The attending psychiatrist as a predictor of length of stay

Although Michalon and Richman (1) found diagnosis to be related to length of stay on a Canadian psychiatnic intensive treatment unit, cliical and demographic characteristics of patients have f ilcd to yield irnpressive results in predicting length of stay. Diagnosis-related groups (DRGs) have failed to sufficiently predict length of stay, and there is little evidence that a patient’s diagnosis is related to the length of psychiatmic hospitalization (2). A patient’s age and cornorbid substance abuse (3) appear to be the only factors consistently related to length of stay. Some data indicate that treatment process variables may be related to length ofstay. Hospitalization costs, based mainly on bed-days, vary widely for different hospitals. Involuntary patients have been shown to stay longer in state hospitals than in pnivate hospitals (4). Depressed patients treated in scatter beds have been meported to stay only half as long as depressed patients on psychiatric specialty units (5). Length of stay in private hospitals varies less than in public hospitals (6).