We report a retrospective study to determine the cost-effectiveness of cranial computed tomography in patients with headache without neurological finding. Five hundred ninety-two neurologically normal patients wars examined between 1990 and 1993 for the complaint of headache. Examination results were reevaluated from written report and image archive systems. Results were divided into three groups. In group P-0, we included patients with normal cranial computed tomography findings. In group P-1, patients showed some minor pathologies like ischemic or atrophic changes. These findings neither explained the reason for headache nor changed the clinical or therapeutic approach. The third group (P-2) was to include patients with gross intracranial pathology like space-occupying lesions or bleeding. Five hundred forty-six of 592 patients were in the P,group (92%), and the remaining 46 patients were in the P-1 group (8%). No patient was found to have serious intracranial pathology detected by computed tomography. Cost of detection of a case with significant pathology was calculated.