The
resi
dent
ass
aulte
d th
e nu
rse
and
bit h
er ri
ght f
orea
rm.
26
£12.
37
Patie
nt
Nea
r mis
s O
ther
D
emen
tia p
atie
nt a
bsco
nded
from
war
d. P
atie
nt e
vent
ually
loca
ted
at a
loca
l ch
emis
t sho
p.
27
£1,2
08.7
0 St
aff
Acc
iden
tal i
njur
y M
anua
l lift
ing/
hand
ling
Porte
r rem
ovin
g a
parc
el fr
om th
e ba
ck o
f a h
igh
top
van
sust
aine
d a
pulle
d ha
mst
ring.
Atte
nded
A&
E fo
r tre
atm
ent a
nd w
as o
n si
ck le
ave
follo
win
g th
e in
cide
nt.
28
£338
.40
Staf
f A
ccid
enta
l inj
ury
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
Po
rter f
ell o
ver c
ardb
oard
box
es w
hile
pic
king
up
blac
k ba
gs fr
om th
e ca
rdbo
ard
stor
age
area
in th
e ho
spita
l. S
usta
ined
a sp
rain
ed a
nkle
and
sore
shou
lder
.
Tabl
e 5
Trus
t E (2
2 In
cide
nts)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £5
.83
Staf
f A
ccid
enta
l in
jury
C
onta
ct w
ith
equi
pmen
t/mac
hine
ry
Porte
r was
mov
ing
trest
le ta
ble
and
was
wou
nded
by
a sp
linte
r in
his l
eft t
hum
b.
Atte
nded
A&
E.
2 £1
4.10
St
aff
Acc
iden
tal
inju
ry
Hot
or c
old
cont
act
Seas
onal
wor
ker (
stud
ent)
was
was
hing
cut
lery
. R
outin
e w
ashe
r bro
ken.
On
usin
g w
ashe
r in
kitc
hen
area
, sca
lded
righ
t han
d/w
rist a
fter i
mm
ersi
ng in
wat
er a
bove
gl
ove
leve
l. 3
£14.
10
Staf
f A
ccid
enta
l in
jury
C
onta
ct w
ith
equi
pmen
t/mac
hine
ry
A c
ater
ing
assi
stan
t sus
tain
ed a
n an
kle
inju
ry a
fter p
ullin
g (tr
aine
d to
pus
h ra
ther
th
an to
pul
l) pl
ate
hold
ers t
rolle
y to
stor
age
area
. Th
e ca
terin
g as
sist
ant c
lippe
d le
ft an
kle
on w
heel
bra
kes.
4 £3
2.90
St
aff
Acc
iden
tal
inju
ry
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
C
ater
ing
assi
stan
t, w
earin
g sa
fety
shoe
s, sl
ippe
d on
kitc
hen
floor
(non
slip
surf
ace)
in
jurin
g rig
ht e
lbow
and
coc
cyx.
5
£2.3
5 St
aff
Acc
iden
tal
inju
ry
Patie
nt li
fting
/han
dlin
g A
por
ter s
usta
ined
a b
ack
inju
ry tr
ansf
errin
g an
ana
esth
etis
ed p
atie
nt fr
om a
thea
tre
trolle
y to
a b
ed u
sing
pat
slid
e in
rece
ptio
n/re
cove
ry a
rea.
The
stra
in w
as su
stai
ned
as a
resu
lt of
poo
r pos
ition
ing
of th
e pa
tient
/bed
/trol
ley.
6
£9.3
5 St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Porte
r sus
tain
ed a
nee
dles
tick
inju
ry fr
om a
clin
ical
was
te b
ag.
The
need
lest
ick
inju
ry w
as su
stai
ned
to th
e rig
ht le
g w
hen
mov
ing
the
bag.
7
£11.
39
Staf
f A
ccid
enta
l in
jury
N
eedl
estic
k/sh
arps
Fo
llow
ing
vena
pun
ctur
e, re
mov
ed c
anul
a an
d pl
aced
in p
ulp
tray.
On
trans
fer f
rom
tra
y to
shar
ps d
ispo
sal b
ox, s
usta
ined
nee
dles
tick
inju
ry.
8 £5
28.7
5 St
aff
Acc
iden
tal
inju
ry
Man
ual l
iftin
g/ha
ndlin
g A
dom
estic
dis
posi
ng o
f rub
bish
into
a c
entra
l cor
e ar
ea.
Rec
epta
cle
alre
ady
over
full,
and
whe
n th
row
ing
rubb
ish
onto
top
of ru
bbis
h pi
le, d
omes
tic p
ulle
d a
173
mus
cle
in h
er b
ack.
9
£4.7
0 St
aff
Acc
iden
tal
inju
ry
Stru
ck a
gain
st so
met
hing
(f
urni
ture
etc
) D
omes
tic se
cond
ed to
out
patie
nts d
epar
tmen
t. H
igh
dust
ing
in c
onfin
ed a
rea
on
step
s. B
ange
d he
ad o
n lig
ht fi
tmen
t.
10
£709
.70
Staf
f A
ccid
enta
l in
jury
C
onta
ct w
ith
equi
pmen
t/mac
hine
ry
Dom
estic
com
men
cing
floo
r cle
anin
g w
ith b
uffin
g m
achi
ne in
adve
rtent
ly tr
od o
n el
ectri
c ca
ble
caus
ing
her t
o je
rk b
ackw
ards
, lea
ding
to a
mus
culo
-ske
leta
l inj
ury.
11
£1
0.39
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Nur
se w
as c
uttin
g ox
ygen
tubi
ng w
hen
her s
ciss
ors s
lippe
d, c
ausi
ng a
pun
ctur
e w
ound
in th
e th
umb
pad
of h
er le
ft ha
nd.
12
£11.
39
Staf
f A
ccid
enta
l in
jury
N
eedl
estic
k/sh
arps
N
urse
dis
posi
ng o
f nee
dle
into
ove
rful
l sha
rps b
ox su
stai
ned
need
lest
ick
inju
ry fr
om
prot
rudi
ng n
eedl
e.
13
£6,8
55.0
0 Pa
tient
A
ccid
enta
l in
jury
Sl
ip/tr
ip o
r fal
l on
the
sam
e le
vel o
r sta
irs
Patie
nt le
ft al
one
on a
com
mod
e sl
ippe
d an
d fe
ll. F
ract
ured
nec
k of
fem
ur su
stai
ned.
14
£6.5
8 Pa
tient
A
ccid
enta
l in
jury
Sl
ip/tr
ip o
r fal
l on
the
sam
e le
vel o
r sta
irs
Patie
nt fe
ll as
leep
in c
hair
whi
le w
atch
ing
tele
visi
on.
On
getti
ng u
p to
go
to b
ed,
slip
ped
and
fell
(no
slip
pers
on
due
to sw
olle
n to
e), h
ittin
g he
ad o
f doo
r and
su
stai
ning
a la
cera
tion
to h
is le
ft ea
r. 15
£1
0.54
Pa
tient
A
ccid
enta
l in
jury
Sl
ip/tr
ip o
r fal
l on
the
sam
e le
vel o
r sta
irs
An
81 y
ear o
ld fe
mal
e pa
tient
in a
hig
hly
conf
used
stat
e at
tem
pted
to v
isit
the
toile
t un
aide
d by
staf
f. F
ell i
n th
e to
ilet a
rea.
Inj
urie
s to
head
and
shin
wer
e su
stai
ned.
16
£5
0.90
St
aff
Acc
iden
tal
inju
ry
Patie
nt li
fting
/han
dlin
g A
uxili
ary
nurs
e su
stai
ned
knee
inju
ry d
urin
g m
anua
l han
dlin
g of
pat
ient
.
17
£17.
00
Staf
f A
ccid
enta
l in
jury
N
eedl
estic
k/sh
arps
C
linic
al su
perv
isor
ove
rsee
ing
proc
edur
e, g
ivin
g in
stru
ctio
n on
cor
rect
pos
ition
ing
of
need
le d
urin
g su
turin
g, re
ceiv
ed n
eedl
estic
k in
jury
. 18
£2
6.70
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Dom
estic
cle
anin
g be
d in
labo
ur w
ard
afte
r clin
ical
pro
cedu
re.
Una
war
e us
ed n
eedl
e w
as m
issi
ng.
Lost
in fo
ld o
f bed
. Su
bseq
uent
ly sc
ratc
hed
her r
ight
arm
, sus
tain
ing
need
lest
ick
inju
ry.
19
£94.
50
Staf
f Ph
ysic
al
viol
ence
Pe
rson
to p
erso
n as
saul
t A
vio
lent
inci
dent
invo
lvin
g su
icid
al p
atie
nt in
war
d.
20
£23.
99
Staf
f A
ccid
enta
l in
jury
N
eedl
estic
k/sh
arps
N
urse
was
rem
ovin
g a
butte
rfly
nee
dle
from
pat
ient
’s sk
in.
On
rem
ovin
g th
e ne
edle
, sh
e st
abbe
d he
r rig
ht th
umb,
nee
dle
pene
trate
d he
r glo
ve.
21
£7
.69
Staf
f A
ccid
enta
l in
jury
N
eedl
estic
k/sh
arps
M
edic
al te
chni
cian
sust
aine
d cu
t fro
m S
tanl
ey k
nife
, whi
ch h
e w
as u
sing
to c
ut
suct
ion
tubi
ng.
Kni
fe sl
ippe
d, c
uttin
g fin
ger o
n le
ft ha
nd.
22
£21.
88
Staf
f A
ccid
enta
l in
jury
N
eedl
estic
k/sh
arps
R
adio
grap
her’
s ass
ista
nt su
stai
ned
need
lest
ick
inju
ry c
lear
ing
away
pro
cedu
res
trolle
y. W
hils
t fol
ding
dis
card
ed p
ad u
sed
by d
octo
r dur
ing
proc
edur
e, n
eedl
e pr
otru
ded
and
pric
ked
left
ring
finge
r.
174
Tabl
e 6
Trus
t F (c
ontro
l) (1
6 In
cide
nts)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £4
.62
Staf
f A
ccid
enta
l inj
ury
Man
ual l
iftin
g/ha
ndlin
g C
ater
ing
Supe
rvis
or su
stai
ned
a sh
ould
er in
jury
sorti
ng m
eat o
n tra
ys b
ecau
se th
e w
rong
ord
er o
f mea
t was
del
iver
ed to
the
Cat
erin
g D
ept (
bags
had
bee
n or
dere
d in
4
kg lo
ts).
2 £1
90.3
5 St
aff
Acc
iden
tal i
njur
y St
ruck
aga
inst
som
ethi
ng
(fur
nitu
re e
tc)
Cat
erin
g A
ssis
tant
cle
anin
g in
staf
f cof
fee
loun
ge tu
rned
to se
rve
a cu
stom
er a
nd
caug
ht h
er le
ft fo
ot o
n a
low
cof
fee
tabl
e, c
ausi
ng h
er to
lose
her
bal
ance
and
fall,
tw
istin
g he
r rig
ht k
nee.
3
£236
.71
Staf
f A
ccid
enta
l inj
ury
Cut
with
shar
p m
ater
ial/o
bjec
t A
coo
k w
as c
uttin
g br
ead
into
cro
uton
s usi
ng a
bre
ad k
nife
. B
ecam
e di
stra
cted
and
cu
t fin
ger o
n le
ft ha
nd.
4 £8
0.18
St
aff
Acc
iden
tal i
njur
y St
ruck
aga
inst
som
ethi
ng
(fur
nitu
re e
tc)
Staf
f Nur
se su
stai
ned
head
inju
ry in
col
lisio
n w
ith a
cei
ling
mou
nted
mic
rosc
ope
in
thea
tre.
5 £1
67.5
5 St
aff
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t St
aff N
urse
sust
aine
d fa
cial
/nec
k in
jurie
s in
viol
ent i
ncid
ent w
ith a
con
fuse
d pa
tient
(p
ost o
pera
tive
hallu
cina
tions
/par
anoi
a).
6 £1
57.4
5 St
aff
Acc
iden
tal i
njur
y M
anua
l lift
ing/
hand
ling
A d
omes
tic a
ssis
tant
sust
aine
d ba
ck in
jury
lifti
ng a
grid
from
a sh
ower
bas
e.
7 £1
49.6
0 St
aff
Acc
iden
tal i
njur
y Sl
ip/tr
ip o
r fal
l on
the
sam
e le
vel o
r sta
irs
Dom
estic
ass
ista
nt sp
rain
ed a
nkle
trip
ping
ove
r vac
uum
hos
e w
hile
wor
king
in
thea
tre.
Did
not
tidy
aw
ay v
acuu
m c
lean
er in
dom
estic
serv
ices
room
. W
ent t
o do
an
othe
r tas
k an
d su
bseq
uent
ly tr
ippe
d ov
er th
e ho
se o
f the
vac
uum
cle
aner
. 8
£40.
00
Oth
er
Acc
iden
tal
prop
erty
lo
ss/d
amag
e
Oth
er
A re
vers
ing
van
driv
er st
ruck
ext
erna
lly p
roje
ctin
g w
ard
win
dow
.
9 £8
1.93
St
aff
Acc
iden
tal i
njur
y St
ruck
by
an o
bjec
t St
aff N
urse
was
dis
conn
ectin
g an
airl
ine
from
the
wal
l in
the
HD
U u
nit.
Bui
ld u
p of
pre
ssur
e w
as su
ffic
ient
to p
ush
the
conn
ecto
r off
its s
eatin
g an
d w
as h
it on
the
chin
by
the
met
al p
oint
. 10
£9
0.56
St
aff
Acc
iden
tal i
njur
y M
anua
l lift
ing/
hand
ling
Cle
rical
Offi
cer s
usta
ined
a b
ack
inju
ry re
cove
ring
note
s fro
m a
hig
h sh
elf.
11
£14.
37
Staf
f A
ccid
enta
l inj
ury
Man
ual l
iftin
g/ha
ndlin
g A
n au
xilia
ry n
urse
was
pus
hing
an
incu
bato
r tro
lley
alon
g th
e co
rrid
or to
a c
lean
ing
room
in th
e N
eona
tal U
nit.
The
nur
se p
ushe
d th
roug
h fir
e do
ors,
mis
judg
ed th
e an
gle
of th
e in
cuba
tor a
nd c
augh
t one
of t
he la
rge
whe
els o
n th
e ed
ge o
f the
doo
r.
The
incu
bato
r rec
oile
d an
d th
e au
xilia
ry n
urse
hit
her h
ead
on th
e Pe
rspe
x ho
od.
12
£265
.68
Staf
f A
ccid
enta
l inj
ury
Patie
nt li
fting
/han
dlin
g St
aff N
urse
sust
aine
d ne
ck in
jury
tran
sfer
ring
a pa
tient
in a
whe
elch
air.
13
£17.
63
Staf
f A
ccid
enta
l inj
ury
Stru
ck a
gain
st so
met
hing
(f
urni
ture
etc
) Po
rter s
usta
ined
hea
d in
jury
bum
ping
aga
inst
incu
bato
r box
han
dle,
whi
ch w
as
sitti
ng o
n a
tabl
e in
the
porte
r’s o
ffic
e.
14
£592
.75
Staf
f A
ccid
enta
l inj
ury
Con
tact
with
Fi
nanc
e D
ept w
as m
ovin
gpr
emis
es a
nd th
e Fi
nanc
ial A
ssis
tant
was
ass
istin
g in
the
175
equi
pmen
t/mac
hine
ry
co-o
rdin
atio
n of
the
mov
ing
of fi
les a
nd e
quip
men
t bet
wee
n of
fices
. Su
stai
ned
a fo
ot c
rush
inju
ry o
n va
n ta
ilgat
e.
15
£17.
01
Staf
f A
ccid
enta
l inj
ury
Man
ual l
iftin
g/ha
ndlin
g Su
perin
tend
ent R
adio
grap
her s
usta
ined
leg
inju
ry m
ovin
g le
ad a
pron
trol
ley.
16
£2
,091
.20
Patie
nt
Fata
lity
Fall
from
hei
ght
Patie
nt su
stai
ned
head
inju
ry in
fall
from
nar
row
scre
enin
g ta
ble.
Tabl
e 7
Trus
t G (c
ontro
l) (1
0 In
cide
nts)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £4
50.8
0 St
aff
Acc
iden
tal
inju
ry
Stru
ck b
y an
obj
ect
Staf
f Nur
se su
stai
ned
head
inju
ry w
hen
stru
ck b
y fa
lling
ligh
t fitt
ing.
Fol
low
ing
the
inci
dent
, Sta
ff N
urse
giv
en T
etan
us in
ject
ion,
whi
ch le
d to
an
adve
rse
phys
iolo
gica
l re
actio
n, re
sulti
ng in
ext
ende
d si
ck le
ave.
2
£22.
00
Staf
f Ph
ysic
Pe
rson
to p
erso
n as
saul
t C
harg
e N
urse
sust
aine
d a
hum
an b
ite to
arm
dur
ing
cont
rol a
nd re
stra
int p
roce
dure
.Tr
ansf
erre
d to
Acu
te T
rust
A&
E D
ept f
or tr
eatm
ent.
3 £1
2.13
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Enro
lled
Nur
se d
ispo
sing
of I
V fl
uids
into
the
sink
of t
he T
reat
men
t Roo
m.
Nee
dle
fell
into
the
sink
and
, on
pick
ing
it up
, the
nur
se su
stai
ned
a ne
edle
stic
k in
jury
to h
er
finge
r. 4
£263
.10
Staf
f A
ccid
enta
l in
jury
C
onta
ct w
ith
equi
pmen
t/mac
hine
ry
Enro
lled
Nur
se su
stai
ned
arm
inju
ry w
hen
a co
mm
ode
topp
led,
afte
r a sc
rew
cam
e ou
t of a
cas
tor a
nd le
ft fr
ont w
heel
snap
ped
off.
5 £1
2.13
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
An
auxi
liary
nur
se w
as fe
edin
g a
patie
nt in
bed
whi
le a
Sta
ff N
urse
was
si
mul
tane
ousl
y ad
min
iste
ring
a flu
vac
cine
. Th
e au
xilia
ry n
urse
’s a
rm w
as ra
ised
an
d, a
s the
nee
dle
was
with
draw
n it
punc
ture
d he
r lef
t for
earm
. 6
£40.
00
Patie
nt
Oth
er
Oth
er
Lear
ning
dis
abili
ties p
atie
nt b
roke
a w
indo
w d
urin
g ra
mpa
ge.
A d
rug
erro
r fol
low
ed
in th
e en
suin
g co
nfus
ion.
7
£2,8
71.7
5 St
aff
Acc
iden
tal
inju
ry
Patie
nt li
fting
/han
dlin
g St
aff N
urse
sust
aine
d in
jury
to b
ack
whi
le a
ssis
ting
patie
nt.
Patie
nt c
ould
n’t b
e tra
nsfe
rred
usi
ng h
oist
/slin
gs a
s she
was
a st
roke
vic
tim a
nd v
ery
limite
d in
abi
lity,
an
d te
nded
to g
rab
and
thro
w h
erse
lf fo
rwar
d.
8 £6
60.0
2 St
aff
Acc
iden
tal
inju
ry
Pers
on to
per
son
assa
ult
Nur
sing
Ass
ista
nt st
ruck
on
brid
ge o
f nos
e by
pla
te th
row
n by
a p
atie
nt.
9 £3
9.36
Pa
tient
A
ccid
enta
l in
jury
Sl
ip/tr
ip o
r fal
l on
the
sam
e le
vel o
r sta
irs
Patie
nt su
stai
ned
hip
inju
ry a
fter f
allin
g w
hils
t bei
ng e
scor
ted
to th
e to
ilet.
10
£2,4
50.0
0 St
aff
Acc
iden
tal
inju
ry
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
Te
leph
onis
t sus
tain
ed fr
actu
red
ankl
e tri
ppin
g on
a h
ole
in p
rote
ctiv
e vi
nyl m
at.
176
PHA
SE 2
Tabl
e 8
Trus
t A (2
7 In
cide
nts)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £2
1.42
St
aff
Ver
bal a
buse
or
thre
aten
ing
beha
viou
r O
ther
C
omm
unity
Nur
se th
reat
ened
by
son
of p
atie
nt.
The
nurs
e be
ing
thre
aten
ed v
erba
lly d
e-es
cala
ted
the
situ
atio
n, b
ut th
e vi
sit w
as e
xten
ded
by a
n ho
ur a
s a re
sult.
2
£16.
08
Patie
nt
Fire
Ex
posu
re to
fire
Ps
ychi
atric
pat
ient
set f
ire to
pap
er to
wel
s, th
us c
ausi
ng fi
re a
larm
to so
und.
3
£280
.72
Patie
nt
Van
dalis
m
Expo
sure
to fi
re
Patie
nt se
t fire
to b
eddi
ng a
nd m
attre
ss in
thei
r roo
m.
Fire
ala
rm a
ctiv
ated
by
smok
e.
4 £1
37.1
0 Pa
tient
A
ccid
enta
l inj
ury
Oth
er
An
old
wou
nd o
pene
d on
fore
arm
dur
ing
stru
ggle
with
pat
ient
. 5
£19.
67
Patie
nt
Fata
lity
Oth
er
Patie
nt fo
und
hang
ing,
usi
ng a
rope
susp
ende
d fr
om a
win
dow
cat
ch.
Iden
tifie
d as
a
high
risk
, but
had
not
yet
bee
n ac
tione
d.
6 £1
9.04
St
aff
Ver
bal a
buse
or
thre
aten
ing
beha
viou
r O
ther
N
urse
Tea
m L
eade
r ver
bally
abu
sed
and
phys
ical
ly th
reat
ened
whe
n he
ask
ed tw
o m
ale
inpa
tient
s to
hand
ove
r alc
ohol
they
wer
e fo
und
to b
e co
nsum
ing.
7
£6.7
2 Pa
tient
O
ther
O
ther
Pa
tient
det
aine
d un
der M
HA
but
hou
sed
in a
n ac
ute
psyc
hiat
ric w
ard,
bec
ame
a nu
isan
ce to
the
neig
hbor
ing
war
d du
e to
thei
r con
tinua
lly le
avin
g th
e w
ard.
De-
esca
latio
n m
easu
res f
aile
d, re
sulti
ng in
pat
ient
rest
rain
t by
3 st
aff a
nd re
turn
ed to
war
d.
8 £1
5.47
Pa
tient
Ph
ysic
al v
iole
nce
Oth
er
A 5
3 ye
ar o
ld fe
mal
e pa
tient
with
lear
ning
dis
abili
ties l
eft w
ard
to w
alk
cons
ider
able
dist
ance
hom
e. R
etur
ned
to w
ard
with
staf
f. A
ggre
ssiv
e on
retu
rn, r
equi
red
to b
e re
stra
ined
. 9
£11.
68
Patie
nt
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t A
53
year
old
fem
ale
patie
nt le
ft w
ard
thre
aten
ing
suic
ide.
Hou
sed
in a
cute
psy
chia
tric
war
d w
here
staf
f do
not r
outin
ely
wor
k w
ith th
is c
lient
gro
up.
Foun
d he
r pro
blem
atic
to
man
age.
10
£9
8.88
Pa
tient
V
erba
l abu
se o
r th
reat
enin
g be
havi
our
Oth
er
Patie
nt w
ith le
arni
ng d
isab
ilitie
s abs
cond
ed fr
om a
cute
psy
chia
tric
war
d. N
o st
aff w
ith
train
ing
in le
arni
ng d
isab
ilitie
s. 11
£9
.13
Patie
nt
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t A
53-
year
old
fem
ale
patie
nt w
ith le
arni
ng d
isab
ilitie
s det
aine
d in
acu
te p
sych
iatri
c w
ard.
12
£23.
80
Staf
f Ph
ysic
al v
iole
nce
Pers
on to
per
son
assa
ult
Nur
sing
aux
iliar
y bi
tten
by p
atie
nt.
Had
rece
ived
no
viol
ence
and
agg
ress
ion
train
ing.
13
£429
.65
Staf
f Ph
ysic
al v
iole
nce
Pers
on to
per
son
assa
ult
Nur
sing
aux
iliar
y le
ft on
her
ow
n in
unf
amili
ar a
rea
assa
ulte
d by
pat
ient
. N
ursi
ng
auxi
liary
not
trai
ned
in v
iole
nce
& a
ggre
ssio
n.
14
£2.6
1 St
aff
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t M
ale
patie
nt w
ith le
arni
ng d
iffic
ultie
s bec
ame
phys
ical
ly a
busi
ve to
war
ds fe
llow
pa
tient
. W
hen
staf
f int
erve
ned
patie
nt b
ecam
e ph
ysic
ally
abu
sive
tow
ards
them
. A
ssis
tanc
e re
quire
d to
rest
rain
the
patie
nt fo
llow
ing
the
atta
ck.
177
15
£5.1
5 St
aff
Phys
ical
vio
lenc
e O
ther
A
pat
ient
requ
ired
rest
rain
ing
afte
r thr
eate
ning
phy
sica
l vio
lenc
e. A
larm
act
ivat
ed fo
ras
sist
ant t
o in
crea
se n
umbe
rs fo
r res
train
t. 16
£3
.85
Staf
f Ph
ysic
al v
iole
nce
Pers
on to
per
son
assa
ult
Occ
upat
iona
l The
rapi
st b
itten
by
patie
nt w
ith le
arni
ng d
isab
ilitie
s.
17
£1,2
85.8
9 St
aff
Oth
er
Oth
er
Wor
k re
late
d st
ress
in n
ursi
ng a
uxili
ary
thou
ght t
o be
rela
ted
to u
nsta
ble
man
agem
ent
arra
ngem
ents
ove
r ens
uing
per
iod
follo
win
g m
ove
to n
ew p
rem
ises
. O
n in
terv
iew
, ot
her s
taff
wor
king
in th
e ar
ea e
xpre
ssed
feel
ings
of w
ork
rela
ted
stre
ss.
18
£3,3
85.0
2 St
aff
Ill h
ealth
O
ther
St
aff a
bsen
t fro
m w
ork
due
to p
rolo
nged
wor
k re
late
d st
ress
. U
nabl
e to
cop
e an
y lo
nger
w
ith p
atie
nt w
ith se
vere
beh
avio
ural
pro
blem
s, ru
nnin
g al
ongs
ide
staf
fing
prob
lem
s pr
ecip
itate
d by
dis
cipl
inar
y ac
tion
resu
lting
in d
ism
issa
ls/re
sign
atio
ns a
s wel
l as a
co
mpl
ete
chan
ge in
env
ironm
ent f
or p
atie
nts a
nd st
aff o
n ne
w p
rem
ises
a y
ear
prev
ious
ly.
19
£3.8
8 St
aff
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t St
aff r
equi
ring
assi
stan
ce to
sepa
rate
2 le
arni
ng d
isab
ilitie
s pat
ient
s fig
htin
g.
20
£7,2
62.7
8 St
aff
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t St
aff N
urse
kic
ked
by p
atie
nt w
hile
tryi
ng to
pro
tect
ano
ther
(Sta
ff N
urse
had
not
had
vi
olen
ce &
agg
ress
ion
train
ing)
. 21
£3
.60
Oth
er
Oth
er
Oth
er
Fire
ala
rm a
ctiv
ated
by
burn
t toa
st.
Kitc
hen
door
hel
d op
en, s
o sm
oke
able
to e
scap
e fr
om k
itche
n. P
atie
nt u
nsup
ervi
sed,
des
pite
pol
icy
that
all
toas
t mak
ing
shou
ld b
e su
perv
ised
. 22
£4
.59
Patie
nt
Oth
er
Oth
er
A 4
5 ye
ar o
ld fe
mal
e de
tain
ed p
atie
nt a
bsco
nded
from
war
d, h
ad to
be
rest
rain
ed to
stop
he
r fro
m g
ettin
g ou
t of m
ovin
g ca
r.
23
£1,5
94.3
2 St
aff
Phys
ical
vio
lenc
e Pa
tient
lif
ting/
hand
ling
Nur
se in
jure
d ne
ck a
nd sh
ould
er d
urin
g pa
tient
rest
rain
t.
24
£120
.00
Patie
nt
Phys
ical
vio
lenc
e O
ther
A
41
year
old
det
aine
d pa
tient
trie
d to
bre
ak d
own
fire
door
with
a fi
re e
xtin
guis
her.
25
£5.4
8 Pa
tient
N
ear m
iss
Oth
er
An
80 y
ear o
ld p
atie
nt a
bsco
nded
from
psy
cho-
geria
tric
war
d.
26
£2.4
3 St
aff
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t St
aff N
urse
ass
aulte
d in
caf
é by
a m
an w
ho se
lf pr
esen
ted,
ask
ing
to se
e a
doct
or.
27
£10.
08
Staf
f A
ccid
enta
l inj
ury
Cut
with
shar
p m
ater
ial/o
bjec
t N
ursi
ng A
uxili
ary
drop
ped
plat
e w
hen
clea
ning
tabl
e af
ter p
atie
nt’s
lunc
h. T
he p
late
w
as b
roke
n. N
ursi
ng A
uxili
ary
cut f
inge
rs w
hen
pick
ing
up b
roke
n pi
eces
.
178
Tabl
e 9
Trus
t B (1
1 In
cide
nts)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £5
.50
Staf
f A
ccid
enta
l in
jury
N
eedl
estic
k/sh
arps
N
urse
suff
ered
nee
dles
tick
whi
le a
ttend
ing
patie
nt a
t hom
e. P
atie
nt m
oved
whi
lst t
ryin
g to
inse
rt ne
edle
and
pun
ctur
ed m
iddl
e fin
ger o
f non
-dom
inan
t lef
t han
d th
at w
as
posi
tione
d on
top
of th
e pa
tient
’s a
rm to
stea
dy it
. 2
£1,5
39.7
8 St
aff
Acc
iden
tal
inju
ry
Man
ual l
iftin
g/ha
ndlin
g Se
nior
Den
tal M
anag
er tr
ansf
errin
g eq
uipm
ent o
ut o
f the
bac
k of
his
car
whi
lst o
n da
y vi
sits
stra
ined
bac
k.
3 £4
24.5
5 St
aff
Acc
iden
tal
inju
ry
Stru
ck b
y an
obj
ect
Spee
ch a
nd L
angu
age
Ther
apis
t hit
on h
ead
by lo
ose
proj
ecto
r scr
een.
4 £9
57.0
0 St
aff
Acc
iden
tal
inju
ry
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
St
aff N
urse
resp
ondi
ng to
em
erge
ncy
alar
m ra
n in
to c
orrid
or a
rea.
As n
urse
ran
alon
g co
rrid
or fa
iled
to se
e w
arni
ng si
gn a
nd sl
ippe
d on
wet
floo
r, su
stai
ning
a tw
iste
d an
kle.
5
£12.
76
Staf
f Ph
ysic
al
viol
ence
Pe
rson
to p
erso
n as
saul
t O
ccup
atio
nal T
hera
py st
uden
t ass
aulte
d w
ith p
ool c
ue b
y cl
ient
with
his
tory
of
aggr
essi
ve b
ehav
iour
, alth
ough
ther
e w
as n
o w
arni
ng th
at th
e ou
tbur
st w
as a
bout
to ta
ke
plac
e. T
he te
nden
cy o
f the
clie
nt to
beh
ave
aggr
essi
vely
was
not
com
mun
icat
ed to
the
stud
ent O
T.
6 £7
,078
.50
Staf
f A
ccid
enta
l in
jury
Sl
ip/tr
ip o
r fal
l on
the
sam
e le
vel o
r sta
irs
Staf
f Nur
se w
ashi
ng h
er h
ands
in tr
eatm
ent r
oom
. W
ater
spill
age
durin
g ha
nd w
ashi
ng
led
to w
ater
on
floor
. N
urse
slip
ped
on fl
oor,
sust
aini
ng in
jury
to k
nee.
7
£3.7
9 St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Enro
lled
Nur
se a
ttem
ptin
g to
dis
pose
of a
use
d sh
arp
follo
win
g ad
min
istra
tion
of in
sulin
. A
s she
trie
d to
ope
n th
e lid
to th
e sh
arps
box
it st
uck.
She
tran
sfer
red
the
need
le fr
om
her r
ight
han
d to
left
hand
so sh
e co
uld
open
the
lid w
ith h
er ri
ght h
and
and
in d
oing
so
sust
aine
d ne
edle
stic
k in
jury
to h
er le
ft ha
nd.
8 £1
3.16
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Whi
lst a
dmin
iste
ring
inje
ctio
n to
pat
ient
, nur
se re
ceiv
ed a
shar
ps in
jury
.
9 £1
92.5
0 St
aff
Acc
iden
tal
inju
ry
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
St
aff N
urse
fell
on st
eps r
espo
ndin
g to
em
erge
ncy
alar
m.
10
£1,0
38.9
8 St
aff
Acc
iden
tal
inju
ry
Patie
nt li
fting
/land
ling
Hea
lthca
re A
ssis
tant
atte
mpt
ed to
bre
ak a
pat
ient
’s fa
ll, su
stai
ning
a b
ack
inju
ry.
11
£5.6
0 St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
GP
taki
ng b
lood
sam
ple
from
pat
ient
. O
n co
mpl
etio
n, p
asse
d us
ed n
eedl
e to
aux
iliar
y nu
rse
to d
ispo
se o
f in
shar
ps b
in.
She
caug
ht h
er 3
rd fi
nger
with
the
need
le.
179
Tabl
e 10
Tru
st C
(13
Inci
dent
s)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £1
08.7
5 St
aff
Acc
iden
tal
inju
ry
Hot
or c
old
cont
act
An
assi
stan
t coo
k w
as sc
alde
d by
soup
whe
n it
sudd
enly
bub
bled
up
and
spla
shed
on
to
her r
ight
han
d, sc
aldi
ng tw
o fin
gers
. 2
£9.7
4 St
aff
Acc
iden
tal
inju
ry
Stru
ck a
gain
st so
met
hing
(f
urni
ture
etc
) D
omes
tic w
as p
ullin
g tro
lley
from
din
ing
room
into
kitc
hen.
Whi
le g
oing
thro
ugh
door
way
she
caug
ht h
er ri
ght h
and
betw
een
the
door
fram
e an
d th
e tro
lley.
Thi
s res
ulte
d in
pai
n to
3rd
and
4th
fing
ers.
3 £4
00.0
0 St
aff
Nea
r mis
s C
onta
ct w
ith e
lect
ricity
C
harg
er fo
r den
tal t
rolle
y w
as n
ot w
orki
ng so
ano
ther
was
bor
row
ed fr
om su
rger
y ne
xt
door
. D
espi
te h
avin
g a
loos
e pi
n it
was
bei
ng ro
utin
ely
used
in th
e ot
her s
urge
ry.
Whe
n th
e de
ntal
nur
se p
lugg
ed in
the
char
ger t
he e
arth
pin
faile
d an
d th
e ch
arge
r ble
w o
ut o
f th
e so
cket
with
a lo
ud b
ang.
RC
D h
ad b
low
n an
d re
set.
4 £1
3.06
St
aff
Acc
iden
tal
inju
ry
Stru
ck b
y an
obj
ect
Secr
etar
y in
staf
f kitc
hen
open
ed a
cup
boar
d (u
pper
uni
t) do
or a
nd a
pla
te sl
ippe
d ou
t, st
rikin
g he
r on
the
uppe
r lip
. 5
£655
.84
Patie
nt
Acc
iden
tal
inju
ry
Fall
from
hei
ght
Nin
ety
year
old
pat
ient
fell
out o
f bed
.
6 £5
.80
Patie
nt
Acc
iden
tal
inju
ry
Fall
from
hei
ght
Stro
ke p
atie
nt o
verc
ame
cot s
ides
and
fell
to fl
oor.
7 £1
,684
.24
Staf
f A
ccid
enta
l in
jury
Sl
ip/tr
ip o
r fal
l on
the
sam
e le
vel o
r sta
irs
Nur
sing
Ass
ista
nt sl
ippe
d on
wet
car
pet,
sust
aini
ng a
hea
d in
jury
.
8 £9
8.40
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Staf
f Nur
se su
stai
ned
need
lest
ick
inju
ry w
hile
dis
posi
ng o
f lan
cet a
fter c
heck
ing
patie
nt’s
gl
ucos
e.
9 £3
28.8
8 St
aff
Acc
iden
tal
inju
ry
Patie
nt li
fting
/han
dlin
g En
rolle
d N
urse
sust
aine
d lo
wer
bac
k in
jury
ben
ding
to a
ssis
t pat
ient
who
had
bee
n si
tting
on
com
mod
e.
10
£10.
28
Patie
nt
Acc
iden
tal
inju
ry
Fall
from
hei
ght
Patie
nt fe
ll to
floo
r fro
m c
hair,
sust
aini
ng h
ead
inju
ry.
Ref
erre
d to
Acu
te T
rust
A&
E fo
r tre
atm
ent.
11
£6.9
9 Pa
tient
O
ther
O
ther
Pa
tient
cho
ked
on si
ngle
text
ure
mea
l (tu
na a
nd v
eg).
12
£9.3
1 Pa
tient
N
ear m
iss
Oth
er
Patie
nt lo
cked
in to
ilet a
fter l
ocki
ng d
oor,
and
coul
d no
t unl
ock
door
. 13
£2
60.4
5 St
aff
Acc
iden
tal
inju
ry
Man
ual l
iftin
g/ha
ndlin
g Po
rter i
njur
ed b
ack
pick
ing
up ta
ilgat
e of
trai
ler.
180
Tabl
e 11
Tru
st D
(34
Inci
dent
s)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £1
0.08
St
aff
Acc
iden
tal i
njur
y N
eedl
estic
k/sh
arps
D
omes
tic A
ssis
tant
was
mop
ping
the
floor
in c
asua
lty ro
om.
She
notic
ed a
pi
ece
of th
read
on
the
floor
and
pic
ked
it up
. Sh
e di
d no
t rea
lise
that
a su
ture
ne
edle
was
atta
ched
and
this
pie
rced
her
glo
ve a
nd h
er sk
in.
2
£91.
12
Staf
f A
ccid
enta
l inj
ury
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
C
ook
slip
ped
on w
et fl
oor s
usta
inin
g ba
ck in
jury
. Th
e w
ater
may
hav
e co
me
from
a n
earb
y di
shw
ashi
ng a
rea.
3
£2.4
0 St
aff
Acc
iden
tal i
njur
y C
onta
ct w
ith
equi
pmen
t/mac
hine
ry
Coo
k su
stai
ned
cut f
inge
r try
ing
to re
mov
e m
etal
stor
age
bin
from
rack
.
4 £1
9.17
St
aff
Acc
iden
tal i
njur
y C
onta
ct w
ith
equi
pmen
t/mac
hine
ry
Ass
ista
nt c
ook
sust
aine
d bu
rn w
hils
t lig
htin
g de
ep fa
t fry
er.
5 £2
.80
Patie
nt
Acc
iden
tal i
njur
y St
ruck
aga
inst
som
ethi
ng
(fur
nitu
re e
tc)
Elde
rly c
onfu
sed
patie
nt b
eing
take
n ho
me
by b
us fr
om th
e da
y ho
spita
l. S
eat
belts
not
fitte
d to
the
bus.
Whe
n bu
s sto
pped
at t
raff
ic li
ghts
, pat
ient
stoo
d up
an
d as
bus
mov
ed o
ff a
gain
pat
ient
fell
back
war
ds, s
triki
ng th
eir h
ead
of th
e lif
t at
the
back
of t
he b
us.
6 £6
7.42
St
aff
Acc
iden
tal i
njur
y C
ut w
ith sh
arp
mat
eria
l/obj
ect
Nur
sing
Ass
ista
nt su
stai
ned
cut t
o fin
ger o
peni
ng ti
n.
7 £3
7.60
St
aff
Acc
iden
tal i
njur
y St
ruck
aga
inst
som
ethi
ng
(fur
nitu
re e
tc)
Dom
estic
Ass
ista
nt w
as v
acuu
min
g in
an
offic
e an
d th
e va
cuum
cle
aner
cau
ght
agai
nst a
fold
ing
tabl
e th
at w
as p
ropp
ed u
p ag
ains
t the
wal
l and
the
tabl
e fe
ll on
to th
e do
mes
tic c
ausi
ng in
jury
to h
er le
g an
d an
kle.
8
£2.3
4 St
aff
Acc
iden
tal i
njur
y C
onta
ct w
ith
equi
pmen
t/mac
hine
ry
Dom
estic
inju
red
thum
b ch
angi
ng b
rush
es o
n “r
otaw
ash”
.
9 £7
.76
Oth
er
Nea
r mis
s O
ther
W
ater
leak
into
off
ice
due
to ta
nk c
lean
ing
proc
ess.
10
£48.
56
Oth
er
Oth
er
Oth
er
Vis
itor d
rivin
g a
high
-topp
ed v
an a
ttem
pted
to d
rive
unde
r an
arch
way
lead
ing
to a
n in
ner c
ourty
ard
of th
e ho
spita
l. T
he v
ehic
le st
ruck
the
over
head
arc
hway
. 11
£8
.72
Patie
nt
Oth
er
Oth
er
A le
gally
det
aine
d pa
tient
abs
cond
ed to
a lo
cal b
ar.
Two
mem
bers
of s
taff
co
llect
ed h
im a
nd d
rove
him
bac
k in
the
hosp
ital c
ar.
12
£9.2
0 Pa
tient
O
ther
O
ther
Le
gally
det
aine
d pa
tient
abs
cond
s to
loca
l bar
aga
in.
13
£4.7
0 St
aff
Acc
iden
tal i
njur
y N
eedl
estic
k/sh
arps
D
omes
tic A
ssis
tant
was
in tr
eatm
ent r
oom
cle
anin
g th
e si
nk.
She
did
not
notic
e a
need
le st
uck
in th
e dr
ain
and
it pu
nctu
red
her r
ight
mid
dle
finge
r th
roug
h th
e gl
ove.
14
£9
.84
Patie
nt
Ver
bal a
buse
or
thre
aten
ing
Oth
er
A p
sych
otic
teen
ager
acc
esse
d a
roof
via
a b
roke
n w
indo
w in
the
bedr
oom
.
181
beha
viou
r 15
£2
7.40
St
aff
Acc
iden
tal i
njur
y N
eedl
estic
k/sh
arps
H
ealth
Vis
itor s
usta
ined
nee
dles
tick
inju
ry w
hen
dist
ract
ed b
y pa
tient
’s
sibl
ings
. 16
£2
,688
.00
Staf
f A
ccid
enta
l inj
ury
Patie
nt li
fting
/han
dlin
g N
ursi
ng A
uxili
ary
sust
aine
d el
bow
inju
ry u
sing
hyd
raul
ic h
oist
(ele
ctric
hoi
st
awai
ting
parts
for r
epai
rs).
17
£42.
00
Staf
f A
ccid
enta
l inj
ury
Patie
nt li
fting
/han
dlin
g N
ursi
ng A
uxili
ary
sust
aine
d ba
ck in
jury
ass
istin
g pa
tient
to u
ndre
ss.
Giv
en th
e hi
gh d
epen
denc
y le
vel o
f the
pat
ient
, a se
cond
nur
se sh
ould
hav
e be
en p
rese
nt
to a
ssis
t. 18
£7
98.0
0 St
aff
Acc
iden
tal i
njur
y Pa
tient
lifti
ng/h
andl
ing
Hea
lthca
re A
ssis
tant
sust
aine
d ne
ck in
jury
dur
ing
patie
nt h
oist
tran
sfer
. Pa
tient
be
cam
e un
coop
erat
ive
durin
g th
e tra
nsfe
r and
the
HA
inte
rven
ed to
supp
ort t
he
patie
nt to
pre
vent
him
from
falli
ng to
the
floor
. 19
£1
43.5
5 St
aff
Acc
iden
tal i
njur
y Pa
tient
lifti
ng/h
andl
ing
Patie
nt b
eing
ass
iste
d in
the
toile
t by
2 m
embe
rs o
f sta
ff.
Patie
nt b
ecam
e w
eak
at th
e kn
ees a
nd w
as su
ppor
ted
by a
mem
ber o
f sta
ff.
Whi
lst a
ttem
ptin
g to
br
eak
the
patie
nt’s
fall,
the
Reh
ab A
ssis
tant
twis
ted
her b
ack.
20
£1
27.6
0 St
aff
Acc
iden
tal i
njur
y Pa
tient
lifti
ng/h
andl
ing
Nur
sing
Ass
ista
nt su
stai
ned
back
inju
ry a
ssis
ting
a pa
tient
out
of t
he b
ath.
The
pa
tient
turn
ed a
nd sl
ippe
d ca
usin
g bo
th n
urse
s to
supp
ort h
is w
eigh
t. 21
£8
.52
Staf
f A
ccid
enta
l inj
ury
Nee
dles
tick/
shar
ps
Nur
sing
Ass
ista
nt sc
ratc
hed
by u
sed
razo
r bla
de, u
sed
by a
n el
derly
stro
ke
patie
nt w
ho w
as sh
avin
g hi
mse
lf. T
he p
atie
nt a
ttem
pted
to c
ontin
ue sh
avin
g as
th
e N
ursi
ng A
ssis
tant
was
dry
ing
his f
ace
with
a to
wel
.
22
£4.7
2 St
aff
Acc
iden
tal i
njur
y N
eedl
estic
k/sh
arps
H
otel
Ser
vice
s Ass
ista
nt su
stai
ned
need
lest
ick
inju
ry fr
om n
eedl
e in
bin
bag
rem
oved
from
pat
ient
’s ro
om.
GP
had
take
n bl
ood
from
the
patie
nt a
nd le
ft sy
ringe
and
nee
dle
on a
tabl
e. T
he p
atie
nt h
ad p
ut th
ese
into
the
bin.
23
£5
66.9
1 Pa
tient
O
ther
O
ther
V
acci
ne a
dmin
istra
tion
erro
r to
infa
nt d
urin
g im
mun
isat
ion
clin
ic.
Staf
f ab
senc
e du
e to
stre
ss o
ver t
he e
rror
in a
dmin
iste
ring
MM
R v
acci
ne in
stea
d of
an
othe
r. 24
£6
04.8
0 St
aff
Acc
iden
tal i
njur
y Pa
tient
lifti
ng/h
andl
ing
Nur
sing
Aux
iliar
y in
jure
d ba
ck w
hils
t fitt
ing
inco
ntin
ence
pad
on
patie
nt.
25
£3.7
9 Pa
tient
N
ear m
iss
Con
tact
with
/exp
osur
e to
ha
rmfu
l sub
stan
ces
Lear
ning
dis
abili
ties p
atie
nt c
over
ed th
emse
lves
with
pai
nt.
26
£1,7
12.1
5 St
aff
Acc
iden
tal i
njur
y Sl
ip/tr
ip o
r fal
l on
the
sam
e le
vel o
r sta
irs
Cle
rical
Offi
cer s
lippe
d on
pol
ishe
d flo
or.
27
£19.
70
Staf
f Ph
ysic
al v
iole
nce
Pers
on to
per
son
assa
ult
Psyc
hotic
pat
ient
pun
ched
Sta
ff N
urse
dur
ing
asse
ssm
ent i
nter
view
.
28
£3.5
9 Pa
tient
O
ther
O
ther
Pa
tient
abs
cond
ed fr
om h
ospi
tal g
roun
ds, r
efus
ed to
retu
rn w
ith st
aff.
Pol
ice
wer
e no
tifie
d an
d re
turn
ed th
e pa
tient
to th
e w
ard
one
hour
late
r. 29
£5
.12
Patie
nt
Oth
er
Oth
er
Det
aine
d pa
tient
abs
cond
ed fr
om fi
re e
scap
e do
or le
ft un
lock
ed.
30
£5
6.80
Pa
tient
A
ccid
enta
l inj
ury
Slip
/trip
or f
all o
n th
e sa
me
Patie
nt fe
ll in
cor
ridor
.
182
leve
l or s
tairs
31
£9
0.51
Pa
tient
A
ccid
enta
l inj
ury
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
Pa
tient
fall
due
to Z
imm
er g
ettin
g ca
ught
on
cont
amin
ated
was
te b
in.
32
£134
.40
Staf
f A
ccid
enta
l inj
ury
Hot
or c
old
cont
act
Hea
lthca
re A
ssis
tant
scal
ded
hand
dur
ing
porr
idge
spill
age
follo
win
g se
rvin
g br
eakf
ast o
n th
e w
ards
from
the
trolle
y.
33
£33.
60
Staf
f A
ccid
enta
l inj
ury
Man
ual l
iftin
g/ha
ndlin
g N
ursi
ng A
uxili
ary
park
ing
a fo
od tr
olle
y ja
mm
ed th
eir f
inge
r bet
wee
n th
e w
all
and
the
trolle
y.
34
£317
.94
Staf
f Ill
hea
lth
Oth
er
Enro
lled
Nur
se c
ontra
cted
infe
ctio
n in
elb
ow fr
om sp
lash
of u
rine.
Tabl
e 12
Tru
st E
(16
Inci
dent
s)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £2
4.97
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Doc
tor s
usta
ined
nee
dles
tick
inju
ry st
retc
hing
to re
ach
shar
ps d
ispo
sal b
ox.
2 £1
8.70
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Nur
sing
Aux
iliar
y tid
ying
up
pick
ed u
p sh
arps
box
to c
heck
if it
was
full
and
sust
aine
d a
need
lest
ick
inju
ry fr
om n
eedl
e st
icki
ng o
ut to
p of
box
. 3
£2.8
9 St
aff
Acc
iden
tal
inju
ry
Stru
ck a
gain
st so
met
hing
(f
urni
ture
etc
) St
aff N
urse
retri
evin
g dr
essi
ng fr
om b
ase
unit
cupb
oard
hit
head
on
wor
ktop
as s
he
stoo
d up
. 4
£30.
12
Staf
f A
ccid
enta
l in
jury
N
eedl
estic
k/sh
arps
A
n an
aest
hetis
t sus
tain
ed n
eedl
estic
k du
e to
a c
ollis
ion
with
a c
olle
ague
wor
king
in
clos
e pr
oxim
ity.
5 £1
9.18
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
A S
enio
r Hou
se O
ffic
er a
ssis
ting
durin
g su
turin
g, su
stai
ned
a ne
edle
stic
k, w
orki
ng in
cl
ose
prox
imity
to th
e su
rgeo
n w
ho w
as st
itchi
ng.
6 £3
4.00
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Con
sulta
nt su
turin
g sk
in in
EN
T th
eatre
sust
aine
d ne
edle
stic
k in
jury
to p
alm
of l
eft
hand
.7
£19.
18
Staf
f A
ccid
enta
l in
jury
N
eedl
estic
k/sh
arps
Se
nior
Hou
se O
ffic
er a
ssis
ting
surg
eon
durin
g su
turin
g, re
ceiv
ed n
eedl
estic
k in
jury
.
8 £2
0.63
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Juni
or H
ouse
Off
icer
sust
aine
d sh
arps
inju
ry w
hils
t sw
abbi
ng w
ound
, ass
istin
g su
rgeo
n.
9 £9
62.5
2 Pa
tient
Fa
talit
y Sl
ip/tr
ip o
r fal
l on
the
sam
e le
vel o
r sta
irs
Con
fuse
d pa
tient
got
up
durin
g th
e ni
ght t
o vi
sit t
oile
t, an
d fe
ll by
her
bed
side
on
retu
rnin
g to
bed
. X
-ray
s mis
plac
ed, 9
day
s bef
ore
radi
olog
ist r
epor
ts o
n X
-ray
s and
re
ports
frac
ture
. 10
£6
.18
Staf
f A
ccid
enta
l in
jury
Pa
tient
lifti
ng/h
andl
ing
Inju
ry to
fing
er su
stai
ned
by st
aff n
urse
mov
ing
patie
nt u
p be
d.
11
£9.9
0 Pa
tient
Ill
hea
lth
Oth
er
Pres
sure
sore
s ide
ntifi
ed in
eld
erly
pat
ient
. N
o pr
essu
re re
lief m
easu
res.
183
12
£4.4
9 St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Nur
sing
Aux
iliar
y cu
t thu
mb
whi
lst w
ashi
ng ta
ble
knife
.
13
£7.9
6 St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Mid
wife
sust
aine
d su
ture
nee
dles
tick.
14
£9.0
8 St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Nee
dle
left
on tr
eatm
ent c
ouch
in O
utpa
tient
s. N
ursi
ng A
uxili
ary
pulle
d th
read
and
ne
edle
spra
ng u
p an
d w
ent i
nto
the
Nur
sing
Sis
ter s
tand
ing
in th
e vi
cini
ty.
15
£20.
19
Staf
f A
ccid
enta
l in
jury
N
eedl
estic
k/sh
arps
N
ursi
ng A
uxili
ary
sust
aine
d la
cera
tion
clea
ning
a k
nife
.
16
£17.
00
Staf
f A
ccid
enta
l in
jury
N
eedl
estic
k/sh
arps
St
aff N
urse
sust
aine
d ne
edle
stic
k du
ring
canu
latio
n of
col
leag
ue.
Tabl
e 13
Tru
st F
(con
trol)
(20
Inci
dent
s)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £2
,679
.84
Staf
f A
ccid
enta
l in
jury
Pa
tient
lifti
ng/h
andl
ing
Staf
f Nur
se su
stai
ned
back
inju
ry.
Lack
of s
lide
shee
t.
2 £1
9.03
St
aff
Phys
ical
vi
olen
ce
Pers
on to
per
son
assa
ult
Patie
nt sl
appe
d St
aff N
urse
in c
onfu
sed/
aggr
essi
ve o
utbu
rst.
3 £3
.82
Staf
f A
ccid
enta
l in
jury
C
ut w
ith sh
arp
mat
eria
l/obj
ect
Tech
nici
an su
stai
ned
a ha
nd in
jury
atte
mpt
ing
to o
pen
a ja
mm
ed d
oor.
4 £2
57.2
6 St
aff
Acc
iden
tal
inju
ry
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
St
erile
Ser
vice
s Tec
hnic
ian
Supe
rvis
or fe
ll du
e to
uns
tabl
e ha
ndra
il.
5 £5
,480
.55
Patie
nt
Acc
iden
tal
inju
ry
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
Tr
ansp
lant
pat
ient
trip
ped
in S
how
er R
oom
, sus
tain
ed a
frac
ture
d hi
p.
6 £8
56.9
2 St
aff
Acc
iden
tal
inju
ry
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
St
aff N
urse
sust
aine
d fr
actu
red
finge
r cha
ngin
g w
ater
can
iste
r on
wat
er c
oole
r mac
hine
.
7 £1
,932
.26
Staf
f A
ccid
enta
l in
jury
Pa
tient
lifti
ng/h
andl
ing
Clin
ical
Sup
port
Wor
ker s
usta
ined
shou
lder
inju
ry m
ovin
g pa
tient
up
bed.
8 £1
0.58
Pa
tient
O
ther
Fa
ll fr
om h
eigh
t B
aria
tric
patie
nt fa
ll. F
ire B
rigad
e re
quire
d to
lift
patie
nt b
ack
into
bed
. Pa
tient
lifte
d ba
ck to
bed
by
4 fir
emen
and
3 p
orte
rs w
ith th
e ai
d of
a sa
lvag
e sh
eet.
A h
oist
cap
able
of
lifti
ng th
e pa
tient
’s w
eigh
t cou
ld n
ot b
e lo
cate
d in
the
hosp
ital.
Wei
ght e
stim
ated
at
>200
kgs
. 9
£504
.05
Staf
f A
ccid
enta
l in
jury
Fa
ll fr
om h
eigh
t B
aria
tric
patie
nt fe
ll ou
t of b
ed.
Nur
se su
stai
ned
frac
ture
d w
rist.
6 F
irem
en a
nd 2
po
rters
man
ually
lifte
d th
e pa
tient
bac
k in
to b
ed.
184
10
£3.1
6 St
aff
Phys
ical
vi
olen
ce
Pers
on to
per
son
assa
ult
Agg
ress
ive
outb
urst
from
pat
ient
suff
erin
g fr
om a
lcoh
ol w
ithdr
awal
.
11
£5.6
0 St
aff
Phys
ical
vi
olen
ce
Pers
on to
per
son
assa
ult
Ban
k N
ursi
ng A
uxili
ary
assa
ulte
d by
pat
ient
suff
erin
g fr
om a
lcoh
ol w
ithdr
awal
.
12
£3.2
5 St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Nur
se su
stai
ned
need
lest
ick
inju
ry fr
om u
sed
lanc
et.
13
£14.
20
Staf
f A
ccid
enta
l in
jury
M
anua
l lift
ing/
hand
ling
Clin
ical
Sup
port
Wor
ker s
usta
ined
bac
k in
jury
pul
ling
out b
ack
rest
to m
ake
patie
nt
mor
e co
mfo
rtabl
e du
ring
the
nigh
t. 14
£5
.95
Patie
nt
Oth
er
Oth
er
Con
fuse
d pa
tient
goe
s mis
sing
from
med
ical
war
d.
15
£164
.70
Staf
f A
ccid
enta
l in
jury
Pa
tient
lifti
ng/h
andl
ing
Trai
nee
Clin
ical
Sup
port
Wor
ker s
usta
ined
bac
k in
jury
dur
ing
patie
nt tr
ansf
er.
16
£735
.90
Patie
nt
Fata
lity
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
C
onfu
sed
patie
nt su
stai
ned
frac
ture
d hi
p in
cor
ridor
of w
aitin
g ar
ea.
Follo
win
g su
rger
y to
the
frac
ture
, pat
ient
did
not
reco
ver f
rom
ope
ratio
n.
17
£4.9
2 St
aff
Phys
ical
vi
olen
ce
Pers
on to
per
son
assa
ult
Nur
se su
stai
ned
shou
lder
inju
ry d
urin
g m
ovin
g an
d ha
ndlin
g of
agg
ress
ive
patie
nt.
18
£5.7
1 St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Staf
f Nur
se su
stai
ned
need
lest
ick
inju
ry fr
om u
sed
need
le in
foil
tray.
19
£2.3
5 St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Trai
nee
Clin
ical
Sup
port
Wor
ker s
usta
ined
nee
dles
tick
inju
ry fr
om la
ncet
dis
card
ed o
n flo
or.
20
£43.
85
Staf
f A
ccid
enta
l in
jury
St
ruck
by
an o
bjec
t St
aff N
urse
sust
aine
d he
ad in
jury
from
falli
ng d
rip st
and.
Tabl
e 14
Tru
st G
(con
trol)
(10
Inci
dent
s)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £4
4.71
Pa
tient
A
ccid
enta
l in
jury
Fa
ll fr
om h
eigh
t Pa
tient
fell
out o
f bed
sust
aini
ng fr
actu
res.
Tra
nsfe
rred
to A
cute
Tru
st fo
r tre
atm
ent.
2 £7
5.20
St
aff
Acc
iden
tal
inju
ry
Slip
/trip
or f
all o
n th
e sa
me
leve
l or s
tairs
Po
rter s
lippe
d on
dar
k st
eps.
3 £1
9.60
St
aff
Acc
iden
tal
inju
ry
Cut
with
shar
p m
ater
ial/o
bjec
t In
the
Fore
nsic
Psy
chia
tric
War
d a
knife
had
gon
e m
issi
ng fo
llow
ing
was
hing
and
N
ursi
ng A
ssis
tant
thou
ght i
t had
falle
n un
der t
he d
ishw
ashe
r. P
ut h
is fi
nger
s und
er th
e m
achi
ne to
feel
if th
e kn
ife w
as th
ere.
Whi
lst d
oing
this
, sus
tain
ed a
cut
fing
er o
n a
shar
p lip
. 4
£116
.00
Patie
nt
Oth
er
Oth
er
Patie
nt a
bsco
nded
thro
ugh
a w
indo
w.
185
5 £1
2.13
St
aff
Acc
iden
tal
inju
ry
Nee
dles
tick/
shar
ps
Staf
f Nur
se su
stai
ned
a ne
edle
stic
k in
jury
dis
posi
ng o
f nee
dle
in fu
ll sh
arps
bin
.
6 £7
3.80
St
aff
Acc
iden
tal
inju
ry
Patie
nt li
fting
/han
dlin
g St
aff N
urse
on
light
dut
ies d
etai
led
to e
scor
t pat
ient
for X
-ray
. O
n tra
nsfe
rrin
g pa
tient
, St
aff N
urse
stra
ined
her
wea
ker a
rm.
7 £2
.46
Patie
nt
Oth
er
Oth
er
Patie
nt is
a sm
oker
and
hab
itual
ly le
aves
war
d to
do
this
. St
aff h
ave
to se
arch
for
patie
nt w
ho is
forg
etfu
l and
con
fuse
d, b
eing
trea
ted
for a
lcoh
ol a
buse
. 8
£37.
00
Patie
nt
Oth
er
Oth
er
Abs
cond
ed p
atie
nt fo
und
in ri
ver.
9 £1
,469
.46
Staf
f Ph
ysic
al
viol
ence
Pe
rson
to p
erso
n as
saul
t Fr
actu
red
wris
t app
lyin
g co
ntro
l and
rest
rain
t.
10
£63.
72
Patie
nt
Phys
ical
vi
olen
ce
Oth
er
Patie
nt re
stra
ined
on
grou
nds a
wai
ting
doct
or a
sses
smen
t.
186
APPE
ND
IX 8
SPE
CIA
L C
ATEG
OR
Y IN
CID
ENT
DES
CR
IPTI
VE D
ATA
PHA
SE 2
Tabl
e 1
Trus
t A (2
1inc
iden
ts)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £2
7.94
St
aff
Oth
er
Oth
er
Car
was
bei
ng d
riven
by
a C
omm
unity
Psy
chia
tric
Nur
se (C
PN) o
n a
road
, whi
ch w
as
cove
red
in sn
ow, b
ut h
ad re
cent
ly b
een
salte
d. C
PN a
ttem
pted
to tu
rn c
ar b
ut d
idn’
t re
spon
d. B
rake
s wor
ked
but t
he ty
res d
id n
ot g
rip a
nd c
ar w
ent o
ff th
e ro
ad.
Spec
ial
cate
gory
: out
with
rea
sona
ble
prac
ticab
ility
– T
rust
has
som
e co
ntro
l of d
rivin
g ab
ility
of
staf
f but
not
ove
r roa
d co
nditi
ons.
2 £7
.26
Patie
nt
Van
dalis
m
Oth
er
Thirt
y-tw
o ye
ar o
ld m
ale
patie
nt w
ith le
arni
ng d
isab
ilitie
s act
ivat
ed fi
re a
larm
out
side
his
ro
om –
no
fire.
Pat
ient
was
und
er g
ener
al o
bser
vatio
ns a
t the
tim
e. S
peci
al c
ateg
ory:
ou
twith
rea
sona
ble
prac
ticab
ility
– u
nder
gen
eral
obs
erva
tion
at ti
me,
whi
ch w
as
reas
onab
le, a
s the
pat
ient
had
not
bee
n pr
oble
mat
ic.
3 £7
.26
Patie
nt
Van
dalis
m
Oth
er
Patie
nt w
ith le
arni
ng d
isab
ilitie
s act
ivat
ed fi
re a
larm
for t
he se
cond
tim
e in
a d
ay.
Spec
ial
cate
gory
: out
with
rea
sona
ble
prac
ticab
ility
– O
nly
2 in
cide
nts o
f thi
s nat
ure
had
prev
ious
ly o
ccur
red
over
the
prev
ious
12
wee
k pe
riod.
Not
dee
med
by
Trus
t to
be
suff
icie
nt g
roun
ds to
war
rant
cos
ts o
f pur
chas
ing
alar
m c
over
s in
this
are
a.
4 £1
3.96
Pa
tient
O
ther
O
ther
Fi
fty-th
ree
year
old
fem
ale
patie
nt w
ith le
arni
ng d
isab
ilitie
s lef
t the
war
d an
d ho
spita
l gr
ound
s. P
olic
e w
ere
info
rmed
. St
aff s
earc
hed
for h
er in
car
, whe
n fo
und
she
initi
ally
re
fuse
d to
retu
rn b
ut w
as p
ersu
aded
. Sp
ecia
l cat
egor
y: C
linic
al ju
dgem
ent –
Clin
icia
n ju
dged
the
patie
nt’s
con
ditio
n w
ith re
fere
nce
to th
e de
gree
of s
uper
visi
on a
nd re
stric
tion
of
free
dom
. 5
£351
.63
Patie
nt
Van
dalis
m
Expo
sure
to fi
re
Patie
nt se
t fire
to a
shee
t in
show
er a
rea.
Als
o se
t fire
to sh
eet a
nd m
attre
ss o
n be
d.
Patie
nt h
ad h
er o
wn
light
er a
nd c
igar
ette
s – u
sed
light
er to
star
t fire
. Pa
tient
had
his
tory
of
self-
harm
but
no
hist
ory
of fi
re ra
isin
g. S
peci
al c
ateg
ory:
ris
k un
fore
seea
ble
– N
o hi
stor
y of
fire
rais
ing.
187
6 £3
,886
.48
Staf
f A
ccid
enta
l in
jury
Pa
tient
lifti
ng/
hand
ling
Nur
sing
Aux
iliar
y (N
A) w
as p
repa
ring
a re
side
nt fo
r bed
. Th
e cl
othi
ng th
e pa
tient
was
w
earin
g w
as ti
ght a
roun
d th
e ne
ck c
ausi
ng d
iffic
ulty
get
ting
it ov
er th
e pa
tient
’s h
ead.
Th
is c
ause
d th
e pa
tient
som
e di
stre
ss a
nd h
e su
dden
ly je
rked
cau
sing
the
NA
to su
stai
n a
brok
en fi
nger
. N
A w
as a
bsen
t fro
m w
ork
with
the
inju
ry.
Spec
ial c
ateg
ory:
lack
of
info
rmat
ion
– N
A in
pro
cess
of c
ivil
clai
m a
gain
st T
rust
so d
eclin
ed to
pro
vide
furth
er
info
rmat
ion
to p
roje
ct re
sear
cher
s. 7
£20.
00
Patie
nt
Acc
iden
tal
inju
ry
Stru
ck a
gain
st
som
ethi
ng
(fur
nitu
re e
tc)
Sixt
y-fiv
e ye
ar-o
ld m
ale
patie
nt st
umbl
ed a
nd fe
ll ba
ckw
ards
bre
akin
g th
e sa
fety
gla
ss in
th
e do
or.
Spec
ial c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– p
atie
nt in
depe
nden
tly
mob
ile, n
o hi
stor
y of
falls
, alre
ady
unde
r gen
eral
obs
erva
tion.
8
£42.
00
Oth
er
Van
dalis
m
Oth
er
Bro
ken
win
dow
s and
forc
ed w
indo
ws i
n su
mm
erho
use
notic
ed b
y si
te p
orte
r. E
stat
es
depa
rtmen
t boa
rded
up
and
secu
red
sum
mer
hous
e. G
lass
not
repl
aced
as h
ospi
tal d
ue to
be
rebu
ilt la
ter i
n ye
ar.
Spec
ial c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– a
s site
was
be
ing
dem
olis
hed
and
rede
velo
ped,
dec
isio
n w
as m
ade
to w
ork
with
pol
ice
to re
duce
risk
s by
pro
mpt
ly c
lean
ing
up a
nd se
curin
g ra
ther
than
incr
ease
secu
rity
sinc
e a
cost
/ben
efit
anal
ysis
had
sugg
este
d th
is w
as n
ot w
orth
whi
le.
9 £3
1.50
Pa
tient
Ph
ysic
al
viol
ence
Pe
rson
to p
erso
n as
saul
t Tw
o pa
tient
s on
acut
e ps
ychi
atric
war
d. F
emal
e pa
tient
thou
ght m
ale
patie
nt w
as la
ughi
ng
at h
er.
She
then
pro
ceed
ed to
atta
ck m
ale
patie
nt ri
ppin
g hi
s shi
rt in
the
proc
ess.
Spe
cial
cate
gory
: out
with
rea
sona
ble
prac
ticab
ility
– a
dequ
ate
cont
rol m
easu
res i
n pl
ace.
10
£5
.80
Patie
nt
Oth
er
Oth
er
Eigh
teen
-yea
r-ol
d fe
mal
e pa
tient
with
susp
ecte
d pa
race
tam
ol o
verd
ose.
Pat
ient
agr
eed
she
had
take
n ov
erdo
se a
nd b
lood
sam
ple
take
n to
labs
by
taxi
for a
naly
sis.
Spe
cial
cat
egor
y:
Clin
ical
judg
emen
t – st
aff h
ad b
een
cons
ider
ing
her d
isch
arge
and
she
was
bei
ng ta
ught
co
ping
stra
tegi
es a
nd h
ad b
een
allo
wed
out
for a
wal
k, a
s par
t of t
his r
ehab
ilita
tion.
11
£1
4.76
Pa
tient
Ph
ysic
al
viol
ence
Pe
rson
to p
erso
n as
saul
t Tw
o m
ale
patie
nts s
harin
g a
flat i
n a
reha
bilit
atio
n un
it. P
atie
nt A
bec
ame
anno
yed
by
Patie
nt B
and
rais
ed th
is w
ith h
im.
Patie
nt B
then
ges
ture
d an
d sw
ore
at P
atie
nt A
mak
ing
him
agi
tate
d an
d ag
gres
sive
. Pa
tient
A th
en lu
nged
at P
atie
nt B
hitt
ing
him
acr
oss t
he fa
ce
and
knoc
king
his
spec
tacl
es o
ff.
Inci
dent
unw
itnes
sed
by st
aff.
Spe
cial
cat
egor
y:
Clin
ical
judg
emen
t – n
ot h
eavi
ly su
perv
ised
, as t
his w
as a
reha
b. u
nit.
No
prev
ious
kn
owle
dge
of p
robl
em b
etw
een
patie
nts.
12
£14.
76
Patie
nt
Oth
er
Oth
er
Thirt
y-on
e-ye
ar-o
ld m
enta
l hea
lth p
atie
nt fo
und
to h
ave
take
n ov
erdo
se o
f par
acet
amol
by
nigh
t-dut
y nu
rse.
Pat
ient
had
his
tory
of s
elf-
harm
and
alc
ohol
abu
se.
Spec
ial c
ateg
ory:
C
linic
al ju
dgem
ent –
risk
ass
essm
ent i
ndic
ated
low
risk
. Th
e pa
tient
seem
ed w
ell t
hat
day
and
was
ther
efor
e on
ly u
nder
gen
eral
obs
erva
tion.
13
£6
.27
Patie
nt
Oth
er
Oth
er
Sixt
y-th
ree
year
old
mal
e pa
tient
adm
itted
2 d
ays p
revi
ousl
y fo
und
to h
ave
mul
tiple
la
cera
tions
infli
cted
with
his
ow
n ra
zor.
Spe
cial
cat
egor
y: r
isk
unfo
rese
eabl
e –
asse
ssed
as
not
bei
ng su
icid
al a
nd fa
mily
repo
rted
no h
isto
ry o
f sel
f har
m.
188
14
£4.2
0 Pa
tient
O
ther
O
ther
Pa
tient
det
aine
d un
der M
enta
l Hea
lth A
ct (M
HA
) abs
cond
ed w
hils
t on
esco
rted
wal
k.
Spec
ial c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– a
ll pr
oced
ures
follo
wed
, pho
ned
war
d im
med
iate
ly.
15
£4.0
2 O
ther
Fi
re
Oth
er
Patie
nt w
as sm
okin
g ci
gare
tte in
his
bed
room
, whi
ch tr
igge
red
the
fire
alar
m.
Patie
nt w
as
rem
inde
d ab
out w
ard
smok
ing
polic
y. S
peci
al c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– c
lient
men
tally
ill b
ut u
nder
stoo
d he
was
not
allo
wed
to sm
oke
in ro
om.
At n
ight
hou
rly c
heck
s wer
e do
ne b
ut p
atie
nt o
ther
wis
e ou
t of s
ight
dur
ing
nigh
t. 16
£1
1,47
7.71
Pa
tient
A
ccid
enta
l in
jury
Sl
ip, t
rip o
r fal
l on
leve
l or s
tairs
N
inet
y-on
e ye
ar o
ld fe
mal
e pa
tient
fell
retu
rnin
g to
bed
sust
aini
ng a
bro
ken
hip.
Pat
ient
ha
d an
ext
ende
d st
ay o
f six
ty-th
ree
days
. Sp
ecia
l cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
no
hist
ory
of fa
lls, w
as to
ld a
bout
usi
ng b
uzze
r whe
n ne
edin
g as
sist
ance
bu
t pat
ient
did
not
use
it.
17
£6.2
5 Pa
tient
V
anda
lism
O
ther
Pa
tient
not
ified
staf
f he
witn
esse
d an
othe
r pat
ient
bre
akin
g th
e gl
ass o
n th
e fir
e al
arm
, ac
tivat
ing
it. P
atie
nt w
as d
istu
rbed
at t
he ti
me
and
psyc
hotic
/ de
lusi
onal
. Sp
ecia
lca
tego
ry: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
. Occ
urre
nce
of o
nly
2 in
cide
nts o
ver t
he
prev
ious
12
wee
k pe
riod
not d
eem
ed to
be
suff
icie
nt to
war
rant
cos
ts o
f pur
chas
ing
alar
m
cove
rs in
this
are
a.
18
£6.0
0 Pa
tient
O
ther
O
ther
Pa
tient
cam
e ba
ck to
war
d af
ter h
avin
g a
pass
allo
win
g he
r to
be o
ut.
War
d st
aff w
ere
enga
ged
for a
n ho
ur in
war
d re
port.
Pat
ient
info
rmed
staf
f hal
f an
hour
afte
r ret
urn
to w
ard
that
she
had
take
n an
ove
rdos
e of
med
icat
ion
and
para
ceta
mol
she
had
boug
ht w
hen
out o
f w
ard.
Cos
t for
blo
ods t
o be
take
n by
taxi
to la
bs fo
r ana
lysi
s. S
peci
al c
ateg
ory:
Clin
ical
ju
dgem
ent –
the
risk
asse
ssm
ent i
ndic
ated
that
the
patie
nt w
as su
itabl
e fo
r lea
ve a
nd
disc
harg
e w
as b
eing
con
side
red.
19
£2
.91
Patie
nt
Oth
er
Oth
er
Fifty
-two
year
old
pat
ient
with
Alz
heim
er’s
Dis
ease
. Pa
tient
act
ivat
ed fi
re a
larm
. Sp
ecia
lca
tego
ry: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
pat
ient
und
er g
ener
al o
bser
vatio
n.
Diff
icul
t to
cont
rol s
uch
beha
viou
r but
fire
ala
rm n
eces
sary
. 20
£5
0.00
Pa
tient
Ph
ysic
al
viol
ence
O
ther
Pa
tient
, ver
y el
ated
, kic
ked
door
off
hin
ges i
n be
droo
m.
Spec
ial c
ateg
ory:
out
with
re
ason
able
pra
ctic
abili
ty –
pol
icy
was
to st
and
back
as l
ong
as n
o ris
k to
pat
ient
/oth
ers.
Pa
tient
in e
xcita
ble
stat
e, la
ngua
ge p
oor,
reas
onin
g di
ffic
ult,
ther
efor
e ap
prop
riate
act
ion
was
take
n.
21
£4.8
4 Pa
tient
O
ther
O
ther
45
-yea
r-ol
d pa
tient
det
aine
d un
der M
HA
on
clos
e 30
min
ute
obse
rvat
ions
atte
mpt
ed
suic
ide
from
cur
tain
rail.
Spe
cial
cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
m
easu
res i
n pl
ace
adeq
uate
and
all
follo
wed
.
189
Tabl
e 2
Trus
t B (5
inci
dent
s)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £6
.53
Staf
f A
ccid
enta
l in
jury
Sl
ip, t
rip o
r fa
ll on
leve
l or
stai
rs
Mem
ber o
f adm
inis
tratio
n st
aff t
rippe
d on
the
stai
rs su
stai
ning
a c
ut to
the
right
kne
e. S
peci
alca
tego
ry: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
con
ditio
n on
step
s goo
d, li
ghtin
g ad
equa
te,
non-
slip
lam
inat
ed fl
oorin
g. S
he sl
ippe
d on
met
al e
dgin
g of
stai
r, an
d no
oth
er c
ontri
buto
ry
fact
or c
ould
be
iden
tifie
d 2
£60.
50
Patie
nt
Phys
ical
vi
olen
ce
Pers
on to
pe
rson
ass
ault
Whe
n as
ked
to ti
dy ro
om p
atie
nt g
ot a
ngry
and
atta
cked
mem
ber o
f sta
ff.
Two
mem
bers
of
staf
f res
pond
ed to
pan
ic a
larm
and
pat
ient
was
con
trolle
d us
ing
cont
rol &
rest
rain
t (C
&R
) te
chni
ques
. D
urin
g th
e ph
ysic
al in
terv
entio
n th
e cl
ient
dam
aged
the
ribs o
f one
of t
he m
embe
rs
of st
aff.
Spe
cial
cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
all
prot
ocol
s wer
e ad
here
d to
an
d cl
ient
was
trea
ted
acco
rdin
g to
pol
icy.
3
£16.
50
Staf
f A
ccid
enta
l in
jury
Pe
rson
to
pers
on a
ssau
lt Pa
tient
in w
ard
beca
me
viol
ent t
owar
ds st
aff.
The
pat
ient
had
to b
e re
stra
ined
, whi
lst
perf
orm
ing
this
a st
aff n
urse
sust
aine
d a
wris
t inj
ury.
Spe
cial
cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
risk
ass
essm
ent i
n pl
ace,
all
staf
f tra
ined
to a
ppro
pria
te st
anda
rd.
4 £1
9.68
St
aff
Phys
ical
vi
olen
ce
Pers
on to
pe
rson
ass
ault
Patie
nt in
war
d be
cam
e vi
olen
t and
ass
aulte
d a
mem
ber o
f sta
ff.
Ano
ther
nur
se re
spon
ded
and
unde
rtook
C&
R to
con
trol t
he p
atie
nt.
Whi
lst d
oing
this
the
nurs
e su
stai
ned
an in
jury
to h
er
finge
r and
requ
ired
treat
men
t at A
&E.
Spe
cial
cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
ris
k as
sess
men
t had
bee
n un
derta
ken,
ala
rm a
ctiv
ated
as p
er p
roto
col a
nd a
ll st
aff w
ere
train
ed
to th
e in
-hou
se st
anda
rd.
5 £1
53.1
2 St
aff
Phys
ical
vi
olen
ce
Pers
on to
pe
rson
ass
ault
Patie
nt in
war
d be
cam
e ho
stile
and
phy
sica
lly a
ggre
ssiv
e, h
avin
g to
be
rest
rain
ed.
Whi
lst
Nur
sing
Ass
ista
nt w
as c
arry
ing
out C
&R
man
oeuv
re p
atie
nt m
anag
ed to
elb
ow h
er in
jurin
g rib
ca
ge o
n rig
ht si
de.
Spec
ial c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– a
ll th
ree
staf
f tra
ined
in C
&R
. A
lthou
gh p
atie
nt h
ad a
his
tory
of v
iole
nce,
an
appr
opria
te ri
sk a
sses
smen
t had
be
en c
arrie
d ou
t and
all
othe
r pro
cedu
res w
ere
follo
wed
app
ropr
iate
ly.
190
Tabl
e 3
Trus
t C (1
3 in
cide
nts)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £1
,537
.20
Staf
f A
ccid
enta
l in
jury
Pa
tient
lifti
ng/
hand
ling
Nur
sing
Ass
ista
nt c
ompl
aine
d sh
e ha
d hu
rt he
r bac
k w
hils
t lea
ning
ove
r bed
tend
ing
to
patie
nt, r
esul
ting
in si
ck le
ave.
Inc
iden
t unw
itnes
sed.
Spe
cial
cat
egor
y: la
ck o
f in
form
atio
n –
No
witn
esse
s to
inci
dent
and
cha
rge
nurs
e di
dn’t
ask
for f
urth
er in
fo w
hen
NA
repo
rted
in si
ck.
2 £7
8.93
St
aff
Acc
iden
tal
inju
ry
Patie
nt li
fting
/ ha
ndlin
g En
rolle
d nu
rse
settl
ing
a pa
tient
for t
he n
ight
was
adj
ustin
g th
e pa
tient
’s p
illow
whe
n sh
e fe
lt a
sudd
en sh
arp
pain
in h
er m
id to
upp
er b
ack.
Spe
cial
cat
egor
y: la
ck o
f inf
orm
atio
n –
nigh
t sis
ter c
ould
not
pro
vide
any
mor
e in
form
atio
n. E
N h
ad le
ft th
e Tr
ust.
3 £6
0.00
Pa
tient
V
anda
lism
O
ther
Pa
tient
bec
ame
aggr
essi
ve a
nd sm
ashe
d a
stoo
l off
a w
indo
w, c
rack
ing
inne
r pan
e of
w
indo
w.
Spec
ial c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– th
is in
cide
nt w
as c
ause
d by
frus
tratio
n on
the
part
of th
e pa
tient
. Diff
icul
t to
prev
ent w
hen
patie
nts h
ave
own
room
s. 4
£6.8
0 Pa
tient
Ph
ysic
al
viol
ence
St
ruck
aga
inst
so
met
hing
Tw
enty
-sev
en y
ear o
ld m
ale
patie
nt w
as to
ld h
e w
as to
be
deta
ined
und
er th
e M
HA
and
be
cam
e an
gry;
he
punc
hed
a w
all a
nd a
win
dow
(win
dow
did
n’t b
reak
). S
peci
al c
ateg
ory:
C
linic
al ju
dgem
ent –
dec
isio
n ta
ken
to le
t him
ven
t his
frus
tratio
n ra
ther
than
inte
rven
e ex
cept
as a
last
reso
rt.
5 £1
4.37
Pa
tient
O
ther
O
ther
Pa
tient
in M
enta
l Hea
lth A
cute
war
d ab
scon
ded.
Sta
ff N
urse
& N
A sp
ent o
ne h
our l
ooki
ng
for t
he p
atie
nt.
Spec
ial c
ateg
ory:
Clin
ical
judg
emen
t – a
form
al se
arch
pro
cedu
re e
xist
s bu
t was
app
lied
to so
me
patie
nts a
nd n
ot to
oth
ers,
base
d on
clin
ical
judg
emen
t of t
he st
aff.
6 £8
.69
Patie
nt
Acc
iden
tal
inju
ry
Slip
, trip
or f
all
on sa
me
leve
l or
stai
rs
Patie
nt w
ho w
as in
depe
nden
tly m
obile
with
his
rola
tor f
ell b
ackw
ards
in c
orrid
or fa
lling
to
floor
. Pa
tient
sust
aine
d a
frac
ture
to h
is sh
ould
er.
Spec
ial c
ateg
ory:
risk
unf
ores
eeab
le –
ris
k as
sess
men
t was
app
ropr
iate
. Th
ere
wer
e no
sign
s tha
t the
pat
ient
was
uns
tead
y, it
had
ne
ver h
appe
ned
befo
re.
7 £7
.39
Patie
nt
Acc
iden
tal
inju
ry
Slip
, trip
or f
all
on sa
me
leve
l or
stai
rs
Patie
nt g
ot o
ut o
f bed
and
fell
to th
e flo
or su
stai
ning
a la
cera
tion
to h
is h
ead,
hae
mat
oma
to
head
and
com
plai
ned
of p
ain
to h
is ri
ght h
ip.
X-r
ay c
onfir
med
frac
ture
to ri
ght n
eck
of
fem
ur.
Spec
ial c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– a
ll th
at c
ould
be
done
had
be
en im
plem
ente
d.
8 £3
5.00
Pa
tient
V
anda
lism
O
ther
El
derly
pat
ient
with
dem
entia
bec
ame
frus
trate
d an
d w
ante
d to
go
hom
e. S
taff
hea
rd
bang
ing
in th
e w
ard.
Whe
n N
A a
ttend
ed, p
atie
nt w
as st
rikin
g a
smal
l win
dow
pane
with
her
zi
mm
er, c
ausi
ng it
to c
rack
. Sp
ecia
l cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
eve
nts
appe
ared
to b
e as
soci
ated
with
pat
ient
’s m
enta
l/beh
avio
ural
pro
blem
s and
staf
f too
k pr
ompt
ac
tion.
191
9 £1
77.4
3 Pa
tient
N
ear m
iss
Oth
er
90 y
ear o
ld p
atie
nt o
n a
soft
food
die
t was
bei
ng in
trodu
ced
to m
ore
text
ured
food
s. W
hile
th
e pa
tient
was
eat
ing
her l
unch
she
chok
ed o
n he
r foo
d. P
atie
nt h
ad to
be
take
n to
A&
E fo
r tre
atm
ent,
as a
irway
cou
ld n
ot b
e cl
eare
d in
war
d. S
peci
al c
ateg
ory:
Clin
ical
judg
emen
t –
clin
ical
dec
isio
n m
ade
to in
trodu
ce m
ore
text
ured
food
s to
diet
. 10
£6
0.00
O
ther
V
anda
lism
O
ther
Pr
actic
e m
anag
er re
porte
d th
at fi
ve w
indo
ws h
ad b
een
brok
en.
Polic
e an
d es
tate
s off
icer
at
tend
ed.
Roo
ms c
heck
ed a
nd g
lazi
er fi
tted
new
win
dow
s. S
peci
al c
ateg
ory:
out
with
re
ason
able
pra
ctic
abili
ty –
ade
quat
e se
curit
y m
easu
res i
n pl
ace
(i.e.
ala
rm sy
stem
and
se
curit
y ca
mer
as).
11
£30.
87
Oth
er
Oth
er
Oth
er
Dis
able
d to
ilet i
n H
ealth
Cen
tre so
iled
with
exc
rem
ent.
So
bad
that
one
NA
refu
sed
clea
n up
mes
s. S
peci
al c
ateg
ory:
risk
unf
ores
eeab
le –
firs
t tim
e th
is h
as h
appe
ned
on th
is si
te.
12
£36.
11
Patie
nt
Acc
iden
tal
inju
ry
Slip
, trip
or f
all
on sa
me
leve
l or
stai
rs
One
pat
ient
trie
d to
ass
ist a
noth
er p
atie
nt w
ho w
as fa
lling
. B
oth
the
patie
nts’
zim
mer
s be
com
e en
tang
led,
resu
lting
in b
oth
patie
nts f
allin
g. P
atie
nt w
ho w
as a
ssis
ting
sust
aine
d a
frac
ture
d fe
mur
and
lace
ratio
ns to
righ
t han
d. S
peci
al c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– ri
sk a
sses
smen
ts a
nd c
are
plan
s wer
e in
pla
ce fo
r bot
h pa
tient
s. A
dequ
ate
mea
sure
s in
plac
e.
13
£2,5
10
Patie
nt
Phys
ical
vi
olen
ce
Oth
er
Var
ious
epi
sode
s ove
r 12-
wee
k pe
riod,
of p
atie
nt w
ho w
as e
asily
ups
et, t
hrow
ing
and
brea
king
item
s. S
trate
gy w
as to
let h
im th
row
the
item
s the
n ca
lm d
own
rath
er th
en
inte
rven
e. S
peci
al c
ateg
ory:
Clin
ical
judg
emen
t – st
rate
gy to
let p
atie
nt th
row
item
s ra
ther
than
inte
rven
e.
Tabl
e 4
Trus
t D (9
1 in
cide
nts)
No
Cost
Staf
f/pat
ient
/ ot
her
Cate
gory
Ty
pe
Des
crip
tion
1 £1
,796
.74
Oth
er
Nea
r mis
s O
ther
C
lient
, liv
ing
at h
ome,
had
tem
pora
ry lo
an o
f airb
ed w
hils
t ow
n w
as in
for r
epai
r.
Req
uest
ed te
mpo
rary
bed
was
left
perm
anen
tly a
s it w
as q
uiet
er th
an h
is o
ld o
ne.
This
was
dec
lined
. Sm
ell o
f bur
ning
repo
rted
from
uni
t on
its fi
rst n
ight
afte
r re
turn
from
repa
ir. T
empo
rary
bed
bro
ught
bac
k un
til n
ew re
plac
emen
t de
liver
ed.
Spec
ial c
ateg
ory:
Lac
k of
info
rmat
ion
– no
evi
denc
e to
supp
ort
susp
icio
n of
crim
inal
dam
age.
2
£54.
31
Patie
nt
Acc
iden
tal i
njur
y Fa
ll fr
om h
eigh
t In
depe
nden
t 89
year
old
mal
e co
mpl
eted
was
hing
han
ds in
toile
t, tu
rned
roun
d to
re
ach
for p
aper
tow
el a
nd fe
ll. A
n ag
ency
nur
se w
as su
perv
isin
g pa
tient
. Sp
ecia
lca
tego
ry: C
linic
al ju
dgem
ent –
NA
cou
ld d
o no
thin
g to
pre
vent
fall
as it
ha
ppen
ed su
dden
ly.
192
3 £1
1.20
Pa
tient
O
ther
C
ut w
ith sh
arp
mat
eria
l/obj
ect
Patie
nt fo
und
to b
e st
agge
ring
on w
ard,
smel
ling
of a
lcoh
ol.
Lock
ed h
imse
lf in
th
e to
ilet.
Re-
appe
ared
hav
ing
cut r
ight
fore
arm
and
sayi
ng th
at h
e ha
d sw
allo
wed
the
blad
e (n
ot b
elie
ved)
. Es
corte
d to
A&
E, b
ecam
e ag
gres
sive
and
A
&E
refu
sed
to tr
eat.
Spe
cial
cat
egor
y: r
easo
nabl
e pr
actic
abili
ty –
staf
f ab
ided
by
drug
& a
lcoh
ol p
olic
y.
4 £6
0.20
St
aff
Phys
ical
vio
lenc
e O
ther
Pa
tient
kic
ked
dow
n pa
rtitio
n do
or in
to fe
mal
e en
d of
war
d (u
nder
rede
cora
tion)
. A
rmed
with
pol
e th
reat
ened
staf
f. A
larm
act
ivat
ed a
nd p
olic
e ca
lled.
Spe
cial
cate
gory
: rea
sona
ble
prac
ticab
ility
– p
atie
nt w
as a
lread
y un
der i
ncre
ased
ob
serv
atio
ns.
5 £2
4.45
Pa
tient
V
anda
lism
O
ther
Pa
tient
smas
hed
door
into
fem
ale
end
of w
ard
(und
er re
deco
ratio
n).
Thre
w p
aint
at
win
dow
. A
larm
act
ivat
ed, s
taff
nur
se d
ecid
ed d
e-es
cala
tion
mos
t app
ropr
iate
ac
tion
cons
ider
ing
envi
ronm
ent.
Spe
cial
cat
egor
y: C
linic
al ju
dgem
ent –
car
e pl
an fo
r pat
ient
did
not
incl
ude
incr
ease
obs
erva
tion
regi
me.
6
£295
.20
Staf
f A
ccid
enta
l inj
ury
Patie
nt
liftin
g/ha
ndlin
g Pa
tient
am
pute
e, e
xcep
tiona
lly h
eavy
, rai
sed
from
bed
usi
ng a
hoi
st.
Two
staf
f un
derta
king
pro
cedu
re. S
taff
Nur
se p
ulle
d ho
ist a
nd su
ffer
ed p
ain
in h
er n
eck.
O
ff w
ork
for o
ne w
eek.
Spe
cial
cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
al
l pol
icy
mea
sure
s in
plac
e.
7 £4
01.5
9 Pa
tient
Ph
ysic
al v
iole
nce
Pers
on to
per
son
assa
ult
Patie
nt re
ceiv
ed in
sulin
in tr
eatm
ent r
oom
, bec
ame
verb
ally
, the
n vi
olen
tly
aggr
essi
ve g
rabb
ing
staf
f nur
se b
y ar
ms.
Nur
se u
nabl
e to
act
ivat
e al
arm
. N
urse
gr
abbe
d ag
ain
and
thro
wn
agai
nst s
ink.
Pat
ient
atte
mpt
ed to
stra
ngle
nur
se a
nd
forc
ed h
er o
ver h
is k
nee.
Nur
se m
anag
ed to
rais
e al
arm
and
pat
ient
rest
rain
ed.
Staf
f Nur
se o
ff w
ork
for s
ix w
eeks
. Sp
ecia
l cat
egor
y: r
easo
nabl
e pr
actic
abili
ty–
gene
ral p
reve
ntat
ive
mea
sure
s in
plac
e, n
urse
was
alo
ne w
ith p
atie
nt b
ut h
ad
not b
een
asse
ssed
as l
ikel
y to
ass
ault.
8
£30.
70
Patie
nt
Oth
er
Cut
with
shar
p m
ater
ial/o
bjec
t Pa
tient
in b
edro
om a
ctiv
ated
em
erge
ncy
buzz
er.
Foun
d to
hav
e cu
t bot
h fo
rear
ms.
Tak
en to
A&
E by
am
bula
nce
for s
utur
ing
then
retu
rned
to w
ard.
R
azor
bla
des f
ound
in ro
om, p
atie
nt h
ad se
lf-ha
rmed
twic
e be
fore
. Sp
ecia
lca
tego
ry: C
linic
al ju
dgem
ent –
car
e pl
an d
id n
ot re
gard
clo
ser o
bser
vatio
n or
ot
her m
easu
res t
o co
ntro
l ris
k ne
cess
ary.
9
£25.
00
Patie
nt
Van
dalis
m
Oth
er
Patie
nt re
peat
edly
slam
med
seat
on
toile
t, sm
ashi
ng it
. Pa
tient
dis
turb
ed, h
earin
g vo
ices
from
the
devi
l in
his h
ead.
Spe
cial
cat
egor
y: C
linic
al ju
dgem
ent –
de
cisi
on to
kee
p pa
tient
on
gene
ral o
bser
vatio
ns d
ue to
pre
viou
s exp
erie
nce.
193
10
£23.
16
Patie
nt
Phys
ical
vio
lenc
e O
ther
D
elus
iona
l pat
ient
pun
ched
a w
all d
urin
g ni
ght.
Sta
ff u
naw
are
until
bre
akfa
st.
Seen
by
doct
or a
nd se
nt fo
r x-r
ay w
ith e
scor
t. N
o fr
actu
re c
onfir
med
and
re
turn
ed to
war
d. S
peci
al c
ateg
ory:
rea
sona
ble
prac
ticab
ility
– n
o in
dica
tion
that
cha
nge
to c
are
plan
was
nee
ded.
11
£20.
00
Patie
nt
Phys
ical
vio
lenc
e O
ther
Pa
tient
hav
ing
brea
kfas
t in
dini
ng ro
om w
hen
loud
noi
se h
eard
. Pa
tient
had
th
row
n pl
ate
and
kniv
es a
t wal
l. P
atie
nt su
ffer
ing
audi
tory
hal
luci
natio
ns a
nd
para
noid
del
usio
ns.
Spec
ial c
ateg
ory:
Clin
ical
judg
emen
t – u
nder
gen
eral
ob
serv
atio
n at
the
time,
no
staf
f mem
bers
pre
sent
. 12
£1
1.59
Pa
tient
O
ther
C
ut w
ith sh
arp
mat
eria
l/obj
ect
Staf
f ale
rted
to n
oise
in m
ale
toile
t. T
oile
t loc
ked
and
patie
nt re
fuse
d en
try to
st
aff.
Ala
rm a
ctiv
ated
, sta
ff p
reve
nted
furth
er la
cera
tion
to a
rm.
Firs
t aid
giv
en.
Spec
ial c
ateg
ory:
Clin
ical
judg
emen
t – se
cond
vio
lent
inci
dent
of t
he d
ay,
reac
ting
to th
e pr
evio
us d
ays a
bsco
ndin
g.
13
£11.
59
Patie
nt
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t Pa
tient
, on
seei
ng N
A, g
rabb
ed h
im b
y th
e ar
m a
nd p
unch
ed h
im o
n th
e fo
rehe
ad.
NA
bro
ke fr
ee a
nd a
ctiv
ated
ala
rm.
Spec
ial c
ateg
ory:
risk
not
fore
seea
ble
– in
cide
nt h
appe
ned
“out
of b
lue”
. 14
£2
3.80
Pa
tient
O
ther
C
ut w
ith sh
arp
mat
eria
l/obj
ect
Patie
nt d
ism
antle
d hi
s raz
or a
nd in
flict
ed a
15c
m la
cera
tion
to a
bdom
en.
Iden
tical
inci
dent
4 w
eeks
pre
viou
sly.
Atte
mpt
at c
losi
ng w
ound
, but
had
to b
e se
nt to
A&
E fo
r tre
atm
ent.
Spe
cial
cat
egor
y: C
linic
al ju
dgem
ent –
bas
ed o
n st
rate
gy to
incr
ease
free
dom
as p
art o
f reh
abili
tatio
n st
rate
gy.
15
£23.
80
Patie
nt
Oth
er
Cut
with
shar
p m
ater
ial/o
bjec
t Pa
tient
requ
este
d an
d w
as g
iven
his
safe
ty ra
zor.
In to
ilet p
atie
nt d
ism
antle
s ra
zor a
nd se
lf in
flict
s 15c
m la
cera
tion
to h
is a
bdom
en.
Take
n to
A&
E fo
r sut
ures
th
en re
turn
ed to
war
d. S
peci
al c
ateg
ory:
rea
sona
ble
prac
ticab
ility
– in
cide
nt
not f
ores
eeab
le (n
o pr
evio
us h
isto
ry).
16
£11.
59
Patie
nt
Phys
ical
vio
lenc
e O
ther
Pa
tient
foun
d sm
okin
g in
dor
mito
ry.
NA
ask
ed p
atie
nt to
ext
ingu
ish
ciga
rette
. Pa
tient
bec
ame
viol
ent.
Cha
rge
Nur
se re
stra
ined
pat
ient
. Sp
ecia
l cat
egor
y:
reas
onab
le p
ract
icab
ility
– a
ppro
pria
te sm
okin
g po
licy
of w
hich
pat
ient
was
aw
are
and
usua
lly c
ompl
ied.
17
£1
1.59
Pa
tient
Ph
ysic
al v
iole
nce
Oth
er
Patie
nt o
vertu
rned
lade
n ta
ble.
Ala
rm a
ctiv
ated
pat
ient
rest
rain
ed a
nd m
edic
atio
n gi
ven.
Spe
cial
cat
egor
y: r
isk n
ot fo
rese
eabl
e –
inci
dent
hap
pene
d w
ithou
t w
arni
ng.
18
£11.
59
Patie
nt
Van
dalis
m
Oth
er
Patie
nt b
ecam
e an
gry
afte
r rec
eivi
ng v
erba
l abu
se fr
om a
noth
er p
atie
nt, t
hen
pick
ed u
p a
chai
r and
beg
an h
ittin
g it
off a
win
dow
in th
e di
ning
room
. A
larm
ac
tivat
ed, p
atie
nt re
mov
ed fr
om d
inin
g ro
om, s
ituat
ion
de-e
scal
ated
qui
ckly
. Sp
ecia
l cat
egor
y: r
easo
nabl
e pr
actic
abili
ty –
all
reas
onab
le c
ontro
l mea
sure
s in
plac
e.
194
19
£11.
59
Patie
nt
Van
dalis
m
Oth
er
Patie
nt in
sitti
ng ro
om p
icke
d up
cha
ir an
d at
tem
pted
to sm
ash
a w
indo
w.
Patie
nt
said
he
was
hea
ring
voic
es te
lling
him
to d
o so
. A
larm
act
ivat
ed a
nd p
atie
nt
rest
rain
ed.
Spec
ial c
ateg
ory:
ris
k un
fore
seea
ble
– in
cide
nt h
appe
ned
with
out
war
ning
. 20
£2
3.16
Pa
tient
O
ther
O
ther
Pa
tient
on
cons
tant
obs
erva
tion,
but
this
did
not
ext
end
to th
e to
ilet f
or p
atie
nt
priv
acy.
Pat
ient
ope
ned
up p
revi
ous a
bdom
inal
wou
nd.
Esco
rted
to a
cute
ho
spita
l for
trea
tmen
t. P
atie
nt c
onst
antly
self-
harm
ing
but n
ot se
lf-th
reat
enin
g.
Spec
ial c
ateg
ory:
Clin
ical
judg
emen
t – P
olic
y on
obs
erva
tion
regi
me
allo
wed
fo
r tim
e fo
r per
sona
l hyg
iene
in p
rivac
y so
no
brea
ch o
f pol
icy.
21
£4
0.00
Pa
tient
Ph
ysic
al v
iole
nce
Oth
er
Patie
nt si
tting
qui
etly
in si
tting
room
pic
ked
up c
hair
and
brok
e a
win
dow
, thi
s w
as re
peat
ed 5
min
utes
late
r. U
ncom
mun
icat
ive
thro
ugho
ut.
Spec
ial c
ateg
ory:
ri
sk u
nfor
esee
able
– b
oth
inci
dent
s hap
pene
d su
dden
ly w
ithou
t any
trig
ger.
22
£40.
00
Patie
nt
Phys
ical
vio
lenc
e O
ther
Pa
tient
in d
orm
itory
und
er g
ener
al o
bser
vatio
n w
as re
stle
ss in
bed
. Pa
tient
di
scon
nect
ed e
nd o
f bed
and
smas
hed
win
dow
. Pa
tient
rem
oved
to si
tting
are
a to
ve
nt a
nxie
ties.
Spe
cial
cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
pat
ient
on
gen
eral
obs
erva
tion.
23
£4
6.69
Pa
tient
O
ther
C
ut w
ith sh
arp
mat
eria
l/obj
ect
Patie
nt, f
or th
e th
ird ti
me,
re-o
pene
d pr
evio
us se
lf in
flict
ed w
ound
. W
ound
cl
eane
d an
d sk
in c
losu
res a
pplie
d. W
ound
bec
ame
infe
cted
, dre
ssin
gs a
pplie
d an
d a
cour
se o
f ant
ibio
tics g
iven
. Sp
ecia
l cat
egor
y: C
linic
al ju
dgem
ent -
re
gard
ing
care
pla
n.
24
£11.
59
Patie
nt
Phys
ical
vio
lenc
e O
ther
Pa
tient
ask
ed to
use
the
offic
e ph
one.
Req
uest
refu
sed.
Pat
ient
pun
ched
cha
rge
nurs
e in
the
face
. A
larm
act
ivat
ed, p
atie
nt re
stra
ined
and
take
n to
sitti
ng ro
om.
Spec
ial c
ateg
ory:
rea
sona
ble
prac
ticab
ility
– p
ayph
one
avai
labl
e fo
r pat
ient
us
e.25
£1
1.20
Pa
tient
O
ther
C
ut w
ith sh
arp
mat
eria
l/obj
ect
Patie
nt in
mal
e do
rmito
ry in
serte
d ho
ok o
f coa
t han
ger i
nto
rece
nt a
bdom
inal
w
ound
. U
nwitn
esse
d, 3
rd v
iole
nt in
cide
nt o
f day
. Pa
tient
sent
to A
&E
for
treat
men
t. S
peci
al c
ateg
ory:
lack
of i
nfor
mat
ion
– un
able
to o
btai
n in
form
atio
n re
gard
ing
obse
rvat
ion
requ
ired.
26
£6
7.64
O
ther
O
ther
C
onta
ct
with
/exp
osur
e to
ha
rmfu
l sub
stan
ces
Staf
f nur
se sm
elle
d ga
s. O
n ca
ll en
gine
er tr
aced
to re
nted
LPG
stor
age
tank
s.
Engi
neer
cal
led
to re
pair
leak
ing
fittin
g. S
peci
al c
ateg
ory:
lack
of i
nfor
mat
ion
–tec
hnic
al re
cord
s una
vaila
ble.
195
27
£26.
31
Patie
nt
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t St
aff n
urse
alo
ne in
trea
tmen
t roo
m w
hen
patie
nt e
nter
ed u
nann
ounc
ed.
Staf
f nu
rse
mov
ed to
war
ds a
skin
g pa
tient
wha
t he
wan
ted.
Pat
ient
repl
ied
that
he
wan
ted
to g
o ho
me.
Whe
n ex
plai
ned
to h
im th
at h
e co
uldn
’t pa
tient
gra
bbed
staf
f nu
rse
by th
e th
roat
resu
lting
in h
er fa
lling
to th
e gr
ound
. Sp
ecia
l cat
egor
y:
outw
ith r
easo
nabl
e pr
actic
abili
ty –
nur
se re
spon
ded
wel
l to
the
situ
atio
n. N
o al
arm
fact
ors t
hat c
ould
hav
e pr
even
ted
inci
dent
. 28
£4
47.2
7 St
aff
Acc
iden
tal i
njur
y Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r sta
irs
Staf
f nur
se tr
ippe
d on
pav
ing
slab
at d
oorw
ay.
No
obvi
ous d
efec
t. R
esul
ting
inju
ries w
ere
bang
on
head
, spr
aine
d w
rist a
nd c
ut fi
nger
s. A
bsen
t fro
m w
ork.
Sp
ecia
l cat
egor
y: r
easo
nabl
e pr
actic
abili
ty –
not
hing
foun
d to
con
tribu
te to
fa
ll.
29
£137
.40
Patie
nt
Oth
er
Oth
er
Patie
nt (d
etai
ned
unde
r MH
A) o
n pa
role
leav
e fa
iled
to re
turn
to w
ard.
Par
ent
tele
phon
ed to
say
he w
as a
t hom
e. T
eam
was
ass
embl
ed a
nd p
icke
d up
pat
ient
fr
om h
ome
and
retu
rned
to w
ard.
Spe
cial
cat
egor
y: r
easo
nabl
e pr
actic
abili
ty –
In
cide
nt d
ealt
with
as b
est a
s it c
ould
hav
e be
en, i
n th
e ju
dgem
ent o
f the
rese
arch
te
am.
30
£4.3
9 Pa
tient
O
ther
C
ut w
ith sh
arp
mat
eria
l/obj
ect
Staf
f wer
e pr
esen
t in
patie
nt’s
room
due
to h
is m
enta
l sta
te.
Patie
nt su
dden
ly
took
hol
d of
gla
ss tu
mbl
er sm
ashe
d it
agai
nst t
he w
all c
uttin
g hi
s han
d in
the
proc
ess.
Stru
ggle
d vi
olen
tly b
efor
e be
ing
rest
rain
ed b
y st
aff.
Spe
cial
cat
egor
y:
reas
onab
le p
ract
icab
ility
– a
ll ris
k co
ntro
l mea
sure
s in
plac
e.
31
£770
.17
Patie
nt
Oth
er
Oth
er
Patie
nt, g
rant
ed 3
0 m
inut
es le
ave
from
war
d, a
bsco
nded
. Po
lice
foun
d pa
tient
at
hom
e w
ith e
vide
nce
of o
verd
ose
– pa
tient
sent
by
ambu
lanc
e to
A&
E an
d de
tain
ed fo
r app
rox.
28
hour
s with
two
staf
f in
atte
ndan
ce o
n sp
ecia
l nur
sing
.Sp
ecia
l cat
egor
y: C
linic
al ju
dgem
ent –
clin
ical
judg
emen
t to
gran
t lea
ve fr
om
war
d.32
£6
.58
Patie
nt
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t Pa
tient
ran
tow
ards
ano
ther
, pus
hing
him
bac
kwar
ds a
nd h
ittin
g hi
s hea
d of
f cu
pboa
rd, s
usta
inin
g a
lace
ratio
n to
hea
d. G
iven
firs
t aid
then
sent
to A
&E.
Sp
ecia
l cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
reas
onab
le m
easu
res i
n pl
ace,
no
hist
ory
of b
ad fe
elin
g be
twee
n pa
tient
s and
no
hist
ory
of v
iole
nce
from
ag
gres
sor.
33
£1,1
28.0
0 St
aff
Acc
iden
tal i
njur
y Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r sta
irs
On
open
ing
door
to c
orrid
or, n
urse
stum
bled
on
a st
one
whe
n pu
tting
ligh
t on.
H
e hi
t his
hea
d of
f fac
ing
wal
l and
was
kno
cked
unc
onsc
ious
. W
as ta
ken
to A
&E
by a
mbu
lanc
e an
d w
as su
bseq
uent
ly o
ff w
ork.
Spe
cial
cat
egor
y: o
utw
ith
reas
onab
le p
ract
icab
ility
– n
o ev
iden
ce to
sugg
est i
ncid
ent w
as c
ause
d by
lig
htin
g le
vels
and
no
way
of k
now
ing
whe
re st
one
cam
e fr
om.
196
34
£36.
40
Patie
nt
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t N
ursi
ng a
ssis
tant
was
bat
hing
pat
ient
. Pa
tient
hit
out a
t NA
cau
sing
her
left
thum
b to
be
bent
bac
k. N
A ta
ken
to A
&E
by ta
xi b
ut n
o fr
actu
re –
retu
rned
to
war
d.Sp
ecia
l cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
risk
s red
uced
to
as lo
w a
s pos
sibl
e, p
atie
nt h
it ou
t im
puls
ivel
y.
35
£355
.19
Patie
nt
Acc
iden
tal i
njur
y O
ther
Pa
tient
got
out
of b
ed a
nd w
alke
d to
war
ds d
oor a
nd n
urse
. Pa
tient
sudd
enly
je
rked
hea
d up
war
ds a
nd st
ruck
nur
se o
n ja
w.
Nur
se o
ff w
ork
for o
ne w
eek
with
w
hipl
ash.
Spe
cial
cat
egor
y: r
isk
not f
ores
eeab
le –
alth
ough
pat
ient
alw
ays
wal
ked
awkw
ardl
y th
e in
cide
nt h
appe
ned
very
qui
ckly
. 36
£3
0.31
Pa
tient
O
ther
O
ther
Pa
tient
agi
tate
d, d
octo
r exp
lain
ed to
him
that
he
was
not
per
mitt
ed to
leav
e th
e w
ard.
Pat
ient
lock
ed w
ard
door
s, pu
lled
dow
n cu
rtain
rails
and
bar
ricad
ed
him
self
in si
tting
room
. A
larm
act
ivat
ed a
nd si
tuat
ion
de-e
scal
ated
. Sp
ecia
lca
tego
ry: C
linic
al ju
dgem
ent –
let p
atie
nt v
ent f
rust
ratio
n an
d ch
ose
to a
dopt
de
-esc
alat
ion
rath
er th
an re
stra
in p
atie
nt im
med
iate
ly.
37
£25.
52
Staf
f V
erba
l abu
se o
r th
reat
enin
g be
havi
our
Oth
er
Supp
ort w
orke
r on
hom
e vi
sit.
Clie
nt b
ecam
e ve
rbal
ly a
busi
ve a
nd p
hysi
cally
th
reat
enin
g. S
W m
ade
deci
sion
to w
ithdr
aw a
nd re
turn
ed to
off
ice
for
coun
selli
ng.
Wen
t hom
e ea
rly.
Spec
ial c
ateg
ory:
risk
not
fore
seea
ble/
outw
ith
reas
onab
le p
ract
icab
ility
– C
lient
had
not
show
n a
pred
ispo
sitio
n to
vio
lenc
e be
fore
. SW
follo
wed
pro
cedu
re b
y w
ithdr
awin
g fr
om si
tuat
ion.
38
£2
56.7
7 St
aff
Acc
iden
tal i
njur
y Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r sta
irs
Gar
dene
r pru
ning
bor
der s
hrub
s tur
ned
to c
ut a
noth
er sh
rub
whe
n he
stum
bled
to
grou
nd tw
istin
g an
kle.
Spe
cial
cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
G
arde
ner w
as fo
llow
ing
corr
ect p
roce
dure
and
usi
ng c
orre
ct e
quip
men
t (fo
otw
ear
etc.
). 39
£1
1.51
Pa
tient
Ph
ysic
al v
iole
nce
Oth
er
Frus
trate
d pa
tient
smas
hed
cup
in m
ain
kitc
hen.
C&
R te
chni
ques
use
d to
reso
lve
situ
atio
n an
d pa
tient
was
put
to b
ed w
ith m
edic
atio
n. S
peci
al c
ateg
ory:
Clin
ical
ju
dgem
ent –
staf
f fel
t med
icat
ion
wou
ld c
alm
pat
ient
dow
n bu
t clo
ser
obs/
supe
rvis
ion
also
nee
ded.
40
£1
9.90
Pa
tient
O
ther
O
ther
Pa
tient
abs
cond
ed a
fter b
eing
adm
itted
nig
ht b
efor
e in
to p
sych
iatri
c as
sess
men
t w
ard.
Abs
cond
ed w
hen
othe
r pat
ient
s wer
e do
wns
tairs
at m
ealti
me
(din
ing
room
on
floo
r bel
ow) –
Spe
cial
cat
egor
y: C
linic
al ju
dgem
ent –
Pat
ient
was
adm
itted
on
a v
olun
tary
bas
is.
41
£11.
20
Patie
nt
Acc
iden
tal i
njur
y Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r sta
irs
Patie
nt w
alki
ng in
to d
ay ro
om tr
ippe
d ov
er a
noth
er p
atie
nt’s
zim
mer
fram
e. N
o br
uisi
ng e
vide
nt a
t tim
e al
thou
gh p
atie
nt h
ad a
lot o
f pai
n in
left
wris
t. S
peci
alca
tego
ry: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
pre
senc
e of
zim
mer
fram
e un
avoi
dabl
e.
197
42
£50.
00
Oth
er
Oth
er
Oth
er
Fire
ala
rm a
ctiv
ated
for n
o ap
pare
nt re
ason
. C
all o
ut fo
r mem
ber o
f sta
ff to
rese
t fir
e al
arm
. Sp
ecia
l cat
egor
y: in
suff
icie
nt in
form
atio
n –
no fu
rther
info
rmat
ion
or w
itnes
ses a
vaila
ble.
43
£9
.79
Patie
nt
Oth
er
Oth
er
Patie
nt fo
rced
ope
n lo
cked
inte
rnal
doo
r at t
he sa
me
time
as a
join
er w
as o
peni
ng
an a
djac
ent e
xter
nal d
oor.
Pat
ient
pus
hed
past
join
er a
nd a
bsco
nded
. Sp
ecia
lca
tego
ry: C
linic
al ju
dgem
ent –
pat
ient
not
judg
ed to
requ
ire a
ny p
artic
ular
ob
serv
atio
n re
gim
e.
44
£18.
20
Patie
nt
Acc
iden
tal i
njur
y Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r sta
irs
Patie
nt fo
und
lyin
g on
floo
r in
corr
idor
and
take
n to
A&
E by
taxi
. W
as re
turn
ed
to w
ard
by ta
xi a
fter x
-ray
. Sp
ecia
l cat
egor
y: r
isk n
ot fo
rese
eabl
e –
first
tim
e pa
tient
had
falle
n.
45
£6.1
4 Pa
tient
O
ther
O
ther
A
bsco
ndin
g pa
tient
det
aine
d in
gro
unds
but
cou
ld n
ot b
e pe
rsua
ded
to re
turn
to
war
d. N
o po
wer
to d
etai
n an
d pa
tient
retu
rned
hom
e bu
t cam
e ba
ck to
war
d tw
o ho
urs l
ater
apo
logi
sing
. Sp
ecia
l cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
ap
prop
riate
con
trol m
easu
res i
n pl
ace.
46
£6
7.16
Pa
tient
O
ther
O
ther
Pa
tient
info
rmed
staf
f she
had
inge
sted
her
hea
ring
aid
and
was
take
n fo
r x-r
ay.
It w
as c
onfir
med
but
no
treat
men
t nec
essa
ry a
nd re
turn
ed to
war
d. P
atie
nt h
as
hist
ory
of m
isch
ievo
us se
lf-ha
rm.
Spec
ial c
ateg
ory:
Clin
ical
judg
emen
t – n
ot
judg
ed to
nee
d sp
ecia
l obs
erva
tion
regi
me.
47
£3
.25
Patie
nt
Oth
er
Oth
er
Patie
nt, g
iven
15
min
utes
una
ccom
pani
ed p
arol
e, fa
iled
to re
turn
. D
octo
r pho
ned
war
d to
info
rm o
f sig
htin
g, c
harg
e nu
rse
then
wen
t to
sear
ch lo
cal a
rea
in c
ar.
Patie
nt p
hone
d 1½
hou
rs la
ter t
o gi
ve lo
catio
n. P
olic
e re
turn
ed p
atie
nt to
war
d.
Spec
ial c
ateg
ory:
Clin
ical
judg
emen
t – a
lthou
gh p
atie
nt u
nder
goin
g pe
rson
al
chan
ges a
nd q
uite
stre
ssed
it w
as ju
dged
app
ropr
iate
to g
rant
par
ole.
48
£1
6.80
Pa
tient
O
ther
O
ther
Pa
tient
faile
d to
retu
rn fr
om O
ccup
atio
nal T
hera
pist
. Pa
tient
seen
boa
rdin
g a
bus
into
tow
n. T
wo
nurs
ing
auxi
liarie
s fol
low
in c
ar b
ut fa
il to
loca
te p
atie
nt in
tow
n or
at m
othe
r’s h
ome.
The
y re
turn
to w
ard
and
call
polic
e to
repo
rt m
issi
ng
pers
on.
Spec
ial c
ateg
ory:
Clin
ical
judg
emen
t – p
atie
nt ju
dged
suita
ble
for
unac
com
pani
ed v
isit
to O
T.
49
£15.
00
Patie
nt
Van
dalis
m
Oth
er
Patie
nt w
ith se
vere
dem
entia
bro
ke fi
re a
larm
gla
ss b
ut a
larm
did
n’t a
ctiv
ate.
Sp
ecia
l cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
pat
ient
diff
icul
t to
man
age
and
refu
sed
to ta
ke m
edic
atio
n.
50
£15.
00
Patie
nt
Van
dalis
m
Oth
er
Lear
ning
dis
abili
ties p
atie
nt se
t off
fire
ala
rm b
y br
eaki
ng g
lass
. Pa
tient
s and
st
aff e
vacu
ated
and
fire
brig
ade
atte
nded
. Sp
ecia
l cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
no
prev
ious
his
tory
of v
anda
lism
so a
dequ
ate
mea
sure
s in
plac
e.
51
£53.
56
Oth
er
Oth
er
Oth
er
Fals
e fir
e al
arm
in c
otta
ge h
ospi
tal.
Ala
rm si
lenc
ed o
n ve
rifyi
ng n
o fir
e in
198in
dica
ted
zone
and
on-
call
engi
neer
atte
nded
to re
set a
nd e
xam
ine
syst
em.
No
faul
t was
foun
d, b
elie
ved
to b
e ca
used
by
pow
er b
eing
off
dur
ing
that
day
. Sp
ecia
l cat
egor
y: r
isk n
ot fo
rese
eabl
e –
all o
ther
con
trol m
easu
res i
n pl
ace.
52
£1
07.1
2 O
ther
O
ther
O
ther
Fi
re a
larm
act
ivat
ed fo
r no
appa
rent
reas
on.
Ala
rm si
lenc
ed a
nd e
ngin
eer c
alle
d ou
t. F
aulty
sens
or re
plac
ed a
nd s
yste
m re
set.
Spe
cial
cat
egor
y: o
utw
ith
reas
onab
le p
ract
icab
ility
– 2
5% o
f sen
sors
che
cked
eve
ry 3
-mon
ths.
Yea
rly
cont
ract
mai
nten
ance
che
ck.
53
£25.
00
Patie
nt
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t Pa
tient
pun
ched
ano
ther
kno
ckin
g of
f the
ir gl
asse
s. F
ram
es re
plac
ed b
y ho
spita
l op
ticia
n. S
peci
al c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– n
o hi
stor
y of
an
tago
nism
bet
wee
n pa
tient
so a
dequ
ate
mea
sure
s wer
e in
pla
ce.
54
£50.
00
Patie
nt
Van
dalis
m
Oth
er
Patie
nt sm
ashe
d w
all m
irror
in b
athr
oom
for n
o ap
pare
nt re
ason
. Sp
ecia
lca
tego
ry: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
not
com
mon
for m
irror
s in
war
d to
be
brok
en, g
laze
d w
ith fi
lm-b
acke
d gl
ass t
o pr
even
t sca
tter o
f gla
ss.
55
£60.
00
Oth
er
Oth
er
Oth
er
Win
dow
in c
ubic
le d
orm
itory
foun
d to
be
brok
en (n
ot h
eard
or w
itnes
sed)
. Pa
tient
in b
ed h
ad h
isto
ry o
f ban
ging
on
win
dow
. Sp
ecia
l cat
egor
y: in
suff
icie
nt
info
rmat
ion
– in
cide
nt w
asn’
t hea
rd o
r witn
esse
d.
56
£85.
00
Oth
er
Van
dalis
m
Oth
er
Pane
of g
lass
foun
d to
be
shat
tere
d in
sing
le b
edde
d ro
om.
No
nois
e an
d in
cide
nt
unw
itnes
sed.
Pat
ient
repe
ated
inci
dent
2 w
eeks
late
r. S
peci
al c
ateg
ory:
out
with
re
ason
able
pra
ctic
abili
ty –
no
prev
ious
his
tory
and
gla
ss ju
dged
by
Trus
t sta
ff
to b
e of
an
appr
opria
te ty
pe fo
r typ
e of
war
d.
57
£2,6
52.9
3 St
aff
Acc
iden
tal i
njur
y O
ther
St
aff n
urse
, with
hel
p fr
om N
A, w
as c
hang
ing
inco
ntin
ence
pad
on
patie
nt.
Patie
nt ro
lled
over
on
side
and
new
pad
put
into
pos
ition
in th
e pr
oces
s sta
ff n
urse
be
nt b
ack
mid
dle
finge
r. S
taff
nurs
e su
stai
ned
a fr
actu
red
finge
r. S
peci
alca
tego
ry: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
staf
f wer
e fo
llow
ing
corr
ect
proc
edur
es.
58
£13.
89
Patie
nt
Oth
er
Oth
er
Patie
nt u
nder
con
stan
t obs
erva
tion
wen
t int
o en
-sui
te to
ilet w
hils
t sta
ff re
mai
ned
outs
ide.
Sta
ff e
nter
ed to
ilet w
hen
nois
e of
key
s hea
rd.
Patie
nt sa
id th
at sh
e ha
d sw
allo
wed
a k
ey a
nd w
as ta
ken
to A
&E
for x
-ray
then
retu
rned
to w
ard.
Spe
cial
cate
gory
: rea
sona
ble
prac
ticab
ility
– u
nder
con
stan
t obs
and
pro
cedu
re o
f pa
tient
priv
acy.
Pat
ient
thou
ght n
ot to
be
in p
osse
ssio
n of
key
s as t
hey
had
been
re
mov
ed.
59
£12.
60
Oth
er
Acc
iden
tal
prop
erty
lo
ss/d
amag
e
Oth
er
Win
dow
in st
aff t
ea-r
oom
ope
ned.
Stro
ng w
ind
caug
ht w
indo
w o
peni
ng it
fully
re
sulti
ng in
the
win
dow
smas
hing
off
inte
rnal
wal
l. S
peci
al c
ateg
ory:
risk
not
fo
rese
eabl
e –
win
dow
not
exp
ecte
d to
be
caug
ht b
y w
ind.
199
60
£324
.24
Staf
f A
ccid
enta
l inj
ury
Patie
nt li
fting
/ ha
ndlin
g Pa
tient
requ
ired
toile
t dur
ing
nigh
t so
com
mod
e br
ough
t to
beds
ide.
Not
qui
te in
po
sitio
n w
hen
patie
nt tr
ied
to si
t on
it. N
A in
stin
ctiv
ely
stre
tche
d ou
t lef
t arm
to
prev
ent p
atie
nt fa
lling
and
sust
aine
d an
inju
ry d
oing
so.
NA
was
off
wor
k fo
r tw
o ni
ghts
with
susp
ecte
d pu
lled
mus
cle.
Spe
cial
cat
egor
y: r
isk
unfo
rese
eabl
e–
patie
nt h
ad n
o hi
stor
y of
falls
, mis
judg
ed d
ista
nce.
61
£1
5.00
Pa
tient
O
ther
O
ther
M
ale
patie
nt w
ith d
emen
tia a
ctiv
ated
a fi
re a
larm
. Pa
tient
is o
vera
ctiv
e an
d th
inks
he
is in
the
arm
y. W
hen
he re
ads “
in c
ase
of fi
re p
ress
bel
l” h
e do
es it
. Fi
re b
rigad
e at
tend
but
no
evac
uatio
n. S
peci
al c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– p
atie
nt h
ad o
nly
done
this
onc
e be
fore
but
not
thou
ght l
ikel
y to
do
it a
gain
. Tr
ust p
lann
ing
inst
all a
larm
cov
ers.
62
£26.
70
Patie
nt
Acc
iden
tal i
njur
y Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r sta
irs
Patie
nt h
ad ju
st b
een
adm
itted
and
was
wal
king
with
zim
mer
fram
e to
see
doct
or,
a st
aff m
embe
r alo
ng si
de.
Patie
nt lo
st b
alan
ce a
nd fe
ll to
floo
r. T
rans
ferr
ed to
ac
ute
hosp
ital.
Spe
cial
cat
egor
y: r
easo
nabl
e pr
actic
abili
ty –
pat
ient
wal
king
w
ell w
ith st
aff m
embe
r. 63
£1
1.87
Pa
tient
O
ther
O
ther
Pa
tient
last
seen
on
war
d ar
ound
10:
30, f
ound
to b
e m
issi
ng a
t 12:
00.
A se
arch
of
the
grou
nds a
nd su
rrou
ndin
g ar
ea, i
n tw
o ca
rs, w
as c
arrie
d ou
t. P
atie
nt fo
und
at
13:5
0 by
pol
ice
and
retu
rned
to th
e w
ard.
Spe
cial
cat
egor
y: r
isk
unfo
rese
eabl
e–
Sect
ion
18 p
atie
nt, b
ehav
iour
nor
mal
. W
ard
alw
ays u
nloc
ked.
64
£2
.80
Patie
nt
Oth
er
Oth
er
Patie
nt fi
t and
abl
e, w
aitin
g on
pla
ce in
car
e ho
me,
left
war
d un
obse
rved
. Fo
llow
ing
exte
nsiv
e se
arch
pat
ient
was
foun
d in
loca
l soc
ial c
lub
and
retu
rned
to
war
d by
the
polic
e. S
peci
al c
ateg
ory:
risk
unf
ores
eeab
le –
no
prev
ious
in
dica
tion
of ri
sk o
f abs
cond
ing.
65
£2,1
04.8
0 Pa
tient
O
ther
O
ther
Pa
tient
out
on
unac
com
pani
ed le
ave.
Whe
n sh
e re
turn
ed in
form
ed st
aff s
he h
ad
take
n an
ove
rdos
e of
par
acet
amol
. B
lood
s tak
en b
y du
ty d
octo
r con
firm
ed
over
dose
. Ta
ken
to a
cute
hos
pita
l for
trea
tmen
t. R
etur
ned
to w
ard
five
days
la
ter.
Spe
cial
cat
egor
y: C
linic
al ju
dgem
ent –
not
judg
ed to
be
a su
icid
e ris
k.
66
£7.9
9 Pa
tient
V
erba
l abu
se o
r th
reat
enin
g be
havi
our
Oth
er
Thirt
y ye
ar o
ld m
ale
patie
nt, w
ith d
epre
ssiv
e co
nditi
on.
Trig
gerin
g fa
ctor
usu
ally
ph
onin
g hi
s mot
her.
Afte
r pho
ne c
all h
e de
man
ded
to g
o ho
me.
Pat
ient
then
be
cam
e ag
gres
sive
, ala
rm a
ctiv
ated
, pat
ient
rest
rain
ed.
Spec
ial c
ateg
ory
of
Clin
ical
judg
emen
t – d
ecis
ion
take
n no
t to
with
hold
acc
ess t
o te
leph
one
to sp
eak
to m
othe
r. 67
£5
.60
Patie
nt
Oth
er
Oth
er
Patie
nt d
ue fo
r dis
char
ge b
y bo
ardi
ng o
ut.
Nex
t day
pat
ient
esc
orte
d ba
ck to
ow
n w
ard,
app
rox.
One
hou
r lat
er st
aff f
ound
an
empt
y as
pirin
pac
ket a
t bed
side
. Sp
ecia
l cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
unl
ikel
y th
at st
aff c
ould
be
exp
ecte
d to
ant
icip
ate
risk
of o
verd
ose.
200
68
£2.1
9 Pa
tient
O
ther
O
ther
Pa
tient
atte
nded
Occ
upat
iona
l The
rapy
(OT)
dep
t une
scor
ted.
OT
phon
ed w
ard
right
aw
ay in
form
ing
them
of p
atie
nt’s
dep
artu
re.
Staf
f Nur
se, i
n ca
r, se
arch
ed
loca
l roa
ds a
nd p
olic
e in
form
ed.
Patie
nt re
turn
ed o
f ow
n fr
ee w
ill 2
½ h
ours
late
r.
Spec
ial c
ateg
ory;
Clin
ical
judg
emen
t - ju
dgem
ent t
o al
low
una
ccom
pani
ed v
isit
to O
T de
partm
ent.
69
£336
.00
Staf
f Ph
ysic
al v
iole
nce
Pers
on to
per
son
assa
ult
Patie
nt a
ttem
pted
to le
ave
war
d. H
isto
ry o
f vio
lenc
e ag
ains
t sta
ff a
nd w
as b
eing
es
corte
d du
e to
pre
viou
s abs
cond
ing
atte
mpt
s. P
atie
nt a
ssau
lted
staf
f and
was
br
ough
t to
floor
by
two
staf
f mem
bers
. In
the
proc
ess,
NA
suff
ered
an
inju
ry
whe
n pu
shed
into
a w
ardr
obe.
Spe
cial
cat
egor
y: C
linic
al ju
dgem
ent –
ju
dgem
ent o
n de
cisi
on to
allo
w p
atie
nt to
leav
e w
ard.
70
£4
6.86
Pa
tient
O
ther
O
ther
Pa
tient
with
long
his
tory
of a
bsco
ndin
g w
as g
iven
par
ole
to a
ttend
ther
apy
depa
rtmen
t. A
gree
men
t was
mad
e w
ith p
atie
nt to
go
unes
corte
d to
bui
ld u
p Tr
ust.
Pat
ient
faile
d to
arr
ive
at th
erap
y de
partm
ent a
nd a
sear
ch w
as in
itiat
ed,
polic
e in
form
ed.
Polic
e re
ceiv
ed c
all s
ayin
g pa
tient
was
at h
er so
licito
rs a
nd h
ad
told
them
she
had
take
n an
ove
rdos
e. P
atie
nt w
as th
en ta
ken
to A
&E
for
treat
men
t. S
peci
al c
ateg
ory
of C
linic
al ju
dgem
ent-
judg
emen
t on
reha
bilit
atio
n re
gim
e.
71
£23.
10
Patie
nt
Oth
er
Oth
er
Patie
nt a
bsco
nded
whe
n sh
ould
hav
e be
en a
ttend
ing
OT
depa
rtmen
t. S
he
pres
ente
d he
rsel
f at p
olic
e H
Q st
atin
g sh
e ha
d ta
ken
an o
verd
ose
so w
as ta
ken
to
A&
E. P
atie
nt th
en a
bsco
nded
from
A&
E an
d w
as se
en in
the
grou
nds o
f ho
spita
l. W
as e
vent
ually
retu
rned
by
car.
Spe
cial
cat
egor
y of
Clin
ical
ju
dgem
ent –
clin
icia
n in
cha
rge
of c
are
wis
hes t
o bu
ild T
rust
by
cont
inui
ng to
al
low
una
ccom
pani
ed v
isits
to O
T de
partm
ent.
72
£8.2
4 Pa
tient
O
ther
O
ther
Pa
tient
on
esco
rted
wal
k in
hos
pita
l gro
unds
abs
cond
s. S
earc
h in
itiat
ed w
ith
polic
e an
d tw
o st
aff a
lso
sear
ch in
car
. Po
lice
loca
ted
the
patie
nt a
nd re
turn
ed
him
to w
ard.
Spe
cial
cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty -
all
reas
onab
le m
easu
res i
n pl
ace.
73
£3
4.71
Pa
tient
O
ther
O
ther
Pa
tient
was
on
two
day
pass
for 1
7th b
irthd
ay.
Thre
e m
embe
rs o
f sta
ff w
ere
requ
ired
to c
olle
ct p
atie
nt fr
om h
ome,
to re
turn
to w
ard,
bec
ause
pat
ient
had
be
com
e in
toxi
cate
d w
ith a
lcoh
ol.
Spec
ial c
ateg
ory:
Clin
ical
judg
emen
t –
judg
emen
t to
allo
w b
irthd
ay p
arol
e.
74
£11.
58
Patie
nt
Oth
er
Oth
er
Patie
nt o
bser
ved
on w
ard
15 m
inut
es b
efor
e te
a bu
t by
teat
ime
patie
nt h
ad
absc
onde
d un
witn
esse
d. A
sear
ch o
f the
war
d an
d gr
ound
s was
car
ried
out.
Po
lice
calle
d to
info
rm w
ard
of p
ossi
ble
sigh
tings
. A
sear
ch in
car
car
ried
out.
Po
lice
loca
ted
patie
nt a
nd re
turn
ed h
im to
the
war
d. S
peci
al c
ateg
ory:
Clin
ical
ju
dgem
ent –
judg
emen
t as t
o le
vel o
f obs
erva
tion.
201
75
£1.4
0 Pa
tient
O
ther
O
ther
Pa
tient
gra
nted
15-
min
utes
tim
eout
but
faile
d to
retu
rn.
No
hist
ory
of
absc
ondi
ng.
Polic
e in
form
ed, a
s the
re w
as c
once
rn fo
r pat
ient
’s sa
fety
. A
m
embe
r of s
taff
saw
pat
ient
in to
wn
and
advi
sed
him
to re
turn
to th
e w
ard,
whi
ch
he d
id.
Spec
ial c
ateg
ory
of C
linic
al ju
dgem
ent –
judg
emen
t to
allo
w p
arol
e.76
£8
.11
Patie
nt
Oth
er
Oth
er
Patie
nt a
bsco
nded
from
unl
ocke
d w
ard.
Not
long
afte
rwar
ds w
ard
rece
ived
a c
all
advi
sing
them
pat
ient
was
on
a br
idge
in to
wn.
Uns
ucce
ssfu
l sea
rch
carr
ied
out
in h
ospi
tal c
ar, p
olic
e w
ere
info
rmed
and
they
retu
rned
pat
ient
to w
ard.
Spe
cial
cate
gory
of C
linic
al ju
dgem
ent –
judg
emen
t tha
t pat
ient
was
suite
d to
an
unlo
cked
war
d.
77
£4.1
5 Pa
tient
O
ther
O
ther
Pa
tient
adm
itted
to w
ard
with
long
his
tory
of a
lcoh
ol a
buse
. G
iven
som
ethi
ng to
ea
t and
seem
ed c
omfo
rtabl
e. S
hortl
y af
ter t
he p
atie
nt a
bsco
nded
. A
sear
ch o
f ho
spita
l and
gro
unds
was
car
ried
out b
ut n
ot fo
und.
Pat
ient
was
foun
d at
his
flat
. Sp
ecia
l cat
egor
y: r
isk n
ot fo
rese
eabl
e –
patie
nt w
as c
omfo
rtabl
e at
tim
e of
ab
scon
ding
. 78
£5
.10
Patie
nt
Oth
er
Oth
er
Patie
nt a
gita
ted,
giv
en m
edic
atio
n an
d pu
t on
15 m
inut
es o
bser
vatio
ns (w
ard
unlo
cked
). Pa
tient
abs
cond
ed a
nd se
arch
car
ried
out i
n ho
spita
l car
. Pa
tient
lo
cate
d an
d re
turn
ed to
war
d. S
peci
al c
ateg
ory:
Clin
ical
judg
emen
t – w
ard
unlo
cked
whe
re p
revi
ousl
y ha
d be
en.
79
£4.3
9 Pa
tient
O
ther
O
ther
Pa
tient
abs
cond
ed w
hile
on
gran
ted
paro
le.
Patie
nt h
ad a
his
tory
of a
bsco
ndin
g,
usua
lly to
loca
l pub
. Se
arch
car
ried
out i
n ho
spita
l car
and
pat
ient
retu
rned
to
war
d.Sp
ecia
l cat
egor
y: C
linic
al ju
dgem
ent –
pat
ient
had
his
tory
but
gra
nted
15
min
ute
paro
le.
80
£51.
73
Patie
nt
Phys
ical
vio
lenc
e Pe
rson
to p
erso
n as
saul
t M
ale
patie
nt u
nder
con
stan
t obs
erva
tion
jum
ped
in a
ir ki
ckin
g fe
mal
e pa
tient
w
ithou
t war
ning
. A
larm
act
ivat
ed.
Patie
nt re
stra
ined
and
giv
en m
edic
atio
n.
Spec
ial c
ateg
ory:
rea
sona
ble
prac
ticab
ility
– a
ll pr
actic
able
mea
sure
s in
plac
e,
inci
dent
hap
pene
d ve
ry q
uick
ly.
81
£20.
31
Patie
nt
Phys
ical
vio
lenc
e O
ther
Pa
tient
mad
e th
reat
enin
g ge
stur
e to
NA
twic
e al
so k
icke
d ou
t at w
indo
w ra
diat
or.
Ala
rm a
ctiv
ated
. Te
am re
stra
ined
pat
ient
and
giv
en m
edic
atio
n. S
peci
alca
tego
ry: r
easo
nabl
e pr
actic
abili
ty –
ade
quat
e m
anag
emen
t of t
he si
tuat
ion.
82
£1
83.0
1 Pa
tient
O
ther
O
ther
Pa
tient
with
a su
spec
ted
over
dose
pho
ned
war
d w
hile
out
on
a pa
ss.
Patie
nt
retu
rned
to w
ard
and
had
bloo
d ta
ken.
Ove
rdos
e co
nfirm
ed a
nd p
atie
nt se
nt to
ac
ute
hosp
ital f
or tr
eatm
ent.
Pat
ient
requ
ired
spec
ial n
ursi
ng w
hils
t aw
ay fr
om
war
d.Sp
ecia
l cat
egor
y: C
linic
al ju
dgem
ent –
judg
emen
t to
allo
w p
arol
e.
202
83
£7.0
0 Pa
tient
O
ther
O
ther
Pa
tient
retu
rned
to w
ard
afte
r tw
o ho
urs p
arol
e in
form
ing
staf
f she
had
take
n 34
pa
race
tam
ol.
Bloo
d an
alys
is c
onfir
med
a la
rge
over
dose
. Pa
tient
tran
sfer
red
to
acut
e ho
spita
l, es
corte
d on
jour
ney
by N
A, f
or tw
o da
ys th
en re
turn
ed to
war
d.
Spec
ial c
ateg
ory:
Clin
ical
judg
emen
t - ju
dgem
ent t
o al
low
par
ole.
84
£4
48.8
5 St
aff
Acc
iden
tal i
njur
y Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r sta
irs
Porte
r par
ked
a va
n go
t out
. H
is fo
ot sl
ippe
d of
f ste
p su
stai
ning
ank
le in
jury
. H
e re
turn
ed to
dep
ot so
ano
ther
driv
er c
ould
com
plet
e m
ail d
eliv
erie
s. W
ent t
o A
&E
afte
r shi
ft w
here
it w
as c
onfir
med
that
he
had
pulle
d lig
amen
ts.
Was
off
for a
pe
riod
of si
ck le
ave.
Spe
cial
cat
egor
y: r
isk u
nfor
esee
able
– fo
llow
ed a
ll pr
oced
ures
. 85
£2
8.20
St
aff
Acc
iden
tal i
njur
y Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r sta
irs
Dom
estic
mop
ping
bat
hroo
m fl
oor s
lippe
d an
d fe
ll, in
jurin
g he
r ank
le.
Spec
ial
cate
gory
: ins
uffic
ient
info
rmat
ion
– no
t eno
ugh
info
rmat
ion
avai
labl
e on
in
cide
nt.
86
£559
.76
Staf
f Ph
ysic
al v
iole
nce
Pers
on to
per
son
assa
ult
Patie
nt a
ssis
ted
from
wet
bed
to c
omm
ode
by 2
staf
f. P
atie
nt b
ecam
e ph
ysic
ally
an
d ve
rbal
ly a
busi
ve a
nd sp
at a
t nur
se.
Nur
se o
ff tw
o da
ys w
ith st
ress
, dut
ies
cove
red
by o
ther
nur
ses.
Spe
cial
cat
egor
y: r
isk u
nfor
esee
able
– fi
rst m
ajor
ag
gres
sive
inci
dent
. 87
£8
.77
Patie
nt
Acc
iden
tal i
njur
y Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r sta
irs
Patie
nt w
alki
ng d
own
corr
idor
turn
ed a
nd st
umbl
ed to
floo
r. E
scor
ted
to A
&E
whe
re fr
actu
red
to ri
ght n
eck
of fe
mur
con
firm
ed.
Frac
ture
pin
ned
in th
eatre
, pa
tient
rem
aine
d in
acu
te h
ospi
tal u
ntil
she
died
thre
e da
ys la
ter.
Spe
cial
cate
gory
: rea
sona
ble
prac
ticab
ility
– n
o pr
ior e
vide
nce
of ri
sk o
f fal
ling.
88
£9
.84
Patie
nt
Acc
iden
tal i
njur
y Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r sta
irs
82 y
ear o
ld fe
mal
e pa
tient
with
nig
ht d
emen
tia lo
st b
alan
ce in
toile
t and
fell
hitti
ng h
er h
ead
on d
oorf
ram
e. P
atie
nt h
ad tw
o kn
ee re
plac
emen
ts b
ut
inde
pend
ently
mob
ile w
ith m
obila
tor.
Pat
ient
seen
by
duty
doc
tor a
nd se
nt to
A
&E
for s
titch
es.
Spec
ial c
ateg
ory:
Clin
ical
judg
emen
t – ju
dgem
ent t
o en
cour
age
mob
ilisa
tion.
89
£5
.60
Patie
nt
Acc
iden
tal i
njur
y Fa
ll fr
om h
eigh
t 90
yea
r old
mal
e pa
tient
, reg
iste
red
blin
d, a
wai
ting
disc
harg
e to
nur
sing
hom
e ha
d fa
ll fr
om b
ed.
Cot
side
s in
posi
tion.
Inc
iden
t see
n by
two
staf
f but
they
wer
e to
o fa
r aw
ay to
inte
rven
e. S
uffe
red
frac
ture
d ne
ck o
f fem
ur, a
nd d
ied
with
in o
ne
to tw
o w
eeks
. Sp
ecia
l cat
egor
y: r
easo
nabl
e pr
actic
abili
ty –
pat
ient
und
erst
ood
conc
ept o
f buz
zer f
or h
elp,
cot
side
s in
plac
e.
90
£2.8
0 Pa
tient
A
ccid
enta
l inj
ury
Slip
, trip
or f
all o
n sa
me
leve
l or s
tairs
82
yea
r old
pat
ient
fell
in to
ilet,
unw
itnes
sed.
Pat
ient
had
his
tory
of b
eing
ra
tiona
l and
safe
with
zim
mer
then
like
“a
switc
h” c
ould
sudd
enly
bec
ome
angr
y,
aggr
essi
ve a
nd d
isor
ient
ated
& a
t the
sam
e tim
e un
stea
dy.
Spec
ial c
ateg
ory
of
outw
ith r
easo
nabl
e pr
actic
abili
ty –
all
reas
onab
le c
ontro
l mea
sure
s in
plac
e.
203
91
£8.5
8 Pa
tient
A
ccid
enta
l inj
ury
Slip
, trip
or f
all o
n sa
me
leve
l or s
tairs
Pa
tient
stoo
d up
from
cha
ir w
ith fo
ot tw
iste
d ro
und
chai
r leg
and
fell
to th
e flo
or.
Frac
ture
d fe
mur
. In
itial
ly re
fuse
d su
rger
y th
en h
as o
ther
com
plic
atio
n. S
pend
s 31
day
s in
acut
e ho
spita
l bef
ore
bein
g re
turn
ed to
war
d. S
peci
al c
ateg
ory:
re
ason
able
pra
ctic
abili
ty –
all
reas
onab
le c
ontro
l mea
sure
s in
plac
e.
Tabl
e 5
Trus
t E (4
inci
dent
s)
No
Cost
Staf
f/pat
ient
/oth
er
Cate
gory
Ty
pe
Des
crip
tion
1 £3
,833
.28
Patie
nt
Acc
iden
tal
inju
ry
Slip
, trip
or
fall
on sa
me
leve
l or
stai
rs
Patie
nt w
as b
eing
trea
ted
in su
rgic
al w
ard
but b
ed w
as n
eede
d so
was
tran
sfer
red
to a
n or
thop
aedi
c w
ard
whe
re sh
e w
as to
stay
ove
rnig
ht th
en b
e di
scha
rged
in th
e m
orni
ng.
The
next
da
y pa
tient
was
foun
d ly
ing
on th
e flo
or h
avin
g st
umbl
ed a
nd fa
llen.
Pat
ient
requ
ired
12 d
ay
exte
nded
stay
in o
rthop
aedi
c w
ard.
Spe
cial
cat
egor
y: r
isk n
ot fo
rese
eabl
e –
patie
nt w
as
inde
pend
ently
mob
ile a
nd d
ue to
go
hom
e, th
eref
ore
she
was
con
side
red
capa
ble
of v
isiti
ng th
e to
ilet a
lone
. 2
£5.3
1 Pa
tient
A
ccid
enta
l in
jury
Sl
ip, t
rip o
r fa
ll on
sam
e le
vel o
r st
airs
84 y
ear o
ld m
ale
patie
nt h
ad b
een
adm
itted
to th
e re
habi
litat
ion
unit
havi
ng b
eing
trea
ted
for a
fr
actu
red
neck
of f
emur
. Pa
tient
trie
d to
rise
from
a c
hair
unas
sist
ed a
nd o
verb
alan
ced
and
fell,
su
stai
ning
lace
ratio
ns to
the
face
. Sp
ecia
l cat
egor
y: o
utw
ith r
easo
nabl
e pr
actic
abili
ty –
pa
tient
kno
wn
to re
quire
ass
ista
nce
and
had
been
co-
oper
ativ
e in
the
past
. St
aff w
ere
clos
e by
bu
t cou
ld n
ot p
reve
nt th
e fa
ll.
3 £5
.12
Patie
nt
Acc
iden
tal
inju
ry
Slip
, trip
or
fall
on sa
me
leve
l or
stai
rs
Patie
nt re
ques
ted
hot d
rink.
Giv
en d
rink
and
seat
ed in
her
recl
iner
with
a b
edsi
de ta
ble
and
the
nurs
e ca
ll be
ll pl
aced
to h
and.
Fiv
e m
inut
es la
ter t
he st
aff h
eard
a n
oise
and
on
inve
stig
atio
n th
ey fo
und
the
patie
nt o
n th
e flo
or b
etw
een
the
bed
and
the
recl
iner
. Pa
tient
sust
aine
d tw
o la
cera
tions
on
her f
oreh
ead
and
the
back
of h
er sk
ull a
nd w
as se
nt fo
r x-r
ays.
Spe
cial
cat
egor
y:
outw
ith r
easo
nabl
e pr
actic
abili
ty –
pat
ient
was
kno
wn
to b
e hi
gh ri
sk, b
ut sh
e w
as n
ot
conf
used
and
refu
sed
to g
o to
bed
. St
aff a
ttend
ed fi
ve m
inut
es b
efor
e in
cide
nt.
4 £7
.84
Patie
nt
Acc
iden
tal
inju
ry
Slip
, trip
or
fall
on sa
me
leve
l or
stai
rs
Into
xica
ted
patie
nt a
dmitt
ed to
A&
E. P
atie
nt le
ft th
e cu
bicl
e, b
ecom
ing
enta
ngle
d in
the
curta
in
as h
e di
d so
, fal
ling
to th
e gr
ound
bef
ore
staf
f cou
ld re
ach
him
. Sp
ecia
l cat
egor
y: o
utw
ith
reas
onab
le p
ract
icab
ility
– m
easu
res w
ere
in p
lace
to p
reve
nt in
cide
nt, b
ut w
as la
rgel
y du
e to
in
toxi
cate
d st
ate
of p
atie
nt.
204
Tabl
e 6
Trus
t F (9
inci
dent
s)
No
Cost
Staf
f/pat
ient
/oth
er
Cate
gory
Ty
pe
Des
crip
tion
1 £1
,289
.32
Oth
er
Nea
r mis
s St
ruck
by
an
obje
ct
Cei
ling
had
been
fitte
d w
ithin
the
last
2 y
ears
. O
n da
y of
inci
dent
ther
e w
as h
igh
win
ds a
nd
wea
ther
con
ditio
ns w
ere
exce
ptio
nally
seve
re.
A w
indo
w in
the
staf
f roo
m c
ould
not
be
prop
erly
clo
sed
and
whe
n th
e w
ind
blew
in it
was
thou
ght t
hat a
cur
rent
of a
ir go
t int
o ce
iling
spac
e. T
he c
eilin
g co
llaps
ed in
secr
etar
ies’
off
ice
brin
ging
dow
n lig
ht fi
tting
s.
Spec
ial c
ateg
ory:
lack
of i
nfor
mat
ion
– no
info
rmat
ion
coul
d be
gai
ned
on th
e st
anda
rds
for f
ittin
g ce
iling
s. H
ad b
een
fitte
d by
con
tract
ors,
who
wer
e no
long
er c
onta
ctab
le.
2 £1
,012
.80
Patie
nt
Acc
iden
tal
inju
ry
Slip
77
yea
r old
pat
ient
foun
d ly
ing
on th
e flo
or.
X-r
ay sh
owed
frac
ture
to h
ip.
Patie
nt
trans
ferr
ed to
orth
opae
dic
depa
rtmen
t for
hip
repl
acem
ent r
esul
ting
in a
n up
grad
ed st
ay o
f ei
ght d
ays (
durin
g st
udy
perio
d).
Spec
ial c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
–
patie
nt n
orm
ally
inde
pend
ently
mob
ile.
3 £3
,798
.24
Staf
f A
ccid
enta
l in
jury
Sl
ip
Clin
ical
Sup
port
Wor
ker s
lippe
d on
urin
e on
floo
r nea
r bed
spac
e. S
he w
as se
nt to
A&
E an
d th
en h
ome.
Sic
k le
ave
for a
ppro
x. th
ree
mon
ths a
nd w
as re
plac
ed b
y an
age
ncy
nurs
e.
Spec
ial c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– th
e w
ettin
g on
the
floor
was
un
avoi
dabl
e an
d ac
cord
ing
to th
e in
form
atio
n av
aila
ble
no o
ne k
new
whe
n th
e flo
or b
ecam
e w
et so
pro
cedu
re fo
r dea
ling
with
spill
age
was
not
initi
ated
. D
im li
ghtin
g w
as a
lso
esse
ntia
l for
pat
ient
s to
slee
p.
4 £7
.10
Patie
nt
Acc
iden
tal
inju
ry
Slip
83
yea
r old
mal
e pa
tient
got
up
to u
se th
e bo
ttle.
Sta
ff h
eard
a lo
ud b
ang
and
they
foun
d th
e pa
tient
lyin
g on
the
floor
by
his b
ed.
He
was
ass
iste
d ba
ck to
bed
by
hois
t. T
he S
HO
st
itche
d la
cera
tion
and
sent
pat
ient
for s
kull
x-ra
y (e
scor
ted
by D
gra
de st
aff n
urse
).
Spec
ial c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– p
atie
nt’s
con
fusi
on w
as a
n in
crea
sing
pro
blem
and
he
pers
iste
d on
tryi
ng to
get
out
of b
ed b
y cl
imbi
ng o
ver c
ot si
des
or g
oing
to b
otto
m o
f bed
. 5
£42.
85
Oth
er
Oth
er
Nee
dles
tick/
sh
arps
W
hile
gra
fting
pat
ient
in c
ardi
ac th
eatre
the
surg
eon
notic
ed th
at o
ne o
f the
sutu
re n
eedl
es
was
mis
sing
. A
thor
ough
sear
ch w
as c
arrie
d ou
t and
the
rout
ine
ches
t x-r
ay w
as re
view
ed
as p
er h
ospi
tal p
olic
y. N
eedl
e w
as n
ot o
bser
ved
on x
-ray
. Sp
ecia
l cat
egor
y: la
ck o
f in
form
atio
n –I
mpo
ssib
le to
asc
erta
in o
n th
e ba
sis o
f the
info
rmat
ion
avai
labl
e.
6 £6
.00
Oth
er
Oth
er
Oth
er
ENT
Con
sulta
nt re
quire
d an
alte
rnat
ive
trach
eoto
my
tray
be b
roug
ht b
y ta
xi fr
om a
noth
er
hosp
ital t
o pe
rfor
m e
mer
genc
y pr
oced
ure.
Spe
cial
cat
egor
y: r
isk u
nfor
esee
able
– th
e st
erili
sing
cen
tre p
repa
red
inst
rum
ent t
rays
in a
ccor
danc
e w
ith n
orm
al re
ques
ts a
nd w
ere
not a
war
e th
at th
is c
onsu
ltant
requ
ired
a sp
ecia
l ins
trum
ent t
o co
nduc
t the
pro
cedu
re.
205
7 £7
.45
Patie
nt
Acc
iden
tal
inju
ry
Patie
nt li
fting
/ H
andl
ing
Patie
nt su
stai
ned
a sh
ould
er in
jury
whi
lst b
eing
rolle
d on
bed
into
hoi
st sl
ing.
Spe
cial
cate
gory
: risk
unf
ores
eeab
le –
clin
ical
con
ditio
n of
pat
ient
not
fully
app
reci
ated
at t
he
time
of th
e in
cide
nt.
Had
staf
f kno
wn
that
the
patie
nt’s
can
cer h
ad m
etas
tasi
sed
to b
ones
th
ey w
ould
n’t h
ave
hois
ted.
8
£238
.65
Oth
er
Fire
Ex
posu
re to
fire
Fr
idge
in w
ard
caug
ht fi
re.
Poss
ible
cau
se sp
ilt m
ilk ru
nnin
g fr
om b
ase
of c
abin
et b
etw
een
case
and
fasc
ia c
over
. Fi
re b
rigad
e w
as c
alle
d al
thou
gh tw
o st
aff n
urse
s wer
e ab
le to
ex
tingu
ish
the
fire.
Spe
cial
cat
egor
y of
risk
unf
ores
eeab
le –
staf
f una
war
e th
at m
ilk h
ad
been
spilt
. 9
£200
.00
Oth
er
Oth
er
Oth
er
Win
dow
was
blo
wn
in b
y w
ind
smas
hing
ove
r an
unoc
cupi
ed b
ed a
nd th
e flo
or in
war
d.
Esta
tes w
ere
cont
acte
d to
secu
re w
indo
w sa
fely
unt
il co
ntra
ctor
s arri
ved
to re
plac
e gl
ass.
Sp
ecia
l cat
egor
y: r
isk u
nfor
esee
able
– b
ased
on
the
fact
s kno
wn
risk
coul
d no
t be
fore
seen
.
Tabl
e 7
Trus
t G (1
6 in
cide
nts)
No
Cost
Staf
f/pat
ient
/oth
er
Cate
gory
Ty
pe
Des
crip
tion
1 £2
9.52
Pa
tient
A
ccid
enta
l in
jury
Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r st
airs
Patie
nt in
depe
nden
tly m
obile
with
zim
mer
usu
ally
man
aged
to g
et a
roun
d on
her
ow
n.
She
was
foun
d on
floo
r at h
er b
edsi
de e
ntan
gled
in h
er z
imm
er fr
ame
suff
erin
g fr
om
pain
in h
er h
ip.
X-r
ay c
onfir
med
frac
ture
. Sp
ecia
l cat
egor
y: r
easo
nabl
e pr
actic
abili
ty –
pat
ient
, alth
ough
eld
erly
, was
inde
pend
ently
mob
ile w
ith h
er z
imm
er
fram
e.2
£15.
78
Patie
nt
Oth
er
Oth
er
Patie
nt le
ft un
lock
ed w
ard
with
out p
erm
issi
on.
Foun
d in
car
par
k an
d re
turn
ed to
war
d by
firm
per
suas
ion.
Pat
ient
had
epi
leps
y/br
ain
dam
age/
chal
leng
ing
beha
viou
r. S
peci
alca
tego
ry: C
linic
al ju
dgem
ent –
pat
ient
not
judg
ed to
nee
d co
nsta
nt o
bser
vatio
n.
3 £4
.92
Patie
nt
Oth
er
Oth
er
Patie
nt tr
ied
to le
ave
war
d bu
t per
suad
ed to
retu
rn.
Patie
nt ra
n of
f whi
le a
ccom
pani
ed a
t fr
ont d
oor (
to sm
oke
a ci
gare
tte).
Ala
rm ra
ised
, pat
ient
foun
d on
road
and
per
suad
ed to
re
turn
to w
ard.
Spe
cial
cat
egor
y: r
easo
nabl
e pr
actic
abili
ty –
all
appr
opria
te m
easu
res
in p
lace
to c
ontro
l ris
k, c
omm
ensu
rate
with
pat
ient
car
e.4
£16.
80
Patie
nt
Acc
iden
tal
inju
ry
Slip
, trip
or f
all
on sa
me
leve
l or
stai
rs
Patie
nt fe
ll ou
t of b
ed w
rapp
ed in
bed
ding
. Pr
one
to w
ande
ring
and
did
not u
nder
stan
d us
e of
buz
zer d
ue to
men
tal s
tate
. Sp
ecia
l cat
egor
y: r
easo
nabl
e pr
actic
abili
ty.
Ris
k as
sess
men
t con
clud
ed th
at th
ere
was
no
need
for c
ot si
des o
r ext
ra su
perv
isio
n as
pat
ient
ha
d no
his
tory
of f
allin
g fr
om b
ed.
5 £4
,130
.08
Patie
nt
Acc
iden
tal
inju
ry
Slip
, trip
or f
all
on sa
me
leve
l or
stai
rs
Patie
nt d
ue to
be
disc
harg
ed fe
ll at
bed
side
. C
omm
ode
tang
led
up in
falle
n kn
icke
rs.
Frac
ture
con
firm
ed.
Add
ition
al 3
4 da
y st
ay o
n w
ard.
Spe
cial
cat
egor
y: r
easo
nabl
e pr
actic
abili
ty –
pat
ient
was
inde
pend
ently
mob
ile &
fit.
206
6 £3
00.0
0 St
aff
Phys
ical
V
iole
nce
Pers
on to
Pe
rson
Ass
ault
Nur
se w
as p
unch
ed b
y re
side
nt w
hen
she
was
tryi
ng to
get
the
resi
dent
to d
ress
her
self.
Th
is c
ause
d da
mag
e to
the
nurs
e’s s
pect
acle
s. S
peci
al c
ateg
ory:
rea
sona
ble
prac
ticab
ility
– P
atie
nt la
shed
out
with
out w
arni
ng.
7 £2
1.50
Pa
tient
V
erba
l abu
se o
r th
reat
enin
g be
havi
our
Oth
er
Doc
tor a
dvis
ed p
atie
nt th
at, b
ecau
se h
is c
ondi
tion
had
dete
riora
ted
he w
as to
be
trans
ferr
ed to
an
inte
nsiv
e ps
ychi
atric
car
e un
it (I
PCU
). P
atie
nt re
acte
d to
the
deci
sion
w
ith h
ostil
ity a
nd a
ggre
ssio
n re
quiri
ng th
e st
aff t
o se
cure
the
situ
atio
n w
ith C
&R
te
chni
ques
. Sp
ecia
l cat
egor
y: r
easo
nabl
e pr
actic
abili
ty. A
ppro
pria
te ri
sk a
sses
smen
t ha
d be
en c
arrie
d ou
t. 8
£150
.17
Patie
nt
Phys
ical
vi
olen
ce
Oth
er
Dec
isio
n ha
d be
en ta
ken
to m
ove
a pa
tient
to a
secu
re w
ard.
On
bein
g to
ld th
is a
nd
whi
lst b
eing
inje
cted
with
med
icat
ion,
pat
ient
star
ted
to th
row
him
self
abou
t and
be
com
e ve
ry v
iole
nt.
Mem
ber o
f sta
ff in
jure
d hi
s han
d w
hils
t try
ing
to re
stra
in p
atie
nt.
Spec
ial c
ateg
ory:
rea
sona
ble
prac
ticab
ility
– R
isk
was
reco
gnis
ed, p
atie
nt w
as in
pr
oces
s of b
eing
tran
sfer
red.
9
£17.
54
Patie
nt
Acc
iden
tal
inju
ry
Slip
, trip
or f
all
on sa
me
leve
l or
stai
rs
Patie
nt (s
troke
reha
b.) a
ssis
ted
to b
edsi
de c
omm
ode
and
left
with
buz
zer.
Whe
n st
aff
cam
e ba
ck to
che
ck fo
und
patie
nt ly
ing
on th
e flo
or.
Frac
ture
con
firm
ed re
sulti
ng in
an
exte
nded
stay
on
war
d. S
peci
al c
ateg
ory:
out
with
rea
sona
ble
prac
ticab
ility
– b
uzze
r to
han
d, p
atie
nt c
apab
le o
f usi
ng it
but
app
ears
she
may
hav
e st
ood
hers
elf u
p an
d th
en
lost
bal
ance
. 10
£3
5.08
Pa
tient
A
ccid
enta
l in
jury
Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r st
airs
Patie
nt re
cove
ring
from
stro
ke w
alke
d un
aide
d to
toile
t to
prac
tice
wal
king
. Se
en to
be
falli
ng a
nd g
ently
low
ered
to fl
oor b
y tw
o m
embe
rs o
f sta
ff.
Patie
nt su
stai
ned
inju
ry to
an
kle
and
was
sent
for x
-ray
. Sp
ecia
l cat
egor
y: C
linic
al ju
dgem
ent –
pat
ient
bei
ng
reha
bilit
ated
inju
red
ankl
e w
hile
wal
king
inde
pend
ently
for s
hort
dist
ance
und
er
obse
rvat
ion
of st
aff.
11
£2.8
9 Pa
tient
O
ther
O
ther
79
yea
r old
pat
ient
with
dem
entia
left
war
d un
obse
rved
by
staf
f. P
atie
nt w
as fo
und
in
grou
nds a
nd re
turn
ed to
war
d. S
peci
al c
ateg
ory:
risk
not
fore
seea
ble
– un
char
acte
ristic
beh
avio
ur d
ue to
exc
item
ent o
f goi
ng h
ome.
12
£4
,515
.00
Patie
nt
Acc
iden
tal
inju
ry
Slip
, trip
or f
all
on sa
me
leve
l or
stai
rs
Patie
nt m
obile
with
use
of z
imm
er w
alke
d al
ong
corr
idor
and
left
zim
mer
out
side
toile
t.
Whe
n fin
ishe
d st
ood
up a
nd fe
ll to
floo
r. T
aken
to a
cute
hos
pita
l and
retu
rned
to w
ard
for e
xten
ded
stay
. Sp
ecia
l cat
egor
y: r
isk n
ot fo
rese
eabl
e/cl
inic
al ju
dgem
ent –
pat
ient
di
d no
t req
uire
ass
ista
nce,
mob
ile w
ith z
imm
er.
Staf
f did
not
fore
see
patie
nt le
avin
g zi
mm
er o
utsi
de to
ilet.
13
£21.
93
Patie
nt
Acc
iden
tal
inju
ry
Slip
, trip
or f
all
on sa
me
leve
l or
stai
rs
Patie
nt w
ith A
lzhe
imer
’s, c
onst
antly
shou
ting
rose
from
bed
and
fell
to fl
oor.
Pat
ient
di
d no
t com
preh
end
use
of b
uzze
r. S
ent f
or x
-ray
con
firm
ing
frac
ture
d hu
mer
us a
nd
retu
rned
to w
ard.
Spe
cial
cat
egor
y: r
easo
nabl
e pr
actic
abili
ty –
pat
ient
und
er g
ener
al
obs.
Inde
pend
ently
mob
ile w
ith z
imm
er b
ut re
quire
d as
sist
ance
get
ting
in &
out
of b
ed.
207
14
£89.
40
Patie
nt
Van
dalis
m
Expo
sure
to fi
re
Patie
nt se
ts li
ght t
o he
r bed
line
n an
d ac
tivat
es fi
re a
larm
. Pa
tient
adm
its to
star
ting
fire
to g
ain
atte
ntio
n. F
ire b
rigad
e at
tend
but
fire
was
ext
ingu
ishe
d by
NA
with
fire
ex
tingu
ishe
r. E
ngin
eer c
alle
d in
to re
set f
ire a
larm
. Sp
ecia
l cat
egor
y: r
isk
unfo
rese
eabl
e –
first
tim
e pa
tient
had
set l
ight
to a
nyth
ing.
15
£9
96.8
0 Pa
tient
A
ccid
enta
l in
jury
Sl
ip, t
rip o
r fal
l on
sam
e le
vel o
r st
airs
Patie
nt ro
se fr
om b
ed w
ith m
inim
um a
ssis
tanc
e. W
alke
d to
toile
t una
ided
, tur
ned
to
clos
e do
or, l
ost b
alan
ce a
nd fe
ll. T
rans
ferr
ed to
acu
te h
ospi
tal f
or x
-ray
, fra
ctur
e co
nfirm
ed a
nd d
etai
ned
2 da
ys b
efor
e re
turn
ing
to w
ard
for e
xten
ded
stay
. Sp
ecia
lca
tego
ry: r
easo
nabl
e pr
actic
abili
ty –
pat
ient
inde
pend
ently
mob
ile a
nd n
o in
dica
tion
of in
crea
sed
risk
of fa
lling
. 16
£3
9.69
Pa
tient
V
anda
lism
O
ther
Pa
tient
retu
rned
to ro
om a
fter l
unch
. N
oise
hea
rd fr
om ro
om, w
indo
w fo
und
to b
e br
oken
. Pa
tient
agi
tate
d fo
r no
appa
rent
reas
on.
Patie
nt w
as re
stra
ined
, rem
oved
from
be
droo
m a
nd m
edic
atio
n gi
ven.
Spe
cial
cat
egor
y: C
linic
al ju
dgem
ent –
no
indi
catio
n to
alte
r obs
erva
tion
regi
me.
208
APPENDIX 9 INDIVIDUAL TRUSTS BREAKDOWN OF TELEPHONE INTERVIEW RESPONSES
Figure 1 All Trusts – workbook holder telephone interviews feedback from three monitoring periods
0
10
20
30
40
50
60
70
80
90
100Ar
e yo
u st
ill th
e ho
lder
of
the
wor
kboo
k?
Hav
e yo
u us
ed th
ew
orkb
ook?
Hav
e yo
u fo
und
the
wor
kboo
k he
lpfu
l?
Hav
e yo
u us
ed th
em
anag
er's
aud
it?
Hav
e yo
u us
ed th
ew
orkb
ook
gene
rally
?
Hav
e yo
u fo
cuss
ed in
on
parti
cula
r sec
tions
?
Hav
e yo
u im
plem
ente
dan
y co
ntro
l mea
sure
s as
are
sult
of a
risk
asse
ssm
ent?
Hav
e yo
u us
ed th
e op
tion
appr
aisa
l?
Hav
e yo
u an
y su
gges
tions
for i
mpr
ovem
ents
?
Question Posed
Perc
enta
ge o
f Wor
kboo
k H
olde
rs th
at R
eplie
d Ye
s (o
ut o
f 177
issu
ed)
1st April to 31st July 2001 1st August to 30th November 2001 1st December 2001 to 31st March 2002
209
Figure 2 Trust B
0
20
40
60
80
100
Are
you
still
the
hold
er o
f the
wor
kboo
k?
Hav
e yo
u us
ed th
e w
orkb
ook?
Hav
e yo
u fo
und
the
wor
kboo
k he
lpfu
l?
Hav
e yo
u us
ed th
e m
anag
er's
aud
it?
Hav
e yo
u us
ed th
e w
orkb
ook
gene
rally
?
Hav
e yo
u us
ed p
artic
ular
sec
tions
?
Hav
e yo
u im
plem
ente
d an
y co
ntro
l mea
sure
sfo
llow
ing
risk
asse
ssm
ent?
Hav
e yo
u us
ed th
e op
tion
appr
aisa
l?
Hav
e yo
u an
y su
gges
tions
for i
mpr
ovem
ents
?Question Posed
Perc
enta
ge o
f Wor
kboo
k H
olde
rs S
urve
yed
that
Rep
lied
Yes
(Out
of 2
1, 2
1 &
11
resp
ectiv
ely)
1st April to 31st July 2001 1st August to 30th November 2001 1st December 2001 to 31st March 2002
210
Figure 3 Trust C
0
20
40
60
80
100Ar
e yo
u st
ill th
eho
lder
of t
hew
orkb
ook?
Hav
e yo
u fo
und
the
wor
kboo
khe
lpfu
l?
Hav
e yo
u us
edth
e w
orkb
ook
gene
rally
?
Hav
e yo
uim
plem
ente
dan
y co
ntro
lm
easu
res
Hav
e yo
u an
ysu
gges
tions
for
impr
ovem
ents
?
Question Posed
Perc
enta
ge o
f Wor
kboo
k H
olde
rs th
at R
eplie
d Y
es (o
ut o
f 19,
25
& 1
9 re
spec
tivel
y)
1st April to 31st July 2001 1st August to 30th November 2001 1st December 2001 to 31st March 2002
211
Figure 4 Trust D
0
20
40
60
80
100
Are
you
still
the
hold
er o
fth
e w
orkb
ook?
Hav
e yo
u us
ed th
ew
orkb
ook?
Hav
e yo
u fo
und
the
wor
kboo
k he
lpfu
l?
Hav
e yo
u us
ed th
em
anag
er's
aud
it?
Hav
e yo
u us
ed th
ew
orkb
ook
gene
rally
?
Hav
e yo
u us
edpa
rticu
lar s
ectio
ns?
Hav
e yo
u im
plem
ente
dan
y co
ntro
l mea
sure
sfo
llow
ing
risk
asse
ssm
ent?
Hav
e yo
u us
ed th
eop
tion
appr
aisa
l?
Hav
e yo
u an
ysu
gges
tions
for
impr
ovem
ents
?
Question Posed
Perc
enta
ge o
f Wor
kboo
k H
olde
rs th
at R
eplie
d Ye
s(o
ut o
f 38,
37
& 2
8 re
spec
tivel
y)
1st April to 31st July 2001 1st August to 30th November 2001 1st December 2001 to 31st March 2002
212
Figure 5 Trust E
0102030405060708090
100
Are
you
still
the
hold
er o
f the
wor
kboo
k?
Hav
e yo
u us
ed th
e w
orkb
ook?
Hav
e yo
u fo
und
the
wor
kboo
k he
lpfu
l?
Hav
e yo
u us
ed th
e m
anag
er's
aud
it?
Hav
e yo
u us
ed th
e w
orkb
ook
gene
rally
?
Hav
e yo
u us
ed p
artic
ular
sec
tions
?
Hav
e yo
u im
plem
ente
d an
y co
ntro
lm
easu
res
follo
win
g ris
k as
sess
men
t?
Hav
e yo
u us
ed th
e op
tion
appr
aisa
l?
Hav
e yo
u an
y su
gges
tions
for
impr
ovem
ents
?
Question Posed
Perc
enta
ge o
f Wor
kboo
k H
olde
rs
that
Rep
lied
Yes
1st April to 31st July 2001 1st August to 30th November 2001 1st December 2001 to 31st March 2002
213
REFERENCES
1. NHS Scotland Partnership Forum. Managing health at work. Edinburgh: Partnership Information Network (PIN) Guideline; 2003.
2. Health and Safety Executive. http://www.hse.gov.uk. In; 2003. 3. Health and Safety Executive. Successful Health and Safety Management. first ed: HSE
Books; 1991. 4. Health and Safety Executive. Successful Health and Safety Management. Second ed.
Sudbury: Health and Safety Executive; 1997. 5. Health and Safety Executive. Reducing error and influencing behaviour. Sudbury: HSE
Books; 1999. 6. Gay AS, New NH. Auditing health and safety management systems: a regulator's view.
Occup. Med. 1999;49(7):471-473. 7. Health and Safety Executive. The costs of accidents at work. Sudbury: HSE Books;
1993. 8. Livingston AD, Jackson G, Priestley K. Root causes analysis: Literature review.
Sudbury: HSE Books; 2001. 9. Niven KJM. Accident Costs in the NHS. The Safety & Health Practitioner
1999;September:34-38. 10. National Audit Office. Health and safety in NHS acute hospital Trusts in England.
Report by the Comptroller and Auditor General. London; 1996. 11. Revitalising Health and Safety. http://www.hse.gov.uk/revitalising/. In; 2000. 12. Securing Health Together. http://www.ohstrategy.net/. 2000. 13. Health and Safety Commission. Health and safety targets: How are we doing? A
supplement to the HSC annual report and HSC/E accounts 2001/02. Sudbury: HSE Books; 2002.
14. Department of Health. An organisation with a memory: report of an expert group on learning from adverse events in the NHS. London: The Stationary Office; 2000.
15. Department of Health. Building a safer NHS for patients. www.doh.gov.uk/buildsafenhs/;2001.
16. Department of Health. Doing less harm: improving the safety and quality of care through reporting, analysing and learning from adverse incidents involving NHS patients - key requirements for health care providers: National Patient Safety Agency; 2001.
17. Department of Health. Preventing accidental injury - Priorities for action. Norwich: The Stationary Office ISBN 0-11-322477-X; 2002.
18. Department of Health. The effective management of occupational health and safety services in the NHS. London: The Stationary Office; 2002.
19. Scottish Executive. Towards a safer healthier workplace: http://www.scotland.gov.uk/library2/doc08/shwm-00.htm; 1999.
20. NHS Executive. NHS Indemnity - Arrangement for handling clinical negligence claims against NHS staff. HSG(96)48: http://www.info.doh.gov.uk/doh/coin4.nsf/Circulars?ReadForm; 1996.
21. Scottish Executive. Clinical negligence and other risks indemnity scheme (CNORIS). NHS HDL(2001)65: http://www.show.scot.nhs.uk/sehd/mels/hdl2001_65.htm; 2001.
22. National Institute for Occupational Safety and Health. Guide to evaluating the effectiveness of strategies for preventing work injuries: how to show whether a safety intervention really works. Cincinnati: NIOSH; 2001.
23. Zwerling C. A review of a guide to evaluating the effectiveness of strategies for preventing work injuries: How to show whether a safety intervention really works. Journal of Safety Research 2001;32:501-503.
214
24. Clarke M, Oxman AD, editors. Cochrane reviewers handbook 4.1.4 [Updated October 2001]. Oxford: Update Software: Updated Quarterly; 2001.
25. Crombie IK, Davies HTO. Research in healthcare: Design, conduct and interpretation of health services research. Chichester: Wiley; 1996.
26. Fagard RH, Staessen JA, Thijs L. Advantages and disadvantages of the meta-analysis approach. J. Hypertens. Suppl. 1996;14(2):S9-12.
27. Crombie IK. The pocket guide to critical appraisal: a handbook for health care professionals. London: BMJ Publishing Group; 1996.
28. Rakel H, Gerrard S, Langford I, Shaw K. Evaluating the impact of contact techniques. Sudbury: HSE Books; 1999.
29. Stufflebeam DL. Evaluation checklists: Practical tools for guiding and judging evaluations. American Journal of Evaluation 2001;22(1):71-79.
30. Easterby-Smith M, Thorpe R, Lowe A. Management research: An introduction. London: Sage Publications Ltd.; 1991.
31. Shannon HS, Robson LS, Guastello SJ. Methodological criteria for evaluating occupational safety intervention research. Safety Science 1999;31:161-179.
32. Smallman C. The reality of "Revitalizing Health and Safety". Journal of Safety Research 2001;32:391-439.
33. DeRoo LA, Rautiainen RH. A systematic review of farm safety interventions. Am. J. Prev. Med. 2000;18(4S):51-62.
34. Chilton S, Jones-Lee M, Loomes G, Robinson A, Cookson R, Covey J, et al. Valuing health and safety controls: A literature review. Sudbury: HSE Books; 1998.
35. ACSNI Study Group on Human Factors. Third report: Organising for Safety. Sudbury: HSE Books; 1998.
36. Nivolianitou Z, Oberhagemann D, Lunn G, Markert F, Nessvi K, Rogers R, et al. SAFETYNET - a European network for process safety. Journal of Hazardous Materials 2001;A87:1-10.
37. Oliver D, Hopper A, Seed P. Do hospital fall prevention programs work? A systematic review. J Am Geriatr Soc 2000;48(12):1679-1689.
38. Krause TR, Seymour KJ, Sloat KCM. Long-term evaluation of a behaviour-based method for improving safety performance: a meta-analysis of 73 interrupted time-series replications. Safety Science 1999;32:1-18.
39. Cooper MD. Towards a model of safety culture. Safety Science 2000;36:111-136. 40. Scotney V. Development of a health and safety performance measurement tool.
Sudbury: HSE Books; 2000. 41. Yassi A, Tate R, Cooper JE, Snow C, Vallentyne S, Khokhar JB. Early intervention for
back injured nurses at a large Canadian tertiary care hospital: an evaluation of the effectiveness and cost benefits of a two-year pilot project. Occup. Med. 1995;45(4):209-214.
42. Haslam RA, Bentley TA. Follow-up investigations of slip, trip and fall accidents among postal delivery workers. Safety Science 1998;32:33-47.
43. Ludbrook A, Godfrey C, Wyness L, Parrott S, Haw S, Napper M, et al. Effective and cost-effective measures to reduce alcohol misuse in Scotland: A literature review. Edinburgh: The Stationary Office; 2001.
44. Harper AC, Cordery JL, de Klerk NH, Sevastos P, Geelhoed E, Gunson C, et al. Curtin industrial safety trial: managerial behaviour and program effectiveness. Safety Science 1996;24(3):173-179.
45. Volinn E. Do workplace interventions prevent low-back disorders? If so, why?: a methodologic commentary. Ergonomics 1999;42(1):258-272.
46. Gressel MG. An engineer's perspective of the intervention research workshop. Am J Ind Med 1996;29:382-383.
47. Guldenmund FW. The nature of safety culture: a review of theory and research. Safety Science 2000;34:215-257.
215
48. Hurst NW, Young S, Gibson H, Muyselaar A. Measures of safety performance and attitudes to safety at major hazard sites. Journal of Loss Prevention in the Process Industry 1996;9(2):161-172.
49. Cabrera DD, Isla R, Viledla LD. An evaluation of safety climate in ground handling activities. In: Soekkha HM, editor. Aviation Safety; 1997 27-29 August; Proceedings of the IASC-97 International Aviation Safety Conference, Netherlands; 1997. p. 255-268.
50. Cox SJ, Cheyne AJT. Assessing safety culture in offshore environments. Safety Science 2000;34:111-129.
51. Flin R, Mearns K, O'Connor P, Bryden R. Measuring safety climate: identifying the common features. Safety Science 2000;34:177-192.
52. Mearns K, Whitaker SM, Flin R. Safety climate, safety management practice and safety performance in offshore environments. Safety Science 2002;In Press.
53. St. Vincent M, Tellier C, Petitjean-Roget T. Accidents that occurred in three hospitals in one year. Safety Science 1999;31:197-212.
54. McGrail MP, Tsai SP, Bernacki EJ. A comprehensive initiative to manage the incidence and cost of occupational injury and illness: report of an outcomes analysis. J. Occup. Environ. Med. 1995;37(11):1263-1268.
55. Bracker A, Blumberg J, Hodgson M, Storey E. Industrial hygiene recommendations as interventions: A collaborative model within occupational medicine. Applied Occupational & Environmental Hygiene 1999;14:85-96.
56. Smallman C, John G. British directors perspectives on the impact of health and safety on corporate performance. Safety Science 2001;38:227-239.
57. Rundmo T, Hale AR. Managers' attitudes towards safety and accident prevention. Safety Science 2002;In press.
58. Harper AC, Gunson C, Robinson L, de Klerk NH, Osborn D, Sevastos P, et al. Curtin industrial safety trial: methods and safe practice and housekeeping outcomes. Safety Science 1996;24(3):159-172.
59. Hauer E. Observational before-after studies in road safety. Estimating the effect of highway and traffic engineering measures on road safety. Oxford: Elsevier Science Ltd. ISBN 0-08-043-053-8; 1997.
60. Vassie LH, Lucas WR. An assessment of health and safety management within working groups in the UK manufacturing sector. Journal of Safety Research 2001;32(479-490).
61. Hayes BE, Perander J, Smeko T, Trask J. Measuring perceptions of workplace safety: development and validation of the work safety scale. Journal of Safety Research 1998;29(3):145-161.
62. Lingard H, Rowlinson S. Behaviour-based safety management in Hong Kong's construction industry. Journal of Safety Research 1997;28(4):243-256.
63. Henderson J, Whittington C, Wright K. Accident investigation - the drivers, methods and outcomes. Sudbury: HSE Books; 2001.
64. Scherer RF, Brodzinski JD, Canty AL. An examination of process and outcome differences in health and safety inspections conducted by state and federal agencies. Journal of Safety Research 1997;28(3):203-211.
65. Cowan J. Achieving consistency in grading adverse clinical incidents: does Doing Less Harm do the job? British Journal of Clinical Governance 2002;7(1):63-67.
66. Spangenberg S, Baarts C, Dyreborg J, Jensen L, Kines P, Mikkelsen KL. Factors contributing to the differences in work related injury rates between Danish and Swedish construction workers. Safety Science 2002;in press.
67. Parker SK, Jackson PR, Sprigg CA, Whybrow AC. Organisational interventions to reduce the impact of poor work design. Sudbury: HSE Books; 1998.
68. Niven KJM. A review of the application of health economics to health and safety in healthcare. Health Policy 2002;61(3):291-304.
69. Niven KJM. Economic principles in occupational health and safety. Occupational Health Review 2000;November/December:13-18.
216
70. Health and Safety Commission, editor. Management of Health and Safety at Work Regulations (MHSWR): The Stationary Office; 1999.
71. Health and Safety Commission. Proposals for a new duty to investigate accidents, dangerous occurrences and diseases. Sudbury: HSE Books (www.hes.gov.uk/condocs/);2001.
72. Centre for Hazard and Risk Management. Offshore safety climate assessment toolkit. In. http://www.lboro.ac.uk/departments/ec/JIP/ ed: Loughborough University; 2000.
73. Kelly DL. How to use control charts for healthcare. Milwaukee: American Society for Quality (ASQ).
74. Carey RG, Lloyd RC. Measuring quality improvement in healthcare: A guide to statistical process control applications. Milwaukee: Amercian Society for Quality (ASQ).
75. Duncan AJ. Quality Control and Industrial Statistics. Homewood, Illinois: Irwin; 1986. 76. Statistical Package for the Social Sciences. SPSS version 10. In. version 10 ed. Woking:
http://www.spss.com/uk; 2001. 77. Emory CW, Cooper DR. Chapter 6 Measurement. In: Business Research Methods:
Irwin; 1991. 78. Siegel S. Nonparametric Statistics for the Behavioural Sciences. In: P.32. New York:
McGraw-Hill; 1956. 79. Anderson NH. Scales and Statistics: Parametric and Nonparametric. Psychological
Bulletin 1961;58(4):305-316. 80. Miles MB, Huberman AM. Qualitative data analysis: an expanded source book. 2nd ed.
London: Sage; 1994. 81. Diamantopoulos A, Schlegelmilch BB. Taking the fear out of data analysis. 1st Edition
ed. London: Business Press; 1997. 82. Neter J, Kutner MH, Nachtsheim CJ, Wassweman W. Applied Linear Statistical
Models. 3rd ed: Irwin; 1996. 83. Kerlinger FN, Lee HB. Foundations of Behavioural Research. 4th ed. Fort Worth, TX:
Harcourt College Publishers; 2000. 84. Health and Safety Executive. A Guide to the Reporting of Injuries, Diseases and
Dangerous Occurrences Regulations 1995 (RIDDOR). Sudbury: HSE Books; 1996. 85. Health and Safety Executive. The Costs of Accidents at Work. Second ed. Sudbury:
Health and Safety Executive; 1997. 86. Slack N, Chambers S, Harland C, Harrison A, Johnston R. Operations Management.
London: Pitman Publishing; 1995. 87. Neder, Wasserman, Kutner. Applied Linear Statistical Models (3rd Edition). 88. Emory CW, Cooper DR. Business Research Methods (4th Edition). 4th Edition ed:
Irwin; 1991. 89. http://www.richland.cc.il.us/james/lecture/m113/post_anova.html. In. 90. De Bono E. Six thinking hats. London: Penguin ISBN 0-14-013784-X; 1985. 91. Levenstein C. Policy Implications of intervention research: research on the social
context for intervention. Am J Ind Med 1996;29:358-361. 92. Adams SJ. Projecting the next decade in safety management: A Delphi technique study.
Professional Safety 2001;October:26-29. 93. Wagenaar WA, van der Schrier J. Accident analysis. The goal and how to get there.
Safety Science 1997;26(1/2):25-33. 94. Culvenor J. Comparison of team and individual judgements of solutions to safety
problems. Safety Science 2003;In Press. 95. Falconer L, Hoel H. Occupational safety and health: A method to test the collection of
"grey data" by line managers. Occup. Med. 1997;47(2):81-89. 96. Davies M, Faulkner D. Evaluating health and safety inspection in EU candidate
countries. Journal of the Institution of Occupational Safety and Health 2002;6(2):65-78.
217
97. Swuste P, Arnoldy F. The safety adviser/manager as an agent of organisational change: a new challenge to expert training. Safety Science 2003;41:15-27.
98. Hawkins J, Booth RT. Safety and health management system guidance. Journal of the Institution of Occupational Safety and Health 1998;2(2):7-24.
99. Wynn PA, Aw TC, Williams NR, Harrington M. Teaching of occupational medicine to undergraduates in UK schools of medicine. Medical Education 2002;36:697-701.
100. Whitaker S, Wynn P, Williams N. Occupational health teaching for pre registration nursing students. Nurse Education Today 2002;22:152-158.
101. Health and Safety Commission. Management of health and safety at work regulations; 1992.
102. Hale AR. Occupational health and safety professionals and management: identity, marriage, servitude or supervision? Safety Science 1995;20:233-245.
103. Hale AR. Conditions of occurrence of major and minor accidents. Journal of the Institution of Occupational Safety and Health 2001;5(1):7-21.
104. Karageorgiou A, Jensen PL, Walters D, Wilthagen T. Risk assessment in four member states of the European Union. In: Wilthagen T, editor. Systematic Occupational Health and Safety Management. Oxford: Pergamon; 2000. p. 251-284.
105. Dorman P. If safety pays, why don't employers invest in it? In: Wilthagen T, editor. Systematic Occupational Health and Safety Management. Oxford: Pergamon; 2000. p. 351-365.
106. Health and Safety Commission Health Services Advisory Committee. Management of health and safety in the health services - information for directors and managers. Sudbury: HSE Books; 1994.
107. Scottish Executive. Minimum dataset for occupational health and safety in the NHS Scotland. NHS HDL(2001)22 2001;http://www.show.nhs.uk/sehd/mels/HDL2001_22.htm.
108. Hale A. Culture's confusion. Safety Science 2000;34:1-14. 109. Goldenhar LM, Schulte PA. Intervention research in occupational health and safety. J.
Occup. Med. 1994;36(7):763-775. 110. Smith D, Hunt G, Green C. Managing safety the systems way. BS8800 to OHSAS
18001 - Implementing a cost-effective management system for occupational health and safety. Second ed. London: British Standards Institution; 2000.
111. Scottish Executive. http://www.show.scot.nhs.uk/isd/. In; 2002. 112. Health and Safety Executive. The costs to britain of workplace accidents and work-
related ill health in 1995/96 (second edition). Sudbury: HSE Books ISBN 0-7176-1709-2; 1999.
113. Health and Safety Executive. HSE Ready Reckoner "Reduce Risks - Cut Costs". In: http://www.hse.gov.uk/costs/; 2002.
114. Wright M, Marsden S. Changing business behaviour - would bearing the true cost of poor health and safety performance make a difference? Sudbury: HSE Books; 2002.
218
BIBLIOGRAPHY
1. National Audit Office. A safer place to work - Improving the management of health and safety risks to staff in NHS trusts. London: The Stationary Office; 2003.
2. Department of Health. Building a safer NHS for patients. www.doh.gov.uk/buildsafenhs/; 2001.
3. Health and Safety Commission. The health and safety at work etc. act. London: HMSO; 1974: ISBN 010 543774 3.
4. Health and Safety Commission. Management of health and safety at work regulations: The Stationary Office; 1992.
5. Health and Safety Commission. Management of health and safety at work regulations: The Stationary Office; 1999.
6. Karageorgiou A, Jensen PL, Walters D, Wilthagen T. Risk assessment in four member states of the European Union. In: Wilthagen T, editor. Systematic Occupational Health and Safety Management. Oxford: Pergamon; 2000. p. 251-284.
7. Department of Health. Working together, securing a quality workforce for the NHS: Managing violence, accidents and sickness absence in the NHS. The performance management process for incidents of violence to staff, accidents and sickness absence. In. www.open.gov.uk.doh/nawnhs/hrstrat.htm; 1999.
8. Scottish Executive. Towards a safer healthier workplace. In: http://www.scotland.gov.uk/library2/doc08/shwm-00.htm; 1999.
9. National Audit Office. Health and safety in NHS acute hospital trusts in England. Report by the Comptroller and Auditor General. London; 1996.
10. Health and Safety Commission Strategic Plan 2001 -2004. http://www.hse.gov.uk/aboutus/plans/hscplans/plan0104.htm. In; 2001.
11. Health and Safety Commission. Health and safety targets: How are we doing? A supplement to the HSC annual report and HSC/E accounts 2001/02. Sudbury: HSE Books; 2002.
12. Health and Safety Executive. A guide to the reporting of injuries, diseases and dangerous occurrences regulations 1995 (RIDDOR). Sudbury: HSE Books: L73; 1996: ISBN 0 7176 1012 8.
13. Department of Health. An organisation with a memory: report of an expert group on learning from adverse events in the NHS. London: The Stationary Office; 2000.
14. Department of Health. Doing less harm: improving the safety and quality of care through reporting, analysing and learning from adverse incidents involving NHS patients - key requirements for health care providers: National Patient Safety Agency; 2001.
15. Department of Health. The effective management of occupational health and safety services in the NHS. London: The Stationary Office; 2002.
16. Department of Health. NHS Indemnity arrangements for handling clinical negligence claims against NHS staff. Health Service Guidelines HSG(96)48. In: http://www.info.doh.gov.uk/doh/coin4.nsf/Circulars/; 1996.
17. Scottish Executive. Clinical negligence and other risks indemnity scheme (CNORIS). In. NHS HDL(2001)65: http://www.show.scot.nhs.uk/sehd/mels/hdl2001_65.htm; 2001.
18. Health and Safety Executive. Successful health and safety management (HSG65). 2nd ed. Sudbury: Health and Safety Executive; 1997.
19. Health and Safety Commission. The health and safety system in Great Britain. 3rd Edition ed. Sudbury: HSE Books; 2002.
20. Wright M, Lancaster R, Jacobson-Maher C, Talwalkar M, Woolmington T. Evaluation of the good health is good business campaign. Sudbury: HSE Books; 2000.
219
21. Monnery N. The costs of accidents and work-related ill-health to a cheque clearing department in a financial services organisation. Safety Science 1999;31:59-69.
22. Revitalising Health and Safety. http://www.hse.gov.uk/revitalising/. In; 2000. 23. Securing Health Together. http://www.ohstrategy.net/. 2000. 24. Wright M, Marsden S. Changing business behaviour - would bearing the true cost of
poor health and safety performance make a difference? Sudbury: HSE Books; 2002. 25. Health and Safety Executive. Successful health and safety management (HSG65). 1st
ed: HSE Books; 1991. 26. Health and Safety Executive. Reducing error and influencing behaviour. Sudbury: HSE
Books; 1999. 27. British Standards Institution. Guide to occupational health and safety management
systems; BS 8800. London: British Standards Institution; 1996. Report No.: BS 8800. 28. British Standards Institution. OHSAS 18001:1999 Occupational health and safety
management systems - Specification. London: British Standards Institution; 1999. 29. British Standards Institution. OHSAS 18002:2000 Occupational health and safety
management systems - Guidelines for the implementation of OHSAS 18001. London: British Standards Institution; 2000.
30. Smith D, Hunt G, Green C. Managing safety the systems way. BS8800 to OHSAS 18001 - Implementing a cost-effective management system for occupational health and safety. Second ed. London: British Standards Institution; 2000.
31. Frick K, Wren J. Reviewing occupational health and safety management - Multiple roots, diverse perspectives and ambiguous outcomes. In: Wilthagen T, editor. Systematic Occupational Health and Safety Management. Oxford: Pergamon; 2000. p. 17-42.
32. Nielsen KT. Organisation theoroes implicity in various approaches to OHS management. In: Wilthagen T, editor. Systematic Occupational Health and Safety Management. Oxford: Pergamon; 2000. p. 99-123.
33. Harper B. A comparison of the national and organisational structures for health and safety in the United Kingdom and Germany. Journal of the Institution of Occupational Safety and Health 2001;5(1):23-42.
34. Niven KJM. Accident costs in the NHS. The Safety & Health Practitioner 1999;September:34-38.
35. Health and Safety Executive. The costs of accidents at work. Sudbury: HSE Books; 1993.
36. Health and Safety Executive. The costs of accidents at work. 2nd ed. Sudbury: Health and Safety Executive; 1997.
37. Davies NV, Teasdale P. The costs to the british economy of work accidents and work-related ill health. Sudbury: HSE Books; 1994.
38. Health and Safety Executive. The costs to Britain of workplace accidents and work-related ill health in 1995/96. 2nd ed. Sudbury: HSE Books; 1999: ISBN 0 7176 1709 2.
39. Kase D, Wiese K. Safety auditing: A management tool. New York: Van Nostrand; 1990.
40. Glendon I. Risk management in the 1990s: Safety auditing. Journal of Occupational Health and Safety, Australia and New Zealand 1995;11(6):569-575.
41. Saunders R. The safety audit: Designing effective strategies. London: Pitman; 1992. 42. Gay AS, New NH. Auditing health and safety management systems: a regulator's view.
Occup. Med. 1999;49(7):471-473. 43. Livingston AD, Jackson G, Priestley K. Root causes analysis: Literature review.
Sudbury: HSE Books; 2001. 44. National Institute for Occupational Safety and Health. Guide to evaluating the
effectiveness of strategies for preventing work injuries: how to show whether a safety intervention really works. Cincinnati: NIOSH; 2001.
220
45. Smallman C. The reality of "Revitalizing Health and Safety". Journal of Safety Research 2001;32:391-439.
46. DeRoo LA, Rautiainen RH. A systematic review of farm safety interventions. Am. J. Prev. Med. 2000;18(4S):51-62.
47. Chilton S, Jones-Lee M, Loomes G, Robinson A, Cookson R, Covey J, et al. Valuing health and safety controls: A literature review. Sudbury: HSE Books; 1998.
48. ACSNI Study Group on Human Factors. Third report: Organising for Safety. Sudbury: HSE Books; 1998.
49. Beahler CC, Sundheim JJ, Trapp NI. Information retreival in systematic reviews - challenges in the public health arena. Am. J. Prev. Med. 2000;18(4S):6-10.
50. Rivara FP, Thompson DC. Systematic reviews of injury-prevention strategies for occupational injuries. Am. J. Prev. Med. 2000;18(4(S)):1-3.
51. Budworth N. Indicators of performance in safety management. The Safety & Health Practitioner 1996;November:23-29.
52. Hale AR, Heming B, Carthey J, Kirwan B. Modelling of safety management systems. Safety Science 1997;26(1/2):121-140.
53. Emslie SV, Lowe RF. Safecode and successful management of health and safety risk in the national health service. Health Informatics 1995:91-100.
54. Health and Safety Commission Health Services Advisory Committee. Management of health and safety in the health services - information for directors and managers. Sudbury: HSE Books; 1994.
55. Chaplin R, Hale AR. An evaluation of the use of the international safety rating system (ISRS) as intervention to improve the organisation of safety. In: Baram M, editor. Safety Management. The Challenge of Change. London: Pergamon; 1998. p. 165-185.
56. Guldenmund FW. The nature of safety culture: a review of theory and research. Safety Science 2000;34:215-257.
57. Hale AR, Hovden J. Management and culture: the third age of safety. A review of approaches to organizational aspects of safety, health and environment. In: Occupational Injury. Risk, Prevention and Intervention P129-165; 1998.
58. Cox SJ, Cheyne AJT. Assessing safety culture in offshore environments. Safety Science 2000;34:111-129.
59. Centre for Hazard and Risk Management. Offshore safety climate assessment toolkit. In. http://www.lboro.ac.uk/departments/ec/JIP/ ed: Loughborough University; 2000.
60. Brown R, Holmes H. The use of a factor-analytic procedure for assessing the validity of an employee safety climate model. Accid. Anal. & Prev. 1986;18(6):455-470.
61. Dedobbeleer N, Beland F. A safety climate measure for construction sites. J. Safety Res. 1991;22:97-103.
62. Niven KJM. A review of the application of health economics to health and safety in healthcare. Health Policy 2002;61(3):291-304.
63. St. Vincent M, Tellier C, Petitjean-Roget T. Accidents that occurred in three hospitals in one year. Safety Science 1999;31:197-212.
64. Rakel H, Gerrard S, Langford I, Shaw K. Evaluating the impact of contact techniques. Sudbury: HSE Books; 1999.
65. Mearns K, Whitaker SM, Flin R. Safety climate, safety management practice and safety performance in offshore environments. Safety Science 2003;41(8):641-680.
66. Cooper MD. Towards a model of safety culture. Safety Science 2000;36:111-136. 67. Flin R, Mearns K, O'Connor P, Bryden R. Measuring safety climate: identifying the
common features. Safety Science 2000;34:177-192. 68. Easterby-Smith M, Thorpe R, Lowe A. Management research: An introduction.
London: Sage Publications Ltd.; 1991. 69. Hurst NW, Young S, Gibson H, Muyselaar A. Measures of safety performance and
attitudes to safety at major hazard sites. Journal of Loss Prevention in the Process Industry 1996;9(2):161-172.
221
70. Cabrera DD, Isla R, Viledla LD. An evaluation of safety climate in ground handling activities. In: Soekkha HM, editor. Aviation Safety; 1997 27-29 August; Proceedings of the IASC-97 International Aviation Safety Conference, Netherlands; 1997. p. 255-268.
71. Jacobs R, Haber S. Organizational processes and nuclear power plant safety. Reliability Engineering and System Safety 1994;45:75-83.
72. Haber S, Metlay D, Crouch D. Influence of organizational factors on safety. Proceedings of the Human Factors Society 1990:871-875.
73. Shannon HS, Mayr J, Haines T. Overview of the relationship between organizational and workplace factors and injury rates. Safety Science 1997;26(3):201-217.
74. Health and Safety Executive. Successful Health and Safety Management. first ed: HSE Books; 1991.
75. Department of Health. Preventing accidental injury - Priorities for action. Norwich: The Stationary Office; 2002: ISBN 0 11 322477 X.
76. Clarke M, Oxman AD, editors. Cochrane reviewers handbook 4.1.4 [Updated October 2001]. Oxford: Update Software: Updated Quarterly; 2001.
77. Elvik R. Assessing the validity of evaluation research by means of meta-analysis. Oslo: Institute of Transport Economics; 1999. Report No.: 430.
78. Crombie IK, Davies HTO. Research in healthcare: Design, conduct and interpretation of health services research. Chichester: Wiley; 1996.
79. Krause TR, Seymour KJ, Sloat KCM. Long-term evaluation of a behaviour-based method for improving safety performance: a meta-analysis of 73 interrupted time-series replications. Safety Science 1999;32:1-18.
80. Oliver D, Hopper A, Seed P. Do hospital fall prevention programs work? A systematic review. J Am Geriatr Soc 2000;48(12):1679-1689.
81. Guastello SJ. Do we really know how well our occupational accident prevention programs work? Safety Science 1993;16:445-463.
82. Rogers B, Goodno L. Evaluation of interventions to prevent needlestick injuries in health care occupations. Am. J. Prev. Med. 2000;18(4(S)):90-98.
83. Rivara FP, Thompson DC. Prevention of falls in the construction industry: evidence for program effectiveness. Am. J. Prev. Med. 2000;18(4S):23-26.
84. Lingard H, Rowlinson S. Behaviour-based safety management in Hong Kong's construction industry. Journal of Safety Research 1997;28(4):243-256.
85. Harper AC, Gunson C, Robinson L, de Klerk NH, Osborn D, Sevastos P, et al. Curtin industrial safety trial: methods and safe practice and housekeeping outcomes. Safety Science 1996;24(3):159-172.
86. Yassi A, Tate R, Cooper JE, Snow C, Vallentyne S, Khokhar JB. Early intervention for back injured nurses at a large Canadian tertiary care hospital: an evaluation of the effectiveness and cost benefits of a two-year pilot project. Occup. Med. 1995;45(4):209-214.
87. Hale AR. Editorial: special issue for the 10th anniversary of the safety science group and the tu Delft. Safety Science 1997;26(1/2):1.
88. Swuste P, Arnoldy F. The safety adviser/manager as an agent of organisational change: a new challenge to expert training. Safety Science 2003;41:15-27.
89. McGuire T, Hanson M, Moody J. Management attitudes throughout Scotland towards the funding, provision and use of mechanical aids for moving and handling purposes. Edinburgh: Lothian NHS Occupational Health Service; 1995.
90. Sheehan JJ. Cost-benefit anaysis: a technique gone awry. In: Rom WN, editor. Legal and ethical dilemmas in occupational health: Ann Arbor Science: The Butterworth Group; 1983. p. 51-215.
91. Gressel MG. An engineer's perspective of the intervention research workshop. Am J Ind Med 1996;29:382-383.
92. Hale A, Swuste P. Safety rules: procedural freedom or action constraint? Safety Science 1998;29:163-177.
222
93. Hale A, Swuste P. Avoiding square wheels: international experience in sharing solutions. Safety Science 1997;25(1-3):3-14.
94. Hillage J, Tyers C, Davis S, Guppy A. The impact of the HSC/E: A review. Sudbury: HSE Books; 2001.
95. Scottish Health Management Efficiency Group. Accidents in the NHS in Scotland. Edinburgh: SCOTMEG; 1993. Report No.: SP54.
96. Bentler PM, Wu E. EQS for Windows: User's Guide. In: Multivariate Software, Inc. Encino, CA.; 1995.
97. Kelly DL. How to use control charts for healthcare. Milwaukee: American Society for Quality (ASQ).
98. Carey RG, Lloyd RC. Measuring quality improvement in healthcare: A guide to statistical process control applications. Milwaukee: Amercian Society for Quality (ASQ).
99. Health and Safety Executive. Successful Health and Safety Management. Second ed. Sudbury: HSE Books; 1997.
100. Niven KJM. Economic principles in occupational health and safety. Occupational Health Review 2000;November/December:13-18.
101. Statistical Package for the Social Sciences. SPSS version 10. In. version 10 ed. Woking: http://www.spss.com/uk; 2001.
102. Emory CW, Cooper DR. Chapter 6 Measurement. In: Business Research Methods: Irwin; 1991.
103. Siegel S. Nonparametric Statistics for the Behavioural Sciences. In: P.32. New York: McGraw-Hill; 1956.
104. Anderson NH. Scales and Statistics: Parametric and Nonparametric. Psychological Bulletin 1961;58(4):305-316.
105. Miles MB, Huberman AM. Qualitative data analysis: an expanded source book. 2nd ed. London: Sage; 1994.
106. Diamantopoulos A, Schlegelmilch BB. Taking the fear out of data analysis. 1st Edition ed. London: Business Press; 1997.
107. Neter J, Kutner MH, Nachtsheim CJ, Wassweman W. Applied Linear Statistical Models. 3rd ed: Irwin; 1996.
108. Kerlinger FN, Lee HB. Foundations of Behavioural Research. 4th ed. Fort Worth, TX: Harcourt College Publishers; 2000.
109. Health and Safety Executive. A Guide to the Reporting of Injuries, Diseases and Dangerous Occurrences Regulations 1995 (RIDDOR). Sudbury: HSE Books; 1996.
110. National Health Service. National Patient Safety Agency. In: http://www.npsa.org.uk/;2001.
111. Slack N, Chambers S, Harland C, Harrison A, Johnston R. Operations Management. London: Pitman Publishing; 1995.
112. Hale AR. Conditions of occurrence of major and minor accidents. Journal of the Institution of Occupational Safety and Health 2001;5(1):7-21.
113. Neder, Wasserman, Kutner. Applied Linear Statistical Models (3rd Edition). 114. Emory CW, Cooper DR. Business Research Methods (4th Edition). 4th Edition ed:
Irwin; 1991. 115. Scottish Executive. Towards a safer healthier workplace:
http://www.scotland.gov.uk/library2/doc08/shwm-00.htm; 1999. 116. Scottish Executive. Minimum dataset for occupational health and safety in the NHS
Scotland. NHS HDL(2001)22 2001;http://www.show.nhs.uk/sehd/mels/HDL2001_22.htm.
117. National Patient Safety Agency. NPSA pilot project evaluation report: National Patient Safety Agency; 2002.
118. Wagenaar WA, van der Schrier J. Accident analysis. The goal and how to get there. Safety Science 1997;26(1/2):25-33.
223
119. Neale G, Woloshynowych M, Vincent C. Exploring the causes of adverse events in NHS hospital practice. Journal of the Royal Society of Medicine 2001;94:322-330.
120. Cox SJ, Janes WH, Walker D, Wenham DG. Tolley's Health and Safety Handbook. Croydon: Tolley Publishing Co. Ltd. ISBN 9 780854 599981; 1995.
121. Culvenor J. Comparison of team and individual judgements of solutions to safety problems. Safety Science 2003;In Press.
122. Falconer L, Hoel H. Occupational safety and health: A method to test the collection of "grey data" by line managers. Occup. Med. 1997;47(2):81-89.
123. Health and Safety Commission. Proposals for a new duty to investigate accidents, dangerous occurrences and diseases. Sudbury: HSE Books (www.hes.gov.uk/condocs/);2001.
124. Scottish Executive. http://www.show.scot.nhs.uk/isd/. In; 2002. 125. Health and Safety Executive. The costs to Britain of workplace accidents and work-
related ill health in 1995/96. 2nd ed. Sudbury: HSE Books; 1999: ISBN 0 7176 1709 2. 126. Health and Safety Executive. HSE Ready Reckoner "Reduce Risks - Cut Costs". In:
http://www.hse.gov.uk/costs/; 2002. 127. Dorman P. If safety pays, why don't employers invest in it? In: Wilthagen T, editor.
Systematic Occupational Health and Safety Management. Oxford: Pergamon; 2000. p. 351-365.
128. Koopmanschap MA, Van Ineveld BM. Towards a new approach for estimating indirect costs of disease. Soc. Sci. Med. 1992;34(9):1005-1010.
129. Wright M, Marsden S. Changing business beahviour - would bearing the true cost of poor health and safety performance make a difference? Sudbury: HSE Books; 2002.
130. Smith M, Cohen H, Cohen A, Cleveland R. Characteristics of successful safety programs. J. Safety Res. 1978;10(1):5-15.
131. Simonds R, Shafai-Sahrai Y. Factors apparently affecting injury frequency in eleven matched pairs of companies. J. Safety Res. 1977;9(3):120-127.
132. Cohen A. Factors in successful occupational safety programs. J. Safety Res. 1977;9(4):168-178.
133. Shafai-Sahrai Y. Determinants of occupational injury experience: A study of matched pairs of companies. East Lansing, Michigan: MSU Business Studies; 1973.
134. Cohen A, Smith M, MH. C. Safety program practices in high versus low accident rate companies - An interim report: National Institute for Occupational Safety and Health (NIOSH); 1975.
135. Hovden J, Tinmannsvik R. Internal control: A strategy for occupational safety and health. Experiences from Norway. J. Occup. Accidents 1990;12:21-30.
136. Ronald L. Identifying the elements of successful safety programs: A literature review. Richmond, British Columbia: Workers' Compensation Board of British Columbia; 1999.
137. Saari J, editor. Successful accident prevention: recommendations and ideas field tested in the Nordic countries. The final report of the Nordic cooperative program of effective accident prevention methods. Helsinki: Institute of Occupational Health; 1987.
138. Goldenhar LM, Schulte PA. Intervention research in occupational health and safety. J. Occup. Med. 1994;36(7):763-775.
139. Wright M, Brabazon P, Tipping A, Talwalkar M. Development of a business excellence model of safety culture: Safety culture improvement matrix. Sudbury: HSE Books; 1999.
140. Ogden J. Health psychology - a textbook. second ed. Buckingham: Open University Press; 2000.
141. Houghton J, Neck C, Manz C. We think we can, we think we can, we think we can: the impact of thinking patterns and self-efficacy on work team sustainability. Team Performance Management 2003;9(1/2):31-41.
142. De Bono E. Six thinking hats. London: Penguin ISBN 0-14-013784-X; 1985.
224
143. Shivers CH. Halos, horns and hawthorne: Potential flaws in the evaluation process. Professional Safety 1998;March:38-41.
144. NHS Executive. NHS Indemnity - Arrangement for handling clinical negligence claims against NHS staff. HSG(96)48: http://www.info.doh.gov.uk/doh/coin4.nsf/Circulars?ReadForm; 1996.
145. McGrail MP, Tsai SP, Bernacki EJ. A comprehensive initiative to manage the incidence and cost of occupational injury and illness: report of an outcomes analysis. J. Occup. Environ. Med. 1995;37(11):1263-1268.
146. Harper AC, Cordery JL, de Klerk NH, Sevastos P, Geelhoed E, Gunson C, et al. Curtin industrial safety trial: managerial behaviour and program effectiveness. Safety Science 1996;24(3):173-179.
147. Chandler DL. Rocket science is not the hard part. New Scientist 2003;20/27 December 2003:14-15.
Printed and published by the Health and Safety ExecutiveC30 1/98
Printed and published by the Health and Safety ExecutiveC1.10 10/04
RR 280
£25.00 9 78071 7 6291 1 4
ISBN 0-7176-2911-2