Medicaid Benefits For Addiction Treatment Expanded After Implementation Of The Affordable Care Act

Abstract

The Affordable Care Act (ACA) established a minimum standard of insurance benefits for addiction treatment and expanded federal parity regulations to selected Medicaid benefit plans, which required state Medicaid programs to make changes to their addiction treatment benefits. We surveyed Medicaid programs in all fifty states and the District of Columbia regarding their addiction treatment benefits and utilization controls in standard and alternative benefit plans in 2014 and 2017, when plans were subject to ACA parity requirements. The number of state plans that provided benefits for residential treatment and opioid use disorder medications increased substantially. States imposing annual service limits on outpatient addiction treatment decreased by over 50 percent. Fewer states required preauthorization for services, with the largest reductions for medications treating opioid use disorder. The ACA may have prompted state Medicaid programs to expand addiction treatment benefits and reduce utilization controls in alternative benefit plans. This trend was also observed among standard Medicaid plans not subject to ACA parity laws, which suggests a potential spillover effect.

Decades of research have established that addiction is a chronic disease thatrequires a continuum of clinical services and medications to manage effectively.^1–3 However, addiction treatment has often been subject to limits on benefits and use that are inconsistent with clinical recommendations and more restrictive than those placed on other medical services.⁴

Such limitations have been a long-standing problem in Medicaid, the nation’s primary health insurance program for low-income Americans and the largest payer for addiction treatment. For example, roughly half of state Medicaid programs limit use of outpatient addiction treatment services regardless of medical necessity.⁵ An even greater number require preauthorization to access Medicaid benefits for many common addiction treatment services and medications.^5,6 Such policies restrict access in meaningful ways, and enrollees subject to such limits are less likely to enter and remain in treatment.^7,8

The Affordable Care Act (ACA) ushered in landmark reforms to Medicaid coverage for addiction treatment in 2010. The legislation established a set of essential health benefits that states must offer within “alternative benefit plans,” which are distinct from traditional Medicaid benefits plans, in states that opted to expand eligibility for Medicaid under the ACA. Because the essential health benefits include coverage of addiction treatment, enrollees newly eligible under Medicaid expansion programs have access to it. The ACA also required that benefits for addiction treatment under an alternative benefit plan be equivalent to or better than those offered by the state’s designated essential health benefits benchmark plan (typically, a small-group insurance plan or government health plan).

The ACA also extended the reach of the Paul Wellstone and Pete Domenici Mental Health Parity and Addiction Equity Act of 2008 to alternative benefit plans, as well as to benefits in Medicaid managed care plans carved out to fee-for-service.^4,9–11 Specifically, the 2008 law prohibits such plans from imposing limits on addiction treatment benefits that are more restrictive than those placed on other medical services.

In 2016 the Centers for Medicare and Medicaid Services (CMS) issued a final rule to provide direction to states on how to implement parity regulations in their Medicaid programs.¹¹ States were required to compare quantitative (for example, annual service limits and copayments) and nonquantitative (such as preauthorization and concurrent review) limits on addiction treatment benefits with those placed on other medical services and to address all compliance deficiencies revealed by their comparison by October 2, 2017. States must monitor their compliance with the parity regulations and submit summary reports to CMS.

Taken together, these regulations have the potential to expand treatment access by enhancing addiction treatment benefits and reducing limits on their use. In 2011, the year after ACA implementation, the proportion of large private insurance plans imposing annual service limits on addiction treatment declined from 26 percent to 3 percent.¹² A more recent study of private health insurance plans found that use of quantitative treatment limits virtually disappeared after the mental health parity law was implemented.¹³

It is important to ascertain the extent to which addiction treatment benefits and the use of utilization controls within the country’s fifty-one Medicaid programs changed after implementation of the ACA. Gaining this understanding is particularly important in light of the nation’s escalating opioid epidemic.¹⁴ In this study we assessed changes in addiction treatment benefits and utilization controls on their use in standard Medicaid and alternative benefit plans. Although standard Medicaid plans are not subject to essential health benefits and parity provisions, we included them in light of early evidence suggesting that some state Medicaid programs have sought to provide similar benefits for enrollees in standard Medicaid and alternative benefit plans.¹⁵

Study Data And Methods

DATA SOURCES

We collected data for this study as part of the National Drug Abuse Treatment System Survey, a nationally representative panel study of addiction treatment programs in the US that has been conducted periodically since 1984. To gather data on Medicaid coverage and use of utilization controls for addiction treatment, we conducted a fifteen-minute, internet-based survey of Medicaid programs in the fifty states and the District of Columbia. The University of Chicago Survey Lab conducted the first wave of the survey in the period November 2013–December 2014 and the second wave in May–December 2017. In each wave, state Medicaid directors were mailed a packet that contained a description of the study, an invitation to participate, and a request to designate a knowledgeable staff person to fill out the survey. To encourage participation, the Survey Lab followed up by phone and email with directors who did not respond. In each wave, forty-seven Medicaid programs responded, for a response rate of 92 percent. Only one state opted not to participate in either wave of the survey.

To reduce burden on respondents, in the first wave the survey was prepopulated to the extent possible using data from the fifty-one Medicaid plans on file with CMS and from state Medicaid program websites. The survey instructions asked respondents to review the data for accuracy and make any necessary modifications. In 2014, information on coverage for methadone, oral and injectable naltrexone, and buprenorphine came from data collected by the American Society of Addiction Medicine. The society used data collected from state Medicaid agencies and published state drug formularies (also referred to as preferred drug lists) to determine which medications were covered and whether limits were imposed.

In the second wave, the survey was prepopulated with each Medicaid program’s responses from the first wave. Respondents were asked to review the survey and make revisions as needed. For states that did not complete the first survey wave, a research team member added information gleaned from a review of publicly available resources on state Medicaid coverage for addiction treatment. Medication data were collected through a review of published state drug formularies using the method employed by the American Society of Addiction Medicine to collect data in 2014. The final study data included the total population of the Medicaid programs in all fifty states and the District of Columbia, which obviated the need for inferential tests.

MEASURES

Each wave of the survey collected data on coverage for the following treatment services: individual outpatient, group outpatient, recovery support, intensive outpatient, short-term residential, long-term residential, and medically managed detoxification. The waves also collected data on coverage for four medications approved by the Food and Drug Administration (FDA) for the treatment of opioid use disorder: methadone, oral and injectable naltrexone, and buprenorphine.

For each service and medication noted above, dichotomous variables indicated whether states reported use of the following utilization control mechanisms: cost sharing, including copayments and deductibles; preauthorization; and annual service limits. These mechanisms were selected because they have frequently been used by state Medicaid programs to monitor and restrict use of addiction treatment,⁵ and because they are frequently the subjects of public attention in policy debates about addiction and mental health parity.¹⁰

Each state Medicaid program responded to questions about benefits and utilization controls separately for its standard fee-for-service Medicaid plans, available to enrollees eligible for Medicaid under criteria set before the ACA, and for its alternative benefit plan for enrollees newly eligible under the ACA’s Medicaid expansion. Medicaid programs in five states—Hawaii, Idaho, Michigan, Tennessee, and Washington—reported that 100 percent of their enrollees participated in contracted managed care entities, which directly set all policies regarding utilization controls. Consequently, these states were not included in this analysis.

LIMITATIONS

Our study had some limitations. First, it did not assess benefits in Medicaid managed care plans. Although all managed care organizations must cover a minimum set of services defined by states, they often have some discretion with regard to benefits and use of utilization controls. Consequently, our data captured regulatory minimums stipulated by state Medicaid programs. Because our data captured minimum benefit policy restrictions imposed by the states, we may have underestimated the extent of restrictive policies on Medicaid benefit coverage.

Second, the data collection method relied on information specified in state Medicaid plans submitted to CMS and other publicly available information, and a review of the accuracy of these data by appropriate staff members in each state’s Medicaid agency. Although this twopronged approach was designed to minimize data error, information provided either through the state plan amendments or from Medicaid agency staff could be inaccurate. Moreover, survey nonresponse necessitated the use of public records for four states in each wave of the survey.

Finally, states may have been planning or anticipating changes to their addiction treatment benefits that would not have been captured in the relatively short time frame of this study. For example, because final parity regulations were issued in early 2016, states might not have completed revisions to their plans by the second wave of the survey, which began in May 2017.

Study Results

STANDARD BENEFIT COVERAGE

Coverage for addiction treatment generally improved during the study period. Coverage grew for individual outpatient (from forty-eight to forty-nine state plans), group outpatient (from forty-nine to fifty), recovery support (from twenty-four to twenty-six), and intensive outpatient (from forty-two to forty-three) treatment services (exhibit 1). More significant gains were observed in coverage for short-term residential (from thirty-one to thirty-six) and long-term residential (from twenty-two to twenty-six) treatment services. The largest increases in coverage were reported for medications, with coverage for methadone increasing from thirty-three to forty state plans and coverage for oral naltrexone increasing from thirty-seven to forty-five. No changes were observed in coverage for detoxification treatment services or injectable naltrexone. All fifty-one plans covered buprenorphine in 2014 and 2017.

EXHIBIT 1. Number of standard Medicaid plans reporting benefits for selected addiction treatment services and medications, 2014 and 2017.

SOURCE Authors’ analysis of data for 2014 and 2017 from the National Drug Abuse Treatment System Survey. NOTES The data shown are for the Medicaid programs of the fifty states and the District of Columbia.

ALTERNATIVE BENEFIT COVERAGE

Trends in coverage for addiction treatment in the alternative benefit plans of the thirty-one states (including D.C.) that had expanded Medicaid by 2017 were similar (exhibit 2). However, this finding resulted, in part, from an increase in the number of states that offered an alternative benefit plan over the study period. More than 90 percent of states aligned coverage in their alternative benefit plans with that in their standard Medicaid plans. Among the few states that did not offer identical coverage, most made coverage in their alternative benefit plans conform to that in their benchmark plan.

EXHIBIT 2. Number of Medicaid alternative benefit plans reporting benefits for selected addiction treatment services and medications, 2014 and 2017.

SOURCE Authors’ analysis of data for 2014 and 2017 from the National Drug Abuse Treatment System Survey. NOTES The data shown are for all alternative benefit plans offered by the twenty-six states and the District of Columbia that expanded eligibility for Medicaid in 2014 and the thirty states and the District of Columbia that had expanded Medicaid by 2017. “In 2014” and “in 2014 and 2017” refer to plans that states offered in those years. “After 2014 and in 2017” refers to plans that states began offering after 2014 and offered in 2017.

COPAYMENTS

The proportion of states requiring copayments within their alternative benefit plans increased from 2014 to 2017. This finding is explained in part by the introduction of new alternative benefit plans in 2017 by states that expanded Medicaid after 2014 and required copays in the new plans. No substantial changes in copays were observed in standard Medicaid plans. Across both types of plans, states were most likely to require copays for medications, including buprenorphine and oral and injectable naltrexone. The proportion of services and medications covered by states that required copays increased from 30 percent to 34 percent in standard Medicaid plans and from 16 percent to 26 percent in alternative benefit plans (exhibit 3).

EXHIBIT 3. Percent of addiction treatment services and medications subject to selected utilization controls, by plan type, 2014 and 2017.

SOURCE Authors’ analysis of data for 2014 and 2017 from the National Drug Abuse Treatment System Survey. NOTES The data shown are for all alternative benefit plans (ABPs) offered by the twenty-six states and the District of Columbia that expanded eligibility for Medicaid in 2014 and the thirty states and the District of Columbia that had expanded Medicaid by 2017.The data shown are also for all standard Medicaid plans of the fifty states and the District of Columbia except those in which all addiction treatment services are contracted to managed care entities (Hawaii, Idaho, Michigan, Tennessee, and Washington).

PREAUTHORIZATION

Use of preauthorization declined from 2014 to 2017 for every addiction service and medication in both standard and alternative benefit plans. The most substantial declines in preauthorization were observed for medications for opioid addiction treatment. In particular, the proportion of state plans requiring preauthorization declined from 48 percent to 40 percent for methadone, 43 percent to 26 percent for oral naltrexone, 72 percent to 51 percent for injectable naltrexone, and 86 percent to 69 percent for buprenorphine (online appendix exhibit A1).¹⁶ Preauthorization for recovery support and short-term residential treatment services also declined substantially during the study period. These trends were more pronounced in the alternative benefit plans (appendix exhibit A1).¹⁶ The overall proportion of services covered by state plans that required preauthorization declined from 56 percent to 46 percent in standard plans and from 55 percent to 45 percent in alternative benefit plans (exhibit 3).

ANNUAL SERVICE LIMITS

Use of annual limits on services sharply declined across most services and medications in both standard and alternative benefit plans. The largest declines were observed for outpatient treatment services. In standard plans, the proportion of states imposing such limits decreased from 49 percent to 23 percent for individual outpatient, from 48 percent to 22 percent for group outpatient, and from 44 percent to 20 percent for intensive outpatient treatment services (appendix exhibit A1).¹⁶ Reductions were of similar magnitude in alternative benefit plans: from 35 percent to 15 percent for individual outpatient, from 33 percent to 14 percent for group outpatient, and from 40 percent to 17 percent for intensive outpatient treatment services (appendix exhibit A1).¹⁶ Overall, the proportion of covered services and medications subject to annual limits dropped from 34 percent to 19 percent in standard plans, and from 25 percent to 13 percent in alternative benefit plans (exhibit 3).

Discussion

Our findings indicate that Medicaid benefits for addiction treatment improved over a relatively short period of time. The proportion of state plans that provided benefits for residential treatment and opioid use disorder medications increased from 2014 to 2017. These changes may be a reflection of states’ efforts to improve benefits for addiction treatment in response to the opioid epidemic. Increasing demand for medication for treatment of opioid use disorder may have spurred decisions by state Medicaid programs to begin covering methadone and naltrex-one. The epidemic has also heightened awareness regarding financial barriers to accessing residential treatment of addiction in many states.¹⁷ Changes in their benchmark plans may also have prompted some states to expand benefits to comply with the essential health benefits provisions of the ACA. States were required to comply with these provisions at the time alternative benefit plans were introduced in 2014 but had the option to change their benchmark plan effective January 1, 2017. Roughly half of the states elected to do so.

The use of utilization controls for addiction treatment declined noticeably after implementation of parity regulations within state Medicaid expansion programs. Such reductions occurred not only in alternative benefit plans (which were subject to the parity regulations) but also in standard Medicaid plans (which were not).

Policy Implications

Although this study cannot explain the underlying political or policy reasons for the observed spillover effect, two mechanisms seem plausible. First, changes that states made to comply with parity regulations for Medicaid managed care and alternative benefit plans may have indirectly reduced utilization controls in standard Medicaid plans. Survey respondents in more than 90 percent of the states indicated that their state offered the same benefits in its standard and alternative benefit plans. Such uniformity may reduce the administrative complexity of operating two plans with different benefit and utilization policies. Thus, achieving compliance with the new parity regulations that applied to alternative benefit plans resulted in the spillover effect we observed, through which enrollees in standard Medicaid plans also benefited.

The ACA’s requirement that states broaden the application of parity requirements within Medicaid managed care plans may have also contributed to a spillover effect. Before the ACA, only utilization controls imposed directly by contracted Medicaid managed care entities were subject to parity regulations.¹⁸ The ACA expanded the reach of the parity regulations to encompass all utilization controls imposed by Medicaid managed care plans.¹⁹ In its final rule on the application of the requirements in the Mental Health Parity and Addiction Equity Act to Medicaid,¹¹ CMS stipulated that if a state uses managed care to provide any health care benefit within a given plan, then parity applies to all benefits in that plan, including those covered under fee-for-service financing. Consequently, states with any managed care involvement in their plans must ensure that all addiction treatment—even when administered through fee-for-service—complies with parity regulations.

The vast majority of state Medicaid programs cover at least some enrollees through managed care plans, and many continue to carve out at least some services for payment through fee-for service. Some states may have elected to reduce the use of utilization controls for addiction treatment services paid for by fee-for-service across all plans, instead of imposing more stringent restrictions on enrollees in plans not subject to parity requirements.

Second, states may have loosened utilization controls on addiction treatment in response to the nation’s opioid crisis. As opioid-related overdoses and deaths have rapidly increased, many states have turned to Medicaid as a major tool to use in responding to the epidemic.¹⁴ The ACA’s massive federal infusion of support for Medicaid has facilitated this action. States have reduced their use of annual service limits that can arbitrarily restrict access to critically important treatment services regardless of medical necessity. Preauthorization for these treatments also dropped substantially over the study period.

Despite improvements in addiction treatment benefits and reductions in controls on their use, important gaps remain. With fifteen states prohibiting coverage for short-term residential treatment, and twenty-five states providing coverage for long-term residential treatment, people with severe addiction who require crisis stabilization and access to a structured, supportive recovery environment are left with few options. The federal Institutions for Mental Diseases (IMD) exclusion bans Medicaid programs from reimbursing services from specialty inpatient behavioral health providers with more than sixteen beds. State and federal lawmakers are exploring ways to modify this policy through waivers and new federal legislation.²⁰

Additionally, our findings reveal substantial variation in benefits and use of utilization controls across the FDA-approved medications for treatment of opioid use disorders. For example, coverage for methadone maintenance continues to lag behind coverage for other medications, despite its lower cost and the strong evidence that the structure of an opioid treatment program is necessary for many patients with an opioid use disorder.²¹ The higher risks related to misuse of methadone compared to other medications, such as naltrexone and buprenorphine,²² may play a role in states’ decisions not to cover methadone. However, stigma is also a likely driver of these disparities in coverage. The specious perception persists among some policy makers, the public, members of the recovery community, and others that these medications “just substitute one addiction for another.” Buprenorphine and injectable naltrexone were more likely to be subject to prior authorization than other FDA-approved medications for treatment of opioid use disorders (appendix exhibit A1).¹⁶ This could be due to cost concerns, as buprenorphine and injectable naltrexone are typically more expensive than methadone and oral naltrexone.²³

Moving forward, states may continue to make improvements in coverage for addiction treatment. Implementation of the ACA’s parity provisions has required state Medicaid programs to make complex and far-reaching changes within their benefit plans. CMS has provided technical assistance and grants to states to ensure effective implementation. Yet some states will require additional time to achieve full compliance with the CMS final rule.¹¹

Conclusion

The study findings provide evidence of an increase in Medicaid benefits for addiction treatment and of a major reduction in utilization controls after implementation of the essential health benefits and parity regulations mandated by the ACA. The findings also point to the success of this policy in increasing the availability of Medicaid benefits across the continuum of treatment for addiction. ■