Getting a grip on guidelines: how to make them more relevant for practice

Evidence-based guideline development can be seen as one of the major achievements in efforts to improve patient care in the last decade. Well developed clinical guidelines provide professionals, patients, and policy makers with information on how to manage health problems appropriately within day-to-day practice. The development of guidelines has made an enormous progress in the last 10–15 years with many guideline development programmes, such as those of NICE, SIGN, and medical colleges, using established methods and procedures, according to AGREE Collaboration criteria.¹

New developments related to searching and grading evidence (SEARCH, GRADE), and adapting guidelines to local context (ADAPTE) aim at making guideline development better and more effective. Guidelines have had a major impact on care in general practice in some countries; for instance, in the Netherlands most problems seen in general practice are covered by national evidence-based guidelines developed by the scientific body with now around 75% of the decisions in line with these guidelines.²

An exciting next step in this development is to link evidence-based guidelines to incentives for performance, as initiated in the Quality and Outcomes Framework (QOF). A worldwide collaboration between guideline developers has been created with over 70 organisations as members (Guidelines International Network). A new world of clinicians, epidemiologists, and professional guideline developers with their own meetings, procedures, and vested interests has grown. This highlights also one of the risks of guideline development, that is, becoming increasingly ‘institutionalised’, top-down driven by research findings, notably randomised controlled trials, imposed upon practice. A balanced view of risks and benefits of clinical guidelines is therefore needed, in which preferences of patients and demands of practice and policy, are matched with the achievements of science.

Against this background, a number of persistent problems can be observed in guideline development that come forward in the paper comparing depression guidelines from seven different countries, in this issue of the BJGP.³ One of the problems identified is that many clinical guidelines still do not yet meet internationally accepted quality criteria, as defined in the AGREE instrument.¹ For instance, the rigour of development scores for the seven studied depression guidelines ranged from 1–64%.³ In addition, experiences with guideline development indicate that there is a risk of substantial bias in guidelines.³ Appropriate evidence is lacking for many recommendations and, even when evidence is available, the final recommendations for practice are often largely a reflection of local culture or personal views of the guideline developers. Guideline developers in different countries often come up with different advice for decision making in practice, partly because they refer to different, not overlapping sources of evidence.^4,5 Normative and cultural opinions about the value of specific performance often play an important role in defining recommendations for practice without making these explicit.^6,7

Many recommendations in clinical guidelines aim at ‘ideal patients’, usually adult patients without any comorbidity, and are not very well tailored to actual patient care and real patients. In normal general practice, many patients, particularly patients with chronic multi-morbidity, have a combination of (interacting) problems, which make appropriate performance and decision making more complex.⁸ There is an obvious development towards integrated care with increasing collaboration between disciplines, while most clinical guidelines do not yet focus on this complexity. A recent publication relating to GPs' management of depression has cast doubts on whether essential aspects of day-to-day care, notably comorbidity, have been considered in evidence-based guidelines and QOF indicators.^9,10 Many clinical guidelines are still mono-disciplinary, often written by specialists from tertiary care, who have a different type of patient in mind.

Guideline development is time consuming and expensive, about €100 000–200 000 per guideline, and the question is, if this is cost-effective. It may be, if guidelines were valid and had a wide impact on health care. However, even well developed evidence-based guidelines are often not used in day-to-day care (estimations range from 25–50%). There are different causes for clinical guidelines not being used, partly related to the guidelines themselves. For instance, some guidelines are written as a handbook on a clinical topic and not as a concise set of concrete recommendations for decisions in day-to-day practice. They often have too many recommendations, making it difficult for the user to identify the key-issues and most important targets. Many do not answer the crucial questions of patients and practitioners related to the health problem well, as outlined by Hegarty et al.³

In a large study (yet unpublished) on the implementation of chronic obstructive pulmonary disease (COPD) guidelines in the Netherlands, we found that improvement failed, largely because patients did not follow the advice of their GPs. The guidelines on COPD do not address this problem of non-adherence sufficiently. Many guidelines also lack the tools that should help to make them work in real practice, such as well-structured care pathways and well-developed indicators to measure performance and change, and focused programmes to support their implementation.

Clinical guidelines are potentially very valuable tools to support decision making in general practice, but some improvements are required in current guidelines and guideline development processes to make them more relevant. They should, for instance, be focused more on key-issues in patient care, with direct relevance for both practitioners and patients; they should take real (comorbid) patients as a starting point; they should be developed in less time-intensive procedures to keep them updated, presented in more concise formats, and be combined with quality indicators and support tools for practice. A priority is better collaboration between all stakeholders — clinicians, scientists, patients, policymakers, and others — to identify jointly the most important questions, assess the available evidence, and draw recommendations that can work under prevailing practice conditions.¹¹ The limitations and importance of drawing guidelines for highly different circumstances under which practitioners encounter their patients should be acknowledged. This is even more important when financial incentives are linked to the evidence-based guidelines. Such improvements should lead to guidelines being able to deliver what they intend: better care for patients in response to their needs.