BACKGROUND Because of the uncertainty about the status of hypochondriasis, the disorder is rarely diagnosed. To address this problem we examined the validity of DSM-III-R hypochondriasis as identified by structured interview. METHODS Patients in a general medicine clinic were screened for hypochondriacal attitudes and symptoms. Those patients who scored above an established cutoff had a structured diagnostic interview, and 50 patients who met DSM-III-R criteria for hypochondriasis and 50 age- and sex-matched controls were thus identified. Information was obtained from both groups on health perceptions, health care utilization, and level of functioning using self-report and physician-rated measures. Additional information on diagnoses and treatment recommendations was obtained from record audits. RESULTS Clinic physicians rated hypochondriacal subjects as having more unrealistic fear of illness (hypochondriasis) and diagnosed psychiatric and functional somatic syndromes more frequently in hypochondriacal than in control subjects. Hypochondriacal subjects viewed their health as worse, had more health worries, and had more severe psychiatric symptoms than control subjects. They also reported poorer physical functioning and work performance, greater health care utilization, poorer response to medical treatment, and less satisfaction with the care received than controls. CONCLUSIONS Results show that, although the diagnosis of hypochondriasis is rarely made, physician recognition is high. They also show that several indicators of internal and external validity of this diagnostic category exist. Findings suggest that if physicians are to reduce the functional impairment and nonproductive health care utilization of these patients, they will need to make the diagnosis of hypochondriasis and intervene appropriately. However, for this to occur, research demonstrating predictive validity and treatment responsiveness of the disorder will be required.