The cost of follow-up examinations for patients having undergone potentially curative surgery for colorectal cancer is considerable. The aim of this study was to provide a thorough assessment of the cost and effectiveness of the follow-up tests used during the 5 years after surgical resection for colorectal cancer and its recurrences. We studied medical and economic data from the records of 256 patients registered in the Herault Tumor Registry who underwent potentially curative surgical resection in 1992. Recurrence, curative recurrence, survival, and the cost of follow-up tests were assessed respectively for at least 5 years. We analyzed the cost and effectiveness of follow-up tests in patients who received either follow-up with carcinoembryonic antigen (CEA) monitoring as advocated by the 1998 French consensus conference recommendations (standard follow-up) or a more minimal follow-up schedule. Nine patients died in the postoperative period. The 5-year survival rates in the standard and minimal follow-up groups were 85% and 79%, respectively (p=0.25). Cost-effectiveness ratios were 2123 in Dukes' stage A patients, 4306 in Dukes' stage B patients, and 9600 in Dukes' stage C patients. Cost-effectiveness ratios for CEA monitoring and abdominal ultrasonography per patient alive in the standard follow-up group were 1238 and 2261.5, respectively. Cost-effectiveness ratios for CEA monitoring and abdominal ultrasonography per patient alive in the minimal follow-up group were 1478 and 573, respectively. There were no survivors 5 years after a recurrence when the recurrence was detected by physical examination, chest X-ray, and colonoscopy in either follow-up group. Dukes' classification is a poor indicator of patient selection. The follow-up tests should only include CEA monitoring and abdominal ultrasonography for the diagnosis of recurrence.