Effect of Insurance Parity on Substance Abuse Treatment

OBJECTIVE—This study examined the impact of parity for substance abuse treatment on use, cost and quality of substance abuse treatment. METHODS—The authors compared substance abuse spending and use from 1999 to 2002 for continuously enrolled beneficiaries covered by Federal Employee Health Benefit (FEHB) plans with similar outcomes from beneficiaries in a matched set of health plans without parity coverage. Logistic regression models of the probability of any substance abuse service use, and conditional on use, linear models of substance abuse total and out-of-pocket spending were estimated. Logistic regression models for three quality indicators for substance abuse treatment were also estimated: identification of adult enrollees with a new substance abuse diagnosis, treatment initiation, and treatment engagement. Difference-in-differences were computed as (post-parity- pre-parity) difference in outcomes in plans without parity subtracted from difference in FEHB plans. RESULTS—There were no significant differences in rates of changes in average use of substance abuse services between FEHB and non-FEHB plans. Conditional on use, rate of substance abuse out-of-pocket spending declined signicantly relative to non-FEHB plans (mean difference=− $101.09, 95 % CI=[−$198.06, −$4.12]), while changes in total spending did not differ. Under parity, more patients were identified with a new substance abuse diagnosis (.10%, [.02%,.19%]). No statistically significant differences were found for initiation and engagement in substance abuse treatment rates. CONCLUSIONS—Findings suggest that for continuously enrolled populations, providing parity of substance abuse coverage improves insurance protection but has little impact on utilization, costs to plans, or quality of care