We developed a structured diagnostic interview for DSM-III-R hypochondriasis (SDIH) that is the first such clinician-administered instrument. The SDIH was administered to 88 general medical outpatients who scored above a predetermined cutoff on a hypochondriacal symptom questionnaire, and to 100 comparison patients randomly chosen from among those below the cutoff. Using the joint assessment method, interrater agreement on the DSM-III-R diagnostic criteria was 88% to 97% and agreement on the diagnosis was 96%. Concurrent validity was suggested by a significant correlation between the interview and the primary care physicians' ratings of hypochondriasis. A measure of external validity was demonstrated in that several clinical characteristics thought to be ancillary features of hypochondriasis were significantly more prevalent in interview-positive patients than in interview-negative patients. Finally, the SDIH appeared to have discriminant validity in that patients diagnosed as hypochondriacal had several other clinical features that distinguished them from the patients who scored above the cutoff on hypochondriacal symptomatology, but failed to be diagnosed as hypochondriacal with the SDIH.