This retrospective review was undertaken to determine the efficacy of routine follow-up in the detection and management of recurrent cancer. The case notes of all women attending a regional cancer center who were diagnosed with cancer in 1997 were reviewed. Of 81 new cancers followed up for a median of 42 months (range 36-48), 14 have recurred after curative treatment and there were six cases of persistent disease. The median number of clinic visits per patient was 3.5 (range 1-16). Eight recurrences (57.1%) were diagnosed at scheduled outpatient appointments, three (2 l.4%) presented to the general practitioner (GP), and three were seen as emergencies in hospital. Seventeen patients with persistent/recurrent disease have died and three are alive with disease. The median time from initial presentation to disease recurrence was 12 months (range 5-25) and the median time from recurrence to death was 5 months (range 1-20). The longest interval between onset of symptoms and diagnosis of recurrence (4 months) occurred in those presenting at scheduled outpatient clinics. This study demonstrates that the current follow-up protocol is associated with delays in diagnosing recurrence, because symptomatic patients postpone seeking help until their scheduled visit. We have therefore commenced a prospective study evaluating other models of follow-up.