State Laws Prohibiting Prior Authorization for Medications for Opioid Use Disorder in Private Insurance, 2015–2023

Abstract

Prior authorization (PA) is a barrier to accessing medications for opioid use disorder (MOUD). While private insurance covered about one-third of opioid use disorder patients in 2023, the understanding of PA prohibitions in private insurance remains limited. We synthesized state laws prohibiting PA for MOUD in private insurance in the U.S. The number of states with such prohibitions grew from 2 in 2015 to 22 in 2023, with variations in the scope of these prohibitions. Seven states fully prohibited PA for all MOUD (“full prohibitions”) since the effective date. Fifteen states adopted legislation that still allows PA under some conditions (“partial prohibitions,” including those covering at least 1 of any MOUD, those covering 1 of each MOUD, and those with limitations based on modality/formulation, generic/branded status, PA frequency, prescription duration, and emergency conditions). Among these 15 states, 4 states transitioned from partial prohibitions to full prohibitions; 11 states maintained partial prohibitions, although many have broadened the scope of prohibitions over time. The proliferation of PA prohibitions for MOUD highlights that this is an area of significant state legislative focus.

Introduction

The opioid epidemic in the U.S. has continued for over a decade with few signs of improvement. Medications for opioid use disorder (MOUD), including methadone, buprenorphine, and naltrexone, are effective treatments for opioid use disorder (OUD). However, substantial treatment gaps persist; about 80% of adults with OUD do not receive these evidence-based medications.^1,2

Prior authorization (PA) is a barrier to accessing MOUD. When insurers require PA, healthcare providers must obtain advance approval from the insurer to cover certain medications or services. Although the medication or service can still be delivered when PA is denied, it will not be covered by insurance and patients will be responsible for all costs. PA creates substantial administrative burdens for healthcare providers and can negatively affect patients.^3,4 A recent national survey found that, on average, healthcare providers spent 13 hours completing PA and addressed 39 PA requests weekly.⁵ Of even greater concern, PA was associated with lower medication adherence, worse clinical outcomes, and serious adverse events.^5,6

In the context of OUD treatment, research has noted that PA requirements for MOUD are largely not supported by evidence⁷ and were associated with reduced provider willingness to prescribe,^8,9 fewer OUD outpatient visits,¹⁰ and shortened treatment duration.¹¹ Despite these concerns, PA remains common across insurers.⁷ In Medicaid, 32 states required PA for at least one formulation of buprenorphine in 2021.¹² In Medicare Part D, requiring PA for MOUD was prevalent but has dropped substantially, from over 70% of formularies in 2017 to under 5% in 2019, after the Centers for Medicare & Medicaid Services restricted plans’ use of PA for buprenorphine.^13–15 A similar trend was observed in private insurance: between 2017 and 2021, more formularies covered at least one buprenorphine product without PA, although PA requirements varied by branded/generic status and formulation.^13,16

As part of their response to the opioid epidemic, many states have enacted laws to prohibit the use of PA for MOUD (hereafter “PA prohibitions for MOUD”). These laws are particularly important for enrollees of private insurance plans since they are more likely to face PA requirements and denials than those enrolled in Medicare or Medicaid.^17,18 In 2023, over one-third of patients with OUD were covered by private insurance;^19,20 however, the landscape of PA prohibitions for MOUD in private insurance remains insufficiently explored. Existing research on PA prohibitions for MOUD has primarily focused on Medicare and Medicaid.^21–26 In Medicare, PA prohibition for MOUD was associated with increased treatment initiation and decreased adverse hospital events.^27,28 In contrast, evidence from Medicaid has been mixed.^24,26 Notably, we identified only one study that included private insurance, and the study reported that no state PA prohibition for MOUD covered by private insurance was in place before 2017.²⁹ To improve understanding of the policy landscape, we synthesized the scope and features of state legislative prohibitions for MOUD in private insurance in all U.S. states and the District of Columbia from 2015 to 2023.

Methods

The goal of our analysis was to identify trends in the scope and features of state laws prohibiting PA for MOUD in private insurance in every U.S. state and the District of Columbia, from 2015 to 2023.

Data

Using the legal research database Westlaw, we searched for state laws prohibiting PA for MOUD in every U.S. state and the District of Columbia. Our methodological approach followed best practices for creating longitudinal datasets of state laws.³⁰ For every state, we searched Westlaw’s specific database of state statutes using a pre-specified search string that contained multiple synonyms for PA (e.g., utilization review, preauthorization, preapproval, prior approval, precertification) and MOUD (e.g., medication-assisted treatment, addiction treatment, opioid use disorder, substance use disorder). See Appendix 1³¹ for all search terms. A legal expert reviewed the Westlaw search results and identified relevant statutes for analysis. Specifically, we included state statutes that placed any restriction on the use of PA by private insurers for MOUD, effective at any time from January 2015 to December 2023. We tracked changes in these laws over the study period. Laws that did not apply to private insurance plans—for example, laws that only applied to Medicaid or to private insurers administering Medicaid managed care plans—were excluded. Additionally, because MOUD are mostly prescribed in outpatient settings, we excluded laws that exclusively applied to inpatient settings. Our search focused on state laws as opposed to regulations, allowing us to understand how states’ political will to address PA for MOUD has evolved in recent years.

Analysis

Our analytic approach included data extraction and coding. When a relevant law was identified, a legal expert retrieved the full text of the current law, along with the complete text of all prior versions of the law that were effective since 2015. A legal expert and a content expert of behavioral health policy then examined the text of each law to identify provisions relevant to the following themes: (1) medications for which PA was prohibited, including specifications regarding modality, formulation, and branded/generic status; (2) the frequency with which insurers can require PA; (3) the duration of prescriptions for which PA was prohibited (e.g., no PA for prescriptions of less than a 5-day supply); and (4) clinical circumstances in which insurers can or cannot require PA (e.g., no PA when “an emergency condition” or a chronic condition is present). In addition to MOUD, we identified whether the state laws extended to PA for naloxone, an opioid overdose reversal agent. Although naloxone is not a MOUD, it is often prescribed for patients with OUD to prevent opioid overdose.³² We provided the legal citations and plain language summary of the laws included for analysis in Appendix 2.³¹

Based on the data extracted from Westlaw, we developed coding rules through discussions among a legal expert, a clinical expert, and two content experts in behavioral health policy. A content expert in behavioral health policy and a research assistant independently applied these rules to code the extracted data and met regularly to reconcile their coding. In instances where multiple PA prohibitions were effective in a state at the same time, we coded the law that was most restrictive of PA (i.e., most protective of patients’ access to MOUD). When questions arose, the two researchers met with a legal expert to revise the coding rules and resolve discrepancies. Disagreements between the coding of two researchers occurred in three states (Delaware, Rhode Island, and New York), but were fully resolved through discussions with the legal expert.

Based on the coded data, we classified state laws into “full prohibitions” and 7 categories of “partial prohibitions.” Full prohibitions are state laws that prohibit private insurers from requiring PA for every MOUD (or for every medication approved by the Food and Drug Administration for OUD, as specified in the law). We defined partial prohibitions as laws that allow PA under certain conditions. We classified partial prohibitions into 7 categories: (1) at least 1 of any MOUD: the law only prohibits PA for at least 1 of any MOUD (i.e., at minimum, these laws prohibit PA for one formulation of either methadone, buprenorphine, or naltrexone, but not necessarily all 3); (2) at least 1 of each MOUD: the law prohibits PA for at least 1 product each of methadone, buprenorphine, and naltrexone; (3) limitations on modality/formulation: the law only bans PA for a certain modality or formulation of MOUD (e.g., no PA for oral medications), (4) limitations on generic/branded status: the law only bans PA for certain MOUD based on generic/branded status, (5) limitations on PA frequency: the law allows insurers to require PA for MOUD but limits the frequency with which it can be applied (e.g., once per year for an enrollee); (6) limitations on prescription duration: the law prohibits PA only for MOUD prescriptions of certain durations (e.g., up to 5 days); and (7) emergency conditions: the law prohibits PA for MOUD only in cases of medical emergencies, while still permitting PA for non-emergency conditions.

This study did not require review by an Institutional Review Board because no human subjects were involved.

Limitations

First, our findings may not apply to all private insurance plans. For example, workers’ compensation plans and limited benefit plans are usually exempt from PA prohibitions. Additionally, state PA prohibitions cannot be enforced on employer self-funded plans, for which state law is preempted by the Employee Retirement Income Security Act. Since 2015, about half of privately insured enrollees have been covered by self-funded plans,^33,34 Nevertheless, it remains crucial to examine state PA prohibitions, as they affect a significant portion of privately insured enrollees and can establish norms that restrict the use of PA. Second, private insurers may discontinue PA for reasons other than state legislative mandates, such as consumer demand, cost-effectiveness, municipal ordinances, or agreements with employers or state regulators; we did not capture these mechanisms in our study of state laws. Lastly, our analysis did not capture data on regulatory interpretation or enforcement, which may warrant further research.

Results

The number of states with any PA prohibition for MOUD in private insurance increased from 2 in 2015 to 22 in 2023 (Exhibit 1). Among the 22 states, 7 had full prohibitions on PA since the laws were first implemented (Massachusetts, Maryland, New Jersey, West Virginia, Illinois, Vermont, and Kentucky); 4 transitioned to full prohibition by the end of 2023 (New York, Arkansas, Colorado, and Missouri); and 11 states maintained partial prohibitions, although many of these had removed some limitations by the end of 2023 (Appendix 3).

Exhibit 1. States with prior authorization prohibitions for MOUD in private insurance, 2015–2023.

Source: Authors’ analysis of data from Westlaw.

Notes: PA=prior authorization. MOUD=medications for opioid use disorder. OUD=opioid use disorder. Full prohibitions are state laws that prohibit private insurers from requiring PA for every MOUD (or for every medication approved by the Food and Drug Administration for OUD, as specified in the law). Partial prohibitions are state laws that allow PA under certain conditions.

Of the 15 states that had a partial PA prohibition at any time, we observed variations in the type of partial prohibitions across states (Appendices 3–4).³¹ First, Colorado (2019), New Hampshire, and Virginia only prohibited PA for at least one of any MOUD. Specifically, they did not specify which drug class or product must be exempt from PA, only that one product, at the insurer’s discretion, must be available without PA. Second, Delaware, Maine, and Washington prohibited PA for at least one product of each MOUD, either by naming the specific drugs (e.g., buprenorphine) or drug classes (e.g., opioid agonist), or requiring one product from each therapeutic class to be free from PA. Third, four states had partial prohibitions with limitations on modality/formulation. Specifically, Arkansas (2019–2023) and Montana excluded injectables from PA; Arizona prohibited PA for “at least one modality” of MOUD; Missouri (2018–2021) excluded buprenorphine other than tablets from PA. Fourth, New Mexico and Oregon allowed PA for branded MOUD, as long as a generic version was available without PA. Fifth, four states had partial prohibitions with limitations on PA frequency. Colorado (January-May 2019) prohibited the first PA in each calendar year. As of 2023, Georgia, New Hampshire, and Rhode Island allowed PA for MOUD once per year. Sixth, five states had partial prohibitions with limitations on prescription duration. Colorado (January-May 2019), Delaware (2018–2019), and New York (2017–2020) prohibited PA for MOUD prescriptions up to a 5-day supply. Oregon prohibited PA for the first 30 days of MOUD prescriptions from 2017 to 2021, then amended the statute to prohibit PA for prescriptions shorter than 60 days. Lastly, Delaware (January-August 2018), New York (2017–2020), and Rhode Island (2012–2022) prohibited PA for emergency conditions.

Notably, many states initially imposed partial prohibitions but moved closer to full prohibitions over the study period (Appendices 3–4).³¹ Rhode Island’s PA prohibition for MOUD was initially restricted to emergency conditions, but the restriction was lifted in 2023. New York expanded its PA prohibitions to cover longer prescriptions and nonemergency conditions in 2020. Delaware expanded its PA prohibition to prescriptions longer than five days in 2019, following removal of the emergency condition restriction in 2018. Arkansas initially excluded injectables from the PA prohibitions, but removed the exclusion in 2023. Missouri had excluded non-tablet formulations of buprenorphine from the PA prohibition and lifted this exclusion in 2021.

The move toward full prohibition was not universal. Two states modified their laws in ways that gave insurers more opportunities to use PA for MOUD (Appendices 3–4).³¹ Oregon’s initial PA prohibition (effective in 2017) only applied to prescriptions of 30 days or less. In 2019, a law was amended to allow PA for branded MOUD prescriptions if a generic version was available without PA. Then the state expanded its PA prohibitions to include prescriptions up to 60 days in 2022. New Hampshire, which since 2017 allowed PA once per year per patient, added another restriction in 2020 by allowing PA for MOUD as long as at least one of any MOUD was available without PA.

The number of states that included naloxone in their PA prohibitions for MOUD increased from 0 in 2015 to 8 in 2023 (Exhibit 2). States that prohibited PA for naloxone as of 2023 were Arkansas, Colorado, Delaware, Maryland, Missouri, New York, Rhode Island, and Washington (Appendix 3).³¹

Exhibit 2. States with prior authorization prohibitions for MOUD in private insurance that include naloxone, 2015–2023.

Source: Authors’ analysis of data from Westlaw.

Notes: PA=prior authorization. MOUD=medications for opioid use disorder.

Discussion

We found a proliferation of state laws prohibiting PA for MOUD in private insurance from 2015 to 2023, with notable variations in the scope of these prohibitions. Among the 22 states with PA prohibitions for MOUD as of 2023, 11 had partial prohibitions that allowed PA under certain conditions. While we identified two states that added new limitations, most states with partial prohibitions gradually reduced limitations or moved toward full prohibitions over time. This trend suggests a promising shift toward reducing barriers to MOUD.

Our findings show that PA for MOUD is an area of significant legislative focus and activity, in which many states have passed new laws and/or amended existing laws over a short period of time. The current moment may be a “policy window” for legislative change, where policymakers may be especially receptive to policy evaluation and recommendations. These legislative actions are responding to public priorities and align with public attention to the role of PA in access to care. Media outlets have helped raise public awareness of the negative impacts PA can have on access to MOUD and other health services.^35–38 Research has shown evidence that PA can hinder access to care, harm treatment outcomes, and increase financial and administrative burdens.^6,39,40 Additionally, numerous organizations have advocated for policy reforms to address these issues of PA across all settings and health services. For example, organizations such as the American Medical Association have ongoing initiatives focused on reducing PA requirements and enhancing the transparency and efficiency of the PA review process.⁴¹ Some states also have implemented programs to reduce the burdens of PA generally (not just for MOUD). For example, states like Texas and West Virginia have implemented “gold card” programs that exempt physicians with a track record of high PA approval rates from additional PA requirements for specific services or medications.^42,43

Future research should prioritize examining the effects of PA prohibitions on access to MOUD, treatment outcomes, and patient preference. Since patients may prefer a specific MOUD^44,45 or benefit more from a specific MOUD,^46–48 states and plans should consider ensuring patients can access all MOUD without PA to maximize the benefits of PA prohibitions. For example, emerging evidence suggests that methadone may be a more suitable treatment for patients with fentanyl-related OUD.^49–52 Some patients may benefit from extended-release buprenorphine injections, particularly those with barriers to storing medications or adherence.⁵³ Additionally, treatment outcomes can also vary based on branded vs generic products among some patients. A recent chart review reported that some patients experienced negative events such as withdrawal symptoms or cravings after switching to generic buprenorphine/naloxone products.⁵⁴ For states that prohibit PA for “at least one of any MOUD,” insurance plans can comply with this law by removing PA requirements for a single MOUD product, while continuing to use PA for all others. Moreover, patients usually need to stay on MOUD for at least 180 days to see effective treatment outcomes.^55–57 When PA prohibitions have limitations on PA frequency or prescription durations, prescribers may still need to address PA requests to have patients stay on MOUD after treatment initiation.

We found 8 states that additionally include naloxone (an opioid overdose reversal agent, but not a MOUD) in their PA prohibitions for MOUD. Naloxone is the standard agent used for reversing opioid overdose and is recommended to be prescribed with MOUD.³² While naloxone became available over the counter in 2023,⁵⁸ the out-of-pocket cost was much lower with insurance (on average, $11.56 per prescription with private insurance in 2022)⁵⁹ compared to the over-the-counter price (approximately $50 for two sprays).⁶⁰ State laws that prohibit PA for naloxone may continue to be an important tool for reducing opioid overdose deaths.

Our findings lay the groundwork for future studies evaluating the impact of PA prohibitions for MOUD, including insurer compliance, MOUD access and treatment quality, and reduced harm from OUD. To our knowledge, no studies have examined private insurance plans’ compliance with these prohibitions. A recent study found that the proportion of private insurance formularies covering at least one immediate-release buprenorphine product without PA increased from 83.3% in 2017 to 94.6% in 2021. However, this proportion varied considerably across buprenorphine’s formulations, modalities, and branded/generic status.¹⁶ These results aligned with our findings on the variations in conditions of PA prohibitions. Future studies can build on our data to examine the extent to which PA requirements in private insurance formularies complied with these state laws. More importantly, while some studies have explored the effectiveness of PA prohibitions in improving access to MOUD within Medicaid^24,26 and Medicare,^27,28 such evaluations are currently lacking for privately insured populations.

Conclusion

State legislative PA prohibition for MOUD in private insurance increased substantially from 2015 to 2023. We observed notable variations in the scope of PA prohibitions across states. While many states with such prohibitions still allowed PA under certain conditions, there was a positive trend toward full prohibitions over time. Our findings lay the groundwork for future research on the impacts of state legislative efforts to address the treatment gap in OUD.