A cohort of 392 patients referred to six outpatient clinics by general practitioners during 1987 with diagnoses of rheumatoid arthritis, osteoarthritis, peripheral vascular disease, psoriasis or eczema, were studied from the time of their first attendance until up to two years later. Six consultant clinics were studied in the three specialties: rheumatology, vascular surgery and dermatology. For each specialty a clinic in both a teaching hospital and a district general hospital were included. The cohort members were predominantly middle-aged or elderly people, with a greater proportion of women, except at the vascular surgery clinic where 65% of patients were men. The 392 patients made a total of 936 visits (median two, range one-eight) during the study period; 91 patients were still attending up to two years after the first visit. Patients referred by their general practitioner for therapy were less likely to be discharged than those referred for other reasons. The principal reason for continuing attendance as perceived by patients, general practitioners and hospital doctors was the necessity for consultant supervision, although agreement was far from complete in individual cases. Junior staff tended to see a higher proportion of patients at follow-up visits than did consultants, and were found to have lower discharge rates than consultants. Analyses of data showed that at the first visit, diagnosis, disease severity and the grade of doctor seeing the patient in the clinic was significantly associated with patient discarge at the P < 0.05 level of significance. Patients considered that their visits had produced improvement in their condition in 38% of cases. Although in most outpatient clinics the ideal of consultant review of all cases at every visit cannot be met, it might still be possible for consultants to retain control over the clinic discharge policy. To help ensure that continued outpatient clinic attendances are appropriate; casenote review with junior staff at the end of a clinic could usefully be performed on patients making a third or subsequent visit.