Randomised trials in general practice--a New Zealand experience in recruitment.

AIM To examine general practitioner (GP) and patient recruitment in a randomised clinical trial to determine the usefulness of brain natriuretic peptide (BNP) in the diagnosis of heart failure in the community. METHODS Different techniques were used to maximise GP recruitment including early consultation with GPs, benefits for participating GPs and patients, and a comprehensive suburb-by-suburb approach to GPs using letters and personal visits. GPs then referred patients. At the conclusion of the study, GPs were given a questionnaire focussing on barriers to referral and reasons for participation. RESULTS Three hundred and twenty seven GPs from 135 practices were sent an introductory letter; 294 were eligible to participate. Of these, 186 GPs (63% of eligible GPs) agreed to participate. Ninety two GPs (31% of eligible GPs) from 62 practices referred 307 patients to the study (range 1-14 patients). There were no significant differences between referring and non-referring GPs with respect to sociodemographic characteristics. Referring GPs were very supportive of GP participation in research and strongly agreed that GPs should be reimbursed for involvement in trials. CONCLUSIONS Patient recruitment by GPs may be aided by the use of a range of strategies including financial reimbursement. GPs who agree to participate will not always recruit patients. Closer collaboration and understanding between primary healthcare professionals and researchers may further enhance recruitment to clinical trials.