Changing the clinical behavior of doctors: a psychological framework.

There is growing concern about the long delays that occur before doctors implement the findings from new research. It is also of concern that the clinical practice of different doctors can vary widely.' Consequently the problem of changing the behaviour of doctors to comply with recommendations of new research or guidelines is one of considerable interest. Several methods have been used to change the clinical behaviour of doctors, including guidelines, reminders, reading materials, feedback, continuing education, and sanctions. The potential role of guidelines has attracted particular attention.2 3 A review of studies of the use of guidelines concluded that they can lead to changes in performance but that the extent of change varies widely and depends on the methods of implementation that are used.4 In a review of 102 studies of strategies for change no single method emerged as effective in changing behaviour in all circumstances and settings, but most strategies were occasionally effective.5 Because of the disappointing findings that single strategies do not necessarily promote change, there is increasing interest in the effectiveness of combinations of strategies.6 7 Success in changing the clinical behaviour of doctors working in a specific setting may be more likely if the change strategy is chosen to fit the setting and circumstances. Several arbitrarily selected strategies used together could be inefficient and not necessarily more effective than a single, appropriate strategy. However, information about which strategy is most effective in different circumstances is limited. Grol produced a classification of change strategies for implementing guidelines which takes into account the different obstacles to change that might be relevant to doctors who are "early adopters", the "majority", or "late adopters".8 9 Grimshaw and Russell also proposed a basic framework for classifying strategies for implementing change.'0 After reviewing the effectiveness of guidelines, they classified strategies into groups with high or low probabilities of effectiveness. The characteristics of high probability methods included participation in development of guidelines, dissemination involving specific educational interventions, and patient specific reminders at the time of consultations. These authors, however, acknowledged that the evidence available on the relative effectiveness of different strategies was sparse. Lomas developed a coordinated implementation model that integrates four approaches to help to understand doctors' responses to implementing guidelines: social influence theory, marketing theories, educational theories, and research studies into the diffusion of innovation." The model is intended as an alternative to traditional passive education to implement change and does not include a stage for the analysis of the specific obstacles to change that might be encountered in different, discrete clinical settings. An ideal model or framework of methods for changing the clinical behaviour of doctors would indicate what obstacles to change might be encountered in different circumstances and which change strategies would then be most appropriate. An analogy that helps to illustrate how obstacles to change might be identified and strategies chosen to overcome them is the approach used by clinicians in diagnosing a disease and selecting a treatment. In diagnosing a disease or syndrome, clinicians will refer to a body of knowledge or "theory" to interpret the symptoms of the patient and to guide the search for additional confirmatory symptoms and signs that indicate presence of the disease; the "theory" also indicates which treatment is most likely to be beneficial. The aim of this paper is to introduce a framework which provides a structure for arranging and applying existing knowledge of behavioural change. This is neither a complete model nor a comprehensive review of all theories of behavioural change. It is an approach that can be used to apply available knowledge to diagnose obstacles to change and then select appropriate treatments or strategies to overcome those obstacles and thus improve the clinical practice of doctors.