Has guideline development gone astray? Yes
BMJ 2010; 340 doi: https://doi.org/10.1136/bmj.c306 (Published 29 January 2010) Cite this as: BMJ 2010;340:c306
- R Grol, director
- 1Scientific Institute for Quality of Health Care, Radboud University Nijmegen Medical Centre, PO Box 9101, 114, 6500 HB Nijmegen, Netherlands
- r.grol{at}iq.umcn.nl
It is a long time since clinical guidelines were seen as cookbook medicine and a threat to professional autonomy. Nowadays, evidence based guidelines are considered one of the major efforts to improve patient care. Development of guidelines has progressed enormously, with many organisations (including the National Institute for Health and Clinical Excellence in the UK) using validated methods such as the AGREE instrument.1 Clinical guidelines are valid if they are developed in a rigorous way, independently of vested interests of their developers, and if they support decision making in practice and affect actual care. But are current guidelines meeting these criteria? I have concerns.
For guidelines to have an impact on actual care, they need to be integrated with other quality improvement initiatives, such as performance measurement and quality improvement programmes. This requires intensive collaboration between the organisations responsible for these tasks,2 which is lacking in most countries. Expert guideline developers, usually clinicians and epidemiologists, often have very different aims and interests from those in the quality improvement world. Stakeholders use guidelines for different purposes, also hindering integration and improvement. For instance, policy makers and authorities want to use them for inspection or to contain costs, whereas professional bodies may use them to strengthen their position in the competition with other disciplines.
Another problem is that many guidelines still do not meet the internationally accepted criteria of the AGREE instrument.1 A recent evaluation of seven depression guidelines from different countries (five of which were issued after the AGREE instrument was published in 2003) showed scores of 25-63% for stakeholder involvement, 1-64% for rigour of development, 0-56% for applicability, and 8-75% for editorial independence.3 New (still unpublished) evaluations of guideline quality show similar findings.
Local influence
Guideline developers are aware that there is a risk of bias in recommendations because appropriate evidence is often lacking. However, even when evidence is available, the final recommendations often reflect personal opinions, local culture, or vested interests of the developers.4 5 6 A recent Institute of Medicine analysis reports many examples of undue industry influence on the development of clinical practice guidelines and makes recommendations to limit conflicts of interests.6 Guideline developers sometimes have close ties with the industry. 7 Another problem is that current methods for translating evidence and professional experiences into recommendations for practice have got less attention and are much less sound and transparent than the methods for finding and grading evidence.
Guideline developers in different countries often come up with different advice, partly because they prefer to use local evidence and partly because normative and cultural opinions have a substantial role in defining good care.8 9 This lack of independence and inconsistencies in different guidelines lower the trust that potential users might place in them.
Lack of applicability
Statements on clinical effectiveness dominate guidelines and assume “ideal patients” without comorbidities. They often do not adequately address issues relevant for everyday care, such as safety and risk management, multidisciplinary collaboration, effect on costs or compliance, and patient self management. The analysis of the seven depression guidelines, for example, showed a focus on drug treatment and limited attention to psychological therapies, suicide risk, and social issues.3 In reality half of patients with a chronic condition have at least two problems that may interact, requiring complex care and support.10 But guidelines are often formulated by mono-disciplinary specialists with specific patient types in mind. Patient and public involvement in guideline development is of growing interest but not standard practice yet, and it is not clear how to incorporate their preferences.11 12
Despite examples of good uptake, audits around the world show that guidelines are, on average, used in only 50-70% of day to day decisions and variation in performance is large.13 At least part of the reason for this is that some of the guidelines have limited relevance to clinicians and patients, are written as a handbook and not as a concise set of recommendations for practice (maximum 5-10), or are incompatible with the norms and values of target users.14 This may make implementation ineffective, particularly when it requires new behaviour and organisational change.15
Value for money
Finally, guideline development is time consuming and expensive (€150 000 (£130 000; $220 000) to €200 000 per guideline in the Netherlands, and more than £400 000 for a NICE guideline). The investments may be worth while if guidelines are clinically relevant and have a wide impact on health care. However, the cost effectiveness of guideline development compared with other methods for improving patient care is unknown.
Guidelines may thus be important for improving patient care, but changes are needed to make them more relevant and effective. Collaboration between all stakeholders—relevant clinicians, scientists, patients, policy makers, quality improvement experts, and others—is essential to identify clinical questions, assess the evidence, draw up workable recommendations and develop related indicators of improved quality of care and programmes for implementation. Procedures must be changed to speed up the development, minimise personal bias in recommendations, and involve patients more actively in the process of both developing and using guidelines. Without such changes, guideline development may increasingly be seen as an expensive but unhelpful and ineffective toy for a happy few.
Notes
Cite this as: BMJ 2010;340:c306
Footnotes
Competing interests: The author has completed the unified competing interest form at www.icmje.org/coi_disclosure.pdf (available from him on request) and declares (1) no financial support for the submitted work from anyone other than their employer; (2) no financial relationships with commercial entities that might have an interest in the submitted work; (3) no spouses, partners, or children with relationships with commercial entities that might have an interest in the submitted work; and (4) no non-financial interests that may be relevant to the submitted work.