Many research projects depend on general practitioners (GPs) to recruit patients into the study. As part of a study of patients' experiences of treatment for menorrhagia, completeness of recruitment and the extent of bias introduced by failure to recruit was assessed. Only 129 (41%) of 315 GPs who had agreed to recruit patients actually did so. A review of notes in six practices revealed that only 40 (20.4%) of 196 eligible patients had been recruited. There was some evidence that the recruited patients had more severe symptoms than those not recruited. There was no difference in recruitment rates between male and female doctors, but those who had received a practice visit from a member of the research team recruited more patients. A survey of participating GPs revealed that forgetfulness and time pressures were the main factors inhibiting recruitment.