How to succeed in a multipractice study.

Abstract

BACKGROUND: In designing a multipractice study on the prescribing habits of general practitioners (GPs) in a Norwegian county, we took into account positive and negative experiences from earlier research on how to improve participation and retention of physician investigators. METHODS: The GP recruitment took place in two phases. The first step was to establish a local network and a sense of community among GPs in the county. The second step consisted of a personal visit by the project leader to all of the practices in the county during the three months before the study started. RESULTS: One hundred and forty-nine (96%) of a total of 156 GPs participated during the first one-month data collection period, registering 43,619 patient contacts. Only six GPs refused to participate, and one GP could not participate because of illness. During the second data collection period, four of the original nonparticipants asked to join, which gave a rate of enrollment of 98% (153 of 156). This time 46,839 contacts with patients were registered. None of the participants dropped out of the study. CONCLUSIONS: Our design for achieving high recruitment and participation rates was successful. We found that in conducting multipractice studies it was essential to spend enough time and resources on establishing a close collaborative relationship with GPs before, during, and after the study.