Quality of ethical guidelines and of ethical content in...

1/24

AGREE-Ethics. A working model(The AGREE instrument modified for the application on ethical guidelines)

SCOPE AND PURPOSE

1. The overall objective(s) of the guideline is (are) specifically describedGuidelines on ethically relevant questions are influential regarding the behaviour of healthcare professionals and have an impact on the health of patients covered by the respective guidance. The overall objective(s) of the guideline with respect to ethically relevant aspects of clinical practices should be described in detail.and the expected benefit from the guideline should be specific to the clinical-ethical problem. For example specific statements would be:• Support of healthcare providers with respect to end of life decision making in intensive care • Assistance of physicians with respect to disclosing bad news to patient with a life limiting disease

The statement should be answered with “1”, “strongly disagree”, either if there is unclarity about the purpose of the guideline or if no objectives have been named.

A rating of “2” requires the naming of objectives in general (e.g. optimization of care at the level of process quality, structural quality or quality of results, support of doctors, patients or family members in dealing with certain illnesses and situations; assistance in the training of those beginning their careers or seeking continuing education).

A rating of “3” requires the naming of specific objectives For the rating “4”, “strongly agree”, the objectives of the guideline should be distinguished from related

objectives which are not pursued by this guideline or the reasons why a further specification of the objectives would not be productive should be explained

Examples for a 4 rating

Example: Renal Physicians Association and American Society of Nephrology (1999). Clinical Practice Guideline on Shared Decision-Making in the Appropriate Initiation of and Withdrawal from Dialysis.

See A GUIDE TO THIS DOCUMENT (p. 5): “This document, Clinical Practice Guideline on Shared Decision-Making in the Appropriate Initiation of and Withdrawal from Dialysis, addresses recommendations concerning withholding or withdrawing dialysis in adult patients with either acute renal failure (ARF) or end-stage renal disease (ESRD).”

See also SECTION 2: BACKGROUND. Scope and Intended Use (p. 13.14): […] The guideline provides recommendations on the targeting, timing, and content of discussions related to either withholding or withdrawing dialysis, and the care of patients who forgo dialysis. The guideline also provides recommendations on when it is appropriate to withhold or withdraw dialysis. The recommendations are not mandatory, but rather flexible guides that warrant tailoring based on particular patient, provider, and geographic circumstances. They allow the renal care team discretion in their application to individual patients.

See also SECTION 2: BACKGROUND. Guideline Objectives (p. 14): The following are the objectives of this guideline:

Synthesize available research evidence on patients with ARF and ESRD as a basis for making recommendations about withholding and withdrawing dialysis;

Enhance understanding of the principles and processes useful for and involved in making decisions to withhold or withdraw dialysis;

Promote ethically as well as medically sound decision-making in individual cases; Recommend tools that can be used to promote shared decision-making in the care of patients with ARF

or ESRD; and Offer a publicly understandable and acceptable ethical framework for shared decision-making among

health care providers, patients, and their families.

2/24

2. The ethical question(s) covered by the guideline is (are) specifically described A detailed description of the ethically relevant questions covered by the guideline should be provided, particularly for the key recommendations (see item 17). Following the examples provided in question 1:• Which are the ethically relevant aspects of end of life decision making in intensive care? • Which are the ethical and communication challenges when disclosing bad news to patients with life limiting diseases?

The statement should be answered with “1”, “strongly disagree”, if the guideline does not specify any ethically relevant questions/problems underlying the key recommendations.

A rating of “2” means that the guideline only specifies underlying ethically relevant questions/problems for some of the key recommendations.

A rating of “3” requires specification of underlying ethically relevant questions/problems for most key recommendations.

A rating of “4”, “strongly agree”, requires that the guideline specifies the underlying ethically relevant questions/problems for all key recommendations. The key recommendations must have a clear (conceptual) connection to the background information.



See SECTION 1: FOREWORD AND RATIONALE FOR GUIDELINE. Ethical Considerations in Dialysis Decision-Making (p. 9-12). This chapter specifies four dimensions of the ethical problem: medical indications, patient preferences, quality of life, and contextual features, that is, the social, economic, legal, and administrative context in which the decision occurs.

See also SECTION 2: BACKGROUND. Scope and Intended Use (p. 13): Shared decision-making--the process by which physicians and patients agree on a specific course of action based on a common understanding of the treatment goals and risks and benefits of the chosen course compared with reasonable alternatives--is recommended. Shared decision-making recognizes the importance of both patient preferences and medical indications. In shared- decision-making, the health care provider is the expert in diagnosis, prognosis, and treatment alternatives, and the patient is the expert in his or her own history, values, preferences, and goals. The two work together to reach decisions that are individualized to the patient’s particular circumstances and preferences. There are limits, however, to the shared decision-making process that protect the rights of patients and the professional integrity of health care professionals. The patient has the right to refuse dialysis even if the renal care team disagrees with the patient’s decision and wants the patient to undergo dialysis. Similarly, the renal care team has the right to refuse to offer dialysis when the expected benefits do not justify the risks. Recognizing that there are circumstances in which patients and renal care teams might disagree about decisions to start, continue, or stop dialysis, this guideline provides recommendations for how to resolve such conflicts.

3/24

3. The patients/other relevant group to whom the guideline is meant to apply are specifically described.There should be a clear description of the target population which is covered by a guideline. The age range,sex, clinical description and other criteria relevant to the ethical questions covered by the guideline may be provided. For example:• A guideline on end-of-life decision making only includes the group of patients in the intensive care setting who lack legal capacity • A guideline on the management of disclosing bad news to adult patients for whom there is no curative treatment option available

There should be an explicit statement if the aforementioned or other characteristics are not relevant for the ethical guideline

The statement should be answered with “1”, “strongly agree”, if the target patient population has not been named or can only be indirectly inferred from the text of the guideline.

A rating of “2” requires a simple statement of the target patient population. A rating of “3” requires the additional description of the characteristics of inclusion or exclusion, such

as the relevant area of care (ambulant or stationary treatment, age, degree of severity). A rating of “4”, “strongly agree”, requires an explanation of the reasons why a further specification of

the target patient population would not be productive.


Example: American Thoracic Society (2007). Clinical Policy Statement: Palliative Care for Patients with Respiratory Diseases and Critical Illnesses

See „Executive Summary“ (p. 912): Although originally conceived and practiced as end-of-life care, palliative care may be applied to all stages of illness, whether terminal or not. Because its members commonly care for children and adults and their families in need of palliative care due to chronic respiratory diseases or critical illnesses, the American Thoracic Society (ATS) approved formation of an ad hoc End-of-Life Care Task Force whose purpose was to develop this statement that focuses on how to better integrate palliative care into standard clinical management (curative and/or restorative care).

See also „Introduction“ (p. 913): Although this statement primarily describes providing palliative care for children and adults with advanced respiratory diseases and critical illnesses, the same professional values and principles and many of the same recommendations also apply to the care of neonates and of patients in intensive care units (ICUs) due to trauma or for perioperative care.

See also: Goals, Timing, and setting for palliative care“ (p. 914): Clearly, palliative care should be available near the end of life. However, it should also be available at any point during the course of a progressive or chronic respiratory disease or critical illness when the patient becomes symptomatic. An important concept is that, in general, palliative care should be available when curative/restorative care begins, while curative/restorative care continues, after life-prolonging treatments are withheld or withdrawn, and, for the patient’s family, after the patient’s death..

4/24

STAKEHOLDER INVOLVEMENT

4. The guideline development group includes individuals from all the groups relevant to the ethical issue at stake.This item refers to all groups relevant to the ethical issue at stake who were involved at some stage of the development process.This may include members of the steering group, the research team involved in selecting andreviewing /rating the evidence and individuals involved in formulating the final recommendations.This item excludes individuals who have externally reviewed the guideline (see Item 13). Informationabout the composition, discipline and relevant expertise of the guideline development group shouldbe provided.

The statement should be answered with “1”, “strongly disagree”, if there is no reference to the composition of the guideline development group or if authors are only mentioned by name and it is not possible to classify them by professional group or function.

A rating of “2” requires, alongside the names of the authors, an unambiguous statement of the professional and interest groups that each represent.

For a rating of “3”, this statement should make it clear that all professional and patient groups addressed by the guideline were actively represented in the guideline development group (e.g., for end-of-life decisions this should generally include: doctors, patients, caregivers).

A rating of “4”, “strongly agree”, requires additional information about how those professional and interest groups not directly addressed by the guideline (e.g. patients’ family, hospital management, additional medical care professions) were involved. This can occur in the guideline itself or in the guideline report (see criterion 29), whereby the existence of a guideline report must be clearly stated in the guideline itself.


Example: Clayton M et al. (2007) Clinical practice guidelines for communicating prognosis and end-of-life issues with adults in the advanced stages of a life-limiting illness, and their caregivers

See „Methods of guideline development“, „Expert advisory panel“ (p. S85): An expert panel was convened, comprising 35 Australian and New Zealand health care professionals and consumers (nine palliative care medical specialists, three medical oncologists, three palliative care nurses, three consumers [including two cancer patients and one caregiver], two general practitioners, two psychosocial experts, two ethicists, two linguists, one oncology nurse, one cardiology nurse, one aged care nurse, one research nurse, one geriatrician, one respiratory physician, one radiation oncologist, one intensive care specialist, and one lawyer).

5/24

5. The views and preferences of patients or other groups affected by the guideline have been soughtInformation about experiences, preferences and expectations of patients or other groups affected by the guideline should inform the development of ethical guidelines. There are various methods for ensuring that these perspectives inform guideline development. For example, the development group could involve representatives of patients, relatives or other relevant groups, information could be obtained from interviews with relevant representatives, literature reviews on relevant socio-empirical research could be considered by the group. There should be evidence that on eliciting preferences and views by patients or other groups affected by the guideline either within the guideline or as part of a separate report on the production of the guideline.

The statement should be answered with “1”, “strongly agree”, if patients or other groups affected by the guideline were not involved or if how and to what extent they were involved can only be guessed at from the statements (not) made.

A rating of “2” requires that representatives of patients or other groups affected by the guideline were indirectly involved (e.g. through a review procedure, with reference to empirical social research in certain key recommendations, or a public discussion of the guideline including modification of its content prior to its going into effect).

A rating of “3” requires direct participation in developing the guideline and reference to empirical social research for all key recommendations. If the relevant external evidence is not available, this must be explicitly explained.

A rating of “4”, “strongly agree”, requires an explanation of how the selection of patient or other groups involved is representative as well as a systematic search of the external evidence. This can occur in the guideline itself or in the guideline report, whereby the existence of a guideline report must be clearly stated in the guideline itself.



With respect to the explanation of how the selection of the patient groups involved is representative see „Methods of guideline development“, „Expert advisory panel“ (p. S85): The panel members were selected to reflect the multidisciplinary nature of care for patients with advanced progressive life-limiting illnesses. Representatives were selected on the basis of either their clinical expertise or a track record of publications on this topic. Consumer bodies, such as Carers NSW and Cancer Voices NSW, were approached and provided consumer representation.

With respect to the systematic search of the external evidence see „Methods of guideline development“ (p. S85): The guidelines were developed using the following process: Literature review: Systematic review of the best available evidence regarding discussion of prognosis and end-of-life issues with patients with a progressive life-limiting illness, and their caregivers; and review of previous related guidelines and expert opinions.

For further information on the systematic search see: „Methods of guideline development“, „Literature review“ (p. S85)

6/24

6. The target users of the guideline are clearly defined.The target users should be clearly defined in the guideline, so they can immediately determine if theguideline is relevant to them. For example, the target users for a guideline on end-of-life decision making in intensive care include physicians, nurses, other healthcare professionals, clinical ethicists and representatives of different congregations.

The statement should be answered with “1”, “strongly disagree”, if the target users have not been defined.

A rating of “2” requires that the target users are clear from the context of the guideline or can be unambiguously inferred.

A rating of “3” requires clear mention of the target users. A rating of “4”, “strongly agree”, requires in addition reference to further (indirect) users (contact

partners) (e.g. users at insurance agencies) or an explanation of the reasons why a further specification of the target users would not be productive.



See „Methods of guideline development“, „Target audience“ (p. S86): These guidelines were developed for use by all members of the health care team involved in the care of patients with progressive life-limiting illnesses, and their families, including GPs, specialist doctors and trainees (including palliative care, medical and radiation oncology, surgical oncology, geriatrics, haematology, respiratory medicine, cardiology, and neurology), junior medical staff, nurses, social workers, psychologists, psychiatrists, physiotherapists and occupational therapists. Some health care professionals (such as junior medical staff, generalist nurses and some allied health staff) may not feel it is their role to be giving detailed prognostic or end-of-life information to patients. However, the general principles in these guidelines are relevant for all health care professionals interacting with this patient population and their families at home or in hospital.

7/24

7. The guideline has been piloted among target users.A guideline should have been pre-tested for further validation amongst its intended end users prior topublication. For example, a guideline may have been piloted in one or several primary care practicesor hospitals. This process should be documented.

The statement should be answered with “1”, “strongly disagree”, if there was no pilot study or no information about any pilot study is provided.

A rating of “2” requires that the guideline refer to a pilot procedure. A rating of “3” requires that the guideline contain information about the procedure for piloting the

guideline (e.g. surveys about its applicability, test application in practice). A rating of “4”, “strongly agree”, requires in addition an explanation of the piloting procedure and

results. This can occur in the guideline itself or in the guideline report, whereby the existence of a guideline (see criterion 29) report must be clearly stated in the guideline itself.

Examples for a 3 rating (no assessed guideline was 4 rated)

Example: Verlato G et al. (2004) Guidelines for resuscitation in the delivery room of extremely preterm infants

See „Methods“ (p. 32): Drafts of the guidelines were distributed to all participants and to all birth centers of the Veneto region. They were later discussed and approved at a plenary meeting.

8/24

RIGOUR OF DEVELOPMENT

8. Systematic methods were used to search for evidence.Details of the strategy used to search for evidence should be provided including search terms used,sources consulted and dates of the literature covered. Sources may include electronic databases(e.g. MEDLINE, EMBASE, CINAHL), databases of systematic reviews (e.g. the Cochrane Library, DARE),handsearching journals, reviewing conference proceedings and other guidelines (e.g. the US NationalGuideline Clearinghouse, the German Guidelines Clearinghouse).

The statement should be answered with “1”, “strongly disagree”, if the guideline contains no mention of any search for evidence.

A rating of “2” requires that a description of method in the guideline at least make it clear that some systematic procedure was used.

A rating of “3” requires that the guideline indicate the methods of search, including the search terms and databases used.

A rating of “4”, “strongly agree”, requires a complete description including the search terms and databases used and if applicable any further sources as well as the results (at least number of hits) and time of search. Any limitations made to the search must be made explicit. This can occur in the guideline itself or in the guideline report (see criterion 29), whereby the existence of a guideline report must be clearly stated in the guideline itself. The description in the guideline report must refer individually to the guideline in question; a description of the general principle is not sufficient.



See „Methods of guideline development“, „Systematic review“ (p. S85): The systematic review was conducted in conjunction with personnel of the systematic review section of the National Health and Medical Research Council (NHMRC) Clinical Trials Centre, in accordance with the principles and processes recommended by the Cochrane Review.4 Eligible studies were those that sampled adult patients who had an advanced progressive life-limiting illness with less than 2 years to live, including but not restricted to cancer, endstage pulmonary disease, end-stage cardiac failure, and motor neurone disease; and/or the caregivers (including bereaved relatives) of such patients; and qualified health care professionals. More specifically for studies where patients formed the study group, the eligibility criteria included: patients in specialist palliative care or hospice settings (inpatient or community), and patients in any care setting with a diagnosis of an advanced lifelimiting illness being treated with palliative intent and with a life expectancy of less than 2 years. Where samples were mixed (ie, potentially curable and advanced disease), studies were included if more than 50% of patients had an advanced life-limiting illness or where the results for such patients were provided separately. Studies where patients formed the study group were excluded if they involved patients with chronic diseases with likely life expectancy of greater than 2 years (such as dementia and multiple sclerosis), patients being treated with curative intent, or patients being cared for in an intensive care unit, or if it was not possible to determine what percentage of the study group was being treated with palliative intent or had a life expectancy of less than 2 years. The medical literature was searched to identify relevant studies and reviews for the period between 1966 and November 2004. Searches were conducted via MEDLINE, CINAHL, EMBASE, Psycinfo and the Cochrane Register of Controlled trials (Central). Out of 4167 abstracts retrieved by electronic searches, and a further 127 abstracts identified by hand searching, 278 full manuscripts were considered for inclusion in the review. One hundred and fifty-one did not meet the eligibility criteria and four were duplicates; hence, 123 articles were included. Individual reviewers extracted data from each study using a standard format. Data extraction was checked by a second reviewer for 10% of studies to ensure consistency. Most studies were descriptive, examining provider practices and patient and/or caregiver views, attitudes, knowledge and behaviour (Level IVa evidence according to the NHMRC 1995 ratings5). Such studies are not considered admissible evidence in the more recent NMHRC rating scale.6 These NHMRC rating scales are designed for research questions regarding interventions, diagnosis, aetiology, screening and prognostic factors, but are not readily applicable to the topic of these guidelines.

Note: The guideline provides an outlink to the specific systematic review: Parker S, Clayton JM, Hancock K, et al. Communication of prognosis and issues surrounding end of life (EOL) in adults in the advanced stages of a life-limiting illness: a systematic review. Sydney: University of Sydney, NHMRC Clinical Trials Centre, 2006. http://www.ctc.usyd.edu.au/ research/publications/articles_2006.htm

http://www.ctc.usyd.edu.au/

9/24

9. The criteria for selecting the evidence are clearly described.Criteria for including /excluding evidence identified by the search should be provided. These criteriashould be explicitly described and reasons for including and excluding evidence should be clearly stated.For example, guideline authors may decide to only include evidence from randomised clinical trials or from specific research designs for surveys and qualitative research. Guideline authors may also decide to exclude articles not written in English.

The statement should be answered with “1”, “strongly disagree”, if the guideline provides no criteria for the selection of evidence.

A rating of “2” required that the guideline provide basic information about the selection of evidence (e.g. only English-language or German-language empirical social research, only quantitative research, only research in MEDLINE, along with indications of explicit criteria for the inclusion and exclusion of studies).

A rating of “3” requires that the guideline contain information about the selection of evidence for key recommendations and describe the reasons for the selection.

A rating of “4”, “strongly agree”, requires that the guideline contain detailed information about the criteria for the selection of evidence for all recommendations. This includes information about the reasons for consideration of evidence as well as for rejection of evidence. A general reason (e.g. “considered/not considered due to the requirements of general practitioner care”) is not sufficient. This can occur in the guideline itself or in the guideline report (see criterion 29), whereby the existence of a guideline report must be clearly stated in the guideline itself. The description in the guideline report must refer individually to the guideline in question; a description of the general principle is not sufficient



See „Selection of Relevant Research Evidence“ (p. 16): Several types of information were deemed relevant to the key questions (see Selection Criteria, Table 2). For information about prognosis in patients with ESRD, large retrospective or prospective cohort studies with at least 100 patients that examined multivariate predictors of mortality or morbidity were selected. For information about prognosis in patients with ARF, smaller retrospective or prospective studies involving at least 20 dialysis patients and reporting mortality outcomes were used. Information relevant to who gets referred for dialysis and when, feasibility, withdrawal frequencies and reasons, patient preferences, shared decision-making, advance directives, and quality of life assessments was taken from descriptive surveys, case-control studies, cohort studies, or randomized trials with at least 20 patients who were receiving or awaiting dialysis. Research evidence from Asian and developing countries was not used because differences in access to dialysis, patients’ values and preferences, and decision-making processes were considered likely to limit generalizability and applicability to patients in the United States.

10/24

10. The methods used for formulating the recommendations are clearly described.There should be a description of the methods used to formulate the recommendations and how finaldecisions were arrived at. Methods include for example, a voting system, formal consensus techniques(e.g. Delphi, Glaser techniques). Areas of disagreement and methods of resolving them should be specified.

The statement should be answered with “1”, “strongly disagree”, if there is no indication of the methods used to formulate the recommendation.

The rating “2” requires clear indications that consensus-formation has been carried out. The rating “3” requires indications that formal consensus-finding methods have been used. This holds

for key recommendations at the least. A rating of “4”, “strongly agree”, requires clear indication that formal consensus-finding techniques

have been used that also contain descriptions of resolution in cases of lack of consensus. Areas where no consensus could be formed are explicitly indicated, as are the methods for resolution.


Example: American Thoracic Society (2007). Clinical Policy Statement: Palliative Care for Patients with Respiratory Diseases and Critical Illnesses

See „Methods“ (p. 913-914): The task force used a principle- and values-based approach to produce this statement. It first identified the core values and principles of the ATS, as a professional society, related to palliative care (Table 1). The task force identified these principles and values based on the mission of the ATS, the positions that ATS has taken in the past related to ethical issues in ist official statements (5, 6), and the ethical and professional duties of its members who are physicians, nurses, or other health care providers. The latter relate primarily to the ethical principles of beneficence and respect for patient and patient autonomy (7). The task force then applied these principles and values to clinical practice for both children and adults with chronic and advanced pulmonary disorders as well as those with critical illnesses. The task force derived its positions and recommendations based on those values and principles, on the expert opinions and experience of members and consultants of the task force, and on a focused literature search. Operationally, the task force used an iterative modified group process by its members and consultants (see list of members and consultants at the end of the statement) (8, 9). This was initiated at the first general meeting of the task force and many of its consultants during which participants offered their opinions of the most important topics in the assigned field. From these, the task force worked by consensus to identify both the substantive ethical and medical issues and related recommendations to be included in the statement. The task force’s two co-chairpersons drafted the initial and subsequent revisions of the statement. Specific medical recommendations and information were first obtained by a combination of group process and a focused review of the medical literature, including standard textbooks of palliative care medicine. […] The task force then used an iterative process to modify early drafts of the statement by circulating drafts among members to be discussed at subsequent meetings. Later drafts were circulated by e-mail to members and consultants for comments and critiques. The drafts were then modified further on the basis of that feedback and the comments of anonymous reviewers to arrive at the final product.

11/24

11. The health benefits, side effects and risks have been considered in formulating the recommendations.The guideline should consider health benefits, side effects, and risks of the recommendations.For example, a guideline on disclosing bad news to patients with life limiting disease may include a discussion on the overall effects on different communication strategies. There should be evidence that these issues have been addressed.

The statement should be answered with “1”, “strongly disagree”, if the guideline contains no indication that the possible benefit, side effects or risks of the recommendations have been considered. Mere mention (e.g. of product information for pharmaceuticals) within the guideline is not sufficient for an assessment higher than 1.

A rating of “2” requires information on the benefits, side effects and risks for the key recommendations in the guideline.

A rating of “3” requires in addition that the indications of the benefits, side effects and risks are supported wherever possible by results from research studies (i.e. studies on the risks, outcome studies, cost-benefit analyses or other sources) and/or linked with figures (e.g. number needed to treat (NNT), number needed to harm (NNH)). If such figures cannot be generated for the recommendations, this should at least be mentioned in the guideline report (see criterion 29).

A rating of “4”, “strongly agree”, requires in addition that the indications for at least the key recommendations allow for a comparison with alternative approaches or the natural course of the illness, and are not limited to a single dimension (e.g., information on mortality rates).



See „methods of development“, „Systematic Review“ (p. S85): Most studies were descriptive, examining provider practices and patient and/or caregiver views, attitudes, knowledge and behaviour (Level IVa evidence according to the NHMRC 1995 ratings5). Such studies are not considered admissible evidence in the more recent NMHRC rating scale.6 These NHMRC rating scales are designed for research questions regarding interventions, diagnosis, aetiology, screening and prognostic factors, but are not readily applicable to the topic of these guidelines.

Possible substantial modifications for a future Ethics-AGREE11a. The normative basis (principles, criteria, theory) have been considered in formulating the recommendations

11b. Other than health realted benefits, side effects and risks (e.g. optimized ized decision making procedures or patient-physician relationship) have been considered in formulating the recommendations

12/24

12. There is an explicit link between the recommendations and the supporting evidence.There should be an explicit link between the recommendations and the evidence on which they are based. Each recommendation should be linked with a list of references on which it is based.

The statement should be answered with “1”, “strongly disagree”, if no connections have been drawn between the recommendations and the underlying literature.

A rating of “2” requires that at least some connections have been drawn between the recommendations and the literature.

A rating of “3” requires at the least that connections have been drawn throughout between all of the key recommendations and the underlying literature. If this literature contains results concerning the benefit and harm of medical measures, the recommendations must at the least indicate their specific grade of recommendation or level of evidence. For empirical social research, alongside the citation additional indications on the properties of the study quality such as sample size, response rate, pre-test etc. should be provided.

A rating of “4”, “strongly agree”, requires in addition that insofar as clinical trials have been considered, all recommendations should indicate the specific level of evidence. Moreover any points at which no clinical evidence could be found and the recommendations are based on expert consensus should be indicated, as should any points at which the authors consciously deviate from recommendations that could be derived from other studies.



See „methods of development“, „Grading of recommendations“ (p. S86): The recommendations detailed in Boxes 1 to 16 are graded as follows: DS = descriptive study or studies; CG = existing consensus guidelines; EO= published expert opinion; and RGP = recommended good practice based on the clinical and consumer consensus opinion of the expert advisory group and chief investigators.

For an example of the grading application see „The guidelines“, „Difficult scenarios such as requests by family members to withhold information, family conflicts and unrealistic requests by patients“ (Boxes 14 to 16) (p. S103): The guideline is not restricted to recommendations in boxes but also discusses different scenarios in the main text. This box illustrates how the guideline links background evidence with the recommendations and indicates the specific level of evidence.

13/24

13. The guideline has been externally reviewed by experts prior to its publication.A guideline should be reviewed externally before it is published. Reviewers should not have beeninvolved in the development group and should include some experts in the field of clinical ethics and somemethodological experts. Patients’ representatives may also be included. A description of themethodology used to conduct the external review should be presented, which may include a list ofthe reviewers and their affiliation.

The statement should be answered with “1”, “strongly disagree”, if there is no indication of a review by external experts or if there is only reference to an external review process to be conducted following publication of the guideline.

A rating of “2” requires indication of a completed external review process or publication of the guideline in a journal with an external review process.

A rating of “3” requires in addition that the reviewers be indicated by name and professional group and/or organization.

A rating of “4”, “strongly agree”, requires in addition that the method as well as the results of the external review process be described. This can occur in the guideline itself or in the guideline report (see criterion 29), whereby the existence of a guideline report must be clearly stated in the guideline itself. The description in the guideline report must refer individually to the guideline in question; a description of the general principle is not sufficient.

Examples for a 3 rating(no assessed guideline was 4 rated)

Example: SIAARTI (2006). End-of-life care and the intensivist: SIAARTI recommendations on the management of the dying patient

See „Methodology “ (p. 930): The work method entailed: 1) individuation of the objectives of thefinal article and examination of the documents listed in the bibliography by a teamof experts including 4 intensivists, a bioethical theorist and a clinical psychologist.2) Sending a draft of the article to the same 30 reviewers of the 2003 Recommendations,3plus 9 reviewers who in the last 3 years have been in contact with SIAARTI’s BioethicalBoard.

For names and professional groups of the reviewers see “Team of reviewers” Appendix 1p. 945

14/24

14. A procedure for updating the guideline is provided.Guidelines need to reflect current research. There should be a clear statement about the procedure forupdating the guideline. For example, a timescale has been given, or a standing panel receives regularlyupdated literature searches and makes changes as required.

The statement should be answered with “1”, “strongly disagree”, if the guideline contains no indication of any procedure for updating the guideline or only states that “the guideline should be updated”.

A rating of “2” requires at the least an indication of timescale of validity for the guideline. A rating of “3” requires in addition that a contact person responsible for updating the guideline be

named. A rating of “4”, “strongly agree”, requires the description of a procedure for updating the guideline

along with a description of the methods or measures used, including the indication of a timescale and the persons responsible. In the revised version of the guideline the updated areas should when possible be recognizable as such.

No assessed guideline was rated higher than 1

There was no information regarding an update of the respective guidelines analyzed in this study

15/24

CLARITY AND PRESENTAT I O N

15. The recommendations are specific and unambiguous.A recommendation should provide a concrete and precise description of which approach isappropriate in which situation of an ethical challenge in clinical practice.• An example of a specific recommendation is: Relatives of competent patients with a life limiting disease should be only informed about the patients’ health state if the patient has given his or her informed consent to such disclosure of information.

However, and especially relevant to ethical challenges in clinical practice the decision about an appropriate approach often is not clear cut and there may be uncertainty about the best management.In this case the uncertainty and the underlying reasons should be stated in the guideline.

The statement should be answered with “1”, “strongly disagree”, if the presentation and/or formulation of the guideline is in general not specific and unambiguous.

A rating of “2” requires that the presentation and/or formulation of the recommendations be at least partially specific and unambiguous.

A rating of “3” requires that the presentation and/or formulation of the guideline’s recommendations be for the most part specific and unambiguous

A rating of “4”, “strongly agree”, requires that the presentation and/or formulation of the guideline’s recommendations are specific and unambiguous throughout. In addition the areas where the recommendations cannot be kept specific or unambiguous are to be explicitly indicated.


Example: Verlato et al. (2003). Guidelines for resuscitation in the delivery room of extremely preterm infants.

See „Ethical Guidelines“ – “Guidelines for the resuscitation in the delivery room of extremely low gestational age infants (p. 31):

17/24

16. The different options for management of the clinical-ethical issue are clearly presented.A guideline should consider the different possible options for approaching an ethical issue in clinical practice. These possible options should be clearly presented in the guideline.For example, a recommendation on the management of a competent patient with a curative disease who rejects life saving treatment may contain the followingalternatives:

a. Information of patients regarding available treatment options and acceptance of patient’s decision without further trial of discussion (informative decision making) b. Information about available options and discussion of values on side of patients and physician relevant to the decision (deliberative decision making)

The statement should be answered with “1”, “strongly disagree”, if the guideline presents no alternative options for the care problem in question and does not explain why no options are provided.

A rating of “2” requires that the guideline contain alternative options (or space for a weighing of alternatives) at least in some areas.

A rating of “3” requires that the guideline contain options (or space for a weighing of alternatives) in most areas with the explicit indication of the decision criteria for those options.

A rating of “4”, “strongly agree”, requires a clear description of the options (or space for a weighing of alternatives) with the explicit identification and justification of the decision criteria for all options. In addition connections should be drawn where possible to the goal criteria such as patient preferences, efficiency, justice or others.


Example: American College of Critical Care Medicine (2008). Recommendations for end-of-life care in the intensive care unit: The Ethics Committee of the College of Critical Care Medicine

See „Patient and Family-Centred Decision Making “ (p. 954): Resolving Conflict. End-of-life decisions are made readily and by consensus when communication among patients, their surrogates, and clinicians is satisfactory. Nevertheless, communication may not resolve all differences, especially when patients or families insist on interventions that clinicians consider inadvisable. In cases of conflict, the first step is for all parties to focus on obtaining clarity about the goals of care. Patients or their surrogates should determine what they hope to achieve—for example, restoring health, extending life, or relieving pain and suffering. For their part, physicians should provide information about the patient’s prognosis and what goals can be accomplished by the use of specific interventions (14). In the small number of cases where agreement is not possible, physicians are not obligated to offer therapies that they believe cannot achieve the goals of care, as mutually defined by the clinicians, patient, and surrogates (15, 16). When this occurs, the American Medical Association (17) and others have recommended a process-based approach to resolving conflicts over end-of-life issues. Central to this process is mediation by hospital ethics committees.

18/24

17. Key recommendations are easily identifiable.Users should be able to find the most relevant recommendations easily. These recommendationsanswer the main ethical questions that have been covered by the guideline. They can be identified indifferent ways. For example, they can be summarised in a box, typed in bold, underlined or presentedas flow charts or algorithms.

The statement should be answered with “1”, “strongly disagree”, if the guideline contains no key recommendations.

A rating of “2” requires that the guideline contain some key recommendations, even if they are not easy to identify.

A rating of “3” requires that the guideline contain key recommendations that are easily and unambiguously identifiable, and are locatable throughout the guideline (e.g. in highlighted text passages or tables).

A rating of “4”, “strongly agree”, requires that the guideline contains key recommendations that are easily and unambiguously identifiable throughout and that the justification for each is easy to find.


Example: European Federation of Critical Care Nursing associations (2007). The presence of family members during cardiopulmonary resuscitation

See „Position statement “ (p. 257) followed by Notes 1-3 on the same page: 1. All patients have the right to have family members present during resuscitation. See Note 1, below. 2. The patient’s family members should be offered the opportunity to be present during resuscitation of a relative. See Note 2, below. 3. Support should be provided by an appropriately qualified health care professional whose responsibility is to care for family members witnessing cardiopulmonary resuscitation. See Note 3, below. 4. Professional counselling should be offered to family members who have witnessed a resuscitation event. 5. All members of the resuscitation team who were involved in a resuscitation attempt when family members were present should participate in team debriefing. 6. Family presence during resuscitation should be incorporated into the curricula of cardiopulmonary resuscitation training programmes. 7. All intensive and critical care units should have multi-disciplinary written guidelines on the presence of family members during cardiopulmonary resuscitation.

19/24

18. The guideline is supported with tools for application.For a guideline to be effective it needs to be disseminated and implemented with additional materials.These may include for example, a summary document, or a quick reference guide, educational tools,leaflets for patients or other groups affected by the guideline, computer support, and should be provided with the guideline.

The statement should be answered with “1”, “strongly disagree”, if the guideline contains no instruments or materials to support its application. This also applies if the guideline contains no clear reference to this material and/or to the ways it can be obtained. A rating of “1” is also called for if all indicated materials are unavailable.

A rating of “2” requires that the guideline contain a clear indication of available material. A rating of “3” requires that the guideline contain a summary of the recommendations including grade

of recommendation and/or an abridged version as well as an indication of the availability of additional materials as the case may be.

A rating of “4”, “strongly agree”, requires that the guideline contain various materials to support its application. These are at the least a summary of the recommendations connected with their grades of recommendation, an abridged version and a patient version and a thorough derivation of the underlying ethical principles. The additional materials must be unambiguously indicated and available.


Example: British Medical Association, Resuscitation Council (UK), Royal College of Nursing (2001). Decisions Relating to Cardiopulmonary Resuscitation: a joint statement from the British Medical Association, the Resuscitation Council (UK) and the Royal College of Nursing.

See „Summary “ (p. 310)

SummaryPrinciples+ Timely support for patients and people close to them, and effective, sensitive communication areessential.+ Decisions must be based on the individual patient’s circumstances and reviewed regularly.+ Sensitive advance discussion should always be encouraged, but not forced.+ Information about CPR and the chances of a successful outcome needs to be realistic.Practical matters+ Information about CPR policies should be displayed for patients and staff.+ Leaflets should be available for patients and people close to them explaining about CPR, howdecisions are made and their involvement in decisions.+ Decisions about attempting CPR must be communicated effectively to relevant health professionals.In emergencies+ If no advance decision has been made or is known, CPR should be attempted unless:+ the patient has refused CPR;+ the patient is clearly in the terminal phase of illness; or+ the burdens of the treatment outweigh the benefits.Advance decision making+ Competent patients should be involved in discussions about attempting CPR unless they indicatethat they do not want to be.+ Where patients lack competence to participate, people close to them can be helpful in reflectingtheir views.[…]

See also „Other published guidance “ (p. 316) Joint Working Party between the National Councilfor Hospice and Specialist Palliative Care Services and the Ethics Committee of the Association forPalliative Medicine of Great Britain and Ireland. Ethical decision-making in palliative care: cardiopulmonaryresuscitation (CPR) for people who are terminally ill. London: National Council for Hospice andPalliative Care Services, August 1997 (available at the Council’s website: www.hospice-spccouncil.org.uk). Resuscitation Council (UK). Resuscitation guidelines 2000. London: Resuscitation Council (UK),2000. Royal College of Paediatrics and Child Health, Royal College of Obstetricians and Gynaecologists.Resuscitation of babies at birth. London: BMJ Publishing Group, 1997 (under revision).

20/24

A P P L I C AT I O N

19. The potential organisational barriers in applying the recommendations have been discussed.Applying ethical recommendations may require changes in the current organisation of care within aservice or a clinic which may be a barrier to using them in daily practice. Organisational changes thatmay be needed in order to apply the recommendations should be discussed. For example:A guideline on breaking bad news may recommend that such discussions should take place in a separate room which can be used by healthcare professionals, and patients to discuss difficult issues with disturbance.

The statement should be answered with “1”, “strongly disagree”, if the guideline does not discuss the barriers to application or they can only be guessed at.

A rating of “2” requires that the guideline indicate possible barriers. A rating of “3” requires that the guideline indicate possible barriers and proposes possible solutions for

most problems related to the most important barriers. A rating of “4”, “strongly agree”, requires that the guideline systematically treat the factors that could

potentially facilitate or hinder implementation and clearly explain them (barrier analysis).


Example: European federation of Critical Care Nursing associations (2007). The presence of family members during cardiopulmonary resuscitation

See „Pre-amble“ (p. 256): Despite changing trends, family-witnessed resuscitation is a controversial issue that is still debated widely [12,13,4]. Concerns in the literature are centred on three areas. The first is the potential for family members' presence to affect the performance of resuscitation staff and increase their anxiety, or that a distressed relative might disrupt the process [14–17].

See also “Note 3” (p. 256): Whilst it is conceded that on some occasions it may not be possible to provide a health care professional whose sole responsibility is to care for the family member, this should not mean the exclusion of the family member from the resuscitation. Rather, an experienced member of the resuscitation team, who is not undertaking a lead role, should be designated primary responsibility for the continued care of the family member.

21/24

20. The potential cost implications of applying the recommendations have been consideredThe recommendations may require additional resources in order to be applied. For example, there maybe a need for staff trained in clinical ethics to be able to facilitate ethical case discussions This may have cost implications for health care budgets. There should be a discussion of the potential impact on resourcesin the guideline.

The statement should be answered with “1”, “strongly disagree”, if the guideline has no information on the potential financial implications that could be relevant in applying the recommendations.

A rating of “2” requires that the guideline indicate at least in selected areas the possible financial implications of applying the guideline.

A rating of “3” requires that the guideline indicate in most areas the possible financial implications of applying the guideline and discuss possible consequences or solutions. Parts of the guideline where sufficiently concrete statements are not possible (e.g. due to insufficient data) should be marked as such.

A rating of “4”, “strongly agree”, requires a detailed discussion of the possible financial implications of applying the guideline, supported wherever possible by data on decision-making analysis (e.g. cost-benefit analysis). Areas with insufficient data should wherever possible provide sample calculations in order to allow for an estimate of the dimensions or for the planning of the necessary data-gathering measures. Parts of the guideline where no statements are possible should be explicitly marked as such throughout. This can occur in the guideline itself or in the guideline report (see criterion 29), whereby the existence of a guideline report must be clearly stated in the guideline itself. The description in the guideline report must refer individually to the guideline in question; a description of the general principle is not sufficient. Parts of the guideline where no sufficiently concrete statements can be made (e.g. due to insufficient data) should be marked as such and the lack of data should be explained with an indication of the literature searched.

Example for a 2 rating (no assessed guideline was rated higher than 2)

Example: European Resuscitation Council (2000). Part 2: ethical aspects of CPR and ECC. Resuscitative

Introducation (p. 27): Efforts may be inappropriate if goals of patient care cannot be achieved. In some instances resuscitation may not be the best use of limited medical resources. Concern about costs associated with prolonged intensive care, however, should not preclude emergency resuscitative attempts in individual patients

22/24

21. The guideline presents key review criteria for monitoring and/or audit purposes.Measuring the adherence to a guideline can enhance its use. This requires clearly defined reviewcriteria that are derived from the key recommendations in the guideline. These should be presented.An example of review criteria relevant to a guideline on ethical aspects of end-of -life decision making isdocumentation of criteria for decisions at the end-of-life decision

The statement should be answered with “1”, “strongly disagree”, if the guideline contains no review criteria or only refers to review criteria describing individual laboratory results or measurements that do not fulfill the additional requirements of clinical or procedural review criteria.

A rating of “2” requires that the guideline contain review criteria that can serve as an orientation, as least, for monitoring the results of care.

A rating of “3” requires that the guideline contain review criteria that have been defined in terms of the key recommendations. Whenever possible the reference value/reference areas corresponding to these review criteria should be defined such that at-risk patients or groups or states can be reliably identified.

A rating of “4”, “strongly agree”, requires that the guideline contain methodologically justified (sets of) review criteria relating to the (key) recommendations that allow for an evaluation of the application of the guideline and the quality of the results. The (sets of) review criteria should be formulated such that whenever possible they also describe the data that need to be obtained for specific contact positions. The review criteria should be embedded into the general concept for evaluation of the guideline.

No assessed guideline was rated higher than 1

There was no mentioning of review criteria in any guideline analysed in this study

23/24

EDITORIAL INDEPENDENCE

22. The guideline is editorially independent from the funding body.Some guidelines are developed with external funding (e.g. Government funding, charity organisations,pharmaceutical companies). Support may be in the form of financial contribution for the wholedevelopment, or for parts of it, e.g. printing of the guidelines. There should be an explicit statementthat the views or interests of the funding body have not influenced the final recommendations.Please note: If it is stated that a guideline was developed without external funding, then you shouldanswer ‘Strongly Agree’.

The statement should be answered with “1”, “strongly disagree”, if the guideline contains no statements or explanations of its funding.

A rating of “2” requires that the guideline indicate the responsible organization (publisher) as well as any additional funding or supporting bodies.

A rating of “3” requires that the guideline indicate the funding and supporting bodies and includes information as to how the financing occurred.

A rating of “4”, “strongly agree”, requires that besides naming the funding/supporting bodies, the guideline indicate the type and scope of funding in the form of an explicit explanation that the funding/supporting bodies did not in any way influence the content of the guideline. This can occur in the guideline itself or in the guideline report (see criterion 29), whereby the existence of a guideline report must be clearly stated in the guideline itself. The description in the guideline report must refer individually to the guideline in question; a description of the general principle is not sufficient.

Example for a 2 rating (No assessed guideline was rated higher than 2)

Example: de Graeff A, Dean M (2007). Palliative sedation therapy in the last weeks of life: a literature review and recommendations for standards.

See Introducation (p. 68) The proposal to develop recommendations for sedation in the last phase of life was raised in 2002 during an Internet discussion on palliative medicine.An international group of palliative care clinicians from Europe (United Kingdom, The Netherlands, Belgium, France, Germany, Switzerland, Finland), Canada, the United States, Argentina, South Africa, Israel, Japan, Australia, and New Zealand, came together, led by the second author of this paper (M.D.). The aim was to develop internationally accepted definitions and recommendations based on the published literature.

and also on p. 68

The finalized papers are presently posted on the European Association for Palliative Care (EAPC) website and discussions are in progress for them to be expanded and published. Members of the group convened at the meetings of the EAPC in The Hague in 2003 to plan the project and in Aachen in 2005 to discuss the papers and the dissemination of the resulting work.

24/24

23. Conflicts of interest of guideline development members have been recorded.There are circumstances when members of the development group may have conflicts of interest.For example, this would apply to a member of the development group whose research on clinical and ethical aspects of palliative sedation is also funded by a company which produces measures necessary for palliative sedation therapy. It would also apply, for example, to a member of the development group for a guideline on abortion who is member of a certain religious community. There should be an explicit statement that all group members have declared whether they have any financial or intellectual conflicts of interest.

The statement should be answered with “1”, “strongly disagree”, if the guideline contains no information on the topic of conflicts of interest.

A rating of “2” requires that the guideline contain a global statement of the lack of any influence by the financial supporters.

A rating of “3” requires that the guideline describe a query about any possible influence with indication of the aspects queried (e.g. with a standard questionnaire form).

A rating of “4”, “strongly agree”, requires the explicit description of the procedure to identify any possible conflicts of interest including the results of the procedure (value conflicts are to be named alongside conflicts of financial interest). This can occur in the guideline itself or in the guideline report (see criterion 29), whereby the existence of a guideline report must be clearly stated in the guideline itself. The description in the guideline report must refer individually to the guideline in question; a description of the general principle is not sufficient.


Example: German Society fort he Study of Pain (2008). Ethic charter of the German Society for the Study of Pain

See p. 205 Conflict of interest. The corresponding author indicates relationships with Mundipharma, Janssen-Cilag, Grünenthal.

Quality of ethical guidelines and of ethical content in...

Documents

Transcript of Quality of ethical guidelines and of ethical content in...