Although medical survey studies often rely on self-reported symptoms to establish the presence or absence of clinical conditions in respondents, recent findings suggest that surveys that assess a broad range of symptoms may have limited sensitivity in detecting specific clinical conditions such as hypertension. The present study evaluated the accuracy of a blood pressure history survey mailed to 800 men and women who had received treatment at a military medical facility in the previous year. Compared with their medical records, patient reports of a previous diagnosis of hypertension exhibited an overall accuracy of 94.2%. This high level of overall accuracy was associated with equally high proportions of correct identifications of high blood pressure histories (sensitivity = 95.4%) and normal blood pressure histories (specificity = 92.4%). Our findings indicate that the Ohio Blood Pressure History Survey is a highly accurate measure of hypertension history among active and retired military personnel.