INTRODUCTION

In 2018, the Department of Health and Human Services began inviting state Medicaid reforms to “increase employment and community engagement” among enrollees by requiring work or similar activities as a condition of Medicaid eligibility. As of September 2019, seventeen states have applied for section 1115 Medicaid work requirement waivers, nine of which have been approved (with three currently on hold by federal courts).1 Such programs oblige individuals to complete a specified number of work, job training, job search, or community service hours to maintain Medicaid coverage, with exemptions for pregnant and disabled individuals.

Many work requirement proposals also make exceptions for individuals with substance use disorders (SUD), though this is often conditional on participation in a SUD treatment program.2 Given the high prevalence of SUD within the Medicaid population,3 the availability of SUD treatment may be crucial to maintenance of Medicaid eligibility for individuals in states pursuing work requirements. We sought to quantify the availability of SUD treatment resources in states with and without Medicaid work requirements.

METHODS

We compared SUD prevalence, all overdose deaths, opioid overdose deaths, and SUD treatment availability across work requirement categories (approved, pending, or no work requirement in the state). Data sources were as follows: work requirement policies (Kaiser Family Foundation,1 National Academy for State Health Policy2); SUD prevalence (National Survey on Drug Use and Health); opioid overdose deaths (CDC WONDER); state population (US Census); SUD treatment facility data, including acceptance of Medicaid payment, offering opioid treatment program, and offering opioid maintenance therapy (OMT) (National Survey of Substance Abuse Treatment Services); number of licensed prescribers with waiver to prescribe buprenorphine for opioid use disorder (Substance Abuse and Mental Health Service Administration [SAMHSA] provider locator). Analyses were conducted at the state level using SAS (version 9.4) in September 2019, with t tests or Wilcoxon tests as appropriate. All data (except work requirement status) pertained to 2017. The study was deemed exempt from approval by the IRB.

RESULTS

Prevalence of SUD was similar across states with and without Medicaid work requirements (Table 1). However, states with approved Medicaid work requirement waivers had a higher burden of both overall and opioid-related overdose deaths, compared with states with pending or no work requirements.

Table 1 Characteristics of Substance Use Disorder Policies and Prevalence in States with Approved, Pending, and no Medicaid Work Requirements
Full size table

Availability of SUD treatment facilities was similar across work requirement categories, and a majority of facilities accepted Medicaid payment irrespective of work requirement status (Table 2). For opioids, the median proportion of SUD treatment facilities with a formal opioid treatment program or offering OMT was low across all states, though was generally lower among states with work requirements.

Table 2 Substance Use Disorder Treatment Availability Among States with Approved, Pending, or no Medicaid Work Requirement Waivers
Full size table

Similarly, when scaled by the burden of opioid-related deaths in each state, states with work requirements had lower numbers of treatment facilities offering opioid treatment programs or OMT (Table 2). The median number of clinicians waivered to prescribe buprenorphine was also lower in states with work requirements compared with states without requirements.

DISCUSSION

We found that states with approved Medicaid work requirements generally have both a higher burden of drug overdose deaths and proportionately fewer treatment resources, compared with states without work requirement waivers. Particularly for opioid use disorder, states with approved or pending work requirements had fewer treatment facilities providing recommended treatments and fewer clinicians waivered to prescribe buprenorphine. Study limitations include limited ability to distinguish between treatment resource availability and true access to care, and known incomplete data on waivered clinicians from SAMHSA.4

Given recent work showing that Medicaid work requirements may disproportionately affect individuals with SUD and other behavioral health disorders,5, 6 states should consider policies that account for the limited availability of SUD treatment to prevent disenrollment and worsening health for individuals with SUD.