Commitment to change statements can predict actual change in practice

INTRODUCTION: Statements of commitment to change are advocated both to promote and to assess continuing education interventions. However, most studies of commitment to change have used self-reported outcomes, and self-reports may significantly overestimate actual performance. As part of an educational randomized controlled trial, this study documented changes that family physicians committed to make in their prescribing and then used third-party data to examine actual changes. METHOD: Following participation in a continuing medical education program using interactive small groups, physicians were asked to identify changes that they planned to make in their practices. For prescribing changes related to four conditions, data from a provincial pharmacy registry were analyzed for 6-month periods before and after the educational intervention. RESULTS: A total of 207 physicians participated in the project, which involved monthly meetings of 30 peer learning groups. Ninety-nine physicians received experimental case-based educational modules +/- personal prescribing feedback, and 91 of these indicated that they planned to make at least one change in practice. Of the 209 intended changes, 71% were directly related to the prescribing messages in the materials. DISCUSSION: In three of four indicator conditions, physicians who expressed a commitment to change were significantly more likely to change their actual prescribing for the target medications in the following 6 months. The percentage of physicians who did change their prescribing varied significantly by condition. Further study of the process of translating commitment to change into real practice change is needed.