https://orcid.org/0000-0002-7750-2397
Figures
Abstract
Background
This study aimed to describe state- and urbanicity-stratified differences in three opioid use disorder (OUD) treatment metrics among Medicaid beneficiaries in 25 states that implemented Medicaid expansion under the Affordable Care Act by the end of 2014.
Methods
Using data from 2019, we identified Medicaid beneficiaries with OUD based on ICD-10 diagnosis codes. We then calculated the percentage of beneficiaries who met criteria for three metrics measuring OUD treatment quality, both overall and stratified by state and urbanicity type. The OUD treatment quality metrics considered were: (1) initiation of medication for opioid use disorder (MOUD) treatment, (2) engagement with OUD services, and (3) retention on MOUD treatment.
Results
Across states, we found that a median of 26.2% of beneficiaries initiated MOUD within 14 days of their OUD diagnosis date in the claims data. A median of 15.8% of beneficiaries engaged with OUD treatment services by initiating MOUD treatment within 14 days of their OUD diagnosis and receiving at least 2 distinct OUD-related services within 30 days of their MOUD initiation date. Among initiators, a median of 30.8% were retained on MOUD treatment for a minimum of 180 days. However, there was considerable heterogeneity in these three metrics across states; New Hampshire and West Virginia were found to have the highest overall performance out of the states considered. With respect to urbanicity, we found that rural and suburban areas had higher percentages of beneficiaries who met our three treatment quality metrics compared to urban areas.
Citation: Andrews RM, Mhasawade V, Maity S, Hung A, Liu R, Ross RK, et al. (2026) Variation in the quality of opioid use disorder treatment in the Medicaid population in 2019. PLoS One 21(3): e0341739. https://doi.org/10.1371/journal.pone.0341739
Editor: David T. Zhu, Virginia Commonwealth University School of Medicine, UNITED STATES OF AMERICA
Received: October 8, 2025; Accepted: January 12, 2026; Published: March 4, 2026
Copyright: © 2026 Andrews et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Data cannot be shared publicly because of restrictions imposed by the Centers for Medicare & Medicaid Services (CMS). Data are available from CMS for researchers who wish to purchase it. Please see https://resdac.org/ for additional information.
Funding: This work was supported by the National Institutes of Health [R01DA053243; R01DA056407]. The funder played no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Opioid use disorder (OUD) is a major risk factor for several adverse outcomes, including fatal overdoses, all-cause mortality, and engagement with the criminal justice system [1]. Medications for OUD (MOUD) effectively reduce risk of relapse, overdose, and death. However, research shows that initiation of and retention on MOUD is low, likely due to a number of substantial structural barriers [2–4]. In response, some burdensome requirements of opioid treatment programs that were relaxed during the COVID-19 pandemic recently became permanent, including telehealth, take-home medication supply, and prescribing flexibilities [5,6]. Nevertheless, access to MOUD treatment has been found to vary across the United States. For example, although studies suggest that federal policy reforms have increased the overall reach of MOUD (particularly buprenorphine) [7,8], other studies have observed state-by-state differences in MOUD prescribing patterns and access, even after these policies went into effect [4,9–16]. Therefore, ongoing research is needed to improve understanding of state-level differences that can facilitate development of tailored policies, programs, or health system infrastructure to address gaps in MOUD treatment.
In addition, studies have shown that despite rural areas experiencing significantly more drug overdose deaths and opioid-related harms compared to urban areas, OUD patients living in rural areas face more barriers to receiving MOUD treatment (e.g., provider shortages and fewer MOUD-dispensing healthcare facilities compared to patients living in more urban areas), even with ongoing efforts to expand OUD-related services (including MOUD) in these areas [17–22]. Taken together, this suggests that aggregate summaries of MOUD treatment patterns may overlook important geographic disparities that could be targeted with future interventions.
Relatedly, Medicaid is the largest payer for OUD services in the United States, covering 40–50% of all non-elderly OUD patients [23,24]. Because Medicaid operates as a federal-state partnership, there are documented differences in access and delivery of MOUD across different states’ Medicaid populations [13,14,25]; there are similar reports of differences in MOUD access and delivery for Medicaid patients residing in urban versus rural areas [15,26]. Therefore, capturing geographic-related heterogeneity in MOUD treatment within the Medicaid system may be policy-relevant.
In this study, we describe state- and urbanicity-stratified differences in three MOUD treatment metrics among Medicaid beneficiaries in 25 states that implemented Medicaid expansion under the Affordable Care Act by the end of 2014. Specifically, we considered three quality metrics related to MOUD treatment from the National Quality Forum that are measures of 1) initiation of MOUD, 2) engagement in care, and 3) retention in MOUD [27,28]. These aspects of OUD treatment are three out of the four components of the treatment cascade within the OUD Cascade of Care Model, which is a public health framework for identifying targets at the local and national level to improve the prevention and treatment of OUD and track the progress of these efforts [28,29].
Methods
Our study included de-identified data from Medicaid T-MSIS Analytic Files (TAF): Demographics, Other Services for 2019. These data include adult, non-dual eligible Medicaid beneficiaries aged 19−64, and we restricted our sample to those diagnosed with OUD (based on ICD-10 diagnosis codes) in 25 states that implemented the Affordable Care Act (ACA) by 2014: AR, AZ, CA, CT, CO, DE, HI, IA, IL, KY, MA, MD, MI, MN, ND, NH, NJ, NM, NY, OH, OR, RI, VT, WA, WV; MD was excluded due to unreliable diagnosis code data, as described in the Medicaid Data Quality Atlas) [30]. We included these states because they covered nearly all non-elderly, low income (< 138% of the Federal Poverty Limit) adults, and the group that was newly eligible for Medicaid under the ACA represents a high-risk group for OUD (e.g., younger, male, and low-wage workers) [31,32]. We classified beneficiaries’ residential county as urban, suburban, or rural using Rural-Urban Continuum Codes (RUCC). Patients missing a RUCC classification were excluded from analyses stratified by urbanicity. All analyses were conducted between March 1, 2025 and July 31, 2025.
To measure OUD treatment quality, we used three metrics: (1) the percentage of patients with an OUD diagnosis who were prescribed buprenorphine, extended-release injectable naltrexone, or methadone within 14 days of their OUD diagnosis (hereafter, “MOUD initiation”); (2) the percentage of patients with an OUD diagnosis who initiated MOUD within 14 days of their OUD diagnosis and received at least 2 services within 30 days of their MOUD initiation (hereafter, “engagement with OUD services”); (3) the percentage of patients initiating MOUD treatment who received at least 180 days of continuous MOUD therapy, where continuous therapy was defined as not having a gap in MOUD of more than 7 days (hereafter, “retention in MOUD treatment”) [27,33,34]. These treatment measures comprise key components of the OUD cascade of care, which tracks progress through an evidence-based treatment episode [28,29].
Each of these percentages were defined as the number of Medicaid beneficiaries who met specific OUD treatment criteria divided by the total number of beneficiaries who were eligible to meet these criteria; Table 1 provides details on the numerator and denominator for each treatment quality metric. For each metric, we calculated both pooled (using all 25 states) and state- and urbanicity-stratified percentages.
This study was approved by the Columbia University Institutional Review Board. All analytic code to replicate the analyses is available at: https://github.com/CI-NYC/OUD-tx-state-year-variability, along with all diagnostic codes used to identify OUD-related Medicaid claims.
Results
The median percentages and the interquartile range (IQR) of each treatment quality metric across states are presented in Table 2. The distributions of the percentage meeting each quality metric across states and urbanicity groups are represented in Figs 1 and 2, respectively. Additional details, including state- and urbanicity-specific cell counts, are given in S1 and S2 Tables, respectively.
OUD = opioid use disorder. * Indicates that data were excluded due to small cell sizes.
OUD = opioid use disorder.
We estimated that a median of 26.2% of Medicaid beneficiaries in our sample with OUD (N = 155,135) initiated MOUD treatment within 14 days of their diagnosis, with an IQR of 16.3–35.6% across all 25 states (Table 1). Iowa was found to have the lowest percentage of MOUD initiators (10.9%), while West Virginia (55.8%) and New Hampshire (56.5%) had the highest (Fig 1). When we stratified by urbanicity, we found that 5,221 (3.4%) beneficiaries had missing RUCC codes and were therefore excluded. Among the remaining 149,914, approximately one third of those living in rural and suburban areas initiated MOUD treatment within 14 days of a diagnosis (34.9% and 32.4%, respectively); and approximately one quarter (24.9%) of those living in an urban area did (Fig 2).
We found that a median of 15.8% (out of 142,746 beneficiaries eligible) engaged with OUD services within 44 days of their diagnosis, with an IQR of 7.7–23.3% across 25 states (Table 2). West Virginia had the highest level of engagement with services (38.7%), while Hawaii had the lowest (3.2%). When looking at engagement by urbanicity, we found that 4,783 (3.4%) beneficiaries had a missing RUCC code and were excluded. Among the remaining 137,963 beneficiaries, those in rural areas had the highest engagement with OUD services (24.8%) compared to those in suburban (20.5%) and urban (15.8%) areas (Fig 2).
When considering retention in MOUD treatment, we excluded data from two states – Hawaii and North Dakota—because of small cell sizes. Among the remaining 23 states, we found that a median of 30.8% of those who initiated MOUD (N = 27,959) were retained on MOUD for at least 180 days, with an IQR of 22.0–36.4% across the 22 states (Table 1). Arizona was found to have the lowest retention percentage (10.1%), whereas New Hampshire (47.1%) and West Virginia (49.1%) were found to have the highest percentages (Fig 1). Stratifying by urbanicity, we found that 682 (2.4%) were missing a RUCC code and were excluded. Among the remaining 27,277 beneficiaries, the percentage of those retained in MOUD treatment was 39.2% for rural areas, 35.4% for suburban areas, and 26.7% for urban areas (Fig 2).
Discussion
This descriptive study aimed to estimate MOUD treatment quality metrics among adult Medicaid patients in 2019 across 25 states that expanded Medicaid eligibility under the ACA by 2014. These metrics captured three out of the four components of the treatment cascade within the OUD Cascade of Care model. We found that approximately one-fifth of beneficiaries engaged with MOUD treatment services per the definition (initiated MOUD treatment within 44 days of their OUD diagnosis and received at least 2 services within 30 days of their MOUD initiation). Percentages of initiation and retention in MOUD treatment were higher, with approximately a quarter of beneficiaries meeting criteria across all 25 states. However, there was noticeable variability across states in the percentage of beneficiaries who met the metrics’ criteria. At the same time, there was moderate to high correlation between the relative ranking of each state according to these percentages, indicating that states that perform well according to one MOUD treatment metric tend to perform well on others. When looking across urbanicity subgroups, we found that rural areas had the highest percentage of beneficiaries meeting MOUD treatment criteria (followed closely by suburban areas), while urban areas had the lowest.
Our overall results are largely consistent with values reported by other studies. One study examining commercially insured patients in 2014 used an Initiation metric identical to ours and reported a slightly higher percentage of 40.9% in its OUD-specific subgroup [35]. This study also reported an engagement metric of 21.6%, slightly higher than in our study. Among Medicaid enrollees across 11 states from 2014 to 2018, another study reported the percentage of MOUD patients being retained on MOUD treatment for a continuous 180 days as 47.8% −57.1%, which was considerably higher than we found [15]. Given the heterogeneity between states that we observed, it is possible that this difference is at least partially driven by the different states considered (comparing [15] with our study, we examined only 5 of the same states – Delaware, Kentucky, Michigan, Ohio, and West Virginia).
We observed state-to-state variability in all three metrics. Similar variability was reported across 11 states by Brown et al., who propose that this variability may be driven by demographic differences [15]. Another explanation for this variability could be state-specific differences in availability of providers and treatment programs. Relatedly, the Drug Addiction Treatment Act of 2000 required that healthcare providers obtain a waiver to prescribe buprenorphine (called the X waiver). This waiver was required for physicians and some advanced practitioners who received training in addiction medicine, and once the waiver was obtained, the number of patients who could be prescribed buprenorphine was capped at 30 for the first year (and up to 275 thereafter) [36]. These administrative hurdles may have differentially affected MOUD across states in 2019, due to, for example, primary care physician shortages. Another explanation for the observed state-by-state heterogeneity in OUD treatment quality could be differences in state-level initiatives targeting OUD. It is notable that West Virigina and New Hampshire consistently were ranked highly for MOUD initiation, engagement, and retention, given that these states also rank among the hardest hit by the opioid epidemic. For example, it is estimated that West Virginia had the highest opioid-related overdose death rate in the United States in 2017 (49.6 per 100,000 residents) and New Hampshire had the third highest rate during this same period (34.0 per 100,000 residents) [37]. Increased awareness of overdose deaths may have brought OUD and MOUD treatment greater attention, which may have resulted in a greater percentage of those with OUD seeking treatment in these states. Relatedly, both West Virginia and New Hampshire introduced policies to combat OUD and expand access to MOUD. In West Virginia, the Comprehensive Opioid Addiction Treatment (COAT) program started in 2005 to provide evidence-based care to those diagnosed with an OUD by combining MOUD management consultations with group-based therapy to treat OUD. Studies have shown that the COAT program has high patient retention, and by treating patients in groups (versus individually), COAT addresses some of the roadblocks to treatment in a predominantly rural state (e.g., provider shortages) [38]. Relatedly, in 2018, West Virgina’s Medicaid began covering methadone for the first time, resulting in a large number of patients receiving OUD diagnoses and MOUD treatment [39]. In New Hampshire, a hub-and-spoke system called “The Doorway” was introduced in 2019 with funding from SAMHSA to improve access to treatment, recovery, and harm-reduction services for individuals with an OUD or other substance use disorder [40]. Overall, The Doorway improved many aspects of OUD treatment in New Hampshire, including expanded access to MOUD. Of relevance to our study, the majority of patients served by The Doorway were Medicaid patients [40]. However, it is important to note that although it is likely that these policies enacted around the time of our data collection helped patients with an OUD in West Virginia and New Hampshire, we have insufficient evidence to claim that these programs are the reason for them being top performers in our study relative to other states.
Our urbanicity-stratified results diverged from previous literature, which had found that rural areas fell behind urban areas in terms of MOUD treatment. For example, a recent systematic review of 18 studies found that those in rural areas had to travel farther to access MOUD and to see OUD treatment providers, which was associated with lower MOUD treatment for individuals with an OUD living in rural areas [18]. Another study found that although rural and urban hospitals had similar rates of OUD screening, those in rural areas were less likely to offer MOUD and other addiction services [17]. In addition, a large study using data from the 2021 National Survey on Drug Use and Health (NSDUH) found that OUD patients residing in non-metropolitan areas had 69% lower odds of receiving MOUD compared to those in metropolitan areas [41]. There could be several potential explanations for why our results diverged from prior studies showing differences in MOUD treatment by urbanicity. One possibility is that primarily rural states with state-wide, OUD-focused programming (e.g., West Virginia and New Hampshire, as mentioned above) led us to find that rural and suburban areas perform better than urban areas in terms of OUD treatment quality. Another possibility is related to our focus on the Medicaid population; rural buprenorphine (a type of MOUD) prescribers are more likely than urban prescribers to accept Medicaid insurance [42]. A third possibility is that, given the barriers to healthcare in rural areas described above, patients living in rural areas with OUD may be less likely to receive an OUD diagnosis than patients living in urban areas with OUD. This urban-rural discrepancy could be especially pronounced for Medicaid-insured patients given that barriers to accessing care would be steeper in the context of socioeconomic challenges. If true, then we would expect the patient populations comprising the denominators of our metrics to differ between urban and rural areas. For example, Medicaid patients in rural areas who have an OUD diagnosis in the claims data may be more treatment-motivated, on average, than otherwise clinically similar Medicaid patients in urban areas who have an OUD diagnosis in the claims data. Thus, our results more accurately describe MOUD treatment quality among rural OUD patients who have been identified and diagnosed with OUD. These conjectures and limitations should be explored in additional studies, including model-based studies that aim to provide empirical evidence for the drivers of MOUD treatment quality by urbanicity.
A strength of our study is that we were able to include Medicaid beneficiaries from 25 states in the US, which allowed us to explore both state and geographic trends in MOUD treatment. Relatedly, because the Medicaid expansion population comprises a high-risk group for OUD (e.g., low-income, childless adults), studying OUD treatment quality within this sample allowed us to capture trends that likely have high relevance for MOUD treatment. Another strength of our study is the reporting of multiple treatment metrics from the OUD Cascade of Care model. Although prior work has used this framework to study OUD and MOUD treatment in electronic health records [43], to our knowledge our study is the first to report on these metrics within a purely Medicaid patient sample.
However, there are several important limitations to our work. First, our results may not generalize to different populations of OUD patients outside of the 25 state Medicaid samples that we used. Second, our results are based on cross-sectional data from 2019, which prevented us from identifying new OUD diagnoses and examining trends in MOUD treatment quality over time. Relatedly, Medicaid accessibility, funding, and policy are constantly evolving, which may limit our ability to interpret 2019 findings as being representative of OUD treatment quality for Medicaid beneficiaries in the present day. This is especially true given key policy changes that took effect after 2019: 1) the Consolidated Appropriations Act, enacted in 2023, eliminated the need for providers to apply for the “X waiver” to prescribe buprenorphine for the treatment of OUD; 2) allowance of MOUD to be prescribed via telehealth, initiated during the Covid pandemic in 2020 and subsequently made permanent; and 3) allowance of take-home methadone doses (up to 28 days for stable patients), also initiated in 2020 and made permanent [5,6]. However, an additional fourth policy, implemented in 2025, increases the relevance of 2019 data, because it rolled back the Covid emergency exemption that paused all disenrollment from the Medicaid due to administrative reasons [44]. Thirty percent of those who lose Medicaid coverage do so for administrative reasons, termed “churn” [44,45] Consequently, generalizability of Medicaid data after 2019 would be compromised for this reason. Finally, our study was by design descriptive, and the percentages we report were not adjusted for any variables; as such, our results should not be interpreted as describing any causal relationships involving MOUD treatment quality, differences in state Medicaid policies, and/or differences in MOUD treatment access across urban, suburban, and rural areas.
Supporting information
S1 Table. Raw counts (numerators and denominators) for the three MOUD treatment quality metrics reported in the main paper by state.
Due to small cell counts, some data are not reported (indicated by an NA). OUD = opioid use disorder; MOUD = medication for opioid use disorder.
https://doi.org/10.1371/journal.pone.0341739.s001
(PDF)
S2 Table. Raw counts (numerators and denominators) for the three MOUD treatment quality metrics reported in the main paper by urbanicity type.
OUD = opioid use disorder; MOUD = medication for opioid use disorder.
https://doi.org/10.1371/journal.pone.0341739.s002
(PDF)
References
- 1. Blanco C, Volkow ND. Management of opioid use disorder in the USA: present status and future directions. Lancet. 2019;393(10182):1760–72. pmid:30878228
- 2. Mauro PM, Gutkind S, Annunziato EM, Samples H. Use of Medication for Opioid Use Disorder Among US Adolescents and Adults With Need for Opioid Treatment, 2019. JAMA Netw Open. 2022;5(3):e223821. pmid:35319762
- 3. Hser Y-I, Saxon AJ, Huang D, Hasson A, Thomas C, Hillhouse M, et al. Treatment retention among patients randomized to buprenorphine/naloxone compared to methadone in a multi-site trial. Addiction. 2014;109(1):79–87. pmid:23961726
- 4. Krawczyk N, Rivera BD, Jent V, Keyes KM, Jones CM, Cerdá M. Has the treatment gap for opioid use disorder narrowed in the U.S.?: A yearly assessment from 2010 to 2019". Int J Drug Policy. 2022;110:103786. pmid:35934583
- 5.
Substance Abuse and Mental Health Services Administration. Medicaid Coverage of Medications to Reverse Opioid Overdose and Treat Alcohol and Opioid Use Disorders. Rockville (MD): Substance Abuse and Mental Health Services Administration; 2024.
- 6. Substance Abuse and Mental Health Services Administration; U.S. Department of Health and Human Services. Medications for the Treatment of Opioid Use Disorder; Final Rule, 42 CFR Part 8. 2024. Available from: https://www.govinfo.gov/content/pkg/FR-2024-02-02/pdf/2024-01693.pdf
- 7. Latsko E, Denham A, Barnett ML. Medicaid and methadone for opioid use disorder: expanded coverage increased distribution in 10 states. Health Affairs. 2025;44(9):1112–21.
- 8. Huebler S, Jones AL, Zhao H, Nelson RE, Hagedorn HJ, Hawkins EJ, et al. Buprenorphine Treatment for Opioid Use Disorder in Non-Addiction Specialty Settings. JAMA Netw Open. 2025;8(11):e2543543. pmid:41231472
- 9. Chen AY, Powell D, Stein BD. Changes in Buprenorphine and Methadone Supplies in the US During the COVID-19 Pandemic. JAMA Netw Open. 2022;5(7):e2223708. pmid:35881394
- 10. Gikoska M, Florio AK, George A, Piper BJ. Analysis of buprenorphine distribution patterns among pharmacies and hospitals in the USA from 2019 to 2023. BMJ Open. 2025;15(7):e094454. pmid:40685232
- 11. Das S, Muhetaer K, Gandotra N, Tomoyasu N. Filling the Treatment Gap: Geographic Expansion of Buprenorphine Providers Across the U.S. AJPM Focus. 2024;4(1):100284. pmid:40225958
- 12. Ali MM, Chen J, Novak PJ. Trends in buprenorphine dispensing by provider type, 2013–2023. Psychiatr Serv. 2025;76(8):767–8.
- 13.
Medicare Payment Advisory Commission. Report to Congress on Medicaid and CHIP. MPAC; 2025 [cited 2025 Dec 15]. Available from: https://www.macpac.gov/wp-content/uploads/2025/06/MACPAC_June-2025-WEB-508.pdf
- 14. Lindner SR, et al. State- and county-level geographic variation in opioid use disorder, medication treatment, and opioid-related overdose among Medicaid enrollees. JAMA Health Forum. 2023;4(6):e231574–e231574.
- 15. Brown E, et al. Use of medications for treatment of opioid use disorder among US Medicaid enrollees in 11 states, 2014-2018. JAMA. 2021;326(2):154–64.
- 16. Yarbrough CR, Cooper HLF, Beane S, Haardörfer R, Ibragimov U, Haley DF, et al. State Medicaid Policies Governing Access to Medications for Opioid Use Disorder (MOUD) and MOUD Treatment Use in a Large Sample of People Who Inject Drugs in 20 U.S. States. Subst Use Misuse. 2025;60(4):531–41. pmid:39741378
- 17. Franz B, Cronin CE, Lindenfeld Z, Pagan JA, Lai AY, Krawczyk N, et al. Rural-urban disparities in the availability of hospital-based screening, medications for opioid use disorder, and addiction consult services. J Subst Use Addict Treat. 2024;160:209280. pmid:38142042
- 18. Lister JJ, Weaver A, Ellis JD, Himle JA, Ledgerwood DM. A systematic review of rural-specific barriers to medication treatment for opioid use disorder in the United States. Am J Drug Alcohol Abuse. 2020;46(3):273–88. pmid:31809217
- 19. Mitchell P, Samsel S, Curtin KM, Price A, Turner D, Tramp R, et al. Geographic disparities in access to Medication for Opioid Use Disorder across US census tracts based on treatment utilization behavior. Soc Sci Med. 2022;302:114992. pmid:35512612
- 20. Lofaro RJ, Bohler RM, Spurgeon R, Mase WA. Rural and urban differences in treatment on demand for substance use treatment involving medications for opioid use disorder. J Rural Health. 2025;41(3):e70076. pmid:40855625
- 21. Reynolds OC, Carlson KF, Gordon AJ, Handley RL, Morasco BJ, Korthuis TP, et al. Receipt of medications for opioid use disorder among rural and urban veterans health administration patients. Drug Alcohol Depend Rep. 2024;14:100311. pmid:39845270
- 22. Kiang MV, Barnett ML, Wakeman SE, Humphreys K, Tsai AC. Robustness of estimated access to opioid use disorder treatment providers in rural vs. urban areas of the United States. Drug Alcohol Depend. 2021;228:109081. pmid:34600256
- 23. Bao Y, Zhang H, Hutchings K, Harris RA, Calderbank T, Schackman BR. Medicaid-Covered Peer Support Services Used by Enrollees With Opioid Use Disorder. JAMA Netw Open. 2024;7(7):e2420737. pmid:38980680
- 24. Saunders H, Rudowitz R. Implications of Potential Federal Medicaid Reductions for Addressing the Opioid Epidemic. 2025 [cited 2025 Dec 16]. Available from: https://www.kff.org/medicaid/implications-of-potential-federal-medicaid-reductions-for-addressing-the-opioid-epidemic
- 25. Besaw RJ, Fry CE. State drug caps associated with fewer Medicaid-covered prescriptions for opioid use disorder, 2017-2022. Health Affairs Scholar. 2024;2(12).
- 26. Andrilla CHA, Woolcock SC, Garberson LA, Graves JM. Differences in the travel burden for care between rural and urban patients with opioid use disorder by insurance type. J Rural Health. 2025;41(4):e70078. pmid:41020666
- 27. National Quality Forum. Behavioral Health 2016-2017 Final Report. 2017 [cited 2025 Nov 14]. Available from: https://www.qualityforum.org/Publications/2017/08/Behavioral_Health_2016-2017_Final_Report.aspx
- 28. Williams A, Nunes E, Olfson M. To battle the opioid overdose epidemic, deploy the ‘cascade of care’model. Health Affairs (blog post). 2017 [cited 2025 Dec 10]. Available from:
- 29. Williams AR, Nunes EV, Bisaga A, Levin FR, Olfson M. Development of a Cascade of Care for responding to the opioid epidemic. Am J Drug Alcohol Abuse. 2019;45(1):1–10. pmid:30675818
- 30. Medicaid data quality atlas. [cited 2025 Nov 14]. Available from: https://www.medicaid.gov/dq-atlas/welcome
- 31.
Kaiser Family Foundation. Status of state medicaid expansion decisions: interactive map. 2020.
- 32. Donohue J, et al. Opioid use disorder among Medicaid enrollees: Snapshot of the epidemic and state responses. The Kaiser Family Foundation 2019 December 17, 2025]. Available from: https://www.kff.org/medicaid/opioid-use-disorder-among-medicaid-enrollees-snapshot-of-the-epidemic-and-state-responses/
- 33. Samples H, Williams AR, Olfson M, Crystal S. Risk factors for discontinuation of buprenorphine treatment for opioid use disorders in a multi-state sample of Medicaid enrollees. J Subst Abuse Treat. 2018;95:9–17. pmid:30352671
- 34. Williams AR, Nunes EV, Bisaga A, Pincus HA, Johnson KA, Campbell AN, et al. Developing an opioid use disorder treatment cascade: A review of quality measures. J Subst Abuse Treat. 2018;91:57–68. pmid:29910015
- 35. Mattke S, Predmore Z, Sloss E, Wilks A, Watkins KE. Evidence for Misspecification of a Nationally Used Quality Measure for Substance Use Treatment. J Healthc Qual. 2018;40(4):228–35. pmid:28933709
- 36.
Substance Abuse and Mental Health Services Administration. Medicaid Coverage of Medication-Assisted Treatment for Alcohol and Opioid Use Disorders and of Medication for the Reversal of Opioid Overdose. Rockville (MD): Substance Abuse and Mental Health Services Administration; 2018.
- 37. Scholl L, et al. Drug and opioid-involved overdose deaths - United States, 2013-2017. MMWR Morb Mortal Wkly Rep. 2018;67(51–52):1419–27.
- 38. Lander LR, Zheng W, Hustead JD, Mahoney JJ 3rd, Berry JH, Marshalek P, et al. Long-term treatment retention in West Virginia’s comprehensive opioid addiction treatment (COAT) program. J Neurol Sci. 2020;411:116712. pmid:32058182
- 39. Harris SJ, Landis RK, Li W, Stein BD, Saloner B. Utilization of Medications for Opioid Use Disorder Among West Virginia Medicaid Enrollees Following Medicaid Coverage of Methadone. Subst Use Addctn J. 2024;45(1):91–100. pmid:38258853
- 40.
New Hampshire Department of Health and Human Services, New Hampshire Doorway Program Evaluation: Final Report. 2023. Concord (NH): New Hampshire Department of Health and Human Services; [cited 2025 Dec 17]. Available from: https://www.thedoorway.nh.gov/sites/g/files/ehbemt566/files/documents/2023-12/nh-dooway-finalreport.pdf
- 41. Jones CM, Han B, Baldwin GT, Einstein EB, Compton WM. Use of Medication for Opioid Use Disorder Among Adults With Past-Year Opioid Use Disorder in the US, 2021. JAMA Netw Open. 2023;6(8):e2327488. pmid:37548979
- 42. Lin LA, Knudsen HK. Comparing Buprenorphine-Prescribing Physicians Across Nonmetropolitan and Metropolitan Areas in the United States. Ann Fam Med. 2019;17(3):212–20. pmid:31085525
- 43. Williams AR, Mauro CM, Chiodo L, Huber B, Cruz A, Crystal S, et al. Buprenorphine treatment and clinical outcomes under the opioid use disorder cascade of care. Drug Alcohol Depend. 2024;263:112389. pmid:39154558
- 44. Schpero WL, Ndumele CD. Medicaid Disenrollment After the COVID-19 Pandemic: Avoiding a New Crisis. JAMA Health Forum. 2022;3(2):e214743. pmid:36218824
- 45. Sommers BD, Gourevitch R, Maylone B, Blendon RJ, Epstein AM. Insurance Churning Rates For Low-Income Adults Under Health Reform: Lower Than Expected But Still Harmful For Many. Health Aff (Millwood). 2016;35(10):1816–24. pmid:27702954