It is frequently argued that insurance for mental health services offers economic incentives to providers and consumers that drive up costs while encouraging clinically inappropriate treatment. It is commonly thought that more substitution of outpatient for inpatient care could occur but that restrictive ambulatory mental health benefits have made the shortening of hospital stays difficult and have contributed to the overutilization of inpatient services. The purpose of this study is to provide empirical verification for this argument by examining patterns of mental health care utilization in the fee-for-service sector. The focus of the research will be on the substitution of outpatient for inpatient care and on how insurance benefit design affects utilization patterns. Insurance claims from three large corporations for a three-year period will provide the data for the analyses. Descriptive analyses will examine who uses services, where they go for treatment and how long they remain in treatment. The relationship between inpatient and outpatient use, given the extent or limitations of the structure of the insurance benefits will also be described. Cox regression techniques will be used to model patterns of utilization over time. Simulations will be performed characterizing how increasing outpatient benefits would affect patterns of utilization and cost. The simulations will be base in part on estimates from the study regarding substitution, and in part on estimates from other studies of own- price response. The issues addressed in the proposed research regarding the substitutability of outpatient for inpatient services and its relationship to insurance benefits have important policy implications for the structuring of insurance benefits to provide appropriate incentives for cost-effective mental health care. We intend to empirically estimate substitution effects and through simulation illustrate the implications for cost.
| Goldstein, J M; Horgan, C M (1988) Inpatient and outpatient psychiatric services: substitutes or complements? Hosp Community Psychiatry 39:632-6 |
