The Effects of Provider Networks on Medicaid Enrollees with Mental Health Conditions

Project Summary More than three in every ten Medicaid enrollees has a diagnosed mental health condition, and many in this group face significant barriers accessing the care they need. In most states, Medicaid enrollees with mental health conditions are covered by managed care organizations, which contract with sets of health care providers and facilities to deliver medical care to its members. The size, composition, and stability of these “provider networks” can affect care continuity and treatment outcomes, particularly in mental health, where provider networks tend to be much smaller than those for physical health. The Centers for Medicare and Medicaid Services requires each Medicaid managed care provider network to provide sufficient and timely access to mental health services. However, little evidence exists to help states understand which mental health network attributes affect patient outcomes and how to design, monitor, and regulate provider networks. Against these knowledge gaps, the overall objective of this proposal is to estimate the effects of mental health networks on health care utilization and outcomes for adult Medicaid enrollees with mental health conditions (including those with serious and mild/moderate mental illness). We propose to use the 2016-2023 Transformed Medicaid Statistical Information System (T-MSIS) Analytic Files (TAF), which provides comprehensive enrollee-level claims data related to service utilization and prescription use for all states and the District of Columbia. We hypothesize that network size, provider churn (exit), and provider panel size (a measure of the degree of plan participation) influence enrollees' mental health care access and outcomes. Specifically, our aims are to 1) identify mental health network attributes associated with plan switching among Medicaid enrollees with mental health conditions; 2) determine the effects of changes in network size and network quality on utilization and patient outcomes; 3) estimate the effects of involuntary network disruption on patient outcomes; and 4) assess how telemedicine affects the attributes and measurement of mental health networks. We apply a number of innovations, including the construction of empirical provider networks from administrative claims data, a focus on psychiatrists and psychiatric mental health nurse practitioners, and a robust study design to evaluate the impact of mental health provider network attributes on patient-level outcomes. In collaboration with an advisory committee of Medicaid and mental health policy stakeholders, results will provide actionable evidence to improve access to high-quality mental health services within a constrained mental health delivery system. 1