Serum tumor markers may be requested inappropriately by clinicians. In this retrospective study, we aimed to investigate the appropriateness of TM requests in our hospital. Patients in the study were identified from the TM requests for 3 months between June-August 2004, using the laboratory database. A total of 2249 patients (1351 men, 898 women) were included in the study and there were 6570 TM requests. The number of requests were 1050 (16%) for Carbohydrate Antigen 19-9, 993 (15.1%) for Cancer Antigen 125 , 941 (14.3%) for Prostate Specific Antigen, 921 (14%) for free PSA, 925 (14.1%) for Cancer Antigen 15-3, 788 (12%) for Alphafetoprotein, 730 (11.1%) for Carcinoembryonic Antigen and 222 (3.4%) for AFP / Human Chorionic Gonadotrophin. Our findings support the idea that for the evidence-based use of TM requests the education of clinical staff is required. Clear clinical guidelines including recommendations about the appropriate use of TM can be useful for this education process. Careful audit studies are also useful to determine the impact of these guidelines on the practice of evidence-based laboratory medicine.