Session number: 1026
Abstract title: Mental Healthcare Utilization and Costs Among Veterans with PTSD: The Impact of Comorbid Disorders
Author(s):
JN Breckenridge - VA Palo Alto Health Care System
JI Sheikh - VA Palo Alto Health Care System
Objectives: The objective of this study was to evaluate the presence of comorbid psychiatric disorders and other variables as predictors of elevated rates mental healthcare utilization and costs among veterans with Post Traumatic Stress Disorder (PTSD).
Methods: The design was a cross-sectional analysis of VA administrative data. Subjects were selected from a 10 percent national random sample of all veterans who had at least one encounter in a VA mental health (MH) clinic during Fiscal Year 1999 and who were assigned an ICD9 diagnosis of PTSD in any care setting (N= 16,168). Subjects were followed for one full year after their first FY99 encounter. Encounter totals were weighted for clinical complexity using the HCFA Resource-Based RVUs. Both Cost Distribution Report (CDR) and DSS-based costs were evaluated. Logistic regression analyses on the top 25% and top 10% of mental health services users were performed employing demographic, geographic, and diagnostic factors as predictive variables.
Results: Utilization and costs varied substantially, with only a third of subjects participating in a PTSD specialty program. High use groups had utilization rates five to six times greater than the average for other participants. Rates varied significantly over geographic areas even when adjusted for all other factors. The adjusted odds (AO) of elevated utilization were higher among subjects who were female, Caucasian, between ages 45 and 65, or had greater than 50% VA service connected benefits eligibility. Comorbid substance abuse and major affective disorders were much more prevalent among both groups of high utilizers, and these comorbidities were associated significantly with increased utilization and costs. In particular, veterans with substance abuse disorders had three times the adjusted odds of consuming exceptional amounts of mental health services (AO= 2.98, (95% CI: 2.61 3.40); AO= 3.29, (95% CI: 2.84 3.81), top 25% and 10%, respectively.)
Conclusions: PTSD treatment protocols must be modified to accommodate psychiatric comorbidities, especially substance abuse. Access to such care may not be equitable across regions of care.
Impact statement: Future development and evaluation of specialized VA treatment programs for PTSD must address the differences in patterns of mental health services utilization observed in this study.