STUDY OBJECTIVE--To assess the scope for reducing unnecessary outpatient reattendances, using a benchmark an acute specialty at a site recognised to have an especially low ratio of repeat to new attendances. DESIGN--This was a survey of the re-attendance workload at general surgery outpatient clinics over a three month period. Patient re-booking and discharge rates for different grades of staff; clinicians' perception of the ability of the GP to have managed the patient; perception of the value of individual re-attendances; reason given for discharging/re-booking; and outcome of attendance for patients in relation to diagnostic category were determined. SETTING--General surgery outpatients clinics with re-attendance rates that were 50% below average, in Taunton and Somerset Hospital, a non-teaching district general hospital. PATIENTS--Altogether 454 patients who made 470 second or subsequent visits (re-attendances) within the same episode of outpatient care. MAIN RESULTS--Thirty eight percent (178/470) of visits were perceived as manageable by the GP, 45% (79, 17% of total re-attendances) of which were also thought to have been of marginal or little value. A substantial group of patients was being followed up largely for reasons of convention and traditional policy. Re-booking rates were higher among junior staff. Subjective views of the value of attendance at the hospital outpatient clinic and the ability of the GP to have seen the patient varied systematically between consultants and junior staff. Judgements varied to some extent according to the diagnostic group. CONCLUSION--The numbers of patients being followed up equivocally at most general surgical outpatient departments will be 50% more an average than those in this benchmark department. A department seeing 2000 new patients per annum will have 3600 reattendances, 25.5% (918) of which may be avoidable on the basis of these results. A variety of approaches can be used to increase the proportion of patients seen appropriately by GPs. In some cases this might be achieved without the intensive commitment required to plan and develop shared care protocols or new formal discharge guidelines, but by encouraging GPs to manage some patients, increasing of hospital clinicians' access to knowledge of local general practices, and internal clinic review of 'routine' follow up policies as shown in this study. This type of review of outpatient practice can also help prioritise conditions likely to repay the effort of developing and implementing clinical management guidelines and local protocols.