Return to 2001 Abstacts List

TH Wagner, VA HSR&D Health Economics Resource Center; S Chen, VA HSR&D Health Economics Resource Center

Objectives: In 1995, VHA established Psychosocial Residential Rehabilitation Treatment Programs (PRRTP) as an alternative to traditional inpatient care. PRRTP programs treat eligible veterans who have psychiatric disorders in a less-intensive and more self-reliant setting. The goal in establishing PRRTP programs was to provide cost-effective care that promotes independence and responsibility. This study used an econometric analysis to determine if VA medical centers with PRRTP programs had lower inpatient psychiatric and specialized substance treatment costs than VA medical centers without PRRTP programs. In addition, we explored whether the cost savings were a one-time savings or whether the savings shrank or grew in the years after implementing a PRRTP program.

Methods: We tracked inpatient utilization using the PTF bedsection, extended care and census files from 1993 to 1999. Over the 7-year period, we also followed costs from the Cost Distribution Report. The utilization data were then merged with the cost data to generate an average cost per day for mental health and substance abuse inpatient bedsections. Medical centers were allowed to apply for a PRRTP program starting in 1996. In our dataset, these medical centers were identified with a dummy variable. The final dataset represents a seven year panel. A Hausman test confirmed our a prior belief that a fixed-effect was more appropriate than random effects. We then used a fixed-effects model, adjusting for medical center size, yearly inflation and unobservable confounders that were fixed over time, to test whether the adoption of PRRTP programs was associated with a decrease in average daily costs.

Results: In 1996, 43 medical centers adopted PRRTP programs. The introduction of PRRTP care was associated with an unadjusted decrease of $101 (22%) per inpatient psychiatric day of stay in 1996. We also find that in the years following 1996, the costs for psychiatric care in non-PRRTP centers is increasing at a greater rate than in those medical centers with PRRTP programs. Similarly, the adoption of PRRTP was associated with a $107 (27%) decrease in the average daily costs of inpatient substance between 1995 and 1996.

Conclusions: PRRTP care was associated with significant cost savings. In the four years since PRRTP was established, an estimated $330.5 million ($1999 dollars) has been saved. With little evidence that inpatient or outpatients utilization rates have changed from PRRTP, these savings appear to be related to organizational changes that affected the costs of providing care. In addition, the savings from having a psychiatric PRRTP program have growing over time.

Impact: The VA allowed medical centers to adopt Psychosocial Residential Rehabilitation Treatment Programs starting in 1996. These programs were used as an alternative to traditional psychiatric and substance abuse inpatient care for less severe patients. While it was expected that the PRRTP programs would save money, we provide empirical evidence that approximately $330.5 million was saved between 1996 and 1999. Moreover, we show that the savings continue to grow for psychiatric PRRTP programs.