Leicester, Leicestershire and Rutland Clinical Quality Audit A RETROSPECTIVE CLINICAL QUALITY AUDIT ACROSS PRIMA RY, SECONDARY AND COMMUNITY CARE (SUMMER 2017)

July 2018

FINAL REPORT

2

Acknowledgements

This report presents a lot of numbers. The team recognise that each number represents a loved one and would like to give their condolences to the families of every person referred to in the

following chapters.

_______

We reviewed patient confidential information relating to many deaths as part of this work. We have used the information therein to help inform our conclusions. It is not appropriate to share

any of this information in the report in order to respect confidentiality. However, we have provided some anonymous information by way of illustration. We are grateful for feedback from

relatives to inform our report.

We would like to thank everyone who assisted in the production of this review. Mazars worked in association with Clarity Informatics who provided software for our work. Particular thanks are due to Dr Gerry Morrow and the team at Clarity Informatics and to Tony Roberts, Deputy Director (Clinical Effectiveness) at South Tees NHS Foundation Trust.

_______

Mazars is an integrated, international audit, tax and advisory firm with a presence in 80 countries. We have over 130 partners and over 1,600 employees working across the UK.

Review team:

Mary-Ann Bruce (RGN, RSCN, RHV, MBA) - Director

Dierdre Dwyer – Associate, RGN, RHV

Dr Jean MacLeod – Consultant Physician

Dr Diane Monkhouse – Consultant in Critical Care

Dr Jane Povey - Associate, GP and Medical Manager

Tracey Sparkes – Manager, RGN

3

Detailed contents List of charts, tables and figures 6

Charts 6

Tables 7

Figures 7

1. Summary 8

Background 8

Scope 8

Key themes 11

Recommendations 18

2. What were relatives’ views? 21

Key findings 21

Is relatives’ feedback consistent with reviewers’ assessment of quality of care? 24

3. Profile of full cohort and reviews undertaken 26

3.1 Comparison of full cohort and reviewed cohort 26

3.2 Key characteristics of the cohort 27

3.3 Detailed profile of reviewed cohort 28

Deaths by CCG 28

Age Range 28

Gender 30

Ethnic Origin 31

Place of Death 32

Place of Residence 33

GP practices 33

Causes of death 34

Type of admission 35

Day of admission 36

Admission time 37

Deaths by Day of the Week 38

Days to Death Post Discharge 39

Admitting Specialty to UHL 41

Discharge Specialty 41

Ward moves 42

Length of stay 42

Patients with a Mental Health Need or with a Learning Disability 43

Palliative and End of Life Care Coding 44

4. Findings 45

4.1 Overall care 46

Number assessed and ratings for Overall Care 46

4

Notable themes 47

Variation in phases of care 47

Quality of patient records and relationship to Quality of Care ratings 47

Memory problems/confusion and care ratings 50

Ratings by age band 51

Overall problem types 51

Problem types by age band 52

Identifying positive and negative lessons from the review 54

4.2 Preadmission care 58

Number assessed and ratings for Preadmission phase 58

Notable themes 60

Use of Admission Avoidance 60

Admission time by age 62

Ambulance provision – preadmission: 64

Dehydration 66

UTI 67

Pressure sores 67

Families struggling to cope 68

Warfarin management 69

Preadmission phase – characteristics of good/excellent care 70

Preadmission phase – characteristics of poor/very poor care 72

4.3 Initial Management and Admission 75

Number assessed and ratings for Initial Management and Admission phase of care 75

Notable themes 76

Initial Management and Admission phase – characteristics of good/excellent care 77

Initial Management and Admission phase – characteristics of poor/very poor care 79

4.4 Ongoing care 80

Number assessed and ratings for Ongoing Care phase 80

Notable themes 81

Fluid Balance monitoring 81

Antibiotic management 81

Diabetic management 81

Weight loss and weight monitoring 82

Family involvement – getting it right 83

Medication reconciliation/review and drug errors 84

Ongoing Care phase – characteristics of good/excellent care 84

Ongoing Care phase – characteristics of poor/very poor care 86

4.5 Procedure Care 88

Number assessed and ratings for Care during a Procedure phase 88

Care During a Procedure phase – characteristics of good/excellent care 89

5

4.6 Perioperative care 90

Number assessed and ratings for Perioperative phase 90

Perioperative phase – points to note across all care ratings 91

4.7 Readmission care 93

Number assessed and ratings for Readmission phase 93

Case examples by care rating 94

4.8 Discharge care 96

Number assessed and ratings for Discharge phase 96

Notable themes 97

Fast track 97

Discharge phase – characteristics of good/excellent care 100

Discharge phase – points to note in cases of poor/very poor care 101

4.9 End of Life Care 102

Number assessed and ratings for End of Life phase of care 102

Notable Themes 103

Engagement with families regarding/at End of Life 103

End of Life for patients with confusion/memory problems 103

Death certification/involvement of Medical Examiner (ME) 103

Organ Donation 104

GP change at end of life 104

Clinical provision at end of life 105

DOLS 106

DNACPR and EoL plans 106

End of Life - characteristics of good/excellent care 107

End of Life – characteristics of poor/very poor care 109

Appendices (separate document) 111

6

List of charts, tables and figures

Charts

Chart 1 Relatives concerns and compliments ............................................................................................................. 22

Chart 2 Full cohort compared with reviewed cohort by age ......................................................................................... 26

Chart 3 Reviews by CCG v full cohort ......................................................................................................................... 28

Chart 4 Number of deaths by age band – full cohort ................................................................................................... 29

Chart 5 Age comparisons between full and reviewed cohort ....................................................................................... 29

Chart 6 Age band by gender – full cohort .................................................................................................................... 30

Chart 7 Age band by gender – reviewed cohort .......................................................................................................... 30

Chart 8 Ethnic origin – full and reviewed cohort .......................................................................................................... 31

Chart 9 Place of death – Community deaths ............................................................................................................... 32

Chart 10 Summary causes of death (edited) ............................................................................................................... 34

Chart 11 Admission route ............................................................................................................................................ 35

Chart 12 Number of deaths (by admission day) full cohort .......................................................................................... 36

Chart 13 Number of deaths (by admission day) reviewed cohort ................................................................................ 37

Chart 14 Admissions by day and time – full cohort ...................................................................................................... 37

Chart 15 Number of deaths by day of the week (full cohort) ........................................................................................ 38

Chart 16 Number of deaths by day of week – reviewed cohort ................................................................................... 38

Chart 17 Days to death post discharge – full cohort .................................................................................................... 39

Chart 18 Days to death post discharge – reviewed cohort .......................................................................................... 39

Chart 19 Length of time between discharge and death by age band ........................................................................... 40

Chart 20 Number of deaths by admitting specialty ...................................................................................................... 41

Chart 21 Discharge Specialty ...................................................................................................................................... 41

Chart 22 Number of cases with Mental Health need or Learning Disability ................................................................. 43

Chart 23 Overall Care ratings ...................................................................................................................................... 46

Chart 24 Quality of care records by Overall Care rating (%) ....................................................................................... 49

Chart 25 Ratings by Age band..................................................................................................................................... 51

Chart 26 Total number of problem types ..................................................................................................................... 52

Chart 27 Problem Types by Age band ......................................................................................................................... 53

Chart 28 Care ratings for Preadmission phase ............................................................................................................ 59

Chart 29 Preadmission ratings by Age Band ............................................................................................................... 59

Chart 30 Admission Avoidance by Age Band .............................................................................................................. 61

Chart 31 Admission time by age band ......................................................................................................................... 63

Chart 32 Initial Management and Admission ratings ................................................................................................... 75

Chart 33 Ongoing Care ratings .................................................................................................................................... 80

Chart 34 Care during a Procedure ratings ................................................................................................................... 88

Chart 35 Perioperative Care ratings ............................................................................................................................ 90

Chart 36 Perioperative Care ratings by age band ........................................................................................................ 91

Chart 37 Readmission ratings ..................................................................................................................................... 93

Chart 38 Readmission ratings by age band ................................................................................................................. 94

Chart 39 Discharge Care ratings ................................................................................................................................. 96

Chart 40 End of Life Care ratings .............................................................................................................................. 102

7

Tables

Table 1 Summary of Care ratings by phase (numbers) ............................................................................................... 12

Table 2 Summary of all care ratings (%) by phase ...................................................................................................... 12

Table 3 Admissions time by age band (full cohort) ...................................................................................................... 14

Table 4 Comparison of reviewer overall rating and relatives views ............................................................................. 24

Table 5 Relatives UHL compliments v reviewer ratings ............................................................................................... 25

Table 6 Relatives UHL concerns v reviewer ratings .................................................................................................... 25

Table 7 Full cohort compared with reviewed cohort to assess representative sample ................................................ 26

Table 8 Average and Median ages at death ................................................................................................................ 28

Table 9 Comparison of age bands by gender between full and reviewed cohort ........................................................ 31

Table 10 Place of death – full cohort ........................................................................................................................... 32

Table 11 Place of death - reviewed cohort .................................................................................................................. 32

Table 12 Distance in miles to LGH .............................................................................................................................. 33

Table 13 Causes of death IA (edited) by location ........................................................................................................ 34

Table 14 Total deaths by admission day ..................................................................................................................... 36

Table 15 Length of time post discharge to death by age ............................................................................................. 40

Table 16 Specialty moves – full cohort ........................................................................................................................ 42

Table 17 Specialty moves – reviewed cohort .............................................................................................................. 42

Table 18 Length of stay ............................................................................................................................................... 42

Table 19 Palliative care coding – full cohort ................................................................................................................ 44

Table 20 Palliative care coding – reviewed cohort....................................................................................................... 44

Table 21 Relationship between quality of care rating (Overall Care) & Quality of Care records ................................. 48

Table 22 Overall care scores by confusion/memory problems .................................................................................... 50

Table 23 Ratio of problem types by age band ............................................................................................................. 53

Table 24 Problem types by age band (%) ................................................................................................................... 54

Table 25 Positive lessons by Overall Care rating ........................................................................................................ 55

Table 26 Negative lessons - Communication by care rating ........................................................................................ 55

Table 27 Positive lessons in relation to Escalation by rating ....................................................................................... 56

Table 28 Negative aspects of Escalation by rating ...................................................................................................... 56

Table 29 Positive lessons in relation to Ceilings of care by care rating ....................................................................... 57

Table 30 Negative aspects of Ceilings of Care by rating ............................................................................................. 57

Table 31 Use of Admission Avoidance Schemes ........................................................................................................ 60

Table 32 Admission time by age .................................................................................................................................. 63

Table 33 Procedure codes for Perioperative Care ....................................................................................................... 91

Figures

Figure 1 Deaths by postcode for full cohort – local view .............................................................................................. 33

8

1. Summary

Background The Learning Lessons Programme across Leicester, Leicestershire and Rutland (LLR) is a joint cross-organisational improvement programme, involving the three CCGs and two Trusts in LLR. The programme was established as a result of the findings of a review, published in 2014, which identified fragmentation of care across the LLR health system. The findings identified actions to be addressed across the system. This programme of work is now embedded in the clinical work streams across LLR, and is overseen by the Learning Lessons to Improve Care System Wide Clinical Task Force (CTF).

The LLR system committed to undertake a further clinical audit and commissioned Mazars LLP to undertake this work.

This review was the first of its kind using Structured Judgement Review methodology across systems instead of individual organisations.

Scope Conventional structured judgement reviews often concentrate on the final episode of care and are typically focused on secondary care. The aim of this review was to provide a more system-wide view of quality of care across organisations for patients in the last weeks of their life by reviewing patients’ notes across secondary, community and primary care.

The review was retrospective and undertaken shortly after the month of death. The period chosen meant that the review focussed on the following cohort of patients:

• All deaths in University Hospitals of Leicester (UHL) from June 21st to July 20th

• All deaths at Leicester Partnership NHS Trust (LPT) Community Hospitals from June 21st to July 20th

• All community deaths in the 30 days after discharge from UHL from July 21st to August 20th

(This excluded babies and children and deaths on mental health wards.)

The review was also designed to include feedback from relatives of the deceased patients. This was undertaken via the UHL Medical Examiners and the UHL Bereavement Support Nurses.

The full cohort that was applicable to the audit amounted to 319 deaths (the full cohort) during the period described above. We reviewed case records from 181 patients (57%) in total with 177 cases being given an overall care rating (the reviewed cohort).

We used an adapted Structured Judgement Review (SJR) methodology for the review with the adaptions being agreed in advance with the review Steering Group. The full detail of the case note review methodology is provided in Appendix 1.

The main additions to the conventional review method was to add a pre-admission phase and an end of life phase of care. This meant that the Overall care rating was an overall assessment of

9

the care across the system and was made up of all the phases throughout the patients care. The phases were:

• Preadmission • Initial Management and Admission • Ongoing Care • Care During a Procedure • Perioperative Care • Readmission • Discharge • End of Life

By reviewing all phases we have been able to identify some key themes for the Learning Lessons Taskforce to consider that affect the overall pathway as well as issues relating predominantly to specific phases of care.

Reflections:

This is the first time that a review using Structured Judgement Review methodology has been attempted. It required considerable engagement and agreement between all parties to facilitate the audit. This effort by all parties should be applauded. Approaches to relatives, access to hard copy records, access to electronic records and systems, provision of secure logins and facilities required co-operation between a wide number of organisations and individuals and were organised by the LLR organisations.

The engagement and co-operation of primary care staff, medical records teams and information governance leads were key to success.

There is much to be learnt from the process from all parties to facilitate such a review in future. Process, engagement and mixed review team are all key.

Lessons are:

• identifying the period for review in advance is critical for the scope • dedicated engagement from medical examiners and bereavement staff to talk to

relatives • collating and storing hard copy records well in advance and ordering them for easy

access • support to ensure information governance protocols were adhered to and patient

identifiable data is protected (no patient identifiable data was downloaded or removed from site)

• secure access to EMIS and SystemOne is complex and upfront engagement with primary care is beneficial

• dedicated medical reviewers with experience in SJR and experience of acute care combined with primary care physicians enables a whole system perspective of good practice across the pathway. A mixed team facilitates a more robust pathway assessment.

• adapting the SJR methodology to suit a pathway review and agreeing with all parties, and

• a protocol for raising concerns throughout the review if needed.

10

Making a judgement across a system of care is subjective and based on the specific review teams’ perspective. It is well documented that various teams rate care differently. Having one team reviewing all cases we consider has gone some way to mitigating this to provide a fair and reasonable assessment of each case and the themes arising for the purposes of overall improvement.

The review team included 2 Consultants experienced in SJR in acute care including a Critical Care Consultant and a Consultant Physician. We had a GP on the team too which was also invaluable in providing primary care input and insight and assessing the quality of care in primary care. The combined team collaborated with 3 nursing reviewers to provide a combined perspective on the quality of care when further team discussion was required. This also enabled a second review to take place where either specialist knowledge was required or an individual team member required a second opinion.

We agreed at the outset that should any case cause immediate concern this would be raised directly on site. Specific cases that highlighted the need for local review outside the audit were also highlighted. This ensured additional case reviews were carried out where appropriate.

The structure of the report is:

• A summary of the key themes (Section 1) • Relatives feedback (Section 2) • An analysis of the full cohort (Section 3) • An analysis of the reviewed cohort with care quality ratings and descriptions of phase

based findings (Section 4)

The key themes arising from our work are described below and we focus on system aspects in this summary. Areas to note and issues arising in specific phases of care are described in the more detailed sections of the report.

11

Key themes Summary of cohort profile

We summarise the analysis of the cohort below:

• 75% of patients were over 71 years of age • The ethnic mix of the patients did not reflect the demographic make-up of the county –

only 11% of the full cohort were from an Asian background • Two thirds of the deaths occurred in the acute hospital setting; one third outside with only

10 patients dying in a hospice • The vast majority of patients had been admitted through the emergency care route from

home or a care home • Tuesdays and Fridays were the most common days for admissions of people who

subsequently die within 30 days of discharge • There were fewer deaths on weekend days than weekdays • The number of deaths on a Monday in the community was double that of any other day of

the week • 60% of patients died within 14 days of discharge – and older people (over 71) died sooner

after discharge • 30% of patients moved specialty between admission and discharge/death • 51% of the reviewed patients had some form of mental health need: 42% of reviewed

patients had confusion (in some cases the cause may not have been from dementia) or memory problems and 9% of the cohort had a specified Serious Mental Illness

• 1 person with a Learning Disability was identifiable amongst the cohort. • One third of the patients had a palliative/end of life code in place

Overall quality of care The overall quality of care across the LLR system w as rated as adequate, good or excellent in 84% (148) of cases. Good or excellent ratings we re given in 91 (51.4%) cases overall.

In individual phases of care the quality of care was assessed as adequate, good or excellent in the majority of cases – ranging from 82.5% in Discharge care to 100% in Peri-operative care. See Table 1 below.

Initial management and admission was the phase rated most highly of all phases with 93.6% of cases rated as adequate, good or excellent amongst the major phases (those highlighted in blue below) and this phase had the highest percentage of good or excellent ratings (65.1%).

Perioperative care (whilst small numbers) was also very highly rated with all care rated as adequate, good or excellent.

12

The tables below summarise the overall care ratings.

Table 1 Summary of Care ratings by phase (numbers)

Table 2 Summary of all care ratings (%) by phase

Major care phases were considered as being Pre-admission, Initial Management and Admission, Ongoing Care, Discharge Care and End of Life Care. Pre-admission included all care before arrival in a secondary care setting and includes primary care, community care and ambulance care settings.

Poor care or very poor care ratings were apparent in all the major care phases ranging from 26 cases (16.7%) at End of Life, 24 cases in preadmission care (15.8%), 21 cases (13.6%) in Ongoing Care to 10 cases (17.5%) in Discharge care. There were only 11 cases (6.4%) in Initial Management and Admission rated as poor or very poor. It is important to note that different phases related to different numbers of cases as not all primary care records were available and not all patients were discharged.

However, 22 cases (12.4%) were rated as Poor care and 7 (4%) as Very Poor care overall.

For very poor care ratings and areas which reviewers felt merited further consideration from the system promptly, we provided anonymised summaries of six cases as well as an additional five more where a rating of very poor had been given.

The tables above suggest four key areas of system care to focus on:

• Preadmission;

Phase of

care

Pre-

admission

Initial

Management

& Admission

Ongoing

Care

Procedure

Care

Peri-

operative

Care

Discharge Readmission End of

Life Overall

Excellent 9.9% 15.1% 10.4% 21.4% 20.0% 8.8% 6.7% 10.3% 5.6%

Good 48.7% 50.0% 51.3% 46.4% 70.0% 42.1% 53.3% 44.9% 45.8%

Adequate 25.7% 28.5% 24.7% 25.0% 10.0% 31.6% 13.3% 28.2% 32.2%

Poor 11.2% 4.1% 12.3% 7.1% 0.0% 14.0% 20.0% 13.5% 12.4%

Very

Poor

Care

4.6% 2.3% 1.3% 0.0% 0.0% 3.5% 6.7% 3.2% 4.0%

13

• Ongoing Care on the wards;

• Discharge; and

• End of Life care.

Relatives views and patient/family engagement and communication

Relatives were predominantly complimentary of the care in all phases. It was notable that the issues that relatives raised were often concerns that would not have been recorded separately in the case records and indicates the value of the combined approach to review in identifying areas for improvement. Key relatives issues were:

• a relative not being seen when requested in primary care

• delays to ambulances or transport

• lack of community provision and care co-ordination

• individual concerns which were incident specific, related to staff attitude or delays

Relatives provided feedback in 50% of cases reviewed (97/181). Reviewers and relatives views were largely consistent. Concerns were raised in 20% of reviewed cases and compliments in 39% of reviewed cases. Relatives raised compliments in 9 cases (31%) where reviewers had given a poor or very rating and concerns in 17 cases (18%) where reviewers had given a good or excellent rating.

Good communication was the most commonly observed feature in the review with well documented conversations on prognosis and treatment plans with the patient and their families being a key feature of care and decision making in all phases.

Whilst there were many examples of diligent care and support in the community in primary care, we observed families expressing their difficulties in coping with caring for their relatives at home. This was apparent in 39 (21.5%) cases.

Relatives commented frequently on the compassion and kindness shown by many staff. This was evident throughout many records we reviewed.

Cumulative impact on quality of care when access is delayed for elderly patients

The most significant theme arising was the cumulative impact of care for the elderly and in particular those with confusion/memory problems. Whilst the cohort had an average age of 77 years, the very elderly (those over 81) tended to fare worse across the system in overall terms.

We describe the detail later in the report but noted a number of clinical characteristics in presenting cases that should be areas for improvement efforts to avoid admissions (UTIs, dehydration, pressures sores, weight loss). We comment on the extent of these issues preadmission in particular in more detail later. System issues were identified that meant care was not provided in the optimum setting or that care was delayed (lack of advance care planning, limited options in the community and cumulative delays in providing care).

14

Families were often struggling to cope with caring for relatives at home due to a lack of support to do so. This was noted in 39 cases (21.5%).

In summary, a picture emerges of elderly patients being admitted necessarily due to deterioration. (143 cases were seen as not clinically appropriate for admission avoidance at the time of presentation). However, the necessity could be mitigated by preventative measures to deal with care pre-admission, providing some clinical solutions in community hospitals or nursing homes, more focussed support for families caring for elderly relatives at home and a recognition of each part of the system acting promptly within their own environment. Cumulatively care is delayed if each part focuses only on their own target.

Initial management and admission is rated very highly against this context with some element of delay in A&E (which is not a focus for this review). However, the delays on both sides of the emergency admission itself can quickly add up. There is a system picture of signs of deterioration occurring at home/in care home, a waiting period before calling a GP who may see a patient after surgery, followed by a waiting period for an ambulance (if called) or a period of ‘wait and see’. The case may not require emergency transfer so it can be a few hours before admission to A&E, a wait in A&E followed by a wait for a bed and then being clerked in the early hours of the morning. This appeared to be particularly evident for the more elderly patient who is also at greater risk of deteriorating further during this transfer period.

The chart below shows the resulting peak in admissions for patients in night time hours with 38% (50/133) of over 81 years olds admitted in these hours compared with other age groups.

Table 3 Admissions time by age band (full cohort)

The dominance of cancer (35%) within the cohort and with 51% of cases having a mental health or confusion/memory problem this adds to the pressures on a system that relies on an emergency route of admission.

Elderly patients experience a higher ratio of problem types as those over 81 experienced 30% more problem types. Clinical monitoring was the most common problem type noted in this age group in 35% of cases followed by treatment and management plan problems.

15

Initial management and Admission

It was notable that this phase of care was the most positive phase of care. There was a predominantly emergency route of access to UHL within this cohort. We did not audit waiting times although we comment above on this and some long waits were observed. However, we observed rapid sepsis assessment, prompt administration of antibiotics and IV fluids, liaison with microbiology and timely access to radiology and CT scanning. We observed 2 specific issues in relation to the need to have clear protocols to stabilise patients needing transfer to another hospital (including UHL) and the complexity of the emergency care records bundle.

Community deaths on Mondays

We observed patterns of admission and discharge during the week that indicate a lack of community services may be impacting acute admissions. This needs further investigation but more admissions occurred on Tuesdays, Thursdays and Fridays than other days, with Friday marginally busiest. Fewer deaths occurred at weekends amongst the hospital deaths however, the significant difference in deaths on a Monday in the community needs further understanding. Out of 103 community deaths, 30 died on a Monday. This was 2 to 3 times higher than all other days of the week. These patients had been discharged an average of 19 days previously and only 1 on the day before. There were no obvious common features of this group of deaths and all 15 that were reviewed were rated as adequate, good or excellent care overall.

Clinical monitoring

Pre-alerts from EMAS to A&E for stroke, cardiac and sepsis cases were good. The pre-alerts focussed on these specific conditions and enabled timely assessment for these critical situations.

Sepsis assessment was clearly an uppermost consideration when infection was apparent. Sepsis assessment forms were evident throughout the records – although sometimes being used as a track and trigger tool with multiple copies being completed. There was good liaison with microbiology teams on management of infections and antibiotic protocols. Prompt administration of antibiotics was apparent.

Having been admitted to a ward a significant proportion of patients changed specialty or ward during their stay. However, there was overall good use of specialists supporting the home team – medical, nursing and therapists.

Diabetic management was a need for at least 24 patients. There was variable practice in glucose monitoring and we highlight in the report a number of examples where improvements may be needed. In one case a hypoglycaemic episode may have occurred due to a cumulative delay.

Fluid balance management was identified in 26 cases in particular. 12 cases highlighted concerns and 7 of these cases appear to relate to one specific ward.

Warfarin management was raised in 5 cases with specific reference in relation to falls risk assessment, monitoring levels and taking account of the effect of antibiotic prescriptions.

16

Weight loss and weight management highlighted 10 cases were the need for monitoring weight on the wards was needed in relation to a mix of issues e.g. pressure care, correct prescription doses and nutrition and fluid management.

We observed medicines reconciliation being effective at identifying the need for changes in dose or antibiotic prescription. Medication errors appeared to be identified early.

Perioperative care formed a small part of the reviewed cohort. All cases were rated as adequate, good or excellent care. The main system issue that one case raised was the need to transfer patients between sites and the possible need for stabilisation protocols where direct admission to Glenfield is not possible.

Psychological support was lacking in a small number of specific cases – in particular these were cases where a prognosis of weeks was given unexpectedly or where a relative was struggling to accept the situation.

Specific patient safety issues in relation to falls (1 case) and infection control were minimal (2 cases related to catheter management).

Quality of records

We observed a clear relationship between the quality of care records (largely based on the hard copy records at UHL) and the quality of care. Record quality was markedly better where care was also rated highly and vice versa.

Discharge and support at home

On discharge fast track arrangements appear to be effective in 62% of cases where fast track was part of the discharge process. However, there are specific issues regarding DNACPR arrangements and a lack of weekend cover for approval which caused delays and uncertainty in some cases.

Whilst occupational therapy/physiotherapy support to get a patient assessed for discharged was efficient with an ability to get equipment in place when needed, community physiotherapy not always provided post discharge for those needing to mobilise which was due to a lack of prioritisation by therapy services. We observed cases where long waits for support to mobilise left patients housebound and at risk.

Escalation problems identified delay in discharge and a lack of an appropriate social care support package in at least 10 cases.

A lack of hospice places and difficulty in finding nursing home placements hampered the ability to secure a preferred place of death at times. Marie Curie support couldn’t be accessed by UHL in one case which may be a system issue.

Funding problems prevented some discharges or caused delays in the last few days of life. This included criteria for, and judgements on, eligibility. In one particular case the insistence on having a DNACPR in place caused a three week delay to discharge.

17

End of Life Care

A lack of clear advance care planning and End of Life plans presented a challenge for ambulance services deciding whether to transfer or not when patients deteriorated.

DNACPR decisions were sought in the majority of cases, however we highlight a number of cases where this did not occur.

End of Life care planning and Advance Care Planning was variable. 55% of cases had a rating of good or excellent. 26 cases (17%) were poor or very poor, with 60% of these cases being for people with confusion or memory problems. People with confusion or memory problems were also not getting access to hospice care.

Additional findings Ethnicity – we saw fewer deaths of people from non-white ethnic backgrounds than would be representative of the local demography. The review could not determine the extent to which this was patient choice or not. The ethnic mix of Leicester was not represented in the cohort of deaths in this period as might be expected. Only 11% of deaths were amongst the ethnic community. Given the significant difference between this and the local demography this needs to be understood better. The numbers were too small to draw conclusions in relation to care but we did note language barriers in 4 cases.

Access to hospice care was limited . 10 patients died in a hospice - 8 had Cancer. Which is similar to the May 2016 findings of the CQC report – A Different Ending. Despite the high levels of dementia only 2 patients had confusion or memory problems. High quality care was noted through the relatives’ feedback on hospice care but it is likely that people with conditions other than cancer and their families are unable to secure this form of care at the end of their lives.

GP changes were noted in situations where patients changed residence to a residential or nursing home. As all these patients are at the end of their lives and the majority die within 2 weeks it means that at this last phase of life they are not known to their new primary care team. Arrangements for continuity at end of life do not appear to be in place.

Organ donation was referred to on one occasion.

18

Recommendations

Pathways

1. Examine the cumulative impact on the timeline for frail, elderly patients when admission is required and identify key pinch points to shorten the elapsed time to ward admission.

2. Examine admission avoidance schemes to establish whether criteria are suitable for very elderly and end of life patients including care coordination, hospice at home, management of acute illness and support to nursing homes

3. Promote a concerted effort to improve advance care planning to support decision making for admission, retaining patients in their preferred place of death and preventing unnecessary admission

4. Promote improved Advance Care Planning across the system in primary care and on

discharge from secondary or community provision

Clinical Management

5. Identify actions to support the prevention of dehydration in the frail elderly patient

6. Identify actions to support the management of UTIs in the frail elderly patient

7. Clinical monitoring issues to focus on include:

a. fluid balance management and recording on wards

b. diabetic management and glucose monitoring/recording throughout the pathway

c. warfarin management including as part of falls risk assessments, monitoring and the additional risks presents on prescribing antibiotics

d. weight management and monitoring particularly in relation to correct medication dose

e. clearer recording of decision making at end of life in regards to completing observations and taking blood glucose reading

f. examining the provision of adequate community therapy services to support

mobilisation on discharge in particular, in patients at risk of pressure sores, with

amputations and at risk of developing chest infections. Look at the prioritisation of

therapy provision in community post discharge to ensure waiting times are

minimised for elderly patients requiring mobilisation

g. examining the availability of TPN in community hospitals

h. securing adequate provision of syringe drivers in the community

i. considering the provision of IV fluid and IV antibiotic administration in community hospitals.

19

8. Stabilisation protocols for transfers to other units (including Glenfield Hospital) should be agreed

9. Cumulative effect elapsed time for elderly patients’ admissions should be reviewed further to include:

a. Management of fluid balance throughout the admission journey b. Monitoring of blood glucose throughout the admission journey c. Reducing late night admissions and identifying any consequent risk factors facing

the older patient d. Examining access to primary care assessment at weekends and early in the

working day 10. Review the support available to ambulance service staff faced with decision making for

admission at End of Life

Process issues

11. Weekend issues to focus on include:

a. Fast track approval processes to ensure decisions are not delayed at weekends

b. Blood taking and blood results being available to GPs/out of hours cover at weekends in community hospitals

12. Ensure DoLs assessments are completed and authorised and capacity assessments are completed for all relevant patients including where DNACPR or best interest decisions are required

13. Criteria for fast track CHC funding should be reviewed to ensure that inappropriate barriers

do not prevent appropriate discharge e.g. DNACPR or perception of lack of imminent

death.

14. Examine ways to reduce the need to change GP practice registration at end of life and consider options for maintaining continuity at end of life

15. Confirm palliative care coding reflects palliative care accurately

16. Clarify the arrangements for seeking and accessing a Marie-Curie service by UHL on discharge

17. Examine ways to prevent ward moves for patients at end of life

Future analysis

18. Undertake to get a better understanding of the use of health care services at the end of life amongst the ethnic population

19. Examine end of life care for people with dementia and their families to secure greater understanding of the specific needs of those caring for relatives at home. This should inform future admission avoidance schemes across health and social care services.

20

20. Examine access to hospice care including those with dementia to establish if there is a

need for greater capacity and choice

21. An approach to clinical governance reviews should be agreed and an agreed model for information sharing if future joint reviews are planned between the CCG, providers and GP practices (including nursing homes or other care settings if possible). Agree a protocol to facilitate future audits by enabling access to GP records (and hospice and care home records) as part of the Learning from Deaths policy.

22. Monitor community deaths to establish if the observation of high levels of deaths on Mondays is replicated in other periods and to understand any specific characteristics

23. An evaluation of the audit process by all parties to seek to improve the process for learning across the NHS and locally should be undertaken.

21

2. What were relatives’ views?

Introduction

This unique case note review across a wide part of ‘the system of care’ has limitations given that it can only rely on what is documented and recorded in case notes and patient records and therefore is based only on the documented view of the professionals providing the care.

Relatives personal experiences may not be documented in full and so the direct view of relatives will provide insight into aspects of care that can be missed in case note reviews – both positive and negative.

Each relative of patients cared for in UHL was provided with an opportunity to give their views on the care their relative received. Relatives are often in a position to experience the full journey of someone’s care which is distinctly different from the organisational view of any given part of a patient’s experience.

We are grateful for the contribution of relatives to this review and welcome the efforts to engage with all families whose loved one has died during the period of the review.

The areas of care that relatives were asked to comment on specifically are in the review form in Appendix 2. For inpatient deaths the Medical Examiners (MEs) at UHL contacted relatives by phone (except where agreed otherwise with the Coroner) and the Bereavement Nursing Service contacted relatives where a death occurred after discharge. Outline guidance was provided to support the Medical Examiners and nurses when contacting relatives to ask questions. Comments were recorded by the MEs and nurses and then uploaded into the reviewers’ data set anonymously.

Key findings Communication with the family is a significant aspect of good care of patients:

• The reviewers specifically noted 52 cases (29%) where there was good communication with the family as part of the positive lessons learnt in communication noted. This was also the most frequently recorded positive aspect.

• The reviewers noted 11 cases (6%) where there was poor communication with the family as part of the negative lessons in communication noted. In these 11 cases only 2 cases were rated as good care overall.

22

The following chart shows how many compliments and concerns were provided by relatives for the review:

Chart 1 Relatives concerns and compliments

In 97 of the 181 cases reviewed, relatives provided some form of feedback on one or more organisations or parts of the system. This covered 89 cases in UHL, 15 regarding GPs, 12 regarding LPT and 18 regarding other organisations. Other organisations included EMAS, LOROS and nursing/residential care homes. Positive feedback showed compassion and high quality care.

Relatives’ feedback demonstrated the care given in primary care was exemplary in a number of cases with regular attendance and good communication. Areas of concern included where there was a lack of attention by GPs including in pain control and delays in admissions or in being seen at home.

Similarly the care at LPT was complimented when staff demonstrated care and compassion including prompt attendance when needed. Issues that caused concern were specific examples including - lack of community provision, a funding issue, lack of care coordination and escalation.

The EMAS concerns related predominantly to transport – either to take people into hospital in an emergency or take them home.

Relatives feedback in relation to care provided at UHL was very positive. Relatives commented on the quality of nursing, medical and all other staff care. Negative feedback included specific concerns that would not be reflected in the case records such as attitudes of staff or observing delays or specific incidents.

01020304050607080

79

43

0 3 1 4 110

39

25

2113

2213 16

7

Compliments and Concerns - full cohort

In-hospital deaths Community deaths

23

UHL – positive and negative feedback from same case s

The comments below are as recorded from relatives

UHL compliments UHL Concerns

All of the departments involved at UHL were excellent. The family appreciated the frankness and honesty.

Patients discharge was delayed due to delays in ordering the oxygen and bed.

The nursing, domestic and catering staff were amazing and gave excellent care.

The manner of a couple of the junior doctors was very unapproachable. They didn't seem to like being asked questions by patient's wife. Had an issue with husband's discharge, but declined to elaborate as felt the phone was not the best medium.

My daughter was treated with dignity and compassion and was not stigmatised due to her lifestyle choices (Which had happened in previous years)

The communicating of blood test and scan results was often delayed.

On mum's last admission - in GH - everyone was lovely and caring.

On previous occasion GP sent mum to admissions unit as he wanted her admitted to UHL. The attitude of the nurse on the admissions unit (can't recall who she was) was absolutely appalling. Her comments were "this is an assessment unit, we don't admit here we look at you and send you home." There was no empathy or compassion. In fact when mum needed admission the last time I insisted we would not allow her to be taken to LRI and fortunately she was taken to GH. 2. We spent some considerable time trying to find a wheelchair to take mum from the car to the assessment unit and this was very distressing for everyone.

All of the staff whenever mum was admitted were very caring and supportive.

Concerned about the appropriateness of repeated admissions to Glenfield in last few weeks of her life rather than being in a community setting appropriate to her needs. Palliative care doctor at [community hospital] said the repeated admissions were not good

Student nurse on the ward was very good and attentive

Daughter felt that the seriousness of the condition was not communicated well - felt that "pneumonia" should have been used instead of chest infection. Were not expecting him to die. Felt the patient was discharged before his full recovery

Compliment to ward 19 staff Concern that treatment for pneumonia was delayed / not sufficiently vigorous - i.e. oral antibiotics

Care was generally good Son felt that there was a lack of nurses on the ward, and therefore his father may not have been optimally monitored.

Described nursing care as 'exemplary' Concerned that normal psychoactive drugs omitted for several days after both admissions leading to paranoid state. Also two failed "disastrous" discharges according to family. They have already raised this as an issue and are in contact with [name].

Nursing care on ward 16 was good - lots of respect and empathy

Family felt that there was a lack of joined-up care with each area only focussing on one group of problems, and not seeing the patient holistically. Also, was the patient moved from ACB too soon?

Commended Dr [name] for his care and time. Ward staff were empathetic and helpful

Difficult and unhelpful interaction with Dr [name] (cons) who appeared not to take on board the various concerns and was pressing for discharge

24

The case study below is an example of the insights from a relative over the journey of care.

We compared relatives’ feedback with the reviewers’ ratings and identified a number of cases where our case note review identified good or excellent practice but where the relatives had identified concerns. We also identified cases where relatives’ feedback was complimentary but where the reviewers had identified poor or very poor care. This is illustrated in the table below:

Table 4 Comparison of reviewer overall rating and relatives views

Overall Care rating of reviewed cohort and relative s views Reviewers rating No of ratings Compliments from relatives Concerns from relatives

Very Poor Care 7 2 3

Poor Care 22 7 5

Adequate Care 57 25 12

Good Care 81 29 14

Excellent Care 10 6 3

Total 177 69 37

Is relatives’ feedback consistent with reviewers’ a ssessment of quality of care? In 17 cases relatives had some concerns where reviewers had rated care as good or excellent overall.

There were 9 cases where relatives were complimentary and reviewers had rated care as poor or very poor.

We were able to map 35 UHL cases of relatives’ concerns to our care ratings. The reviewers considered ALL cases as adequate, good or excellent at Initial Management and Admission phase. In the Ongoing phase of care the reviewers considered the care adequate or poorer in 15 of these cases. Relatives’ concerns tended to relate to the Ongoing phase of care as many of the comments related to ward based care.

The more significant observations from relatives were reflected in the reviewers’ ratings. Areas of concern from relatives in cases of good or excellent care ratings were issues that would not typically be documented in case notes and often related to specific experiences or observations.

System case study (one case)

The GP team went above and beyond in their care. The clinical team at [Community Hospital] were wonderful. So caring and supportive. Mum was in the palliative care suite at [Community Hospital] when she was told by the Matron that she couldn't stay there anymore because she didn't meet the continuing health care criteria and therefore rather than a nursing home the family would need…. There were no trained nurses to attend to mum's nursing needs [at Residential care home]. The paramedics whenever they came out to mum were excellent, so caring and patient. All of the staff whenever mum was admitted were very caring and supportive. Concerned about the appropriateness of repeated admissions to Glenfield in last few weeks of her life rather than being in a community setting appropriate to her needs. Palliative care doctor at [community hospital] said the repeated admissions were not good.

25

The table below provides reviewers ratings for each phase of care for all cases where relatives had been complimentary. There was a clear correlation between reviewers and relatives. Relatives still complimented aspects of care even where reviewers were less positive.

Table 5 Relatives UHL compliments v reviewer ratings

Com

plim

ent

case

s U

HL

-

69

Pre

adm

issi

on

IM&

A

Ong

oing

Pro

cedu

re

Per

iope

rativ

e

Rea

dmis

sion

Dis

char

ge

EoL

Ove

rall

Very Poor Care 3

2

Poor Care 9 3 7

2 1 6 5

Adequate Care 11 18 13 1

1 8 18 24

Good Care 28 33 32 7 2 3 9 32 30

Excellent Care 6 11 7

1 3 7 5

The table below provides reviewers ratings for each phase of care for all cases where relatives had concerns. Overall there were fewer concerns fed back than compliments. Reviewers did rate care positively at a similar rate to the cases where relatives had complimented care. However, a review of the types of comments from relatives’ begins to reveal that many of the concerns from relatives would not have been documented in the case notes.

Table 6 Relatives UHL concerns v reviewer ratings

Con

cern

case

s U

HL

-

35

Pre

adm

issi

on

IM&

A

OG

C

Pro

cedu

re

Per

iope

rativ

e

Rea

dmis

sion

Dis

char

ge

EoL

Ove

rall

Very Poor Care

1

1 2 3

Poor Care 3

8

3 2 3 6

Adequate Care 13 15 6 2

7 12 10

Good Care 11 14 13 2 2 3 2 10 13

Excellent Care 2 5 3 1

1

5 3

Some specific examples were:

• The manner of a couple of the junior doctors was very unapproachable.

• Would have liked earlier access to a single room but understand this was a space issue

• Difficult and unhelpful interaction with Dr [name] (cons) who appeared not to take on board the various concerns and was pressing for discharge

• IV had become disconnected on 2/7 and was left in a puddle of liquid on the floor. The nurses simply connected the line up again without cleaning it, or the floor

26

3. Profile of full cohort and reviews undertaken

This section of the report presents the profile of the full cohort of deaths (full cohort) and an analysis of the actual reviews (reviewed cohort) undertaken to show how these compare to the full cohort. The following comparison indicates that the reviewed cohort was representative of the full cohort.

3.1 Comparison of full cohort and reviewed cohort

The full cohort size was 319 i.e. the number of deaths of patients in hospital, in community hospitals and within 30 days of discharge for the period. The number of cases where records were available and could be reviewed was 181, which represents 57% of the full cohort. The figures in the table below show that the records reviewed were representative by age, gender and location of death of the full cohort.

Table 7 Full cohort compared with reviewed cohort to assess representative sample

Deaths in the community

Acute Hospital deaths

Average age Gender

Full cohort 117 (37%) 202 (63%) 76.7 years (range 27 – 113)

Male – 174 (55%) Female – 144 (45%) Unknown - 1

Reviewed cohort

64 (35%) 117 (65%) 77.1 years (range 36 – 102)

Male – 100 (55%) Female – 81 (45%)

The chart below provides a comparison of the full cohort with the reviewed cohort by age.

Chart 2 Full cohort compared with reviewed cohort by age

0

2

4

6

8

10

12

14

16

27

36

41

44

47

49

51

53

55

57

59

61

63

65

67

69

71

73

75

77

79

81

83

85

87

89

91

93

95

97

99

11

3

Nu

mb

er

of

de

ath

s

Age in years

Age comparison between full cohort and reviewed cohort

Full cohort Reviewed Cases

27

3.2 Key characteristics of the cohort

The detailed findings from the review of the patient cohort are in Section 3.3 below. We summarise the main themes here:

• 75% of patients were over 71 years of age • The ethnic mix of the patients did not reflect the demographic make-up of the county –

only 11% of the full cohort were from an Asian background • Two thirds of the deaths occurred in the acute hospital setting; one third outside with only

10 patients dying in a hospice • The vast majority of patients had been admitted through the emergency care route from

home or a care home • Tuesdays and Fridays were the most common days for admissions of people who

subsequently die within 30 days of discharge • There were fewer deaths on weekend days than weekdays • The number of deaths on a Monday in the community was double that of any other day of

the week • 60% of patients died within 14 days of discharge – and older people (over 71) died sooner

after discharge • 30% of patients moved specialty between admission and discharge/death • 51% of the reviewed patients had some form of mental health need: 42% of reviewed

patients had confusion or memory problems and 9% of the cohort had a specified Serious Mental Illness

• 1 person with a Learning Disability was identifiable amongst the cohort. • One third of the patients had a palliative/end of life code in place

28

3.3 Detailed profile of reviewed cohort

Deaths by CCG The number of deaths within each of the three CCGs in Leicestershire are shown below. More deaths were reviewed in Leicester City CCG (LCCCG) and West Leicestershire CCG (WLCCG) due to an extension agreed by these CCGs to the project.

65% of WLCCG deaths were reviewed, 57% of those in the LCCCG area and 42% of East Leicestershire CCG (ELCCG) patients.

Chart 3 Reviews by CCG v full cohort

Age Range The age range of the full cohort broken down by age band is shown in the chart below. The age range was between 27 and 113. 73% of patients were over 71; 45% over 81 years. The average and median ages are shown in the table below. The average and median ages of the reviewed cohort was slightly older than the full cohort. Table 8 Average and Median ages at death

Full Cohort Reviewed Cohort Average Age 76.75 years 77.06 years Median Age 78 years 79 years

The deaths that occurred in LPT care (in Community Hospitals) were slightly older. Average age was 80 and median age was 82 years.

89

104110

37

59

72

0

20

40

60

80

100

120

East Leicestershire CCG Leicester City CCG West Leicestershire CCG

Nu

mb

er

of

de

ath

s

Number of cases in full cohort and reviewed cohort by CCG

Full cohort Reviewed

29

Chart 4 Number of deaths by age band – full cohort

The ages of the reviewed cohort (see chart below) ranged from 36 to 102. 76% (138) were over the age of 71; 45% (82) were over the age of 81.

Chart 5 Age comparisons between full and reviewed cohort

3

14

23

45

89

95

47

1 1 1

0

10

20

30

40

50

60

70

80

90

100

31-40 41-50 51-60 61-70 71-80 81-90 91-100 21-30 101-110 111-120

Nu

mb

er

of

de

ath

s

Age Band

Number of deaths by age band - full cohort

1 3

14

23

45

89

95

47

1 12

812

21

56 57

24

1

0

10

20

30

40

50

60

70

80

90

100

21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100 101-110 111-120

Nu

mb

er

of

de

ath

s

Age Band

Age comparison between full cohort and reviewed cohort

Full cohort Reviewed cases

30

Gender The full cohort gender mix included more men (174) than women (144). The chart below shows the age range of both sexes:

Chart 6 Age band by gender – full cohort

As shown in the chart below there were also more male patients in the reviewed cohort than female patients in particular in the 50-90 age bands. There was a similar spread in the reviewed cohort to the full cohort throughout the age ranges. 50% of the female patients were over 80:42% of the male patients were over 80.

Chart 7 Age band by gender – reviewed cohort

As shown in the table below most age bands were represented in the reviewed cohort for both males and females. The age band where the reviewers had least case records available were

31

for the 61-70 age group where we reviewed 85% of all female deaths and under 40% of male deaths.

Table 9 Comparison of age bands by gender between full and reviewed cohort

Age Band Female

(Full)

Female

(Reviewed)

Male

(Full)

Male

(Reviewed)

Total

(Full)

Total

(Reviewed)

%

reviewed

21-30 1 1 0%

31-40 2 2 1 3 2 66%

41-50 8 5 6 3 14 8 57%

51-60 7 3 16 9 23 12 52%

61-70 18 9 27 12 45 21 47%

71-80 36 22 53 34 89 56 63%

81-90 46 28 48 29 94 57 67%

91-100 26 11 21 13 47 24 51%

101-110 1 1 1 1 100%

111-120 1 1 0%

Grand Total 144 81 (56%) 174 100 (57%) 318 181 (57%O

* 1 unknown gender

Ethnic Origin The ethnic mix of all deaths is shown in the chart below.

Chart 8 Ethnic origin – full and reviewed cohort

The ethnic origin of 167 patients in the reviewed cohort was available from UHL. The 14 community hospital patients did not have an ethnic origin stated. Of the 167 patients 82% (137) of patients were White British. 11% were from an Asian background.

In the reviewed cohort there were very few patients of a non-white British background. Only 15 (9%) patients were of an Asian background. This raises questions about why this might be given

1

1

1

2

2

2

3

4

4

10

15

29

245

1

0

0

0

0

2

3

2

2

7

0

13

137

0 50 100 150 200 250

BLACK/BLACK BRITISH AFRICAN

MIXED WHITE & BLACK CARIBBEAN

MIXED WHITE AND ASIAN

ASIAN/ASIAN BRITISH PAKISTANI

BLACK/BLACK BRITISH CARIBBEAN

WHITE IRISH

ANY OTHER ETHNIC GROUP

ANY OTHER ASIAN BACKGROUND

NOT STATED

WHITE OTHER WHITE BACKGROUND

(blank)

ASIAN/ASIAN BRITISH INDIAN

WHITE BRITISH

Number of deaths

Ethnic Origin

Reviewed Cohort Full Cohort

32

the demographic makeup of the Leicestershire population. The 2011 Census indicates that 45% of the population identify as White British and 28.3% as Asian/Asian British Indian.

An analysis of the reviewer’s narrative shows that there were 4 cases where language barriers were apparent – one Polish speaker and 3 Asian/Asian British Indian.

Place of Death For the full cohort the majority of deaths occurred in an acute hospital bed.

Table 10 Place of death – full cohort

Place of death – community deaths (n=118); acute hospital deaths (n= 201)

In acute hospital (UHL) 201

Community (home/care home/hospice) 103

Community Hospital (LPT care) 15

Of the 181 cases reviewed the place of death is shown below. The 64 community deaths reviewed included those in community hospitals, hospices, nursing and residential homes and at home, broken down as follows:

Table 11 Place of death - reviewed cohort

Place of death – community deaths (n=64); acute hospital deaths (n=1 17)

Place of death Total cohort Reviewed cohort

In acute hospital (UHL) 201 117

Nursing / Residential Care Home 103

20

Home 17

Hospice 10

Community Hospital 15 16

Unanswered 1

Location of deaths in the Community 17 patients died at home and 10 in a hospice in the reviewed cohort. The remaining deaths occurred in a community hospital or nursing/residential care home.

Chart 9 Place of death – Community deaths

33

Place of Residence Patient’s usual place of residence is represented below with size of ‘bubble’ increasing with patient numbers. There was a wide geographic spread from Grimsby, Peterborough and Coventry.

Figure 1 Deaths by postcode for full cohort – local view

We examined the average distance by postcode for patients who travelled for care. The reviewed cohort reflected the full cohort in average distance using LGH as the admission point. Average distance travelled was 9.6 miles.

Table 12 Distance in miles to LGH

Cohort Average Distance to LGH (LE5) Maximum Distan ce to LGH (LE5)

Not reviewed 9.5 76.3

Reviewed 9.7 31.5

Total 9.6 76.3

GP practices

Not all practices in Leicestershire (135) had a registered patient that died during this review period. The full cohort of deaths were registered to 106 practices. 167 patients in the reviewed cohort related to 70 GP practices. We did not receive the practice details for the LPT care patients.

34

Causes of death Causes of death were captured from a data set provided by UHL. Cause of death was available in 190 cases. This covered 184 of the 202 (91%) inpatient deaths. Causes of death for 6 community cases were recorded by UHL. The remaining inpatient cases were still waiting a cause of death or a coroner outcome. Edited cause of death uses part 1A from death certificates only. Pneumonia was the predominant cause in 27% of cases. ‘Other infections’ accounted for another 12%.

Chart 10 Summary causes of death (edited)

The following table shows the broad causes of death (1A) for the 190 cases where this information was available.

Table 13 Causes of death IA (edited) by location

Summary cause of death –

Cause of Death 1A

Community In-hospital Total

Pneumonia 1 51 52

Cancer 2 35 37

CHF

21 21

Liver disease

10 10

Sepsis

10 10

Infection 1 9 10

Stroke/TIA

9 9

MI 1 5 6

Bowel

4 4

Respiratory Failure

3 3

Perforated bowel

3 3

Other CV

3 3

PE

3 3

Parkinsons/dementia 1 2 3

35

COPD

2 2

Other brain

2 2

Other respiratory

2 2

IHD

2 2

MOF

2 2

Biliary peritonitis

1 1

Cholecystitis

1 1

Kidney failure

1 1

Haemoptysis

1 1

Peritonitis

1 1

Renal failure

1 1

Grand Total 6 184 190

Type of admission The majority of admissions were through the emergency care route via A&E.

Chart 11 Admission route

6

95

2

195

6

0

50

100

150

200

250

Day Case Emergency Inpatient

Nu

mb

er

of

de

ath

s

Admission route

Community deaths In-hospital deaths

36

Day of admission The following charts show the number of deaths (by admission day) for the full cohort.

Chart 12 Number of deaths (by admission day) full cohort

Table 14 Total deaths by admission day

Day Monday Tuesday Wednesday Thursday Friday Saturday Sunday

Full cohort 38 52 44 53 56 34 42

Reviewed cohort

24 33 28 25 33 20 18

Tuesdays, Thursdays and Fridays were the busiest days for admissions for people who die in hospital or within 30 days of discharge.

Further work is needed to establish if these busy periods are influenced by local factors or community provision. This would include whether busy primary care services result in delayed visits after the weekend. The reviewed cohort did not show a marked difference on weekdays but there were fewer admissions at the weekend amongst the cases we reviewed.

37

Number of deaths (by admission day) reviewed cohort

Chart 13 Number of deaths (by admission day) reviewed cohort

Admission time We then looked at time of admission. The chart below shows admissions by day and time for the full cohort. We split this into day (06:00 to 22:00) and night periods (22:00-06:00). 30% of admissions across the week were during the night hours – a rate similar to the day time hourly admission rate.

There was a marked difference during the day on Thursdays compared to other days of the week with more admissions on Thursdays through to Friday. Thursday night/Friday morning saw the least number of people being admitted during the night. This could be a local effect of bed availability at this time of the week or other local factors in this specific time period.

Chart 14 Admissions by day and time – full cohort

38

We analyse further the admissions at night (between 22:00 hrs and 06:00hrs) by age in the Preadmission section of this report.

Deaths by Day of the Week In the full cohort (see chart below) there were more deaths on a Monday than all other days largely due to deaths in the community being at least double on a Monday. There were no obvious characteristics to this group that would be useful here. Cause of death was not available but 80% died more than 8 days after discharge. This effect should be examined further to understand why this is so marked and whether this is replicated throughout the year or not.

There are fewer inpatient deaths at weekends than on all other weekdays.

Chart 15 Number of deaths by day of the week (full cohort)

The reviewed cohort showed a similar trend although there were more deaths in the early part of the week in acute care.

Chart 16 Number of deaths by day of week – reviewed cohort

39

Days to Death Post Discharge Full cohort The chart below indicates that 37% within 7 days of discharge and 60% of patients in the full cohort died within 2 weeks. Chart 17 Days to death post discharge – full cohort

40% of patients died more than 2 weeks after discharge. The Fast Track arrangements are discussed later in this report but were largely seen to be effective. The reviewers however identified some individual difficulties with the Fast Track arrangements and securing funding/agreement where end of life was not considered imminent enough and where it was not possible to organise at the weekend.

Reviewed cohort

Similar findings were observed in the reviewed cohort as in the chart below. Of the 50 patients discharged into the community 38% died within 7 days and 70% within 2 weeks.

Chart 18 Days to death post discharge – reviewed cohort

40

The following graph analyses the length of time between discharge and death by age band

Chart 19 Length of time between discharge and death by age band

101 people were discharged from UHL care in the full cohort.

We looked at the effect of age on days to death post discharge. In the table below it shows that just over a third of people died within a week of being discharged and 59% died within 14 days. This is slightly higher in the older age group on both counts. Older people appeared to die sooner after discharge than those under 71.

Table 15 Length of time post discharge to death by age

No of days post discharge All Over 71 years Under 7 1 years Die within 7 days of discharge 36% 39% 29%

Die within 14 days of discharge 59% 64% 48%

Die after 14 days of discharge 41% 36% 52%

11 12

3

1

32

3 3

10

4

6

11

88

3

12

8

45

1

0

2

4

6

8

10

12

14

1 2-7 8-14 >14

Nu

mb

er

of

de

ath

s

Days to death after discharge

Days after Discharge by Age Band

21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100 101-110

41

Admitting Specialty to UHL 92 patients were admitted under general medicine, 49 (53%) of whom went to other specialties.

Number of deaths by admitting specialty

Chart 20 Number of deaths by admitting specialty (Full cohort)

Discharge Specialty The figures indicate that the most frequent move between specialty is into geriatric medicine.

Chart 21 Discharge Specialty (Full cohort)

42

The number of patients that moved speciality – full cohort – during their admission was 109 (36%) in UHL.

Table 16 Specialty moves – full cohort

SPECIALTY BETWEEN ADMISSION AND DISCHARGE Community In-hospital LPT Care Total

DIFFERENT 38 71

109

SAME 65 130

195

N/A

15 15

Total 103 201 15 319 This was replicated in the reviewed cohort (35.9%)

Table 17 Specialty moves – reviewed cohort

SPECIALTY BETWEEN ADMISSION AND DISCHARGE

Community In-hospital LPT Care Total DIFFERENT 22 38

60

SAME 28 79

107

N/A

14 14

Total 50 117 14 181

Ward moves We analysed ward moves within UHL

• In total of the 167 UHL patients we reviewed 137 (82%) patients moved wards at least once during their stay in acute care.

• 30 patients stayed in the same ward. Patients who stayed on the same ward had an average length of stay of 4 days (3 days excluding 2 longer stays)

• Those who moved wards at least once had an average length of stay of 12 days.

The extent of moves between wards may be clinically appropriate or as a consequence of the demand for beds. However, further analysis may be needed to understand the cumulative impact this has as part of the total patient journey.

Length of stay The following table shows that the average length of stay in UHL was 11 days before discharge home and 10 days before death. The longest a patient stayed in hospital before being discharged home was 30 days and the longest a patient stayed in hospital before dying in hospital was 97 days. Length of stay where the death was in a Community Hospital was 13 days.

Table 18 Length of stay

Location of death Average length of stay * Minimum length of stay Maximum length of stay

Home 11 0 30

Acute hospital 10 0 97

Community hospital 13 2 25 *In UHL

43

Patients with a Mental Health Need or with a Learni ng Disability Learning Disability

One of the patients reviewed was identified as having a diagnosed learning difficulty. It was not readily apparent in the notes when patients might have a learning disability and the one patient we identified was because this was noted in the GP records only. Further analysis of Learning Disability patients may be required to ensure patients are flagged.

Mental health

People with a diagnosed serious mental health problem (that could be determined from the care notes) accounted for 9% (17) of the reviewed cohort.

Confusion or Memory problems

76 patients (42%) had confusion or memory problems; 48 (63%) of these had clear diagnostic definition of confusion/memory problems. The remaining 28 (37%) with confusion or memory problems did not have a clear diagnostic definition of the cause but it was evident from case notes that this was a characteristic of their presentation (although some cases may have non-mental health causes). The management of the patient with confusion still creates similar challenges however and this figure is an indication of the extent of confusion in totality. We analyse the quality of care for this group by ratings later in the report.

In total people with a mental health need of some form accounted for 51% of the reviewed cohort – see chart below.

Number of cases with evidence of a Mental Health ne ed or a Learning Disability

Chart 22 Number of cases with Mental Health need or Learning Disability

We examined the place of death for those with confusion or a memory problem. Of the 10 deaths in hospice care 2 people had a confusion/memory problem. No one with a mental health condition died in hospice care in this reviewed cohort. Further analysis of hospice admissions to establish better if people with dementia and their families are able to access hospice care if needed would be necessary.

44

Palliative and End of Life Care Coding 102 (33.5%) patients had a palliative/end of life care code in the full cohort as shown in the table below. The latest SHMI guidance (2013) indicates that Trusts should record all cases where palliative care is part of their needs. Even if there is variable practice within Trusts in terms of SHMI coding it is a helpful way to understand the extent of palliative care in the system.

Palliative care coding full cohort

Table 19 Palliative care coding – full cohort

Palliative and End of Life Care Coding

Community deaths In-hospital Total

No 74 128 202 (66.5%)

Yes 29 73 102 (33.5%)

Total 103 201 304

Of the reviewed cohort there was a similar proportion but this would appear to be under representative based on the extent of end of life care observed in our review. The team rated 155 cases at end of life – of which 109 were in-hospital.

Palliative care coding reviewed cohort

Table 20 Palliative care coding – reviewed cohort

Palliative and End of Life Care Coding

Community deaths In-hospital deaths

Total

Coded - No 37 75 112

Coded - Yes 13 42 55 (33%)

Total 50 117 167

45

4. Findings

This section of the report presents the reviewers’ findings on the 181 reviews completed. This includes a summary of the overall scores and their scores for each phase of care. The scores followed the same format as the RCP structured judgement review tool adapted for the review:

1 = Very Poor Care 2 = Poor Care 3 = Adequate Care 4 = Good Care 5 = Excellent Care

In the sections that follow we examine the key positive characteristics of each phase drawing out the key issues that resulted in excellent and good ratings. We also identify the key negative characteristics and identify areas for improvement from the poor/very poor judgements of care.

46

4.1 Overall care

Number assessed and ratings for Overall Care

At the end of each review, the reviewers were asked to rate the overall care for that patient and provide a summary of their reason for that rating. 177 cases were rated overall (4 were unanswered for Overall Care). 84% were rated as adequate, good or excellent.

The chart below shows the ratings given for the care overall. Half (50%) of the patients reviewed were assessed as having received “good” (45%) or “excellent” (5%) overall care. At the other end of the scale, 29 (16%) patients were rated as having received “poor” (12%) or “very poor” (4%) overall care.

Chart 23 Overall Care ratings

7

22

57

81

10

4

0

10

20

30

40

50

60

70

80

90

Very Poor Care Poor Care Adequate Care Good Care Excellent Care Unanswered

Nu

mb

er

of

De

ath

s

Care rating

Overall Care ratings (n=181)

47

Analysis of the phases of care that contributed to the overall care rating indicates that the care received in each phase throughout the overall experience was consistently good to excellent for those marked as excellent.

The phase of care ratings were very poor to adequate for the very poor Overall Care scores.

The only phase for which this pattern differed was for the initial management and admission phase. This was consistently more highly rated even in the poor/very poor Overall Care ratings.

Notable themes

Variation in phases of care The overall care assessment was given based on an assessment of care at each phase.

Preadmission assessments were made on 152 cases. 89 cases (58.5%) were assessed as good or excellent. The preadmission phases included primary care, community care and ambulance services.

Initial management and admission was the most positively rated phase of care. 112 cases (65%) were good or excellent.

Ongoing Care was assessed as good or excellent in 95 cases (62%).

Procedure Care and Peri-operative care were assessed as good to excellent in 68% and 90% of cases respectively.

Discharge care and End of Life care were the least positively rated aspects that impacted on the Overall care rating.

Discharge was only apparent in 57 cases – of these 29 (50%) were assessed as good to excellent.

End of Life Care was assessed in 156 cases – of these 86 (55%) were assessed as good to excellent.

Quality of patient records and relationship to Qual ity of Care ratings We analysed the Overall Care ratings against rating of the quality of the care records and there was a clear relationship between the two. Good quality patient records appeared to be a good indicator of the quality of care.

There was a direct correlation between the Overall Care rating and the assessment of the quality of care records. This was particularly the case in the Initial Management and Assessment phase of care. The table below shows this.

48

Table 21 Relationship between quality of care rating (Overall Care) and Quality of Care records

Care rating Good Excellent Total

Overall care rating of good or excellent 80 10 90

Number with good or excellent records 53 6 59

% good or better with good or better records 66% 60% 66%

Include adequate record quality 20 2 22

% good or better with adequate or better records 91% 80% 90%

Care rating Very poor Poor Total

Overall care rating of poor or very poor 7 22 29

Number with good or excellent records 0 6 6

% good or better with good or better records 0% 27% 21%

Care rating Very poor Poor Total

Overall care rating of poor or very poor 7 22 29

Number with very poor or poor records 5 7 12

% poor or very poor records 71% 32% 41%

For good or excellent care ratings up to 90% of records were considered to be of adequate, good or excellent quality. It is not possible to say to what extent this helped the reviewers give a better rating but it is clear that high quality care records make case review more effective at assessing care quality.

For very poor care only 21% have good or excellent care records and 41% were poor or very poor. For excellent care 60% of records were good or excellent.

The following chart shows the proportion of records rated very poor to excellent by Overall Care rating.

49

Chart 24 Quality of care records by Overall Care rating (%)

Examples of some of the recording issues are described below:

Record keeping issues

Case 1

Reviewing hospital and GP notes it seemed to the reviewer that there were many excellent assessments and elements of her care, but the notes were somewhat chaotic and a more streamlined approach to agreeing and documenting the evolving plan might have enabled the workload of all involved and the patient's experience.

Case 2

Filing of loose scraps of paper with unattributed clinical information - paperclip/Post It rather than formal handover of concerns verbally, recorded handover or within notes - must have patient identifier and staff identifier when recording concerns

Case 3

Better handover notes about end of life care would have helped the staff in the accepting nursing home and the GP

Case 4

Poor notes filing - difficult to see the chain of events.

Case 5

SALT risk feeding protocol was left in medical notes for doctors to sign but not signed.

29%

9% 2%4% 0%

43%

23%

18%5%

20%

29%

41%

46%

25%

20%

0

27%32%

60%

30%

0% 0% 2% 6%

30%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Very Poor Care Poor Care Adequate Care Good Care Excellent Care

Overall care rating

Quality of care records compared with Overall Care rating

Very Poor Record Poor Record Adequate Record Good Record Excellent Record

50

Case 6

The number of admissions mean that there are a lot of notes and this made it difficult to work through them in an orderly fashion. However, there was little reference throughout to how the patient felt, or how his wife was coping. Some of the nursing notes had become detached and it appeared that some nursing entries (particularly around the time of death) were completed sometime after the event.

Memory problems/confusion and care ratings We analysed the overall care scores by a range of characteristics including the extent to which people with memory problems/confusion experienced care. The table below shows the results.

Table 22 Overall care scores by confusion/memory problems

Presence of memory problem or confusion

Very Poor Care

Poor Care

Adequate Care

Good Care

Excellent Care

Total

No confusion/memory problem

3 10 31 48 8 100

% of care score ratings for people without confusion or memory problems

43% 45% 56% 59% 80%

Unanswered

1

Yes - but without clear diagnostic definition

3 4 11 10

76 Yes - clear diagnostic definition of the confusion/memory problems

1 8 14 23 2

% of care score ratings for people with confusion or memory problems

57% 55% 44% 41% 20%

Total 7 22 57 81 10

The care ratings for the 76 people identified as having a confusion or memory problem evident were lower than for patients without a confusion or memory problem. There was a declining level of better care ratings amongst this group; the opposite was apparent for the care group without a memory or confusion problem.

51

Ratings by age band The table below shows the number of Overall Care ratings by age band.

Chart 25 Ratings by Age band

In patients under 70 years, 70% (30/43) of Overall Care ratings were good or excellent. In those aged 71 and over, 44% (61/138) of Overall Care ratings were good or excellent. This may reflect the greater complexity of care of older patients across the system.

Overall problem types Reviewers were asked to log specific problem types during their case note review. The problem types were:

• Assessment, Investigation or Diagnosis

• Re-admission care

• Treatment and Management plan

• Problems with operations/invasive procedures

• Clinical Monitoring

• Infection Control

• Medication/IV fluids/Electrolytes/Oxygen

• Resuscitation

• Discharge

The chart below shows which problem types were most predominant.

52

Chart 26 Total number of problem types

The areas where reviewers logged most problems were in areas that were predominantly related to Ongoing Care and Discharge. Clinical monitoring and the management of medication, IV fluids, electrolytes and oxygen taken together are the biggest area for problems being logged. We draw on some of the main themes in the Ongoing Care section later in the report. Infection control issues were noted in only 6 cases of which 4 cases were in patients over 80 years. There were 4 resuscitation problems. We discuss this further in the Ongoing Care section.

Problem types by age band

We analysed problem types by age band. The table below provides the results of the analysis. Problem types were more prevalent amongst patients in the over 80s age group where all problem categories were recorded.

53

Chart 27 Problem Types by Age band

The tables below show the ratio of all problem types by age band.

Table 23 Ratio of problem types by age band

Number of patients 2 8 12 21 56 57 24

Ratio - problems per patient 2.5 1.0 0.3 1.2 1.1 1.7 1.6

Age Band 31-40 41-50 51-60 61-70 71-80 81-90 91-100

Assessment, Investigation or Diagnosis 1 2 1 3 12 11

Medication/IV Fluids/Electrolytes/Oxygen

1

2 7 12 5

Treatment and Management Plan 1 1

3 10 15 7

Infection Control

2

3 1

Operation/Invasive Procedure

1 1 2 2

Clinical Monitoring 1

4 7 20 6

Resuscitation

1

3

Discharge 1

4 10 10 6

Re-admission Care

1

2 3 5 4

Other 1 3

5 9 14 10

Total 5 8 3 26 60 95 39

Patients over 81 had a higher ratio of problem types than other age bands. The table below shows the percentage of times a particular problem occurred in each age band. People over 81 had a higher percentage than other age groups in nearly all categories.

0

10

20

30

40

50

60

70

80

90

100

31-40 41-50 51-60 61-70 71-80 81-90 91-100

Nu

mb

er

of

pro

ble

ms

Age Band

Problem Types by Age Band

OTHER

RE-ADMISSION CARE

DISCHARGE

RESUSCITATION

CLINICAL MONITORING

OPERATION / INVASIVE PROCEDURE

INFECTION CONTROL

TREATMENT and MANAGEMENT

PLAN

MEDICATION / IV FLUIDS /

ELECTROLYTES / OXYGEN

ASSESSMENT, INVESTIGATION or

DIAGNOSIS

54

Table 24 Problem types by age band (%)

Age Band 31-40 41-50 51-60 61-70 71-80 81-90 91-100

Assessment, Investigation or Diagnosis 3% 7% 3% 10% 40% 37% 0%

Medication/IV Fluids/Electrolytes/Oxygen 0% 4% 0% 7% 26% 44% 19%

Treatment and Management Plan 3% 3% 0% 8% 27% 41% 19%

Infection Control 0% 0% 0% 33% 0% 50% 17%

Operation/Invasive Procedure 0% 0% 17% 17% 33% 33% 0%

Clinical Monitoring 3% 0% 0% 11% 18% 53% 16%

Resuscitation 0% 0% 25% 0% 0% 75% 0%

Discharge 3% 0% 0% 13% 32% 32% 19%

Re-admission Care 0% 7% 0% 13% 20% 33% 27%

Other 2% 7% 0% 12% 21% 33% 24%

Identifying positive and negative lessons from the review In 77% of cases there were positive aspects of care noted; in 72% of cases there were negative aspects of care noted.

Reviewers were asked to record aspects of care in three specific areas:

• Communication • Escalation • Ceilings of care

Lessons relating to Communication This was the most positively rated aspect of care throughout the review. There were 115 positive lessons recorded overall. Three positive aspects stood out in relation to communication.

• Good communication - 52 • Attitude of staff – 39 • Clear management plan present – 15

Good communication and attitude of staff were clear characteristics in Good Care. There were few positive lessons in relation to Communication in Poor and Very Poor Overall Care ratings.

55

Table 25 Positive lessons – Communication by Overall Care rating

OVERALL CARE RATING Very Poor Care

Poor Care

Adequate Care

Good Care

Excellent Care

Total

Good communication with family 1 4 15 30 2 52

Attitude of staff

3 6 27 3 39

Clear management plan present

2 5 6 2 15

Patient's stated wishes were followed

2 2

4

Good quality of documentation

1 1

2

DNACPR in place

1

1

Good coordination of clinical care / senior input / advanced decision making

1 1

Quality of handover

1

1

Total 1 9 30 67 8 115

There were half the number of negative lessons in Communication recorded – 62. There was no predominant area that stood out. However, more negative issues were logged for poorer care ratings. A lack of a clear management plan was an aspect of Poor Care. The table of negative lessons is shown below by Overall Care rating.

Table 26 Negative lessons - Communication by Overall Care rating

OVERALL CARE RATING Very Poor Care

Poor Care

Adequate Care

Good Care

Excellent Care

Total

Poor communication with family 3 2 4 2

11

Lack of clear management plan 1 7 2 1

11

DNACPR not in place or invalid or ignored and CPR undertaken

2

3 3

8

Poor coordination of clinical care / lack of senior input / advanced decision making

1 3 2

6

Patient's stated wishes not followed

2 2

1 5

Poor quality of documentation

1 3 1 5

Attitude of staff

2 1 1

4

Quality of handover

1

1

Other

1 3 6 1 11

Total 6 15 20 18 3 62

Lessons relating to Escalation The use of EWS was the most commonly recorded positive lesson noted across the review in 21 cases.

The table below shows the areas where positive lessons were recorded – there were more positive lessons recorded in good Overall Care ratings.

56

Table 27 Positive lessons in relation to Escalation by Overall Care rating

OVERALL CARE RATING Very Poor Care

Poor Care

Adequate Care

Good Care

Excellent Care

Total

EWS recorded 1 3 6 9 2 21

Palliative care instituted in timely manner

1 3 10 1 15

Good standard of care over weekend

2 3

5

Rapid readmission following earlier (inappropriate?) discharge

2 2

4

Timely discharge into community/ appropriate supportive care package in place

1

3

4

No delay in treatment/surgery

1 2 3

Physiological observations / deterioration escalated appropriately

1 1 1

3

Test results / tests being undertaken timely

2

2

Timely and appropriate clerking

1

1

Total 1 6 16 30 5 58

Amongst the negative lessons in Escalation it was clear that Poor and Very Poor care ratings had some of these negative characteristics. The table below shows the distribution of issues recorded by reviewers. Delays in discharge and in instituting palliative care were the most commonly noted issues. Table 28 Negative aspects of Escalation by Overall Care rating

OVERALL CARE RATING Very Poor Care

Poor Care

Adequate Care

Good Care

Excellent Care

Total

Delay in instituting palliative care 1 4 7 2

14

Delayed discharge into community/lack of appropriate supportive care package

1 1 5 2 1 10

Incomplete physiological observations / deterioration not escalated

4 2 2

8

Delay in test results / tests being undertaken

2 3 2

7

Possible poorer standard of care over weekend

1 2

2

5

Delay in referral to critical care 2 2

4

Delay in treatment/surgery due to staff shortages/equipment failure

1 1

2

Patient fall not escalated appropriately

1 1

2

Rapid readmission following earlier (inappropriate?) discharge

1

1

Total 5 15 20 12 1 53

57

Lessons relating to Ceilings of Care In lessons relating to Ceilings of Care positive lessons were noted most commonly as a cumulative aspect of Good care.

Table 29 Positive lessons in relation to Ceilings of care by Overall Care rating

OVERALL CARE RATING Very Poor Care

Poor Care

Adequate Care

Good Care

Excellent Care

Total

Appropriately aggressive treatment

1 1 5 1 8

Appropriate admission from nursing home / community hospital / community setting

2 5

7

Appropriate referral to Critical Care

1 1 1 3 6

No wait in A&E for a bed 1

1 4

6

Treatment of complex medical patient on surgical ward

1 2

3

Appropriate bed (non ICU) available

2

2

Total 1 2 6 19 4 32

Negative aspects in relation to Ceilings of Care were very few with no overall theme.

Table 30 Negative aspects of Ceilings of Care by Overall Care rating

OVERALL CARE RATING Very Poor Care

Poor Care

Adequate Care

Good Care Total

Inappropriate admission from nursing home / community hospital / community setting

1 3

2 6

Availability of appropriate bed (non ICU) compromising care

2 1 3

Long wait in A&E for a bed

3

3

Inappropriately aggressive treatment

1 1

Treatment of complex medical patient on surgical ward

1

1

Total 1 3 6 4 14

58

4.2 Preadmission care

Number assessed and ratings for Preadmission phase

The reviewers made 152 assessments of preadmission care. They were unable to gain access to preadmission information for 23 of the 181 reviews completed, due to a lack of access to the GP records for these cases. However, even without access to GP records, reviewers were able to access information about preadmission care through the hospital notes in a few cases. In some cases, reviewers felt that an assessment of preadmission care was not applicable, for example when a patient had a sudden stroke or heart attack.

The majority of admissions in the cohort were through the emergency department and a small number of elective admissions.

Overall, 85% of ratings were assessed as adequate, good or excellent care.

The chart below shows the scores for this phase of care. Over half (58%) of the patients reviewed were assessed as having received “good” (48%) or “excellent” (10%) preadmission care. At the other end of the scale, 24 (16%) patients were assessed as having received “poor” (11%) or “very poor” (4%) preadmission care.

59

Chart 28 Care ratings for Preadmission phase

Preadmission ratings by age band

The following graph shows preadmission ratings by age band.

Chart 29 Preadmission ratings by Age Band

Of the poor and very poor ratings 21 of 24 (88%) were for people over 71 years.

7

17

39

74

15

22

7

0

10

20

30

40

50

60

70

80

Very Poor Care Poor Care Adequate Care Good Care Excellent Care Unanswered Not Applicable

Nu

mb

er

of

De

ath

s

Care rating

Preadmission Care ratings (n=152)

0

5

10

15

20

25

31-40 41-50 51-60 61-70 71-80 81-90 91-100 101-110

1 = Very Poor Care 2 = Poor Care 3 = Adequate Care 4 = Good Care 5 = Excellent Care

60

Notable themes

Use of Admission Avoidance Table 31 Use of Admission Avoidance Schemes

Admission Avoidance Scheme

Age bands Not clinically

indicated

Scheme

considered

Scheme

considered & not

used

Scheme not

considered

Scheme

used

31-40 2

41-50 8

51-60 12

61-70 18

1

2

71-80 43

1 8 4

81-90 45 3 2 5 2

91-100 15

3 3 2

101-110

1

Total 143 3 7 17 10

Admission avoidance was only used in a small number of cases (10) before admission. In most cases (79.5%) an admission avoidance scheme was considered as not clinically indicated at the actual time of admission. It is may be earlier admission avoidance intervention could have helped.

Reviewers commented on:

• a lack of apparent care coordination;

• a possible lack of skills in the community (e.g. TPN, bloods at weekends) to prevent unnecessary admission (in care homes and community hospitals as well as availability at home); and

• a lack of signposting out of hours predominantly due to lack of service provision/known alternatives.

The following graph shows admission avoidance scheme by age band. In the small number of cases where admission avoidance schemes were used or considered these were for older people and predominantly over 80.

61

Chart 30 Admission Avoidance by Age Band

Reviewers documented where better community support might have prevented an admission. Some case examples are shown below:

Examples of cases preadmission where better communit y support or facilities might have prevented

admission and improve patient experience

Case 1

Once patients become frail, for example being housebound, requiring regular personal care, there is an

opportunity to discuss and agree an ACP . This lady would have preferred to stay at home rather than being

admitted and although her outcome might have been the same, since she was admitted following a fall, a

multidisciplinary , joined up approach to her care planning , discussed with her and her daughters might have

enabled her to remain at home and avoid unnecessary investigations/admission.

Case 2

The patient's son, wife and OT asked for a GP to visit him, explaining he had become bed bound, and was not

eating or drinking well. It would have been good medical practice to assess the patient earlier and this might have

enabled him to receive treatment at home at least in the first instance rather than him requiring immediate

admission to hospital.

Case 3

ACP could have improved this patient's end of life experience and there was an opportunity to start this 12m prior

to her death when her increasing frailty became apparent.

Case 4

The reviewer questions whether the Community Hospital needed to readmit the patient rather than being able to

actively manage his ongoing respiratory infection

Case 5

Community therapy services placed patient on waiting list despite breathlessness of very elderly person post

discharge. Mobilisation may have prevented a chest infection developing

Case 6

The reviewer questions the need to admit a patient with advanced malignancy in receipt of palliative care to an

acute hospital setting for blood transfusion?

Case 7

62

Patient discharged from hospital with no home assessment and so no suitable bed or pressure care arrangements

in place

Community nursing referral not documented, and arrangements for home assessment delayed and so delay in

getting home support to family which if it had been in place might have prevented admission in the first place or

facilitated discharge when patient requested

Case 8

Although it would not have affected the outcome, it is noted that the GP referral to the MacMillan nurse was not

accepted due to a lack of information. There is a lesson to learn about how to improve this for future patients,

especially when they are in the last days of their lives.

Case 9

Perhaps ICS should revisit their admissions criteria so that poorly older people can be cared for at home for

longer.

Case 10

Poor discharge planning and assumption that existing package of care likely to continue to be appropriate without

review within a few days of discharge results in delay to rehabilitation post discharge of frail elderly

Case 11

Waiting times in community rehabilitation can result in poor mobility which may result in reduced recovery and

repeat chest infections in susceptible patients

Case 12

More proactive monitoring by the DNs and involvement of GP at earlier stage might have prevented the

deterioration he made requiring hospital admission. His end of life might then have been managed at home.

Case 13

Insufficient package of care preadmission that might have facilitated early discharge - although in this case it

would have had to be intensive to prevent readmission

Case 14

Admission avoidance options should be considered to support nursing care home manage a patient with advance

care plan stay in care home including assessment of any acute care needs. E.g. U&Es being accessible promptly

and diabetic assessment could potentially have been delivered in nursing care setting.

Case 15

The deterioration over 2 weeks in weight loss, resulting in pressure sores and possible lack of early recognition of

sepsis in [community hospital] resulted in an emergency admission which might have been avoided. The reviewer

questions: whether community wards can manage the acutely ill patient, the arrangements for medical cover at

weekends for the deteriorating patient. Some thoughts as to whether the ability to do and secure bloods and

results is possible in a community ward. Ultimately the patient experience was excellent in the acute setting and

this may have been the best place for care in the circumstances but a 999 admission is distressing at end of life for

patient and family.

Avoidance schemes should review whether the issues faced by the very elderly and their families in this report adequately support avoid admission. This would include suggestions such as hospice at home for confusion/memory problems, management of acute illness and support to care homes.

Admission time by age As shown in the table below there is an indication that older people (over 71) are being admitted on to the wards very late at night and in to the early hours of the morning. Case review evidence would suggest a cumulative system effect of being seen in primary care, waiting for ambulance arrival and a period in A&E sometimes with delays before being admitted on to a ward.

63

Table 32 Admission time by age

Full cohort Reviewed cohort

Age Band

06:01 - 21:59 22:00 - 06:00 Total 06:01 - 21:59 22:00 - 06:00 Total

21-30 1 1

31-40 2 1 3 2 2

41-50 9 4 13 5 2 7

51-60 19 4 23 10 2 12

61-70 37 8 45 16 5 21

71-80 62 22 84 37 14 51

81-90 53 35 88 31 20 51

91-100 30 15 45 13 9 22

101-110 1 1 1 1 1

111-120 1 1 1

Total 215 89 304 115 52 167

There appears to be a particular issue with very elderly people as shown in the chart below. The chart indicates the extent of the peak in the reviewed cohort. The same picture was replicated for the full cohort.

Chart 31 Admission time by age band

Admission very late at night and into the early hours was apparent in at least 22 cases. In all cases patients were over 70 and 16 were over 80 years old. We could not accurately track the cases from time of initial request for ambulance attendance to admission but the cumulative effect risks being significant for the frail elderly patient in particular. During this time the increased risk of poor fluid intake, lack of continuity of care, stress and discomfort can impact on recovery and trigger deterioration. Delayed admission to wards with disturbed rest over the admission period and for clerking in the early hours presents a risk to frail elderly patients. These issues are reflected in the case study examples below:

0

5

10

15

20

25

30

35

40

31-40 41-50 51-60 61-70 71-80 81-90 91-100 101-110

Nu

mb

er

of

ad

mis

sio

ns

Age Band

Admissions by age band reviewed cohort06:00-22:00 and 22:00-06:00 hrs

06:01 - 21:59 22:00 - 06:00

64

Case studies – late night admissions

Knock on system delay:

Admission by ambulance took 8 hrs and seen early hours of the morning by SHO which resulted in a delay in

being seen for a frail elderly lady with dementia

Cumulative delay:

Lady admitted to A&E [midday] with abdominal pain and distension with metastatic cancer. Insulin diabetic noted

early on. DNACPR noted early as being applicable across all care settings. Seen within 1 hour of arriving with

blood results.

Handed over to AMU - Medical admissions at early evening. Some delay and had a hypoglycaemic episode

shortly after. Early assessment but delays in securing surgical opinion - frequent bleeping. Assessed in AMU at

night by surgical registrar. Medication review by pharmacist.

Quality of care impacted with some delays in admission to AMU and surgical assessment and lack of blood

glucose monitoring in the initial period leading to hypoglycaemia.

Clerking frail elderly early hours:

Admission to A&E managed promptly with assessment within 30 mins. Sepsis considered promptly ….Early

recognition of likely need for EoL. Treatment of sepsis prompt.

Delay in admission to ward from A&E with transfer at [midnight], clerking [in early hours] whilst patient asleep and

examination [around breakfast time]

Whilst care delivered was appropriate there were delays in transfers late at night for EoL frail elderly.

Early hours admission 1:

Presented at A&E via ambulance [late evening] and admitted to stroke unit [in the early hours]. Appropriate tests

done in A&E including CT of head.

Early hours/delayed admission 2:

A&E notes very scant but appears patient presented at A&E [at around midday]. He was unwell and needed a

view from oncology. Seen by a doctor [late afternoon]. Needed IV antibiotics and fluids and a PET scan to assess

Cancer spread. Admitted to a ward at [nearly midnight]that day.

Early hours/delayed admission 2:

Admitted to A&E on [date] with cough - diagnosed as bilateral PEs and a chest infection. Treated with IV

antibiotics and warfarin. In A&E for nine hours awaiting bed on the clinical decisions unit. Finally admitted to the

CDU [just before midnight].

Early hours admission 3:

Admitted via A&E with weepy legs as a result of oedema. Appeared to be in A&E for some time. Arrived [late

evening] on [date] and not found a bed until [early hours] on [next day]. Then clerked and did not get rest until

5.30am.

Planned admission – very late at night:

Admission to [Community Hospital] was very late at night and given this was a planned admission this was not

acceptable

Ambulance provision – preadmission :

The reviewers noted a number of points in relation to ambulance provision:

• Good use of ambulance services in avoiding admissions if possible but some examples of inappropriate use of resources which would need to be considered further. See case in box below.

65

• There was evidence of detailed assessments and in the main clear documentation in the care records when admission was by ambulance. There were however a number of instances where the carbon copy was illegible and the quality of the paper should be reviewed. The reviewers commented on the illegible transfer information on 3 occasions.

• Pre-alerts were evidenced in particular for suspected strokes with estimated time of arrival to A&E clearly recorded on a pre-alert sheet. There were 13 cases noted by reviewers of pre-alerts to A&E. All cases related to suspected strokes, cardiac arrest or red/amber sepsis risk.

• Waiting times were a concern on a number of occasions. Reviewers commented on delays between call and arrival at A&E in 10 cases. In one case the ambulance did not arrive, in another a second call to chase was required.

• Specific examples of compassionate care were:

o Spouses being able to accompany relatives in the ambulance in cases of extreme emergency – e.g. cardiac arrest and status epilepticus.

o Ambulance staff working with a GP over 2 days to persuade a patient to get admitted and then pre-alerting the A&E department

The following 2 cases highlight the difficult decisions about admission avoidance at End of Life for paramedics on the ground, in particular when no end of life care plan provides guidance.

The challenges for ambulance staff decision making on admission at End of Life

Case 1

Admission avoidance would have been inappropriate in this case. Ambulance service sent pre-alert to hospital as

patient had red flag for sepsis and cellulitis. However there was a complaint made by the DN team about the

ambulance because the ambulance team would not immediately take the patient to hospital, saying that she was

too ill and would benefit from dying at home. The DN felt that this delay contributed to the patient's deteriorating

condition.

Case 2

Pre-alert from EMAS to A&E for attendance, seen very quickly in A&E and red flag sepsis noted. Full advance

care plan in notes which is clear in terms of patient’s wishes and actions in the event of anticipated emergencies.

Clear message to EMAS and OOH that place of death should be home or hospice. However, the care plan also

states there were no anticipatory medications available. The ambulance had been called on previous occasions

and been able to settle her when she panicked.

Patient very clear they didn't want to come in but EMAS called GH respiratory technician and decided to take to

CDU, CDU unable to accept so went to A&E Resus instead.

Inappropriate use of ambulance resource – what alte rnative was there/is there?

Lives alone with wife, who is frail and his main carer. Paramedics called twice already on day of admission, to

help her get him out of his bed and chair. Third time, they took him into hospital.

66

Patient had personalized care plan fully documented and in place for end stage COPD. Deteriorated and panicky

in early hours of the morning with sudden deterioration. Unclear if patient was alone and how ambulance called

but decision made to admit despite clear indications that she didn't want to be including from the patient. Admitted to LRI as GH refused her admission. There was no signed DNACPR available, so A&E staff rang the practice to ascertain that aligned to her ACP, DNACPR was appropriate. The staff in LRI were aware she didn’t want to be in hospital but document that she did want their help on arrival. Spent last 24hrs in an acute care bay in LRI where she died. Son was involved in last stages and discussed and was with her when she died. Clear frustration throughout the notes and in relative feedback about the final place of death despite clear plan otherwise. The lack of anticipatory medications seems to have been the key problem here and the lack of a final end of life plan despite having an advance care plan.

Dehydration 8 cases referred specifically to dehydration as a major aspect of the reason for admission. Of these, 5 had a clear diagnosis of confusion or memory problems. At least 4 of these patients were admitted from care homes. Some examples are provided below:

Case 1

Lady with advance dementia. Repeated admissions in the 6 months pre death due to falls, dehydration, delirium,

infections. Lived in Nursing home. 6 months before death advance care planning beginning to be discussed but no

evidence of it being actioned or implemented. Repeat DNACPRs in place over a 2 year period. Clearly identified

likelihood to repeat admissions but no alternative options considered in light of advanced dementia.

Admitted by ambulance as a readmission and 'failed discharge'. Took 8 hours from nursing home call to arrive in

ED at midnight. The admission of an advanced dementia patient at midnight is unacceptable given a call from GP

in the afternoon.

Case 2

Known dementia in 24 hour care. Falls on [date] requiring OOH assessment diagnosed cystitis and treated

appropriately. In assessment unit due to worsening of aggression with history of fall and reduction in food / fluid

intake. Apparently given sub-cutaneous fluids in community as dehydration recognised but admitted with

significant dehydration and hypernatraemia.

Case 3

Patient had congestive cardiac failure and was being managed by GP and hospital outpatients. Awaiting surgery.

Admitted on [date] with dehydration, worsening AKI and fluid overload.

Case 4

He was readmitted from the care home end of [date] to A&E and then to AFU. Appropriate tests done. UTI

diagnosed. UTI treated and was fit for discharge but care home refused to have him back because of dehydration.

LRI staff did not agree with this. However, discharge was delayed because of high BM and also UTI appears

unresolved.

Case 5 Gentleman admitted from care home with possible sepsis. On admission realised this patient was dehydrated and delirious. However, had previously had full capacity. Reviewer has some concerns about indications that care in nursing home may have been weak given admission with dehydration and referencing to not using a toilet and enabling faecal incontinence. Had not been eating or drinking for 4 days before admission. All acute medical care appropriate with investigations, IV fluids and antibiotics however no evidence of weekend medical care apparent. Patient had very little nutrition for a week according to charts (refusing) but no dietician involved or alternatives sought. Patient had little nutrition for 10 days in total…[]

67

UTI 23 cases (13%) referred to UTIs with 20 of those referenced in the first three phases of care. 11 cases (48%) showed confusion/memory problems (6 had a clear diagnosis of such).

Pressure sores 7 patients were identified as having pressure sores at preadmission phase. These cases suggest a need for more intensive physiotherapy/mobilisation at home and support to carers with managing their relatives at home.

Case 1 Patient had been sent to [Community Hospital] 4 weeks before death for rehabilitation. [With] cancer diagnosis. Whilst there he deteriorated with postural problems resulting in being bed bound. He developed 3 grade 2 pressure sores. He developed [infection] and was treated for this. He lost [several] kgs in previous 2 weeks with [a very low admission weight] . Sudden deterioration and 999 called to admit with OOH doctor not attending.[..] … patient was weighed daily in [the unit] and noted as being able to eat and drink well. Case 2 Patient presented at A&E at the end of June having had seizures in the home. The family were contacted to discuss end of life as patient remained very frail and had advanced dementia. Admitted to the acute frailty unit. Patient had a grade 2 pressure sore obtained in the home Case 3 Preadmission known cancer patient bed bound at home and unable to attend Outpatient appointments so admitted on request. Lack of planning for bed bound patient requiring acute oncology treatment and review. Care at home from services appears lacking as husband trying to cope alone and not accepting that his wife is dying. No Macmillan intervention; some DN intervention but urostomy bag in place and pressure sores developing. Diabetic patient but no evidence of this in admission documentation or clear care plan in nursing notes of management of diabetes in terminally ill patient. Case 4 Admission LRI May 2017 …Discharged home A week later GP visit and diagnosis of [Lower Respiratory Infection] made, antibiotics given. 2 days later - Crisis Response Referral - assessed for intensive community support plan (ICSP) since in need of greater input. 5 days later discharged from ICSP with note that LPT team member was to discuss this with GP but no note of discussion with GP in GP records. Referral made for community therapy to enable her to progress with outdoor mobility and letter to patient explaining there will be a 10 week wait 2 weeks later GP visited since patient had hallucinations overnight. No infection identified so agreed with [relative] to wait and see how things went. Next day contact with daughter from LPT therapy team and daughter said she didn't think her mum was well enough to go outdoors so agreed to review in 2 weeks. Hospital bed requested - since developing pressure sores and in bed most of the time 2 weeks later LPT therapy made contact again and daughter said mum again not well enough to go outside and wondered if it would ever be possible for her to. Appointment made to assess her a week later. In the meantime seen by OOH service and admitted ? LRTI - acutely unwell, weak , confused, respiratory symptoms Case 5

Very elderly gentleman with bilateral amputations 5 months pre death and a recent history of rectal cancer. Unclear why but had recent diagnosis of Diabetes Mellitus and Peripheral Vascular Disease. The GP records indicate that he was cared for at home but struggling to become mobile and developed a Grade 3 Pressure Sore and recurrent chest infections. There was no physiotherapy provided post amputations. Wheelchair arrangements unsatisfactory and whilst there was a lot of activity there also appears to have been poor communication between professionals. There is no clear evidence of a care coordinator which may have prevented episodes of lack of equipment e.g. continence and dressing supplies, confusion over provision of oxygen in home and discharge arrangements. Patient readmitted to hospital in May with infection and sepsis. Discharged after 1 month to nursing home. Changed GP on change of residence. After 2 days in new nursing home deteriorated further and records indicate that he acquired a further chest infection in hospital and may have been discharged prematurely. New GPs were prompt in seeing him in care

68

home and wanted to readmit him given his condition. Some evidence that previous GPs had considered anticipatory care plan and admission avoidance schemes in the past but not indicated here. DNACPR in place from previous admission but reason given was for Sepsis and CCF/frailty which were not the presenting symptoms on readmission. No End of life care plan in place and new GP did not have sufficient time from discharge to arrange Advance Care Plan.

Families struggling to cope In 39 (21.5%) cases reviewers commented on families or carers struggling to cope with the care of their relative. 8 case studies are summarised below:

Case 1

Home visit by GP - patient living with son who along with partner are sharing care for patient. Struggling to cope

since patient's mobility declining since recent TIAs […]. GP diagnosed chest infection, and since patient and son

preferred for him to remain at home, provided oral antibiotics and made urgent referral to ICS for District Nurse

support over the weekend. Referral to ICS states DNACPR needs to be discussed.

Case 2

Patient has advanced dementia. Has package of care and community nursing. Seen by GP the day before

admission as was feeling unwell - referred to DN for blood tests …. Therefore it seems likely that when he was

admitted with "advanced dementia" he had acute on chronic confusion. Wife had been struggling to cope at home

with him. Seen by GP on [date] - dehydrated, general decline - bloods arranged. DN took some blood [next]

morning but was going to return for others that afternoon by which time he had been admitted…. Regular DN visits

and assessments ahead of that so could his deterioration and need for medical assessment/fluids have been

identified earlier?

Case 3

…from the GP records it is clear they proactively asked for help on a regular basis, explaining they were

struggling, communication was sometimes delayed (e.g. when messages being left with wrong daughter) and the

reviewer considers that more active communication with the daughters and care planning might have [improved]

their mother's end of life experience. [Date] daughters struggling to care for Mum with her declining mobility -

sharing 24hr shifts with her- OT with equipment, DN administering insulin, Sertraline for depression. [Date]

daughters talked to GP re her declining mobility - "unable to mobilise" - referred to community physio. [Date]

further decline so referred urgently to Intermediate Care Services who assessed her that day, made plan to put in

support for her at home but then she was admitted to hospital that evening via A&E- had been prescribed oral

Augmentin by GP which she hadn't been managing to take. To AFU 0014hr [date] and discharged home 1800

same day on oral [antibiotics] for UTI; Gliclazide having been restarted (GP had recently stopped it due to HbA1C

[result]); OT during this admission noticed her choking so assessed by SALT in hospital who referred for SALT in

community. Readmitted when daughter called 111 and advised to call 999 Reviewers question whether the patient

was medically fit for discharge home on [date] - would ongoing hospital treatment and a more proactive discharge

plan have enabled her to recover from rather than succumb to the infection?

Also no evidence of ACP although she had become frail 12 months ahead of her death. Although a number of

services involved including LHP, Parkinson’s Nurse, the GP records do not demonstrate active coordination and

forward planning of her care during her final year of life. ACP might have enabled this. Examples of difficulties in

communication between professionals and daughters at times over the year.

Case 4

Excellent preadmission care… Referred by GP to DN team for treatment on cellulitis pathway and oral antibiotics.

Nursing notes show that they also identified stress in the household and possible aggression [from a relative], who

was struggling to cope. Referred to adult safeguarding team. Also referred to social services for package of care.

During the time the DN were involved, they provided personal care when they could and also referred patient to

69

OT and continence nurse. Thorough OT assessment done and equipment ordered to make it easier for husband

to move patient and safer for her to get up out of the chair. DN notes state that one of the objectives of the input

was admission avoidance. DN discharged after ten days.

GP referred patient to CMHT for advice about how to manage her medication. She suffers from anxiety and was

on Temazepam but began sleeping during the day. GP reduced the dose but she then couldn't sleep at night.

Case 5

Admitted after fall down 6 stairs. Lost consciousness. Has moderate dementia and severe frailty. PMH of COPD.

Registered blind. Symptoms on admission - neck pain and acute tenderness,?fracture. Possible exacerbation of

COPD. History of multiple falls.

Investigations and treatment on admission - CT of head and neck and immobilise neck until have results. Suture to

head wound, slow IV fluids, bloods and senior review. Notes say ?social input as struggling to cope at home.

Son - main carer, who reports that he is struggling to cope with his mum. Have had carers in the past but son was

not happy with them.

Case 6

Admitted to LRI for blood transfusion (transfused) and treatment of LRTI ( antibiotics) , being discharged [date]

thinking he was fit enough to return home but he and his wife struggled so seen by OOH GP evening [5 days

later].

Patient found by OOH GP to be in bed, dehydrated with 3 day history of lethargy, reduced mobility, poor oral

intake, reduced responsiveness and increased confusion. Mostly bedbound. Wife struggling to care. Admitted to

community hospital

This therefore appears to have been an inappropriate discharge - either the patient needed more support/care at

home or discharge from LRI to community hospital.

Case 7

On [date] it is noted that his wife was struggling with "the bigger picture", i.e. his deterioration.

The reviewer noted that although the patient had been terminally ill for some time, and had been noted to be

declining towards end of life [date], it was not until [9 days later] that an end of life pathway was agreed. There is a

sense that this could have been initiated earlier.

The medical aspects of his care appear to have been very proactively managed, but the reviewer raised a

question about whether a more formal, holistic end of life plan might have prepared the patient's wife and family

more effectively for the last days of his life.

There are comments in the notes to suggest his wife struggled with adjusting to his decline towards his last few

days of life. Examples are that she wanted sedative medications, being used to reduce his agitation, reduced

when he became drowsy and perhaps more explanation of what might be needed would have helped her.

Case 8 Admitted from home as daughter (main carer) not coping - has MS. History of terminal Ca and liver metastases. Lived with daughter. Had become increasingly restless - daughter struggling to cope and in need of respite. OT visited re: discharge planning on day 1. Notes say - for palliative care. Had been receiving DN care - as had syringe driver for pain. Known to Macmillan service. GP reported that they made every effort to keep patient at home but patient wanted to come to hospital. Daughter reports to want her at home but can't cope and GP concerned that she would be passed from service to service and that a plan needed to be in place.

Warfarin management We noted 10 cases (5%) where warfarin management was noted in a variety of contexts. There were 5 examples that suggest there could be better warfarin management – in falls risk assessments, monitoring and the additional risk of antibiotics. There is an INReach Anticoagulation service which is referenced in one case.

70

Preadmission phase – characteristics of good/excell ent care

We analysed the narrative describing the quality of care in the preadmission phase. We were able to assess preadmission in 84% of cases.

Excellent care (10%) in 15 cases was characterised by a range of issues. Excellence did not mean perfect but in the circumstances and context of the presenting situation the reviewer considered a rating of excellent as appropriate. Excellence usually involved a notable difference being made or implemented, and included:

• Having a single senior GP partner to coordinate care

• Being on the Gold Standards Framework early (3 cases)

• Having an advance care plan in place early

• Specialist palliative care and other specialist teams involved e.g. Heart Failure Nurse

• GP monitoring and engagement with family; seen promptly at home by GP

• Care packages in place and supporting patient and family successfully

• Good transfer of care between GP practices where a change of GP was required

• Involvement of LOROS (hospice care)

Good care was rated in 74 (48%) of cases. Good care was characterised by:

• GP involvement often when also well known to GP

• Palliative care and LOROS involvement

• Patient’s wishes being respected

• Clear reasons for admission and identification of deterioration

• Early conversations with families about End of Life plans, DNACPR discussed and in place

• Applying the Gold Standards Framework

• Good communication with patients and families discussing care options

• Prompt referral and testing

• GPs seeing patients in care homes

• Personalised care plans

• People identifying and dying in their preferred place

• Specialist team involvement e.g. ICS, Liaison psychiatry, tissue viability, Marie Curie, Pulmonary nurse specialist

• Compassionate persuasion

71

• Daily/very regular attendance with elements of additional activity e.g. IV antibiotics in the community

Case studies of excellent care are described below:

Excellent preadmission care 1.

Patient has advanced metastatic cancer of unknown primary - first diagnosed through 2 week wait cancer referral

on [date], so very recent. GP notes show that patient was on the Gold Standard Palliative Care Framework. The

GP visited the patient to discuss the diagnosis with him and his wife and also talk about what support and advice

he needed. Weekly contact, including a home visit the day before he was admitted as wife was worried about him.

Prescribed pain control and actively discussed impact so that this could be altered.

Excellent preadmission care 2. Excellent preadmission care by GP. GP discussed prognosis with patient and asked about his understanding. Patient said he wished to die at home so admission avoidance plan commenced and referral made to MacMillan. Notes say patient on gold standard palliative care framework. DNACPR put in place by GP and flagged in GP records. Admission avoidance failed as MacMillan nurse said referral wasn't sufficiently detailed in terms of whether he needed more than what a DN could provide. This meant that admission avoidance would not be successful. However, in the event patient was admitted with query sepsis. Ambulance sent red flag for sepsis notification to ED.

Excellent preadmission care 3. This patient had metastatic ovarian cancer and her symptoms had been getting worse recently. Preadmission notes show that she was regularly seen by the GP (up to weekly), who referred her to LLR specialist palliative care team. The team provided ongoing care. GP put in place advance care plan - using Gold Standard Framework. Also put in place a DNACPR in discussion with patient. Patient's chemotherapy was delayed as she was admitted to private hospital with a query bowel obstruction

Characteristics of Adequate ratings

There were 41 ratings of adequate Preadmission care (26.5%). Adequate ratings where given where an aspect of care was lacking but not to the extent of providing poor care. Adequate ratings included:

• Evidence of respecting the families wishes, good long term conditions monitoring, GP and nursing support, discussions with families, End of Life care plans in place, choice of community hospital, CHC funding sorted

However the issues that arose for improvement were:

• Capacity issues arose e.g. ICS not having capacity, District Nursing unable to provide two people for a visit

• District Nurses unable to get a syringe driver at the time so admitted

• MH assessment and support for bereavement and depression; referral to Turning Point not followed up

• Care home issues - Nutritional supplements requested by care home but not provided, a lack of GP input to nursing home patient, patients being dehydrated, concerns about care in care homes

• DNACPR not discussed

72

• Best interest decisions potentially not appropriate e.g. warfarinisation

Preadmission phase – characteristics of poor/very p oor care

The narrative of the reviewer judgements was analysed to identify recurring issues. There were 24 cases of preadmission care rated as Poor (17) and Very Poor (7).

Poor care included the issues below:

• Weight loss in the community

• Deterioration not acted upon

• Lack of community support (ambulance attending to get patient out of chair), family not coping at home

• Lack of plan for coordinating the needs of frail elderly

• Warfarin/INR monitoring (reviewer considered there was a need for extra monitoring due to antibiotics)

• Not visiting frail elderly at home (asked to attend surgery) and assessment by phone

• No Advance Care Plan, End of Life Plan or DNACPR

• Lack of active engagement by GP

• GP not available to speak with EMAS

• Delays – X-rays, Ambulance attendance, delay in admission, in getting community therapy for mobilisation, waiting list for therapy, wheelchair problems

• Lack of investigations

• Paperwork – MacMillan forms/criteria, Marie Curie could not be commissioned by UHL

• Late admissions to A&E after a call earlier in the day from primary care (sometimes resulting clerking in early hours of the morning)

Very poor care was characterised by:

• Deterioration not being acted upon

• Bed bound patients

• Pressures sores

• Failure of OOH services to attend/GP not attending

• Family struggling to cope with declining mobility

• Inappropriate discharge

Specific cases included:

• No Bank Holiday cover for Total Parenteral Nutrition

• The failure of a detailed Advance Care Plan resulting in death in A&E shortly after arrival

73

• Deteriorating renal impairment not followed up in primary care

Six cases of preadmission care assessed as “very poor”.

In the case below, the reviewers questioned the classification of the Hospital Acquired Infection being treated as Community Acquired Pneumonia (CAP) as this gentleman had been in hospital for 4 weeks before his death. A possible reason for this may be GP medical cover at the Community Hospital that might be creating this discrepancy in classification.

The reviewers also questioned the appropriateness of the admission to acute care and whether this was because the community hospital could not administer IV antibiotics or treat the acutely ill patient.

Very poor preadmission care 1 Patient had been sent to [Community Hospital] 4 weeks before death for rehabilitation. [With] cancer diagnosis. Whilst there he deteriorated with postural problems resulting in being bed bound. He developed 3 grade 2 pressure sores. He developed [infection] and was treated for this. He lost [several] kgs in previous 2 weeks with [a very low admission weight] . Sudden deterioration and 999 called to admit with OOH doctor not attending.[..] … patient was weighed daily in [the unit] and noted as being able to eat and drink well.

In the case below the reviewer raised a concern that declining renal function had not been identified earlier.

Very poor preadmission care 2 Unfortunately no access to GP notes despite permissions having been given.

Review of ICE results however demonstrated declining renal function over previous months, with significant renal impairment by [date], warranting medication review and more frequent follow up of renal function.

In one case the reviewer commented that it seemed inappropriate for a patient in a care home with an agreed EoL plan and anticipatory medications to be readmitted to hospital for the final days of his life. In this case below, the reviewers questioned whether the patient was medically fit for discharge home or would ongoing hospital treatment and a more proactive discharge plan have enabled her to recover from rather than succumb to the infection. The reviewers found no evidence of an ACP although she had become frail twelve months ahead of her death. Although a number of services were involved including community services and the Parkinson’s Disease Nurse, the GP records do not demonstrate active coordination and forward planning of her care during her final year of life. An ACP might have enabled this.

Very poor preadmission care 3 Patient had type 2 diabetes and Parkinson’s Disease. … daughters struggling to care for Mum with her declining mobility - sharing 24hr shifts with her - OT with equipment, DN administering insulin, [meds] for depression June - patient’s daughters talked to GP re her declining mobility and she was referred to community physiotherapy.

74

22 days later - further decline. Referred urgently to Intermediate Care Services who assessed her that day, made plan to put in support for her at home. However, she was admitted to hospital that evening via A&E. She had been prescribed oral Augmentin by GP which she hadn't been managing to take. Admitted and discharged from Acute Frailty Unity a week later with oral trimethoprim for UTI. OT during this admission noticed her choking so assessed by SALT in hospital who referred for SALT in community. Readmitted 2 days later after daughter called 111 and advised to call 999. A&E letter states likely urinary sepsis. Patient died 2 days later.

In this case, the reviewer noted a system issue as no-one was available to set up TPN feeding at home over a bank holiday. This meant that the patient was discharged before the bank holiday, before having an OT and physiotherapy assessment.

Very poor preadmission care 4 This was a readmission following a failed discharge. The patient had been in following a fall and with a leaking stoma dressing. He was on parental feeding and his notes immediately before discharge say that he needed to get home before the bank holiday so that the DNs could set up his Total Parental Nutrition (TPN). However this meant he had not been assessed by the physiotherapist or OT. Patient said he felt he needed rehabilitation before going home as he was not strong enough.

75

4.3 Initial Management and Admission

Number assessed and ratings for Initial Management and Admission phase of care

The chart below shows the ratings for this phase of care. This was the most positive phase of care throughout the pathways reviewed. In total, nearly 94% of patients were assessed as having adequate, good or excellent care. 65% of the patients reviewed were assessed as having received “good” (50%) or “excellent” (15%) in this phase. At the other end of the scale, 11 (6%) patients were assessed as having received “poor” (4%) or “very poor” (2%) initial management and admission care.

Chart 32 Initial Management and Admission ratings

47

49

86

26

5

0

10

20

30

40

50

60

70

80

90

100

Very Poor Care Poor Care Adequate Care Good Care Excellent Care Unanswered

Initial Management and Admission ratings (n=177)

76

Notable themes

Cases presenting in A&E included a range of significant issues discussed elsewhere in the report that reflects an elderly group of patients requiring emergency care. This includes:

o presentation of patients with UTIs and dehydration from the community (See Preadmission section);

o a significant amount of delirium/confusion/memory problems in some form (See Overall Care and profile sections)

o families presenting with patients expressing difficulty in being able to cope with their relatives

Themes arising:

• Antibiotic management – at least 73 patients were observed to be receiving antibiotics with 51 referred to in this phase of care. On the whole we saw rapid sepsis assessments with, in the main, sepsis paperwork completed in a timely fashion. The practice did vary at times with both extremes observed of no assessment through to several forms being completed in a short period of time. Antibiotics were given promptly once prescribed although there were cases of delay to administration.

• IV fluid administration, where needed, was promptly prescribed and administered in the majority of cases.

• Timeliness – delays in A&E were seen with waits over 4 hours in at least 14 cases. The 14 cases were for patients who were all over 71. Of concern is that 11 (78%) were people over 82 with 5 patients in their 90s. Full analysis of waiting delays was not possible to undertake reliably for the reviewed cohort from the notes available.

• Transfer from A&E (see also preadmission comments on timeliness of admission to wards late at night). Whilst there were some A&E delays the cumulative effect of patients presenting later in the day should be noted.

• The cumulative impact of delays through the system in preadmission and on to a ward for elderly patients, with elements of confusion, dementia, dehydration or UTI needs to be monitored. We have also highlighted above the greater risk of problems in the very elderly group.

• Access to radiology - reviewers noted timely access to radiology. At least 29 patients (16%) had CT scans in the reviewed cohort. Access to X-rays and ultrasound were available in a timely fashion. There were a small number of cases of delay or delayed results being shared with patients in other phases of care.

• One case was reviewed and discussed with the local team at UHL during our review which triggered a possible concern. It was unclear from the notes what triage mechanisms were in place generally prior to transfers being made. In this case concerns were raised by the receiving team triggering an internal review so there may be a new system in place by now. Some Standard Operating Procedures to stabilise patients for transfer before and after travel are to be recommended with the focus towards stabilisation on the first site.

77

• Reviewers commented on the complexity and potential confusion of the ED record bundle. Whilst it was easy to find for case reviewers (and others) as the bundle was a different colour, there is a great deal of information requested but rarely completed. This could lead to confusion or difficulty finding the key information needed.

Initial Management and Admission phase – characteri stics of good/excellent care

Good or excellent care was rated in 114 cases (66%).

Excellent care was characterised by:

• Prompt attention

• Respecting patient’s wishes

• Appropriately aggressive care

• Advance care plans being available

• Excellent documentation

• One specific case included the completion of a Get to Know Me Profile for a patient with dementia

Good care included:

• Stroke team involvement once patient referred or team notified

• CT scans being available promptly

• Discussions with families on management plans and compassionate explanations and information provision

• Prompt Sepsis assessments and administration of antibiotics

• Prompt administration of IV fluids

• Medication review/reconciliation (which identified need for changes/corrections)

• Use of Early Warning Scores (EWS)

• Efficient referrals to other teams – SALT, Palliative Care team, alcohol liaison team

• Involvement of specialists – Stroke, Heart Failure, Oncology

• Good liaison with other Trusts – e.g. in Nottinghamshire and Birmingham

• Direct admission to Glenfield when needed

• End of Life care recognised and provided in this phase of care facilitated by Advance Care Plan

• A specific case of proactive decision making in a care home

• A specific case of a patient’s wife (with dementia) being cared for in the emergency department because the patient was her main carer

78

Case studies:

Excellent care 1 Known Breast Cancer with metastatic disease and pleural effusions Assessed in Oncology clinic [date]: breathlessness and pain not controlled despite introduction of steroid therapy on previous admission. CXR no worsening of pleural effusion. Potential for PE considered and CTPA done for exclusion - appropriate not to do D-Dimer and was covered with treatment dose Dalteparin in line VTE guidance. Seen quickly on ward by Palliative nurse day of admission - pain management plan agreed. Reviewed following day pain control better on-going pain management plan agreed. Discharged with pain medication and treatment for chest infection - raised WCC acknowledged possibly due to steroids Follow up agreed with Oncology team (parent team) Discharged next day Excellent care 2 Bloods taken and results available within 30 mins of arrival at ED. Investigations ordered and completed within 90mins of arrival in ED. Neuro bleed identified promptly. Discussion with daughter on management contemporaneously and referred to stroke nurse in under 3 hours post investigations and verbal reporting from radiology and discussion with neurosurgeon. Seen and assessed by stroke nurse within 3.5 hrs. Discussed management plan and completed DNACPR with daughter in under 4 hours. Admitted to Stroke Unit Excellent care 3 Patient found on the floor … by his neighbour …. On admission, EWS 5 and crackles in lungs. Sepsis 6 bundle commenced. Later that evening - patient had improved - EWS 2. Patient has a diagnosis of Alzheimer's disease - lives alone and has carers three times a day. Has integrated care plan and advance care plan in place - noted that not for resuscitation since [date] but no DNACPR form seen at that time (noted). Also has personalised care plan and dementia care plan. Admitted to AFU for CXR and IV antibiotics. On admission to AFU, Get to Know Me Better profile completed so that staff know what to call him, what makes him happy and sad, and a bit about his life.

Characteristics of Adequate ratings

In 50 cases (29%) the reviewers rated care as adequate. As with other phases there were elements of the care that were good and elements that were not. Adequate ratings were given when an aspect of care was felt to affect the delivery of the care. These included:

• Delays in aspects of care – provision of investigations, ambulance wait, available porter, 8 cases of delays in excess of 6 hrs for admission to a ward

• Clinical monitoring – lack of medical oversight, fluid balance not complete, MUST not completed, labelling error, records not with patient

• Lack of beds – specific cases included: Acute Care Bays limited, refused by ITU (and unclear documentation why), no community beds which meant admission to Acute care, no capacity in Oncology

• Inability to implement care – unable to sign off EoL plan at weekend, UHL unable to commission Marie Curie care

79

Initial Management and Admission phase – characteri stics of poor/very poor care

As stated this phase of care was the most positively rated phase through the review. Of 172 cases rated, 7 (4.1%) were rated as Poor and 4 as Very Poor (2.3%). These points therefore relate to 11 cases only:

• Sepsis assessment not completed or antibiotic administration delayed • Poor monitoring – no fluid balance completed, likely inaccurate recording of EWS/no ABG

taken, gap in observations; E-obs not undertaken overnight and not therefore escalated • An inappropriate admission to Glenfield, not appropriate to condition and pre-transfer

checklist blank • Notes not available for last admission as lost • Delay in handover from ED to AMU with delay for surgical opinion – patient had

hypoglycaemic attack due to lack of blood glucose monitoring and possibly due to cumulative delays

• Busy department resulting in lack of management of continence and pressure areas • Long delay in admission to ward for a very elderly patient

Case studies

Very poor care 1 Stayed in for less than a day. EWS on admission was 3. Discharge letter showed that the only treatment was that the hospital stopped his [type] acid. Reviewer queries whether the sepsis bundle checklist was filled in correctly as he had symptoms of an infection (cellulitis) and new confusion but considered low risk for sepsis. His cause of death was a UTI five days later. Falls checklist completed. Bloods taken. No fluid balance recorded. Very poor care 2 Admitted short of breath to ED sepsis due to community acquired pneumonia. NEWS 9 with paramedics 6 on arrival ED - antibiotics administered within 25 minutes of arrival. NEWS Sepsis 6 followed except no ABG (venous gas done 1647 Lactate 1.4) or referral to outreach from ED. Observations taken show unlikely improvement in oxygenation ? accurately calculated 0 at 2028 in ED. Also no urine output documented in ED department (patient PU at home noted 1100) again contributing to miscalculation. No comment on renal impairment Creat 172 - assume AKI but no previous result for comparison. Inadequate fluid resuscitation. 3 hour gap in e-observations recorded . Note nursing record 1920 sats dropped to 64% improved to 97% with 10 L delivered via 60% Venturi mask. These not logged within e-observations. CRP 304 Decision to transfer to Glenfield for ongoing care not appropriate to condition. Pre transfer checklist blank.

80

4.4 Ongoing care

Number assessed and ratings for Ongoing Care phase

133 (86%) of cases were rated adequate, good or excellent in Ongoing Care. 21 (14%) were rated poor or very poor. 62% of cases were rated as having good or excellent care.

Chart 33 Ongoing Care ratings

2

19

38

79

16

0

10

20

30

40

50

60

70

80

90

Very Poor Care Poor Care Adequate Care Good Care Excellent Care

Nu

mb

er

of

De

ath

s

Care Rating

Ongoing Care ratings (n=154 )

81

Notable themes

Fluid Balance monitoring Fluid balance monitoring was referred to specifically in 26 cases.

There were 12 cases of fluid balance monitoring in a negative context. Most of the patients moved ward during their short stays. However, Ward 33 LRI had 7 cases of negative fluid balance management identified as the admitting ward.

There were 14 cases of fluid balance monitoring referred to in a positive context. There was no particular pattern to these positive practices but they were spread across 8 units – 7 cases in ESM, 4 in CHUGGS and 2 in RRCV. 1 case was in LPT care.

Antibiotic management

We observed 8 cases where the microbiology team were involved in this phase of care in the decision making in relation to the most appropriate antibiotic protocol.

Engagement of Microbiology 1

This gentleman was admitted via ED with a 2 day history of vomiting on a background of bladder cancer,

hypertension, AF, ischaemic heart disease and aortic valve replacement. He was dehydrated with evidence of

AKI. A tentative diagnosis of viral gastroenteritis was made and fluid resuscitation was commenced. The fluid

balance chart was only partially completed. …. Sepsis 6 screening was only partially completed. The reviewer

notes a normal white cell count with raised CRP and low temperature. Empirical antibiotics were given.

Following senior medical review it was felt that he may have pneumonia and empirical antibiotics were modified

accordingly. A sepsis screen was sent. It later transpired that he had positive blood cultures. After discussion with

microbiology antibiotics were changed to match the sensitivities and further investigations were requested to rule

out infective endocarditis.

Engagement of Microbiology 2

Good subsequent record of engagement of Microbiology registrar to agree antibiotic regime as patient had penicillin allergy. Clear consideration of all options for antibiotic plan…

Medication management – the value of engaging with Microbiology and Pharmacy

Admitted and treated promptly with antibiotics but pharmacist notes clearly that microbiologist needed to be

involved as antibiotics not being given in accordance with UHL policy

Patient’s anticoagulation dose too high and pharmacist requested reduction due to patients low body weight.

Effective medication reconciliation review in place

The reviewer cannot comment on whether the correct medication and antibiotics would have made any difference

to outcome but sudden deterioration in frail elderly due to sepsis and failure to follow policy should be reviewed.

Medication reconciliation completed and identified incorrect medication - no Datix apparent in records

Diabetic management We observed 24 patients as having some form of diabetic care need. However, there were 11 cases in the Ongoing Care narrative regarding diabetic care. This showed very variable practice in blood glucose monitoring. 5 cases involved regular blood glucose monitoring/response and

82

care. 6 cases highlighted below suggests improvements in including blood glucose as part of EWS is required.

Blood glucose monitoring 1 – in patient with langua ge barrier

Blood sugars increased throughout admission. The blood glucose monitoring was not a clear part of care plan. On

EWS despite being diabetic the blood sugar was not referenced as part of the score

Blood glucose monitoring 2 – in patient with infect ion

Note diabetes control sub-optimal during admission influencing risk of infection

Diabetic care 3 – lack of dietary input

Pressure monitoring, fluid balance well completed, Pt rounds checklist, MRSA screen blood glucose, falls and

cannula care checklists on all 4 occasions. Weight chart kept and maintained.

Only obvious service lacking was dietary input - despite referral and weight loss there appeared to be no dietetic

input although the patient was on supplements.

Blood glucose monitoring 4

Nursing records very poor. No care plan in place, no reference to wheelchair requirements and double amputee in

care plan including e.g. suggestion that safe environment was not an issue when he depended on side rails at

home to manoeuvre. No recognition of end of life.

No blood glucose monitoring despite being diabetic.

No medical care over most of weekend until death.

Very poor fluid balance monitoring and no evidence of what IV fluids were given - if any.

Blood glucose monitoring 5

Patient died the following morning after admission after sudden deterioration overnight. The reviewer notes that very little reference to blood glucose was made in care plan given patient diabetic. No evidence of blood glucose being monitored. EWS states blood glucose not clinically indicated. Blood glucose monitoring 6

AKI and electrolyte disturbance managed

Documented interaction with patient and husband

Fluid balance / weight not recorded completely - unclear management of heart failure

Diabetes control poor in context of infection and newly diagnosed

Weight loss and weight monitoring There were 19 (10%) cases where weight loss was commented on across all phases. 10 cases were in the Ongoing Care phase. There was variable practice in relation to weight related aspects of care. The cases highlighted below indicate the need for monitoring weight in relation to pressure care, correct prescriptions, nutrition and fluid management.

83

Case 1 Aggressive fluid management in frail underweight patient. Best practice would recommend daily review U&Es but relevant that patient refusing some interventions. Questionable suggestion of use of Amiloride to manage low potassium in patient admitted dehydrated and hypokalaemia iatrogenic…Community dietitian referral to In-hospital Dietetic team - dietitian visited ward [date] to alert that no referral placed on ICE and did not then review patient despite MUST triggering for review and clear history of poor intake for preceding days. No evidence of review during admission - referral on ICE [4 days later]. Case 2 Management of heart failure explored wide range of options medically with regular review and management by heart failure team. Management of pain balanced with constipation kept under review….Heart failure team observed poor fluid balance documentation and failure to weigh daily - no weight chart found in records …. risk assessments completed but not kept up to date e.g. high risk WATERLOW not reviewed for 4 weeks of admission Case 3 Admitted and treated promptly with antibiotics but pharmacist notes clearly that microbiologist needed to be involved as antibiotics not being given in accordance with UHL policy …Patients anticoagulation dose too high and pharmacist requested reduction due to patients low body weight. Case 4 Comprehensive nursing notes available with assessments all completed and upto date and reviews - bed rail, BEST, pressure monitoring, fluid balance well completed, Pt rounds checklist, MRSA screen blood glucose, falls and cannula care checklists on all 4 occasions. Weight chart kept and maintained. Only obvious service lacking was dietary input - despite referral and weight loss there appeared to be no dietetic input although the patient was on supplements. Case 5 Supportive care given with attention to fluid intake and nutrition - did have repeat weights recorded but MUST not [taking] change in weight into account. Potential weight loss of 10% during admission in keeping with behavioural issues and food chart records. No record of dietetic involvement. No comment on nutrition from Dementia support team. Case 6 During her time on the ward she was referred to and seen by the specialist teams - the heart failure and palliative care teams both saw the patient and assessed….other evidence in the notes that the nurses were not always thinking through the care they were giving to this patient. The doctor noted one day that the lady had nasal cannulae in place but the Oxygen was not switched on. Another doctor noted that although this lady had a lot of oedema and was on diuretics, no weight chart was in place. However, this lady was on end of life care, so this may not have been so beneficial.

Family involvement – getting it right Efforts to engage families were extensive and was the norm in most cases.

Case 1

As the patient's condition deteriorated the team asked the patient about contacting the family. The patient was

keen for them not to be involved and the team abided by his wishes, but then the patient relented and the family

were called. The team had sympathetic and honest discussions with the family.

Case 2

Another positive finding was that the team seem to have respected the wishes of patient and family along the way,

for example enabling his wife to continue with all his personal care whilst in [Community Hospital].

Case 3

Patient's wife was called early in the morning and asked her to come in to discuss a plan. Medics were planning

for end of life. Nursing notes explicitly state that nurses new on shift went over to introduce themselves to the

patient and his family.

84

…And not so right

Case 1

The son wanted an update on his father's condition, stating he was not clear what was going on. The reviewer

wonders whether the family might have been more actively communicated with over the previous 24 hrs from

when the patient decided to decline [treatment] This might have better prepared the family for the possibility of

their father's decline and death.

Case 2

DNACPR and "for escalation" documented in notes but until this point no note of discussion with wife/son/daughter

about this or what the latter meant

Case 3

Son very unhappy about end of life care - as his mother was told her prognosis at 3am in a busy, noisy ward.

Medication reconciliation/review and drug errors Medication reconciliation was apparent throughout the case review with clear evidence of pharmacy review. It was noted as a positive lesson in 3 cases in particular. These cases demonstrated that this was working well with the review noting dosage errors (usually to do with weight issues) being corrected or picking up cases where antibiotics needed amending.

Ongoing Care phase – characteristics of good/excell ent care

Excellent care characterised by:

• Multidisciplinary approach • Communication, information sharing with patient and relatives • Diligent close and attentive clinical monitoring • Comprehensive nursing records and use of checklists • Senior input with an early management plan • Regular and continuous communication with the family • EoL plan updated on a daily basis with detailed care planning • Treatment options discussed with Microbiologist • Family wishes respected • Clinical monitoring

Case examples are shown below:

Excellent care 1

Care was responsive, timely, systematic and communication between multidisciplinary team, patient and cousin

was good

Patient remained under care of respiratory team with involvement of other team – oncology, palliative, and

community nursing. Good practice in diagnosis, sharing information and taking her views into account

Timely transfer to cath lab and intervention on arrival. Managed subsequently in ITU with evidence of good team

work in care but scan and EEG evidence of hypoxic brain damage. Good evidence of discussions with family

members. Preparation for end of life and discussion of potential for organ donation. Evidence of best practice in

care and documentation

85

Excellent care 2

Admitted to Ward after 2 days where the care was diligent and attentive. regular medical review meant monitoring

was close and responsive to signs of deterioration. CT scan, repeat CXRs to monitor effusion, draining ascites.

Some chasing of bloods at times and the odd result but on the whole efficient. Discussed prognosis and situation

with relatives when available.

Sepsis carefully monitored with 4 checklists in admission. Comprehensive nursing notes available with

assessments all completed and upto date and reviews - bed rail, BEST, Pressure monitoring, fluid balance well

completed, Pt rounds checklist, MRSA screen blood glucose, falls and cannula care checklists on all 4 occasions.

Weight chart kept and maintained.

Only obvious service lacking was dietary input - despite referral and weight loss there appeared to be no dietetic

input although the patient was on supplements.

Good care was characterised by:

• EWS regularly monitored, nursing assessments completed • Discussions on options with families and patients • DNACPR in place after discussion with patient and/or family • Advance Care Planning introduced early (see description below) • Senior review with MDT involvement and Critical Care Outreach Team (CCOT)

engagement • Detailed investigation • Palliative care team involvement • Discharge planning considered early • Clear documentation and management plans • Hydration and fluid management monitored • Mouthcare • Engagement of SALT and other members of the multi-disciplinary team • Fast track arrangements in place early • DoLs assessments in place • Early engagement with LOROS once need identified with families and patient

Good care

This 80+ year old lady was appropriately cared for by the whole MDT on the ward. The physio treated her

regularly. Seen by the doctors daily. Nursing input was caring and all appropriate tools were used to monitor this

patient. The MUST score was checked weekly; Best shot was completed daily. Strict fluid management was

undertaken appropriately because of her impaired kidney function.

During her time on the ward the patient had a number of tests to try and find a clear diagnosis. This included CT

scans (when patient's renal function had improved slightly) and frequent blood tests.

A number of discussions were had with the family. Once a clear diagnosis of cancer had been made these

changed from updating on progress to prognosis and plans for end of life. As well as her team, the consultant

personally spoke to the family on two occasions. All of this indicates compassionate care. Questions were

answered honestly and clearly. The family were clear that they had had conversation with their mother in the past

and knew that she would not want invasive treatment. This was clearly noted in the medical notes and also made

clear in a MDT meeting.

86

Good advance care planning Daughters were consulted throughout. The patient had put in place an advance care directive and therefore there was a detailed discussion with the family about next steps. It was agreed that the patient did not want any more treatment in hospital. A CT was cancelled because the daughters said the procedure made her very distressed. A plan to treat in the nursing home was set out with explicit notes of the circumstances in which an admission to hospital would be appropriate (e.g. a broken bone). IV antibiotics were rejected because it would mean remaining in hospital. Oral medication was prescribed instead.

Adequate care issues:

• Weekend care lacking • Confusion over DNACPR • Lack of ACP resulting in unnecessary work e.g. home assessments/fast track app • Repeated NBM due to cancelled surgery • Confused monitoring due to lack of clarity in EoL situations • No DNACPR in notes or lacking clarity/absence of discussion

Ongoing Care phase – characteristics of poor/very p oor care

Poor care and Very Poor care

• Side room obs/night time obs not done/gaps in e-obs • Failure to weigh daily where required • Poor fluid balance monitoring or no fluids /nutritional status concerns • Weekend facilities lacking in community hospitals or no weekend cover for some teams • Unable to give IV antibiotics anywhere but acute care • Development of pressure sores • Failure to listen to or discuss care with relatives • Lack of MDT approach

Case examples are shown below:

Poor care

Good treatment of the presenting problem. Unclear about how dysphagia first became noted and possible cause.

Poor communication between ward and SALT team. The reviewer asks - were there early enough efforts to get the

patient eating and drinking again (delays to feed at risk forms, delays of SALT review, no decision re: PEG)?

Areas to note:

Deprivation of Liberty Safeguards

There was variable practice in place with some DoLs clearly in place, others delayed and others not in line with best practice

Complexity of nursing records

Reviewers commented on the complexity of the nursing records which at times meant that a holistic view of care was difficult to see. This appeared to result in unnecessary repetition. At times reviewers also commented on the sepsis bundle being used as a track and trigger too

87

Clinical monitoring

A number of instances were noted regarding observations. Side room observations (mainly at night) appear not to be completed contemporaneously, e-obs at times had to be repeated within a few minutes (which could be a training issue) and EWS not completed fully.

Dietetic support

Whilst in many cases the dietician and SALT team were involved (in particular in the Stroke unit) reviewers commented on cases where this hadn’t occurred and should have been.

Investigation delays

Delays in getting some investigations e.g. U/S for DVT were noted for 2 inpatients. This included a 3 day mid-week wait.

Psychological support

Reviewers noted a number of cases where psychological support to the family or immediate carer would have been appropriate. These were in circumstances where an unexpected diagnosis that would be imminently terminal had been made or where relatives were not able to accept end of life. Lack of 121 support when needed was noted in 1 case.

Falls

We noted the lack of falls as a patient harm in the cohort.

Infection Control

We also noted a lack of reference to infection control problems. Whilst 6 cases were logged, only 2 were thought to have led to harm and both related to catheter management (indwelling and suprapubic)

88

4.5 Procedure Care

Number assessed and ratings for Care during a Proce dure phase

Care during a Procedure was rated in 28 cases. 92% (26) were adequate, good or excellent shown in the chart below.

Chart 34 Care during a Procedure ratings

2

7

13

6

0

2

4

6

8

10

12

14

Poor Care Adequate Care Good Care Excellent Care

Nu

mn

er

of

de

ath

s

Care Rating

Care during a Procedure (n=28)

89

The 28 procedures observed included:

• ascitic drains/taps (including CT guided), • chest drains/pleural aspiration, • biopsies, • NG tube placement, • gastrografin enema, • oesophageal stent insertion, • videofluroscopy, • PICC line, • coronary scan and angiogram, • central venous catheter insertion, • ICD management at End of Life, • blood and platelet transfusions, • catheter insertion and intubation • cannula care (not all patients with a cannula had a rating).

Care During a Procedure phase – characteristics of good/excellent care

• Consent taken and recorded

• Good record keeping and clear procedure notes

• Documented results

An example is below:

Excellent care example Pleural aspiration completed. Excellent practice demonstrated in reasons for procedure. Pain killer prescribed, verbal consent gain and documented and plan clearly stated in terms of response to outcome of tap. Documentation completed timed, signed and dated. Clear documentation of outcome subsequently and decision making.

Specific issues of note raised in the Care during a Procedure phase

• No consent on file for an ascitic drain • No consent for the gastrografin enema • Delay in CT, coronary scan and angiogram for 10 days • Best interest decision making at end of life and DoLs application

23 out of the 28 patients recorded as having a procedure died in hospital; 5 were discharged.

The 2 poor care ratings were in patients over 81. These related to a cannula not being changed for 8 days and a question from the reviewer as to whether a Central Venous Line was in the best interests of a patient at end of life.

90

4.6 Perioperative care

Number assessed and ratings for Perioperative phase

The overall ratings for the Perioperative Care Phase are shown in the chart below. All (100%) care was rated as adequate, good or excellent.

Chart 35 Perioperative Care ratings

1

7

2

0

1

2

3

4

5

6

7

8

Adequate Care Good Care Excellent Care

Perioperative Care (n=10)

91

10 patients had surgery out of the reviewed cohort. 7 were emergency admissions and 3 were from the waiting list. All care was rated as adequate, good or excellent. 3 patients died in the community after discharge; 7 died in hospital.

We do not know how long the waiting list patients had waited – 1 was intestinal surgery and 2 were cardiac surgery. All died within 6 days of admission.

Poor care overall (though not in this phase) was the conclusion for one surgical patient which has been highlighted in advance of this draft report and the details provided for internal review.

The age bands of the patients subject to surgery are show in the chart below with the care rating.

Chart 36 Perioperative Care ratings by age band

The procedures codes that the 10 cases related to are in the following table:

Table 33 Procedure codes for Perioperative Care

GC17A - Non-Malignant, Hepatobiliary or Pancreatic Disorders, with Multiple Interventions, with CC Score 9+ ED28A - Standard Coronary Artery Bypass Graft with CC Score 10+ FZ74F - Complex Large Intestine Procedures, 19 years and over, with CC Score 0-2 EY41A - Standard Percutaneous Transluminal Coronary Angioplasty with CC Score 12+ DZ17Q - Respiratory Neoplasms with Single Intervention, with CC Score 6-9 FZ74F - Complex Large Intestine Procedures, 19 years and over, with CC Score 0-2 ED12A - Complex Repair of Aortic Root with CC Score 7+ HT13A - Major Hip Procedures for Trauma with CC Score 12+ GB06E - Intermediate Therapeutic Endoscopic Retrograde Cholangiopancreatography with CC Score 6+ FZ73C - Very Complex Large Intestine Procedures with CC Score 9+

Perioperative phase – points to note across all car e ratings

• Surgical checklists were completed and on file (In one case the reviewer noted the CCU checklist had not been completed)

• Surgical support on the ward was good

• Twice daily medical review in ITU was evident

1 1 1

4

2

1

0

1

2

3

4

5

Adequate Care Good Care Excellent Care

Nu

mb

er

Periperative Care by Age Band

41-50 51-60 61-70 71-80 81-90

92

• CCOT outreach post ITU discharge was provided

• Comprehensive procedure notes and post operation observations were in place

• Good communication with GP was observed

• Informed consent was gained

• Effective transfer between theatre, ward and ITU

• Team working in response to deterioration with timely intervention

• A specific example of the DNACPR being suspended for surgery in line with patients wishes (although this caused some confusion as to the timing of reinstatement)

Under problem types identified, 2 issues were noted which related to the surgical patients in these 10 cases – in one case pressure sores developed in theatre – the patient had very thin skin and a very low body weight. The other problem related to a delay in surgery.

Areas noted for consideration:

• Possible over reliance on pre-assessment for elective cases with a lack of clerking/medical review at point of admission and documentation of risk

• Assessment of competence and capacity when getting consent

93

4.7 Readmission care

Number assessed and ratings for Readmission phase

Readmission was difficult to rate looking across the pathway as there was evidence of readmission in several cases but not necessarily on the last admission. The chart below shows those cases the reviewers assessed in this episode of care but under-represents the extent of repeated admissions. Most readmissions (73%) appeared appropriate.

Chart 37 Readmission ratings

94

The chart below provides readmission ratings by age band. 9 cases (60%) were over 81 years).

Chart 38 Readmission ratings by age band

Case examples by care rating

There were 9 cases of good to excellent care. Case studies below describe some of these cases.

Excellent care

Patient readmitted on [date]. In the interim, she attended outpatient appointment with cardiology, where it was

decided with the patient and family that she would not have the surgery. Admission avoidance attempted by family

and GP this time - family called 111. GP referred to DNs to visit to assess. Patient too unwell to leave at home and

was taken to hospital by ambulance. (DNs had also been visiting regularly to assess her bloods before this time).

On readmission, seen by palliative care team. Prognosis discussed (weeks) and discussed where patient would

like to be cared for. Preferred home but thought her children wouldn't cope. Subsequent discussions with patient

and her family led to decision to discharge to a Nursing Home. Fast tracked.

Good care (5 case examples)

• Readmission was well managed. Patient was very poorly and had immediate care.

• Patient seen by emergency stroke team in A&E and readmitted to the stroke unit very quickly.

• Appropriate tests were done and planning for discharge happened as soon as the patient arrived on the ward each time.

• Readmission in [date] was well managed with a quick admission to the AFU (4 hours).

• Readmission was well managed on this occasion, with reasonably fast admission to AMU.

95

There were 4 cases rated as poor or very poor – the narrative is below. The main concern is that these admissions could have been avoided.

Case 1

Patient was readmitted to LRI but stayed less than 24 hours so not recorded as an admission. No bloods taken.

Notes (from community hospital) did not accompany patient. No name band on patient.

Case 2

Care home admitted out of hours despite reassurance from family members and then refused re-admission -

raises questions regarding initial placement and their understanding of needs on previous discharge. No attempt

by out of hours primary care team to prevent readmission.

Acute setting inappropriate for her needs despite her receiving good care.

Case 3

This was a readmission as the patient had been discharged a few weeks earlier with the same problems. Not clear

what support she and her family had in the intervening period given her diagnosis and prognosis.

Case 4

The readmission in July was handled well in A&E but the fact that the A&E doctor noted that although the lady was

ill and that her oedema was no worse than usual suggests that this readmission could have been avoided and

managed in the community, especially as this is where the patient wanted to be.

This is rated as poor care. Second reviewer and I both feel that this admission could have been avoided or at

least reduced to possibly an overnight stay with the care home and the primary health care team providing care in

the community.

96

4.8 Discharge care

Number assessed and ratings for Discharge phase

There were 57 patients who were discharged from acute care before death. 82.5% (47) cases were rated as adequate, good or excellent. Over half (29) were good or excellent. 10 cases were rated poor or very poor.

Chart 39 Discharge Care ratings

2

8

18

24

5

0

5

10

15

20

25

30

Very Poor Care Poor Care Adequate Care Good Care Excellent Care

Nu

mb

er

of

de

ath

s

Care rating

Discharge (n=57)

97

An analysis on days to death following discharge is shown in the Profile section of the report.

Problems with aspects of discharge were recorded in 31 cases (17%) in total. This was due to a variety of issues which are described below. Discharge issues were noted in cases that had not been discharged so the issues identified below are drawn from a wider set of cases than just those discharged from hospital.

Notable themes

Fast track Fast track appeared to be working well on weekdays. Fast track arrangements were referenced in at least 42 discharges (74%). On the whole we saw good use of fast track arrangements (26 cases). Planning for discharge appeared to start early in most cases or happened very quickly when OTs and equipment could all be arranged during the week. Examples are described below:

Case 1

CHC Fast track application made; prompt OT Home assessment completed, equipment delivered and patient

discharged home within a week as per patient wishes

Case 2

Fast track care plan includes advance care plan. Hospice at Home put in place for support and package of care to

include 4 visits per day. This took 1 day from completion of fast track form to discharge.

Case 3

The ward had a clear management plan from the start and discharge planning started as soon as the lady was fit

enough to consider options. These were discussed with son early on. The discharge team assessed for fast track

as soon as was appropriate and plans for equipment to be sent to the care home were expedited.

Case 4

Patient got into LOROS quickly - within 2 days of fast track being set up. He told the Palliative Care Nurse that he

was happier once he had made the decision to stop treatment.

Although the following case sums up the impact of n ot planning for discharge soon enough:

Case 5

Discharge planning started 8 days after admission which could have been started earlier given diagnosis and

known to live alone and need support. Referral to OT and PT made after 8 days too.

Palliative care provided and patient wished to die at home. Discharge sister engaged 3 weeks post admission and

still looking at options for discharge

Sudden decision to discharge home seems to catch the team out and equipment not ready but promptly executed

with access visit undertaken same day. Once discharge sister involved referrals appear to be done quickly for

equipment

Reviewer considers that discharge could have been less hurried if discharge planning had been acted upon earlier

and discharge sister involved earlier/home visit completed given patients stated wish to go home. However this

was achieved.

98

The exceptions are highlighted further below which provide insight into some refinements that need to be made. We would note that:

• fast track discharge processes were delayed over weekends • in some cases fast track appears to be hurried putting pressure on all concerned • whilst the fast track process appears to work well there is an absence of an Advance

Care plan alongside in some cases • packages of care do not appear to include night time cover (and Marie Curie cannot be

commissioned by UHL). • the need for a DNACPR appeared to be a requirement for CHC funding in fast track

cases in Leicestershire in one case – this needs to be reviewed as CHC care should not necessarily be commensurate with not resuscitating someone unless part of a clear EoL plan and Advance Care Plan. In the case referred to this resulted in a 3 week delay to discharge

• there is a lack of available nursing home or hospice places which prevented some discharges to a preferred place of death.

Fast track not possible at weekends

A lack of weekend processing of fast track applications results in delays in discharge to preferred or more appropriate place of death. See below:

Case 1

Recognition of worsening dementia and move to palliative. Fast Track discharge [date - a week later] with IMCA and

Care home agreement noted on [date]. Transfer not arranged [Sunday] as "Discharge team felt too complex to sort

on a Sunday". Discharged [Monday] to care home

Case 2

Two days later discharge team note that he meets the criteria for fast track and doctor and palliative care nurse talk

to family. Again, family say they want to take him home. However, this is now Friday and fast track won't be able to

respond until after the weekend. Notes say cannot be discharged unless supported by Hospice at Home and SPA

but they have no capacity today or over the weekend.

Case 3

Discharge planning in place once prognosis conversations held and package of care being planned for day time but

night time care will not be covered

Patient will have care at home but bed bound and no night time cover given state of patient poor this is poor

Home assessment not completed which would have identified equipment problems

No full community assessment in place to ensure that MacMillan involved to provide psychological support to patient

and husband

Commencement of fast track application delayed over weekend

In at least 8 cases fast track was being planned/in place but the patient deteriorated and died before discharge – sometimes due to:

• a lack of hospice or nursing home bed • lack of advance care planning • funding problems.

99

Lack of nursing home/hospice beds

A lack of nursing home beds or hospice places results in an inability to discharge to preferred place of death. See below:

Case 1

The patient remained in hospital and did not get discharged. The discharge team visited and did do a fast track

referral but no beds were available at the hospice.

Case 2

Discharge was planned for fast track to a nursing home. Patient was seen by the discharge co-ordinator and a place

was arranged, but patient died before discharge. However, because of his other medical problems, the patient could

not have been discharged earlier.

Case 3

Referred to palliative care team and discharged planned for 2 days (fast track). However, this was delayed as she

experienced chest pain and a chest infection. On [dates] patient still keen to get home. OT had arranged equipment.

Patient died [next day] before she could get home.

Case 4

Discharge to NH discussed on [date]. Residential home he previously lived in no longer had a room. [4 days later]

not eating or drinking at all. Wants to fade away. Notes say on end of life care pathway. Doctor met with son and

agreed fast track discharge to Nursing Home. However, this was not put in place in time and the patient died on the

ward 8 days later.

Lack of Advance Care planning

Advance care planning was not in place for some fast track discharges. See below:

Case 1

There was a lot of effort expended in not achieving a great deal - e.g. fast track applications, OT and home

assessments which could have been reduced had a clear advance care planning process been in place

Case 2

Patient discharged to nursing home 4 days before he dies - sent with DNACPR and emergency health care plan.

No advance care plan sent with patient. System One demonstrated a letter from hospital to DNs … with anticipatory

EOL medications but although DNACPR in place the absence of ACP was noted on admission to nursing home (

their admission form)

Funding problems

Funding problems prevented some discharges or caused delays in last days of life. This included difficulties with criteria for funding and judgements on eligibility for CHC funding. All patients in this cohort however died in hospital or within 30 days of discharge. See examples below:

100

Case 1 (did not get discharged)

… Not discharged today 'due to funding problems'…

Case 2 (died 8 days after discharge)

However, the funding for the continuing health care package would not be released by the CCG until a DNACPR

had been put in place. The decision to fast track this gentleman was taken on [date] and a DNACPR was finally

discussed with the family and patient and completed [8 days later]. The patient was then finally discharged on [12

days later]. This was 20 days after the decision was taken to discharge. The questions here are why did the ward

not know the CCG rules about continuing health care? Why did it take eight days to implement a DNACPR? In

addition, why did it then take the CCG another 12 days to put a fast track package of care in place?

Case 3 (did not get discharged)

Discharge planning in place during admission with OT attempts to organise package of care. Some limitations though

to speed and availability of package of care as husband struggling to care for her and one morning visit to wash and

dress not sufficient

CHC checklist not met and so not funding available for additional care either

Discharge phase – characteristics of good/excellent care

Excellent and good care (24 cases) was characterised by:

• Respecting patient’s wishes • Preferred place of death • Planned discharge processes occurring smoothly

Excellent care example

Prompt CHC assessment with early discharge planning that was effective and promptly executed by OT and home

services in line with family and patient wishes. GP discharge letter and all notifications to community nursing

prompt

Good care examples

Case 1

Good quality of care leading up to discharge to community. Early discussions with family, efforts made to keep

patient comfortable re: SALT, end of life care discussed. No end of life care plan put in place in acute hospital

however there was a clear community advance care plan in place. Could have been discharged sooner - although

discharge planning began on admission, it took seven days for the patient to get a nursing home bed although the

nursing home he was discharged to was the preferred place of death if home death not possible as indicated 7

months previously in discussion with family.

Case 2

Discharge was organised a day earlier than plan because of the patient's rapid deterioration. There were

discussions with the wife about delays to access to carers. The wife was happy to bridge this gap so that the

101

patient could come home. The ward did all they could to ensure the patient was discharged by ambulance that

night and the patient was discharged later in the evening. Oxygen and the bed were in place at discharge.

Case 3

Discharge was well planned with hospice at home in place. The lady went home with a syringe driver. The DN

team was informed.

Case 4

The patient's family decided to take him home without the package of care in place (OT was due to visit that day).

This was against advice but it was agreed that the OT would visit the next day instead. This was flexible and

responding to patients' and relatives' needs. The patient's safety was assured, with family feeling able to cope with

his care until he was assessed by OT and nursing teams following day.

Case example 5

Patient was discharged home after an admission in March 2017. On this occasion good plans were made. The

specialist pulmonary care nurse referred to the nursing and OT teams. This in turn ensured that the necessary

equipment; mattresses and end of life drugs were prescribed. She also referred to social services for support for

the family. The family reported that they were happy that the gentleman was peaceful and not in discomfort.

Discharge phase – points to note in cases of poor/v ery poor care

These points relate to 10 cases in total

• No record of discussion with family • Patient wishes were not respected • Lack of care coordination on discharge • No GP follow up post discharge • Therapy services not in place • Inappropriate discharge – ongoing infection and swallowing problems • No Advance Care Planning • Funding problems • Family unable to cope on discharge • No night time PoC/insufficient package of care • Fast track delayed over weekend • Delay in finding care home bed/transport/late referral to discharge team/late discharge

planning/restart if not discharged • Use of fax or letters by post to communicate rather than electronic means • One particular case highlighted an issue about the provision of TPN support. It would

appear that District Nurses can’t provide TPN in rehabilitation wards but are able to do so in a nursing home. This should be examined further.

102

4.9 End of Life Care

Number assessed and ratings for End of Life phase o f care

There were 155 End of Life ratings. 83% were adequate, good or excellent. 26 cases (17%) were rated as poor or very poor.

Chart 40 End of Life Care ratings

5

21

44

70

16

25

0

10

20

30

40

50

60

70

80

Very Poor Care Poor Care Adequate Care Good Care Excellent Care Not Applicable

Nu

mb

er

of

De

ath

s

Care Rating

End of life care (n=155)

103

Notable Themes

Engagement with families regarding/at End of Life Talking to the patient about prognosis and families to consider implementing End of Life plans was apparent throughout and a very positive aspect overall.

Lack of, or late, discussion with a patient and the family was apparent at times. Four examples are given below:

Case 1 End of life discussion appeared to be held late with the family so notes only indicate discussion a few hours before death. Discussing palliative care review an hour before which the reviewer notes appears very late in the admission. No discussion about advance care planning evident despite DNACPR being signed 5 days pre-death. Case 2 Although the palliative care team offered a great deal of support both in the community and in the hospital, there is no evidence of an advance care plan or a discussion with the patient and his relatives about his end of life care choices until very late on. Case 3 Patient had a DNACPR in her notes, which was signed during her previous admission. End of life care planning was absent. The notes do not indicate any discussions with the patient herself - other than telling her she needed to be nursed in a side room due to her blood results. Although palliative care and end of life care are noted, these are not followed by any clear plan. No referral to palliative care team.

End of Life for patients with confusion/memory prob lems Of 76 cases where confusion/memory problems were present 67% were rated as good or excellent compared with 77% of cases without confusion/memory problems. 60% (16) of very poor/poor ratings were for patients who had confusion/memory problems.

Of 10 cases of deaths that occurred in hospice – 9 cases involved malignancy; 1 COPD. Two cases had confusion or memory problems. There is a lack of hospice provision which may also make it even more difficult for palliative care provision in hospices for the more complex (confused) patient.

Death certification/involvement of Medical Examiner (ME) There was clear evidence that death certification is discussed with the Medical Examiner and clearly documented as such in the notes in a significant number of cases. In one case this process picked up an unlikely contributory factor and the death certificate was re-issued.

In one case the reviewer questioned the cause of death as being incorrect and in another not in line with ONS best practice.

Conversations post death with families – on two occasions reviewers noted that the ME or a Consultant had spent time answering questions post death from families.

104

We noted that death certification did not always reflect the extent of some conditions when analysed as a whole. We were provided with the cause of death information on 114 of the reviewed UHL cases (63%).

In the case review, for example, Type 2 diabetes was noted as part of the patient’s condition in 23 cases. Of those 23 cases, 19 had a cause of death logged and Type 2 Diabetes/Diabetes/Diabetes Mellitus was referred to on Part 2 of the death certificate in only 8 cases. When looking at the whole reviewed cohort it was referenced only 10 times on the death certificate.

We did a similar analysis for the prevalence of Cancer. In at least 64 cases (56%), Cancer (using various terms) was clearly identified in the reviewers’ narrative as part of the condition of the presenting patient. 34 patients had Cause of Death information. Of these 30 matched and the remaining 4 appeared to be historical cases of Cancer. Cancer appears to be reflected accurately on death certificates.

If examining the cause of death, contributing factors or the prevalence of a condition, then the death certificate information may under represent some major conditions.

Organ Donation We noted there was only one reference to Organ Donation in the reviewed cohort. Unfortunately it was declined by the recipient centre but this may be an area for further work.

GP change at end of life A change of GP is enforced on many patients when being discharged to a care home – as these patients have all died within 30 days of discharge few of them were therefore known to their new GP some having had long standing relationships in primary care beforehand.

We noted at least 8 cases where the GP practice changed as a result of a move to a care home on discharge. Given all the patients in the cohort died within 30 days of discharge the impact of a GP change at end of life should be examined further. Whilst we comment on some good handover practice on GP change there was also a theme of limited/no primary care involvement in last weeks or days of life. Some examples are shown below:

Case 1 – continuity from new GP

Patient with dementia cared for attentively in primary care with overall care co-ordinated by one practice partner

clearly with good, trusting, relationship with family. Patient regularly seen and bloods monitored and responded to

when needed. Respite care provided, day centre care provided and package of care provided to support wife

caring for him at home. Advance care planning commenced 7 months prior to death and this was revisited several

times including by the new GP on transfer to nursing home. Named Nursing home was second choice of preferred

place of death and wishes respected. Good coordination with CMHT. Best interests assessments completed with

family.

Case 2 – out of catchment area for GP

96 year old lady living alone with a carer once a day. Previous admission to UHL for a fractured pelvis in April and

discharged to [Community] Hospital LPT for Rehab. Home assessment requested the same day as discharge and

this never happened. Patient called to chase therapy to help her mobilise at home but put on waiting list. Therapist

do not appear to have visited at all. DN requests to change Butrans patch and check BP - sparse evidence of

visiting. Patient also appears to have told the GP she was at home following discharge from [Community] Hospital.

105

Develops a chest infection 2 weeks later and repeat infection a month later. Readmitted after referral from GP for

this chest infection that wasn't responding to antibiotics.

The reviewer is concerned that the lack of community rehabilitation reduced the ability of this lady to successfully

mobilise and reduce the risk of chest infections developing.

Her short stay in a residential home out of area may have disrupted the medical oversight - but this was difficult to

ascertain from the records as she was out of catchment area for the GP practice.

Case 3 – change in GP on admission to nursing home/o nly seen by OOH

Recent discharge from [an out of area acute] Hospitals and change of GP when moved to nursing care home in

[town] following a 2 month stay in Stroke Unit at [town] following Ischaemic stroke care. Had been well with sudden

change in condition.

OOH doctors called to care home [midweek lunchtime] which the reviewer notes as strange mid-week - but could

be a lunch time closure of the GP practice

New patient letter sent 2 weeks prior due to change in GP, offered a health care assistant check. Reviewer raises

whether a protocol for ensuring new patients are seen on admission to Nursing Care should be in place. As a letter

sent to a patient with dementia may not be responded to if sent to a Nursing Home.

Case 4 – change of GP but lack of time to organise l ack of ACP

Very elderly [patient] with bilateral amputations some months pre death and a recent history of cancer….Patient

readmitted to hospital with infection and sepsis. Discharged after 1 month to nursing home. Changed GP on

change of residence. Shortly after admission to new nursing home deteriorated further and records indicate that

patientacquired a further chest infection in hospital and may have been discharged prematurely. New GPs were

prompt in seeing patient in care home and wanted to readmit given condition. Some evidence that previous GPs

had considered anticipatory care plan and admission avoidance schemes in the past but not indicated here.

DNACPR in place from previous admission but reason given was for reasons which were not the presenting

symptoms on readmission. No End of life care plan in place and new GP did not have sufficient time from

discharge to arrange Advance Care Plan.

Case 5 – lack of GP review on admission to new nursi ng home

Discharged from Glenfield after pacemaker inserted one month before death to Rutland Community Hospital and

then quickly into [name] Care Home. Advanced dementia and on delirium pathway.

The GP changed as a result of the patient’s new residence. The care home did not request a medical review for

their new patient and it was 2 weeks before GP review of a new patient [resulting from a] request for medications.

Started to develop a Urinary Tract Infection and given antibiotics but deteriorated.

Care home not proactive in seeking management plan and advance care planning for very frail patient.

Patient admitted with Pressure sores which initially started at Glenfield and then developed worse to Grade 4 at

Rutland Community Hospital.

DOLS applied for - not authorized pre death by Local Authority. Had only been in care home for three weeks.

Between discharge from Glenfield and readmission in ED the patient had lost 13kg in weight.

I

Clinical provision at end of life There were a group of cases which highlighted the decision making for clinical care at end of life. One case highlighted the need for clarity over withdrawing fluids at EoL when oral fluids could not be administered.

Equally there was a need to balance decisions for comfort care with end of life – this included:

• whether aspiration risk was of concern at this stage and withdrawing oral fluids as a result

106

• mouth care appeared to be variable when oral fluids were not being administered • the extent of disturbing patients with nursing observations when escalation was not

appropriate • clarity over blood sugar monitoring at end of life • best interest decisions e.g. a Central line being inserted

A specific issue was that medication reviews may not be happening well enough at End of Life.

DOLS DOLS applications were being completed in line with the MCA either in place in advance of admission or an urgent authorisation being applied. However, it was not apparent whether full authorisation was being received by the Local Authority in a timely manner. In one case DoLs reporting to the Coroner appeared not to have been in line with guidance.

DNACPR and EoL plans In the 68 cases rated with very poor care to adequate end of life care there were 9 cases (13%) of no DNACPR in place/in place in a timely fashion, 20 cases with no clear End of Life plan and 15 cases of no evidence of Advance Care Planning. These were more predominant factors in the poor and very poor ratings.

Whilst there is a focus on getting a DNACPR in place with the patient or family in the large majority of cases there are instances of this being too late in the day. There were 3 cases where CPR was commenced (2 with a DNACPR and 1 where not DNACPR had been considered but not discussed with family)

End of Life planning

There were compassionate and well documented discussions with patients and families when it was considered appropriate to move to end of life care. In some End stage Liver disease cases this was particularly evident. There were positive moves for people at end of life with engagement with the palliative care teams appropriately.

However, end of life care planning was not always put in place. Some examples are given below:

Case examples – End of Life Care Planning Case 1 Family was communicated with well on the whole, the DNACPR and discussions about prognosis. Family was listened to in terms of concern about relative getting back home. However, more clarity on options and likely prognosis could have been improved as the patient had conflicted messages on prognosis as investigations still planned despite terminal situation. No psychological support to family or patient whilst in hospital. Given this patient was only diagnosed 4 weeks before death more could have been done to explain the options. No EOL care plan in place so that the conversations were unstructured and reactive. Case 2 No end of life care plan in place until choking episode at weekend and readmission to LRI. Reviewer questions is there an end of life plan for people with end stage liver disease? Discussion with patient re: DNACPR notes that he wants full resuscitation. Not clear if this was reviewed with him or whether he then lacked capacity.

107

Case 3 DNACPR considered but not discussed with patient or relatives so attempts at resuscitation made and failed when patient had arrest. No end of life care plan in place or discussed. Patient on IV fluids throughout but appears to have eaten very little for a week and no dietician seen so not at risk of aspiration

Advance Care Planning

There were a small number of clear, well-structured advance care plans apparent. Advance Care Plans could not be identified in the records in some cases. Some examples of missed opportunities for advance care planning are described below:

Case examples – no Advance Care Plan – missed opport unities

Case 1

No ACP ahead of admission and there had been a number of opportunities over recent months to consider this

when as she became increasingly frail with acute episodes or respiratory infection/ cholecystitis

End of life care provided in hospital for final days was adequate but over last months of life not proactive,

personalised or systematic

Case 2

There was no ACP evident including DNACPR/preferred place of death despite clear documentation that the patient

had mounting frailty and his wife had talked to the community team about increasing difficulties they were having.

Case 3

DNACPR was put in place when patient had been in earlier in the year with a gastric bleed. Discussion with patient

documented at the time. Patient was depressed in the last days of his life (GP prescribed antidepressants) and said

he wanted to die. No ACP on file and no record of a discussion about his needs and wishes.

End of Life - characteristics of good/excellent car e

Excellent care included:

• Fast track used and implemented promptly • Patients on GSF and regular engagement from GP with patient and family • LOROS involvement as preferred place of death • Respecting patient wishes • End of Life recognised • Clinical staff spending time with patients • Preferred place of death possible • Personalised care plans

Good care characterised by:

• Early conversation in line with GSF • Good GP engagement – regular contact with family and patient • ACP in place • Spiritual care provided

108

• GP visiting at home • District Nurse care coordination • Helping family and relatives to manage at home • Handover from GP to new GP on changed residence • LOROS involvement

Excellent care examples

Case 1

Excellent end of life management in acute hospital. Doctors and family discussed the developing situation and

likely prognosis in a timely manner. Clear plan if patient deteriorated and when this happened clear actions

taken on EoL in line with expected EoL care. MacMillan and Palliative care team involved, visited on ward.

Side room found for EoL. Considered the best needs of patients in deciding to keep on the ward until final

stages, Daughter present and treated with respect in final hours. Discussion with Medical Examiner on cause

of death documented.

Case 2

Discussions with patient about his wishes and concerns recorded and all attempts made to discharge him

home to die according to his wishes. This included OT involvement and sensitive handling of his wants to use

a commode but being too weak to do so the OT let him try and make the decision for himself). Fast track care

plan includes advance care plan. Hospice at Home put in place for support and package of care to include 4

visits per day. This took 1 day from completion of fast track form to discharge.

Contingency put in place while awaiting fast track - doctor notes that end of life care plan should be put in

place in hospital if he deteriorates before they can get him home.

Case 3

Excellent support for care home from GP and community nursing team - home visit very quickly after

discharge, medication review and amendment of tablets to liquids, equipment checked, daily contact. Peaceful

death.

Good care example – change of GP

On change of GP at admission to Nursing home the new GP reviewed the Advance Care Plan and assessed

patient on admission so there was good continuity of care from a situation where the patient had been looked

after well by a long standing GP and the new GP continued on.

Good care examples – Preferred Place of Death

Case 1 End of life care commenced by GP at home. Seen by palliative care team daily ... Discussed end of life care with patient and wife. Decisions and wishes documented. Patient wishes to die in hospital. Plan - to discharge to LOROS or Loughborough Community Hospital, which was closest to patient's home. Fast track discharge put in place. Individualised end of life plan put in place and updated daily. Advance care plan put in place by GP service.

109

In community hospital, comprehensive verbal handover sheets completed daily in community hospital, which

clearly state that patient is for end of life care. Describe patient's ability to use call bell, skin integrity,

communication, elimination, nutrition, emotional wellbeing and meds. End of life care medications noted as

prescribed but patient not having symptoms yet. Care in the last days of life form completed, which notes

location of DNACPR.

Case 2

Patient deteriorated and had daily input from the DN team and lots of input from the GP. The carers

telephoned the team when the lady was in pain and the DN team responded well. A syringe driver was put in

place. A mattress was ordered when the lady couldn't get up any more. She died at home in her own bed

with the carers and her neighbours. A good end of life experience.

Case 3 Full discussions throughout with family as patients prognosis changed and agreement with active management plan discussed. DNACPR agreed. End of Life medications provided and discussions with family face to face during the night with medical team when deterioration apparent.

End of Life – characteristics of poor/very poor car e

Poor Care

• DNACPR in place too late • No Advance Care Plan • Lack of EoL Plan or implementation of plan • Change in GP when transferred to Nursing Home (through a lack of communication, or

lack of involvement of primary care with a change in residence) • LOROS – confusion on fast track funding if death not imminent • Lack of weekend medical care/lack of community facilities at Community Hospital • Anticipatory medications not given • No psychological support/MacMillan/palliative care in sudden/recent diagnosis of illness

with imminent terminal prognosis • 121 not possible due to staff shortages/bed pressures in oncology • Lack of access to specialist help – no beds at hospice, delays for MacMillan

support/incorrect MacMillan paperwork • Delays – syringe drive/fast track delay • Inappropriate monitoring once at EoL stage – e.g. disturbed during the night for

observations or confused observation regime but not needing to respond to EWS • Good care in the wrong place • Death not noticed

Very Poor Care

There were 5 cases of very poor end of life ratings. These cases, whilst individual in their own rights, were poorly managed in terms of recognition and planning for end of life. This included issues relating to DNACPR, talking to families, EoL plans and ACPs.

• 2 cases where resuscitation was attempted

110

• In both resuscitation cases the DNACPR had been either considered and not discussed with patient/family or not discussed. In a further case lack of engagement with family was raised via Datix

• A case where a patient arrested twice and DNACPR not in place first time and then not put in place before the second time.

• Lack of explicit recognition of end of life and so end of life plans not in place and advance care plans not considered.

• In one case retrospective entries in nursing and medical records • Poor monitoring e.g. EWS, blood sugars, fluid balance/nutrition/lack of medical review

111

Appendices (separate document)

Appendix 1 – Methodology and Approach

Appendix 2 – Review Form

112

This report ("Report') was prepared by Mazars LLP at the request of West Leicestershire Clinical

Commissioning Group for and on behalf itself and of East Leicestershire & Rutland Clinical

Commissioning Group, Leicester City Clinical Commissioning Group ("Authority'), University Hospitals of

Leicester and Leicestershire Partnership NHS Trust and terms for the preparation and scope of the

Report have been agreed with them. The Report was prepared solely for the use and benefit of the

Authority and to the fullest extent permitted by law Mazars LLP accepts no responsibility and disclaims

all liability to any third party who purports to use or rely for any reason whatsoever on the Report, its

contents, conclusions, any extract, reinterpretation, amendment and/or modification. Accordingly, any reliance placed on the Report its contents, conclusions, any extract, reinterpretation, amendment and/or modification by any third party is entirely at its own risk.

Mazars LLP is the UK firm of Mazars, an international advisory and accountancy organisation, and is a limited

liability partnership registered in England with registered number OC308299. A list of partners’ names is

available for inspection at the firm’s registered office, Tower Bridge House, St Katharine’s Way, London E1W

1DD.

Registered to carry on audit work in the UK by the Institute of Chartered Accountants in England and Wales.

Details about our audit registration can be viewed at www.auditregister.org.uk under reference number

C001139861.

© Mazars 2018