6533b7d7fe1ef96bd1268c94

RESEARCH PRODUCT

Declaration and Handling of Conflicts of Interest in Guidelines

subject

description

Clinical guidelines are developed in order to support physicians and patients in specific clinical situations when decisions concerning diagnosis and treatment are made. Recommendations provided in guidelines are based on the findings of clinical studies and on expert opinion. Identical study findings may be evaluated differently depending on whether or not guideline authors have conflicts of interest (1). A conflict of interest is defined as a circumstance that gives rise to a risk that professional judgement or actions concerning a primary interest may be inappropriately influenced by a secondary interest (2, 3). A conflict of interest is therefore a state of affairs, not a biased evaluation or the result of a particular act (3, 4). The primary interest is in line with an individual’s professional activities, in the case of doctors the best possible patient care. Secondary interests may be financial, psychological, or social in nature (5). An example of a financial secondary interest is a financial link with manufacturers of drugs or medical devices—accepting gifts or fees for consultancy or lectures, for instance. However, no system for remunerating physicians’ work can help but create secondary interests, and these inevitably conflict with primary interests. Nonfinancial interests may be, for example, the adoption of a particular treatment-related conviction or school of thought. Nonfinancial conflicts of interest often go hand in hand with financial ones (3, 5). Studies conducted in various countries show that many guidelines (36 to 98%) contain no information on conflicts of interest. When conflict of interest statements are made, multiple conflicts of interest are often declared (6– 17). In Germany, the Association of Scientific Medical Societies (AWMF, Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen Fachgesellschaften) coordinates specialist societies’ development of guidelines on diagnosis and treatment. Member societies’ guidelines are divided into three tiers according to how they are developed: S1 guidelines are developed by a group of experts who come to an informal consensus. S2 guidelines are based on either the structured consensus of a representative committee or a systematic analysis of the scientific evidence. S3 guidelines meet both the criteria above (18). The AWMF has developed recommendations on handling conflicts of interest, which it published in April 2010 (19). These are based on recommendations made by the US Institute of Medicine (IoM) which have also been discussed and adapted in Germany (2, 20, 21). The AWMF’s regulations include, among others, the following: Conflicts of interest must be declared and published using a form. Authors’ declarations must be evaluated by the steering committee and the guideline coordinators. Collaborators assessed as being biased must be excluded from the evaluation of evidence and development of consensus. Care should be taken that authors are free of significant conflicts of interest. Guidelines in which conflicts of interest of individual collaborators are not transparent must not be included in the AWMF register. The AWMF provides a template form for download on its homepage for conflict of interest statements (22). This contains nine points that inquire into financial and nonfinancial conflicts of interest in the last three years. There are two studies available in Germany concerning the handling of conflicts of interest when guidelines are compiled. An analysis of dermatology guidelines dating from 2010 shows that information on guideline funding and authors’ conflicts of interest is insufficient (14). A study of S2 and S3 guidelines of German specialist societies dating from 2009 to 2011 shows, in particular, significant shortcomings in handling conflicts of interest: although the practice of disclosing conflicts of interest has become established, the duty to disclose them does not lead to discernible countermeasures (10). However, the majority of all AWMF guidelines, approximately 60%, are S1 guidelines (23). These provide recommendations which may sometimes even have legal consequences if breached (24). S1 guidelines are developed as a result of informal consensus reached by a group of experts. This makes it particularly important that authors’ conflicts of interest be handled transparently. Because there are no studies as yet on conflicts of interest among authors of S1 guidelines, this study addressed the following questions: How frequently are conflicts of interest declared in S1 guidelines, and what information about them is provided? Are authors of guidelines on the use of drugs more likely to have financial conflicts of interest than those of guidelines on other subjects? Do any consequences result from declared conflicts of interest? Are the AWMF’s April 2010 recommendations on handling conflicts of interest implemented?