Using a structured instrument, 325 elderly patients admitted to a general hospital for an acute medical problem were evaluated daily in order to detect symptoms of delirium. Patients were scored for 'hyperactive' or 'hypoactive' symptoms, and then the 125 patients with DSM-III delirium were rated as 'hyperactive type' (15%), 'hypoactive type' (19%), 'mixed type' (52%), or 'neither' (14%). There were no statistically significant differences between the groups with respect to age, sex, place of residence, or presence of dementia. These definitions of subtypes should be studied further.

To study delirium in hospitalized elderly, a delirium symptom interview (DSI) was developed by an interdisciplinary group of investigators. This interview was administered in an acute care hospital to 50 patients who were over the age of 65 years. Results from the interview were compared to assessments of major symptoms of delirium made independently by a neurologist and a psychiatrist. This interview had good validity and reliability. The sensitivity of the DSI was .90 and the specificity was .80, when compared with the clinical judgment of a psychiatrist and neurologist. Interrater reliability, using lay interviewers, was .90 for the detection of major symptoms of delirium. These results indicate that the DSI could be used by lay interviewers to assess reliably the symptoms of delirium.