995 Howe.qxd

Systematic reviews are an important part of the current move towards evidence-based practice. Independent reviewers use a variety of search strategies to identify and assess relevant articles in the field of concern. Criteria for quality must be agreed and articles evaluated accordingly. This study systematically reviewed educational interventions targeting physicians in primary care (excluding hospital clinic and academic settings) to determine their effectiveness in changing behaviour and to investigate whether studies gave information about the resource implications of the interventions described and their rationale for choosing a particular target group. Studies in English, French, or German language journals were included. The review applied the criteria of the Cochrane Collaboration for methodological quality of studies (but was not conducted under the auspices of the Cochrane Collaboration). The results showed that relatively few studies had occurred in primary care compared with academic and hospital clinic settings. Many articles did not fit the criteria for rigour of method, and those that did were very heterogeneous in method and target group. Only two studies assessed resource implications, and one study also calculated economic benefits. The review suggests that future studies should either target geographical areas or doctors with an identifiable learning need associated with patient outcome, and that studies should be evaluated on their ‘intention to educate’. Evaluations of educational initiatives need to describe the resource implications versus measurable benefits of the intervention to make their studies useful to policymakers and planners of educational provision.