Variation among VA hospitals in length of stay for treatment of depression

OBJECTIVE: The purpose of the study was to determine the degree of interhospital variation in length of stay for patients treated for depressive disorders in Department of Veterans Affairs (VA) medical centers and to identify the number of hospitals with significantly longer or shorter than average lengths of stay (outlier hospitals). METHODS: The sample included 6,374 veterans discharged from acute psychiatric inpatient units at 107 medical centers in 1992 with a primary diagnosis of major depressive disorder or depressive disorder not otherwise specified. To identify statistical outliers, the average length of stay at each medical center was compared with the overall sample mean while controlling for differences in case mix between hospitals. RESULTS: Patients' demographic characteristics, treatment history, and severity of illness all significantly predicted length of stay at the patient level. After case mix factors were controlled for statistically, approximately 29 percent of the medical centers were found to have mean lengths of stay significantly different (p < .01) from the sample mean. Specifically, 15 percent of the medical centers were short-stay outliers, and 14 percent were long-stay outliers. CONCLUSIONS: Despite the fact that the treatment regimen for depression is relatively standardized and all VA medical centers operate under the same administrative model and reimbursement system, a high degree of interhospital variation was found in treatment duration for depression. Results led to the tentative postulation that variations in treatment duration reflect differences in physicians' practice styles. Substantial opportunities may exist for reducing expenditures for the treatment of depression within the VA health care system.