The Mental Health Parity and Addiction Equity Act (MHPAEA) Evaluation Study: Impact on Mental Health Services Access, Utilization and Expenditures

The Paul Wellstone and Pete Domenici Mental Health Parity and Addiction Equity Act of 2008 (MHPAEA), which took effect for plans renewing from July 2010 on, significantly changed the law and regulations governing the design and management of mental health and substance abuse (MH/SA) benefits for private insurance products sold to employers with 50 or more employees. MHPAEA represents a sea change in the law, as together with the interim and final regulations, its requirements go well beyond prior federal and state parity legislation. MHPAEA is a federal law, so it applies to self-insured plans that are exempt from state mandates, but unlike the previous federal parity law, MHPAEA affects non-quantitative treatment limits (NQTLs) such as utilization management (e.g., medical necessity review) in addition to benefit design features such as cost-sharing requirements and quantitative treatment limits (e.g., number of visits covered). MHPAEA also covers substance use disorders in addition to mental disorders. The implications of this landmark piece of legislation are not yet known and could be substantially greater than the effects of earlier parity laws. Our NIH-funded study is documenting the impact of the MHPAEA on behavioral healthcare benefit design, care management, utilization, and expenditures by analyzing administrative databases from one of the largest managed behavioral health organizations in the country. The study design uses pre-, transition and post-MHPAEA data to compare changes in study outcomes over time among plans, employer groups and patients expected to experience differential impact of the legislation (e.g., comparing states with strong, weak or no pre-existing parity laws; large vs. small groups; self-insured vs. fully insured firms).  The proposed conference presentation would present results from patient-level analyses examining the association of MHPAEA with changes over time in access (penetration rates), utilization and expenditures of mental health services, using linked insurance claims, eligibility data and plan/employer group information from 2008 to 2013. We currently have these data for 2008-2012 and will update with 2013 data early in 2014.  We use a difference-in-differences approach with regression modeling to adjust for patient, plan, and employer group characteristics.  Logistic regressions will be used for dichotomous dependent variables and two-part models for limited-dependent variables; marginal effects with confidence intervals based on Taylor series expansions will be reported.  Sensitivity analyses will focus on the continuously enrolled subpopulation to estimate longitudinal fixed effects models.  The statistical analyst is currently finalizing the analytic file and running preliminary analyses, so we should have final estimates by June.  Findings from our evaluation will help to inform policymakers considering future modification to the MHPAEA to supplement its provisions (e.g., mandates aimed at preventing employer groups from dropping coverage altogether) and provide an evidence base for employer groups making choices about MH/SA coverage.