Abstract
OBJECTIVE: To assess the association between the implementation of 1115 Medicaid substance use disorder (SUD) waivers and changes in Medicaid-paid use of residential treatment and other types of services. STUDY SETTING AND DESIGN: We compared 20 states with SUD waivers to 14 non-waiver states using a staggered difference-in-differences design. Primary outcomes were Medicaid-paid opioid-use disorder (OUD) related residential treatment stays and length of stay (LOS). Secondary outcomes included admissions and LOS for all-cause and OUD-related inpatient stays, psychiatric hospital admissions, emergency department (ED) visits, outpatient visits, and primary care visits. DATA SOURCE AND ANALYTIC SAMPLE: We used the 2016-2021 Transformed Medicaid Statistical Information System (T-MSIS) Analytic Files (TAF). The analytic sample included Medicaid enrollees ages 18-64 with OUD. PRINCIPAL FINDINGS: On average, waiver implementation was associated with an increase in residential treatment stays (estimate: 0.4%; 95% CI: 0.1%-0.7%), OUD-related inpatient visits LOS (estimate: 0.3 days; 95% CI: 0.0%-0.5%), psychiatric hospital LOS (estimate: 1.0 days; 95% CI: 0.6 days-1.4 days), primary care visits (estimate: 3.0%; 95% CI: 1.2%-4.7%), and OUD-related primary care visits (estimate: 2.7%; 95% CI: 0.9%-4.4%); and a decline in all-cause inpatient visits (estimate: -0.9%; 95% CI: -1.9% to -0.0%) and OUD-related inpatient visits (estimate: -0.8%; 95% CI: -1.6% to -0.0%). Results for psychiatric hospital LOS and OUD-related primary care visits were sensitive to adjusting for pre-trends. Among four early-adopting states (Indiana, Louisiana, New Jersey, Virginia), Medicaid-paid residential treatment increased 1-4 years following waiver implementation (e.g., 2-year estimate: 2.8%, 95% CI: 2.5%-3.0%), and inpatient visits declined 1-4 years following waiver implementation (e.g., 2-year estimate: -3.1%, 95% CI: -3.5% to -2.6%). CONCLUSIONS: SUD waivers were associated with a small increase in Medicaid-paid residential treatment and a decline in inpatient visits across states, with changes being concentrated among early-adopting states.