Open Access to Antipsychotics in State Medicaid Programs: Effect on Healthcare Resource Utilization and Costs among Patients with Serious Mental Illness

TL;DR

This study found that open access to antipsychotics in Medicaid reduced hospitalizations and costs for patients with serious mental illness compared to states with restrictive policies.

Contribution

The study provides new evidence on the impact of Medicaid antipsychotic access policies on healthcare utilization and costs.

Findings

Patients in Michigan had significantly lower SMI-related hospitalization rates compared to non-OA states.

Total SMI-related costs were higher in all non-OA states except Colorado.

Pharmacy costs were higher in Michigan, but inpatient costs were lower.

Abstract

Background: The restrictive consequences of Medicaid formulary restriction policies on antipsychotic medications may lead to higher healthcare utilization and costs among beneficiaries with serious mental illness (SMI). Objectives: This study compared outcomes among patients with SMI accessing antipsychotic medications through state Medicaid programs with open access (OA) policies (Michigan) vs 5 states without Medicaid OA policies (California, Colorado, Florida, Illinois, Wisconsin). Methods: A retrospective analysis was conducted using Kythera Labs Medicaid data (Jan. 1, 2016–Dec. 31, 2023). Outcomes were assessed for patients with SMI (>18 years of age, ≥1 antipsychotic medication claim during the identification period (Jan. 1, 2017–Dec. 31, 2022), ≥1 SMI claim in the 12-month baseline). Continuous medical and pharmacy benefits were required for 12 months pre- and post-index date. Outcomes included SMI-related hospital admissions, length of hospital stay, emergency department and outpatient visits, and associated costs. Results: A greater proportion of beneficiaries with SMI resided in Michigan than in the other states. After matching, significantly more antipsychotics users experienced SMI-related hospitalizations in California (18.25% vs 9.47%, P < .0001), Colorado (11.41% vs 7.33%, P  =.0004), Florida (19.70% vs 10.17%, P < .0001), Illinois (23.57% vs 8.79%, P < .0001), and Wisconsin (15.21% vs 10.02%, P = .0046) than in Michigan. Length of stay was lower in Michigan than in California, Colorado, and Illinois. Inpatient costs related to SMI were significantly lower in Michigan, yet pharmacy costs were higher. Total SMI-related costs were higher in all non-OA states than in Michigan, except Colorado. Discussion: State Medicaid programs without OA to antipsychotics were associated with higher rates of SMI-related resource utilization and costs vs Michigan. Conclusions: Policy makers should consider the potential downstream cost implications of restrictive access policies and evaluate whether OA could result in improved health outcomes for patients and savings for Medicaid programs.

Figures