The anatomy and relations of evidence-based medicine

Current tensions between evidence-based medicine (EBM) and some clinicians are counterproductive and unnecessary. The most contentious issues concern (a) the limitations of efficacy data from randomised trials as evidence; (b) differences in attitudes to medical diagnosis and clinical judgement; and (c) political concerns about the use of the concept of clinical evidence and guidelines to restrict physician autonomy. Health services research has evolved in response to a bureaucratic need to study health care, including clinical practice, in order to improve its effectiveness (defined mainly in terms of technological interventions), and to contain costs. Its perspective is from the top-down representing the interests of bureaucracy and managed care, and articulates with political demands for professional accountability and cost-containment. EBM has established its place as an important contributor to the methodological toolbox for health services research. There is a need for a corresponding coherent programme of clinical practice research which would locate EBM in the clinical environment beside quality assurance, the study of the appropriateness and effectiveness of interventions, and multidisciplinary research related to the art of medicine and supportive aspects of clinical care. EBM would then be seen as one organ in relation to many others making their contribution to the body of knowledge needed for clinical decisions and policy making. A 'centre for the study of clinical practice' would be an appropriate structure to support such a comprehensive programme of clinical practice research in a tertiary hospital. The bottom-up perspective of clinical practice research would complement the current top-down perspective of most health services research, providing information to doctors, patients and administrators concerning local quality of care and health outcomes, information which could also be aggregated for guidance of health policy makers. It would also represent the voice of the clinician in policy debates. Such a programme, located in the practice environment, would also foster mutual understanding, respect and cooperation between workers from different backgrounds.