This study evaluates the usefulness of routine follow-up of breast cancer patients. In all, 416 patients who were treated with curative intent for breast cancer were followed according to a fixed follow-up schedule for a minimum of 2.5 years and a maximum of 13.5 years (mean about 5 years). During the 4533 routine out-patient visits, 4116 chest radiographs, 3889 pelvic radiographs and about 17,000 laboratory tests were carried out routinely. In the follow-up period, 148 patients were found to have distant recurrence of whom 34 (23 per cent) had asymptomatic metastases and 114 symptomatic metastases. Of the 8005 routinely performed radiographs, 24 (0.3 per cent) revealed asymptomatic metastases, and the 17,000 laboratory tests led to the discovery of six asymptomatic bone and four asymptomatic liver recurrences. Screening for metastases did not result in a reduction of the lead time to the diagnosis of asymptomatic metastases; the disease-free interval was equal in both symptomatic and asymptomatic patients. Of the 46 locoregional recurrences 42 were found by physical examination during a routine follow-up visit and 37 had not been noticed by the patient. Seventeen second primary breast cancers were diagnosed, six of which were in stage I (less than 2 cm). Mammography was not a part of the routine follow-up scheme. It is concluded that routine follow-up of breast cancer patients by history and physical examination is sufficient to detect local recurrence and a second primary tumour as well as giving the opportunity to track signs and symptoms of distant recurrence at an early stage. Performing annual or biannual mammography is advisable, but the use of other costly routine investigations in the follow-up is not justifiable, as no therapeutic advantages can be expected.