PLoS Medicine
Public Library of Science
image
Medications for opioid use disorder among pregnant women referred by criminal justice agencies before and after Medicaid expansion: A retrospective study of admissions to treatment centers in the United States
Volume: 17, Issue: 5
DOI 10.1371/journal.pmed.1003119
  • PDF   
  • XML   
  •       
Abstract

Why was this study done?There has been a 4-fold increase in the number of pregnant women with opioid use disorder (OUD). Medications such as methadone and buprenorphine are standard of care for OUD and are recommended during pregnancy, but only 50% of pregnant women receive such medication.Pregnant women with OUD who are involved in the criminal justice system are at high risk of poor outcomes, but data regarding the use of medications for OUD in this population are limited.What did the researchers find?From 1992 to 2017, pregnant women in the US who were referred to treatment for OUD by a criminal justice agency (versus other referral sources) were half as likely to receive medication as part of their treatment plan.After implementation of the Affordable Care Act’s Medicaid expansion, medication for OUD increased significantly more among pregnant women referred to treatment by criminal justice agencies in Medicaid expansion states compared with nonexpansion states.What do these findings mean?Pregnant women referred to treatment for OUD by criminal justice agencies were consistently less likely to receive evidence-based treatment, which increases their risk of overdose and poor maternal and neonatal outcomes.Improving access to Medicaid for justice-involved individuals may increase the rate at which pregnant women receive evidence-based treatment for OUD.

Winkelman, Ford, Shlafer, McWilliams, Admon, Patrick, and Song: Medications for opioid use disorder among pregnant women referred by criminal justice agencies before and after Medicaid expansion: A retrospective study of admissions to treatment centers in the United States

Introduction

Overdose deaths among women increased 260% between 1999 and 2017, which was largely driven by a dramatic increase in deaths related to fentanyl, heroin, and prescription opioids [1]. At the same time, the prevalence of opioid use disorder (OUD) among pregnant women more than quadrupled [2,3]. Pregnant women with OUD are more likely to experience severe maternal morbidity and mortality relative to women without OUD at the time of delivery [2]. Treatment of OUD with medication has consistently been shown to improve pregnancy outcomes and reduce the risk of overdose-related death [46] compared with non-medication-based treatment, which would typically include individual/group counseling services, withdrawal management, or referrals to other social services. Although effective treatment is available, only half of pregnant women with OUD who enter treatment ultimately receive medication for OUD [7].

Criminal justice involvement (e.g., interaction with police, courts, community supervision, jail, or prison) is common among pregnant women with OUD, with nearly 1 in 5 affected women referred to treatment through criminal justice agencies in 2012 [8]. Since 1980, the incarceration rate among women has increased at a rate two times larger than among men [9]. Criminal justice involvement confers additional risk during pregnancy [10,11] and is associated with significant morbidity and mortality among individuals with OUD [12,13]. Although pregnant women with OUD are increasingly referred to treatment by criminal justice agencies, such as court-mandated treatment or treatment as a condition of probation/parole, there are limited data on trends in medications for OUD for this population [7,8]. Treatment data are critical for leaders in healthcare and correctional settings who have an opportunity to reduce morbidity and mortality among mothers and their infants. Furthermore, since passage of the Affordable Care Act (ACA), insurance rates among individuals involved in the criminal justice system and access to medications for OUD have increased [1416]. Although the ACA did not alter enrollment criteria for pregnant women, improved Medicaid enrollment strategies, new linkages between community-based healthcare systems and the criminal justice system, and mandated coverage of substance use treatment in states that chose to expand their Medicaid programs could impact treatment patterns for justice-involved pregnant women [1719].

To assess trends in receipt of medications for OUD among pregnant women with OUD, we examined 25 years of national data from substance use treatment facilities. We compared pregnant women referred to treatment facilities by criminal justice agencies to those with individual referrals (e.g., self-referral to treatment, referred by a friend or family member) and those referred from other sources (e.g., referral from a healthcare provider or employer). We also estimated treatment rates with medications for OUD among pregnant women referred to treatment facilities by criminal justice agencies by state Medicaid expansion status. We hypothesized that pregnant women referred by criminal justice agencies would have lower rates of medications for OUD and that, among this population, Medicaid expansion would be associated with larger increases in medications for OUD relative to nonexpansion states.

Methods

Data source and sample

We used all available years (1992–2017) of the Treatment Episode Data Set-Admissions (TEDS-A), an annual national survey of substance treatment facility admissions conducted by the Substance Abuse and Mental Health Services Administration [20]. TEDS-A includes pregnancy status and demographic, substance use, and treatment data that publicly funded treatment facilities are required to report. Data elements are obtained from information included in the referral itself (e.g., information provided by a criminal justice agency or healthcare provider at the time of referral) and during an interview that is conducted with the patient at the time of intake to the treatment facility. The unit of analysis within TEDS-A is an admission to a treatment facility, not an individual. Therefore, some individuals may be represented more than once. TEDS-A does not contain identifiable information, so it is not possible to link admissions for the same individual. Furthermore, TEDS-A contains information from admissions to treatment centers that accept public funding, and it does not include information about treatment that occurs in other facility types (e.g., primary care offices, in health clinics inside jails and prisons, treatment centers that do not accept public funding). Initial admissions to each center are included; transfers between facilities are excluded.

We restricted the sample to women who were pregnant at the time of admission to treatment and whose primary reason for treatment was related to opioid abuse (heroin, nonprescription methadone, other synthetic opioids). Substance Abuse and Mental Health Services Administration staff corresponded that reporting changes in Florida between 2010 and 2017 made it difficult to compare data across years, so we excluded Florida from the sample. Less than 5% of admissions were missing a variable used in this study. Admissions with missing referral source were excluded from all analyses (2.7%), whereas admissions with missing covariates were excluded from regression models.

Key independent variable—Treatment referral sources

We examined longitudinal trends in receipt of medications for OUD among pregnant women by referral source to a substance use treatment facility. A referral source is defined as the agency or person referring an individual to treatment. The referral process may vary by jurisdiction. We classified referrals into three categories: criminal justice, individual, and other. Criminal justice agency referrals included referrals to treatment centers from police, probation officers, judges, prosecutors, DUI/DWI court, and parole boards. Criminal justice referrals do not include treatment received within a correctional facility. Individual referrals included admissions that were initiated by the patient, their family, a friend, or an individual who was not included in another category. Other referrals included those from an alcohol or drug abuse care provider, another healthcare provider, schools, employers, or other community referrals (e.g., Alcoholics Anonymous, shelter, or religious organization).

Key dependent variable—Medications for OUD

Receipt of medications for OUD, such as methadone or buprenorphine, as part of a treatment plan was the primary outcome measure. TEDS-A data do not distinguish between medications. Opioid-related treatment episodes that do not include medications for OUD typically involve individual, family, or group services; withdrawal management; and/or transitional housing. Referrals to treatment do not necessarily ensure an individual will receive medication for OUD and could, instead, result in counseling or other behavioral treatment only.

Medicaid expansion

Beginning in 2014, states could choose whether to expand their Medicaid programs under the ACA. For purposes of this analysis, we categorized a state as “expanded” in a given year if they had been expanded for more than 6 months that year. We considered most expansion states to have expanded in 2014, with the exception of New Hampshire (2015), Alaska (2016), Montana (2016), and Louisiana (2016).

Sociodemographic characteristics and service setting

We examined differences between pregnant women with OUD by referral source. We assessed age, race/ethnicity, educational attainment, employment, census region, and service setting. Service setting indicates the location at which an individual received treatment and could be classified as a detoxification, residential, or ambulatory center. We controlled for these characteristics in multivariable models to assess trends in treatment by year.

Statistical analysis

We prespecified our analysis plan to examine trends in medications for OUD over time among pregnant women who were referred by criminal justice agencies or other sources (S1 Text). After these initial analyses, we conducted an additional difference-in-differences (DID) analysis using previously described specifications [21] to investigate the extent to which the increase in medications for OUD after 2014 was associated with Medicaid expansion and hypothesized that this increase would largely be explained by Medicaid expansion. We first assessed sociodemographic characteristics and service settings of our study population by referral source (i.e., criminal justice, individual, or other referral) between 1992 and 2017. Next, we examined the number of pregnant women with OUD referred by criminal justice agencies or other sources who did and did not receive medications for OUD during our study period. We then used multivariable logistic regression models to estimate the proportion of women who received medications for OUD by referral source overall and in each study year, adjusting for the covariates described above. We used postestimation predictive margins to depict and compare adjusted proportions between referral sources and years. Comparisons are presented as adjusted rate ratios (ARRs).

Finally, we examined changes in receipt of medications for OUD among pregnant women referred by criminal justice agencies by Medicaid expansion status between 2011 and 2017. We first described unadjusted trends by expansion status. Next, we used a DID framework to compare changes in medications for OUD receipt in states that did and did not expand Medicaid. We interacted a state-specific time variable that indicated whether the admission was before or after implementation of the ACA’s Medicaid expansion with a variable that indicated whether the admission was in a state that expanded Medicaid during the study period. We used a linear model with robust standard errors clustered at the state level and adjusted for covariates. We did not control for census region in our DID model because we clustered our standard errors at the state level. We excluded 2014 from our adjusted DID model to allow for a washout period [22,23]. We used a linear model with the same covariates to estimate unadjusted and adjusted 2017 rate differences in medications for OUD receipt among pregnant women in states that did and did not expand Medicaid. The primary assumption of a DID design is that trends in comparison groups are parallel prior to the intervention (i.e., Medicaid expansion). Therefore, we assessed trends in medications for OUD by expansion status in the pre-Medicaid expansion time period (2011–2013) by interacting a pre-ACA linear time trend with expansion status in our multivariable linear regression model. Per reviewer request, we also compared our DID estimate among pregnant women referred by criminal justice agencies to women referred by other sources.

We conducted analyses between September 2019 and October 2019. This study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (S1 STROBE Checklist). We used Stata 15.1 for all analyses and considered P < 0.05 to be statistically significant. This study of publicly available, deidentified data was not considered human subjects research and was exempt from review by the Hennepin Healthcare Research Institute Institutional Review Board.

Results

From 1992 to 2017, we identified a total of 131,838 pregnant women with OUD; 17,563 (13.3%) were referred by criminal justice agencies, 64,246 (48.7%) were individual referrals, and 50,029 (38.0%) were referred by other sources. Women referred by criminal justice agencies were younger than those referred by an individual or by another source (ages 18–24: 34.1% versus 27.2% versus 29.3%, respectively), were less likely to be Black, non-Hispanic (8.8% versus 16.7% versus 16.2%, respectively), and were more likely to receive treatment in a residential setting (26.2% versus 9.8% versus 20.9%). Women referred by different sources did not differ substantially in education, employment status, or census region (Table 1). During the 25-year study period, more pregnant women referred to treatment for OUD received medications for OUD (N = 67,937; 51.6%) than did not (N = 63,623; 48.4%).

Table 1
Demographic characteristics of pregnant women with opioid use disorder by referral source, US, 1992–2017.
N (%)
CharacteristicCriminal Justice
(N = 17,563)
Individual
(N = 64,246)
Other
(N = 50,029)
P value
Age<0.001
    12–17133 (0.8)169 (0.3)219 (0.4)
    18–245,996 (34.1)17,491 (27.2)14,669 (29.3)
    25–296,027 (34.3)21,182 (33.0)17,106 (34.2)
    30–343,363 (19.2)14,772 (23.0)11,017 (22.1)
    35+2,044 (11.6)10,632 (16.6)6,964 (13.9)
Race/Ethnicity<0.001
    White, non-Hispanic12,704 (72.8)41,7301 (65.3)33,199 (66.8)
    Black, non-Hispanic1,540 (8.8)10,674 (16.7)8,070 (16.2)
    Hispanic832 (4.8)3,206 (5.0)2,175 (4.4)
    Native American/Alaskan Native539 (3.1)1,339 (2.1)1,531 (3.1)
    Other1,825 (10.5)6,923 (10.8)4,739 (9.5)
Education<0.001
    Less than high school6,599 (38.3)23,588 (37.3)18,870 (38.4)
    High school complete7,181 (41.7)26,549 (42.0)19,962 (40.6)
    Some college or more3,462 (20.1)13,115 (20.7)10,371 (21.1)
Employment<0.001
    Not employed15,480 (89.3)56,173 (88.6)44,554 (90.3)
    Employed1,853 (10.7)7,229 (11.4)4,770 (9.7)
Census Region<0.001
    Northeast5,520 (31.4)19,665 (30.6)18,569 (37.1)
    Midwest3,391 (19.3)11,286 (17.6)10,234 (20.5)
    South3,952 (22.5)17,111 (26.6)12,211 (24.4)
    West4,700 (26.8)16,184 (25.2)9,015 (18.0)
Service Setting<0.001
    Detox708 (4.0)7,947 (12.4)3,461 (6.9)
    Residential4,587 (26.2)6,291 (9.8)10,444 (20.9)
    Ambulatory12,229 (69.8)49,841 (77.8)35,985 (72.1)
“Criminal Justice” includes referral from police, probation officers, judges, prosecutors, DUI/DWI court, or parole board. “Individual” includes referral from patient, family, or friends. “Other” includes referral from alcohol/drug abuse care provider, healthcare providers, school, employer, or community referral.Abbreviations: DUI, driving under the influence; DWI, driving while intoxicated

Medications for OUD by treatment referral source

The number of pregnant women with OUD referred for treatment increased substantially between 1992 and 2017. Pregnant women with OUD referred by criminal justice agencies increased from 318 in 1992 to 1,491 in 2017. Among pregnant women with OUD referred by criminal justice agencies, treatment without medications for OUD was more common than treatment with medications for OUD. Pregnant women with OUD referred by other sources also increased, from 2,325 in 1992 to 9,402 in 2017. In contrast to women referred by criminal justice agencies, pregnant women with OUD referred from other sources were more likely to receive medications for OUD than not (Fig 1).

Number of pregnant women who did and did not receive medications for OUD by referral source, US, 1992–2017.
Fig 1
OUD, opioid use disorder.Number of pregnant women who did and did not receive medications for OUD by referral source, US, 1992–2017.

Adjusted receipt of medications for OUD was significantly lower across the entire study period (1992–2017) for women referred by criminal justice agencies (26.3%, 95% CI 25.7–27.0, P < 0.001) compared with other sources (51.3%, 95% CI 50.8–51.7, P < 0.001; ARR 0.51, 95% CI 0.50–0.53) and individual referrals (59.1%, 95% CI 58.8–59.5, P < 0.001; ARR 0.45, 95% CI: 0.43, 0.46). Longitudinal trends in receipt of medications for OUD among pregnant women also differed by treatment referral source. For example, between 1992 and 2005, the proportion of pregnant women receiving medications for OUD continuously declined over time for those who were referred by criminal justice agencies, whereas similar declines were not seen among other referral sources (Fig 2). Rates of medications for OUD among pregnant women referred by criminal justice agencies remained lower throughout the study period compared with women referred by individual and other sources and never returned to levels from the early to mid-1990s. Adjusted rates of medications for OUD by referral source for each study year are available in the S1 Table of the online supplement.

Adjusted proportion of pregnant women receiving medications for opioid use disorder by referral source, US, 1992–2017.
Fig 2
Adjusted for age, race/ethnicity, educational attainment, employment, census region, and service setting.Adjusted proportion of pregnant women receiving medications for opioid use disorder by referral source, US, 1992–2017.

Pregnant women referred by criminal justice agencies were less likely to receive medications for OUD in 2007 (ARR 0.72, 95% CI 0.57–0.88, P < 0.001) and equally likely in 2017 (ARR 0.98, 95% CI 0.81–1.15, P = 0.80) compared with 1997. Disparities in receipt of medications for OUD between women referred by criminal justice agencies and those referred by other sources changed over time. In 1997, pregnant women referred by criminal justice agencies were less likely to receive medications for OUD than women who were individual referrals (ARR 0.69, 95% CI 0.58–0.79, P < 0.001) or referred from another source (ARR 0.73, 95% CI 0.62–0.85, P < 0.001). This disparity grew by 2007 (criminal justice versus individual referral: ARR 0.44, 95% CI 0.37–0.50, P < 0.001; criminal justice versus other referral: ARR 0.47, 95% CI 0.40–0.54, P < 0.001) and then began to decrease by 2017 (criminal justice versus individual referral: ARR 0.52, 95% CI 0.48–0.55, P < 0.001; criminal justice versus other referral: ARR 0.59, 95% CI 0.55–0.64, P < 0.001). ARRs are available in 5-year increments in the S2 Table of the online supplement.

Medications for OUD among pregnant women referred by criminal justice agencies by state Medicaid expansion status

We examined changes in medications for OUD receipt among pregnant women with OUD referred by criminal justice agencies in states that did and did not expand Medicaid through the ACA (Fig 3). In Medicaid expansion states, rates of medications for OUD were relatively consistent between 2011 and 2013. Medicaid nonexpansion states had lower rates of medications for OUD among pregnant women referred by criminal justice agencies than expansion states. Unadjusted rates of medications for OUD by Medicaid expansion status are available in the S3 Table of the online supplement.

Proportion of pregnant women referred by criminal justice agencies receiving medications for OUD by state Medicaid expansion status, US, 2011–2017.
Fig 3
OUD, opioid use disorder.Proportion of pregnant women referred by criminal justice agencies receiving medications for OUD by state Medicaid expansion status, US, 2011–2017.

Between 2011–2013 and 2015–2017, medications for OUD increased to a greater degree among pregnant women with a criminal justice referral with OUD in states that expanded Medicaid (DID 11.7 percentage points [pp], 95% CI 0.5–22.9; P = 0.04), and this difference remained significant after adjusting for covariates (adjusted DID 12.0 pp, 95% CI 1.0–23.0; P = 0.03). The DID estimate among pregnant women referred by criminal justice agencies was similar to individual referrals (adjusted DID 11.9 pp, 95% CI 2.3–21.4; P = 0.02) and other referral sources (adjusted DID 13.4 pp, 95% CI −1.3 to 28.0; P = 0.07). In 2017, pregnant women referred by criminal justice agencies were significantly more likely to have received medications for OUD in states that expanded Medicaid compared with women in states that did not expand Medicaid (unadjusted rate difference: 26.6 pp, 95% CI 10.2–43.0; adjusted rate difference: 27.2 pp, 95% CI 11.3–43.0). Adjusted rates of medications for OUD by Medicaid expansion status are available in the S4 Table of the online supplement. Trends in pre-ACA rates of medications for OUD did not vary significantly by expansion status (interaction coefficient: 0.1%, 95% CI −4.7 to 4.9).

Discussion

From 1992 to 2017, pregnant women with OUD referred to substance use treatment facilities by criminal justice agencies were consistently less likely to receive medications for OUD than pregnant women referred by other sources. OUD during pregnancy is associated with low birth weight, preterm labor, fetal death, and increased severe maternal morbidity [2,4,24]. Medications for OUD, the recommended first-line treatment for OUD in pregnancy, reduces overdose risk and improves maternal and infant outcomes regardless of criminal justice involvement [5,25]. Persistent disparities in medications for OUD among women referred by criminal justice agencies suggest structural and philosophical barriers to evidence-based treatment that negatively impact the health of mothers involved in criminal justice agencies and their infants [25].

Our finding of low rates of medications for OUD among pregnant women referred by criminal justice agencies is consistent with studies in other justice-involved populations with OUD [12,26,27]. For example, medications for OUD are available to fewer than half of drug court participants [26]. Barriers to medications for OUD in drug courts include cost, lack of provider, concerns about diversion, and stigma [26]. Similar barriers also likely contribute to the disparities in medications for OUD found in our study population. Poor linkage to evidence-based OUD treatment within the criminal justice system is a critical public health issue because many individuals with OUD interact with the criminal justice system and, subsequently, have a high risk of death from opioid overdose upon release [13].

To address the opioid crisis, researchers have called for the implementation of a cascade of care that effectively links high-risk populations, especially individuals with criminal justice involvement, to OUD treatment [28]. In addition, sequential intercept mapping, a systems-based model to divert individuals with behavioral health issues from the criminal justice system to community-based treatment, is a promising approach to identify points within the criminal justice system at which alternative strategies could improve care for individuals, including pregnant women, with OUD [29]. Policy responses to substance use during pregnancy that incorporate both criminal justice and public health interventions have been associated with higher levels of treatment, whereas approaches focused solely on criminal justice policy have not improved treatment rates [30,31]. Thus, our findings, in conjunction with previous data, indicate that a robust cross-sector approach to opioid use during pregnancy is required to increase medications for OUD among pregnant women with OUD. Improving medications for OUD rates for pregnant women referred by the criminal justice system could provide benefits to both public safety and public health [30,31].

Although the ACA did not alter Medicaid coverage during pregnancy [32], we found that the ACA’s Medicaid expansion was associated with a significant increase in medications for OUD among pregnant women with a criminal justice agency or individual referral in expansion states compared with nonexpansion states. There are several possible explanations for this finding. First, the ACA improved coverage for women of reproductive age [33]. Thus, pregnant women may be more likely to start their pregnancy, and subsequently treatment, with insurance. Insurance status may alter referral patterns and increase referrals to centers that provide medications for OUD. Second, beginning in 2014, there were numerous efforts to link individuals to health insurance upon release from correctional facilities [17]. Such linkage could increase insurance levels among recently incarcerated women who qualified for insurance coverage prior to the ACA but were not covered because of difficulties with the enrollment process. Medicaid is often terminated during incarceration, and individuals must reenroll upon release [34]. Low-income women enrolled in Medicaid are likely to have better access to medications for OUD than women who are uninsured [35]. Finally, Medicaid expansion is associated with increased uptake of medications for OUD in the general population, partly because Medicaid expansion was mandated to cover behavioral health services [19]. Therefore, although coverage did not change during pregnancy, the treatments covered by Medicaid expanded, which could have increased medications for OUD among this population [36].

Limitations

The findings from this study are not without limitation. First, though TEDS-A is the most comprehensive survey of treatment admissions in the US, some states only report clients whose care was publicly funded, though most states report all eligible admissions [37]. Omission of some privately funded admissions could potentially alter our results if reporting varied substantially between expansion and nonexpansion states. Second, our data set includes medications for OUD only at treatment facilities. The number of providers who prescribe buprenorphine from an outpatient setting increased over 600% between 2006–2008 and 2012–2014, but visits with these providers are not available through TEDS-A [38]. Furthermore, TEDS-A only includes data from treatment centers in the community, and individuals who are receiving treatment inside of a correctional facility (prison or jail) are not included. Third, payer is missing for most admissions, and thus we could not assess changes in payer for treatment admissions. Fourth, TEDS-A only accounts for individuals referred to treatment by criminal justice agencies and thus is a conservative estimate of criminal justice involvement in this population. Many individuals with OUD who are not linked to treatment have involvement with the criminal justice system, and some individuals with criminal justice involvement will self-refer to treatment without an actual referral from a criminal justice agency. Fifth, our data do not allow us to infer why pregnant women referred by criminal justice agencies receive medications for OUD less often than those referred by other sources. For example, it is possible that those referred by criminal justice agencies are referred to different facilities because of state/county contracts with treatment centers or that those with criminal justice involvement refuse medications because of conditions of community supervision or mistrust of the healthcare system. Finally, our DID model can identify associations between policy and treatment utilization but is not an experimental design.

Conclusion

Pregnant women with OUD referred by criminal justice agencies between 1992 and 2017 were consistently less likely to receive evidence-based treatment than women referred through other sources. The ACA’s Medicaid expansion was associated with significant improvements in receipt of medications for OUD among pregnant women referred by criminal justice agencies. Expansion of Medicaid in states that have not yet taken it up would likely further improve OUD treatment among pregnant women. A cross-sector approach that links pregnant women with OUD to care is needed to stem rising rates of opioid-related morbidity and mortality and address persistent disparities between pregnant women with and without involvement in the criminal justice system.

Abbreviations:

ACA

Affordable Care Act

ARR

adjusted rate ratio

DID

difference-in-differences

OUD

opioid use disorder

pp

percentage point

STROBE

Strengthening the Reporting of Observational Studies in Epidemiology

TEDS-A

Treatment Episode Data Set-Admissions

References

1 

JP VanHouten, RA Rudd, MF Ballesteros, KA Mack. . Drug overdose deaths among women aged 30–64 years—United States, 1999–2017. MMWR Morb Mortal Wkly Rep. 2019;68: , pp.1–5. , doi: 10.15585/mmwr.mm6801a1

2 

LK Admon, G Bart, KB Kozhimannil, CR Richardson, VK Dalton, TNA Winkelman. . Amphetamine- and Opioid-Affected Births: Incidence, Outcomes, and Costs, United States, 2004–2015. Am J Public Health. 2019;109: , pp.148–154. , doi: 10.2105/AJPH.2018.304771

3 

SC Haight, JY Ko, VT Tong, MK Bohm, WM Callaghan. . Opioid Use Disorder Documented at Delivery Hospitalization—United States, 1999–2014. MMWR Morb Mortal Wkly Rep. 2018;67: , pp.845–849. , doi: 10.15585/mmwr.mm6731a1

4 

E. Jones Hendrée, E. O’Grady Kevin, Debbie Malfi, Michelle Tuten. . Methadone maintenance vs. methadone taper during pregnancy: maternal and neonatal outcomes. Am J Addict. 2010;17: , pp.372–386. , doi: 10.1080/10550490802266276

5 

Committee on Obstetric Practice. . Committee Opinion No. 711: Opioid use and opioid use disorder in pregnancy. American College of Obstetricians and Gyncecologists. Obstet Gynecol. 2017;130: , pp.e81–94. , doi: 10.1097/AOG.0000000000002235

6 

M Terplan, HJ Laird, DJ Hand, TE Wright, A Premkumar, CE Martin, et al. Opioid detoxification during pregnancy: a systematic review. Obstet Gynecol. 2018;131: , pp.803–814. , doi: 10.1097/AOG.0000000000002562

7 

VL Short, DJ Hand, L MacAfee, DJ Abatemarco, M Terplan. . Trends and disparities in receipt of pharmacotherapy among pregnant women in publically funded treatment programs for opioid use disorder in the United States. J Subst Abuse Treat. 2018;89: , pp.67–74. , doi: 10.1016/j.jsat.2018.04.003

8 

CE Martin, N Longinaker, M Terplan. . Recent trends in treatment admissions for prescription opioid abuse during pregnancy. J Subst Abuse Treat. 2015;48: , pp.37–42. , doi: 10.1016/j.jsat.2014.07.007

9 

The Sentencing Project. Incarcerated Women and Girls. Washington, DC: The Sentencing Project; 20196 [cited 2020 Apr 20]. Available from: https://www.sentencingproject.org/publications/incarcerated-women-and-girls/.

10 

C. SufrinJailcare—finding the safety net for women behind barsOakland, CA: University of California Press; 2017.

11 

J. Csete. Criminal justice barriers to treatment of opioid use disorders in the United States: the need for public health advocacy. Am J Public Health. 2019;109: , pp.419–422. , doi: 10.2105/AJPH.2018.304852

12 

N Krawczyk, CE Picher, KA Feder, B Saloner. . Only one in twenty justice-referred adults in specialty treatment for opioid use receive methadone or buprenorphine. Health Aff (Millwood). 2017;36: , pp.2046–2053. , doi: 10.1377/hlthaff.2017.0890

13 

IA Binswanger. . Mortality After Prison Release: Opioid Overdose and Other Causes of Death, Risk Factors, and Time Trends From 1999 to 2009. Ann Intern Med. 2013;159: , pp.592, doi: 10.7326/0003-4819-159-9-201311050-00005

14 

TNA Winkelman, EC Kieffer, SD Goold, JD Morenoff, K Cross, JZ Ayanian. . Health insurance trends and access to behavioral healthcare among justice-involved individuals—United States, 2008–2014. J Gen Intern Med. 2016;31: , pp.1523–1529. , doi: 10.1007/s11606-016-3845-5

15 

CD Knapp, BA Howell, EA Wang, RJ Shlafer, RR Hardeman, TNA Winkelman. . Health insurance gains after implementation of the Affordable Care Act among individuals recently on probation: USA, 2008–2016. J Gen Intern Med. 2019, doi: 10.1007/s11606-019-04900-3

16 

B Saloner, J Levin, H-Y Chang, C Jones, GC Alexander. . Changes in buprenorphine-naloxone and opioid pain reliever prescriptions after the Affordable Care Act medicaid expansion. JAMA Netw Open. 2018;1: , pp.e181588, doi: 10.1001/jamanetworkopen.2018.1588

17 

SN Bandara, HA Huskamp, LE Riedel, EE McGinty, D Webster, RE Toone, et al. Leveraging the Affordable Care Act to enroll justice-involved populations in Medicaid: state and local efforts. Health Aff (Millwood). 2015;34: , pp.2044–2051. , doi: 10.1377/hlthaff.2015.0668

18 

SA Somers, E Nicolella, A Hamblin, SM McMahon, C Heiss, BW Brockmann. . Medicaid expansion: considerations for states regarding newly eligible jail-involved individuals. Health Aff (Millwood). 2014;33: , pp.455–461. , doi: 10.1377/hlthaff.2013.1132

19 

A Meinhofer, AE Witman. . The role of health insurance on treatment for opioid use disorders: Evidence from the Affordable Care Act Medicaid expansion. J Health Econ. 2018;60: , pp.177–197. , doi: 10.1016/j.jhealeco.2018.06.004

20 

Substance Abuse and Mental Health Services Administration (SAMHSA). In: Treatment episode data set (TEDS) 1992–2017 [Internet]. [cited 2020 Apr 20]. Available from: https://www.datafiles.samhsa.gov/study-series/treatment-episode-data-set-admissions-teds-nid13518.

21 

R Mojtabai, C Mauro, MM Wall, CL Barry, M Olfson. . The Affordable Care Act and opioid agonist therapy for opioid use disorder. Psychiatr Serv. 2019;70: , pp.617–620. , doi: 10.1176/appi.ps.201900025

22 

BD Sommers, MZ Gunja, K Finegold, T Musco. . Changes in self-reported insurance coverage, access to care, and health under the Affordable Care Act. JAMA. 2015;314: , pp.366–374. , doi: 10.1001/jama.2015.8421

23 

BD Sommers, RJ Blendon, EJ Orav, AM Epstein. . Changes in utilization and health among low-income adults after Medicaid expansion or expanded private insurance. JAMA Intern Med. 2016;176: , pp.1501–1509. , doi: 10.1001/jamainternmed.2016.4419

24 

CA Fullerton, M Kim, CP Thomas, DR Lyman, LB Montejano, RH Dougherty, et al. Medication-assisted treatment with methadone: assessing the evidence. Psychiatr Serv. 2014;65: , pp.146–157. , doi: 10.1176/appi.ps.201300235

25 

M Peeler, K Fiscella, M Terplan, C Sufrin. . Best practices for pregnant incarcerated women with opioid use disorder. J Correct Health Care. 2019; 107834581881985. , doi: 10.1177/1078345818819855

26 

H Matusow, SL Dickman, JD Rich, C Fong, DM Dumont, C Hardin, et al. Medication assisted treatment in US drug courts: Results from a nationwide survey of availability, barriers and attitudes. J Subst Abuse Treat. 2013;44: , pp.473–480. , doi: 10.1016/j.jsat.2012.10.004

27 

C Sufrin, L Sutherland, L Beal, M Terplan, C Latkin, JG Clarke. . Opioid use disorder incidence and treatment among incarcerated pregnant women in the United States: results from a national surveillance study. Addict Abingdon Engl. 2020, doi: 10.1111/add.15030

28 

AR Williams, EV Nunes, A Bisaga, FR Levin, M Olfson. . Development of a Cascade of Care for responding to the opioid epidemic. Am J Drug Alcohol Abuse. 2019;45: , pp.1–10. , doi: 10.1080/00952990.2018.1546862

29 

N Bonfine, MR Munetz, RH Simera. . Sequential intercept mapping: developing systems-level solutions for the opioid epidemic. Psychiatr Serv. 2018;69: , pp.1124–1126. , doi: 10.1176/appi.ps.201800192

30 

SW Patrick, DM Schiff, SUA Prevention C on. . A public health response to opioid use in pregnancy. Pediatrics. 2017; , pp.e20164070, doi: 10.1542/peds.2016-4070

31 

KB Kozhimannil, WN Dowd, MM Ali, P Novak, J Chen. . Substance use disorder treatment admissions and state-level prenatal substance use policies: evidence from a national treatment database. Addict Behav. 2019;90: , pp.272–277. , doi: 10.1016/j.addbeh.2018.11.019

32 

JR Daw, BD Sommers. . Association of the Affordable Care Act dependent coverage provision with prenatal care use and birth outcomes. JAMA. 2018;319: , pp.579, doi: 10.1001/jama.2018.0030

33 

JR Daw, BD Sommers. . The Affordable Care Act and access to care for reproductive-aged and pregnant women in the United States, 2010–2016. Am J Public Health. 2019;109: , pp.565–571. , doi: 10.2105/AJPH.2018.304928

34 

JR Daw, LA Hatfield, K Swartz, BD Sommers. . Women in the United States experience high rates of coverage ‘churn’ in months before and after childbirth. Health Aff (Millwood). 2017;36: , pp.598–606. , doi: 10.1377/hlthaff.2016.1241

35 

M Olfson, M Wall, CL Barry, C Mauro, R Mojtabai. . Impact of medicaid expansion on coverage and treatment of low-income adults with substance use disorders. Health Aff (Millwood). 2018;37: , pp.1208–1215. , doi: 10.1377/hlthaff.2018.0124

36 

B Saloner, KB Stoller, CL Barry. . Medicaid coverage for methadone maintenance and use of opioid agonist therapy in specialty addiction treatment. Psychiatr Serv. 2016;67: , pp.676–679. , doi: 10.1176/appi.ps.201500228

37 

Treatment Episode Data Set Admissions (TEDS-A), 2014—Codebook. Rockville, MD: Center for Behavioral Health Statistics and Quality. Substance Abuse and Mental Health Services Administration; [cited 2020 Apr 20]. Available from: https://www.datafiles.samhsa.gov/study-dataset/treatment-episode-data-set-admissions-2014-teds-2014-ds0001-nid16950.

38 

H Wen, TF Borders, JR Cummings. . Trends in buprenorphine prescribing by physician specialty. Health Aff (Millwood). 2019;38: , pp.24–28. , doi: 10.1377/hlthaff.2018.05145


3 Feb 2020

Dear Dr. Winkelman,

Thank you very much for submitting your manuscript "Medicaid expansion and receipt of medications for opioid use disorder among pregnant women referred by the criminal justice system" (PMEDICINE-D-19-03648) for consideration at PLOS Medicine.

Your paper was evaluated by a senior editor and discussed among all the editors here. It was also discussed with an academic editor with relevant expertise, and sent to independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that addresses the reviewers' and editors' comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We expect to receive your revised manuscript by Feb 24 2020 11:59PM. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see http://journals.plos.org/plosmedicine/s/submission-guidelines#loc-methods.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

We look forward to receiving your revised manuscript.

Sincerely,

Caitlin Moyer, Ph.D.

Associate Editor

PLOS Medicine

plosmedicine.org

---------------------------------------------------------

The Academic Editor raised a point and we would ask you to discuss in the manuscript:

The Medicaid result is quite strong. However, the other result about criminal justice referral is less strong, because there are potentially unobserved attributes of people referred by one mechanism or another that are also correlated with their receipt of medications for OUD. It is not possible to control for all of those things, so I worry about the issue of endogeneity (of confounding) there.

Editorial points and queries:

Please revise your title according to PLOS Medicine's style. Your title must be nondeclarative and not a question. It should begin with main concept if possible. "Effect of" should be used only if causality can be inferred, i.e., for an RCT. Please place the study design ("A randomized controlled trial," "A retrospective study," "A modelling study," etc.) in the subtitle (ie, after a colon). Also please add the country in which the study is set.

Abstract- Please add summary demographic information of the women included in the study; please add p values (and elsewhere I tables / text) where 95%Cis are given; the final sentence of the Methods and Findings section should include a description of the studies limitations.

Data – the URL provided returns ‘page not found’. PLOS Medicine requires that the de-identified data underlying the specific results in a published article be made available, without restrictions on access, in a public repository or as Supporting Information at the time of article publication, provided it is legal and ethical to do so. Please see the policy at

http://journals.plos.org/plosmedicine/s/data-availability

and FAQs at

http://journals.plos.org/plosmedicine/s/data-availability#loc-faqs-for-data-policy

At this stage, we ask that you include a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract. Please see our author guidelines for more information: https://journals.plos.org/plosmedicine/s/revising-your-manuscript#loc-author-summary

Please avoid bold font in main text for emphasis

Did your study have a prospective protocol or analysis plan? Please state this (either way) early in the Methods section.

a) If a prospective analysis plan (from your funding proposal, IRB or other ethics committee submission, study protocol, or other planning document written before analyzing the data) was used in designing the study, please include the relevant prospectively written document with your revised manuscript as a Supporting Information file to be published alongside your study, and cite it in the Methods section. A legend for this file should be included at the end of your manuscript.

b) If no such document exists, please make sure that the Methods section transparently describes when analyses were planned, and when/why any data-driven changes to analyses took place.

c) In either case, changes in the analysis—including those made in response to peer review comments—should be identified as such in the Methods section of the paper, with rationale.

Screen reader support enabled.

Please ensure that the study is reported according to the [STROBE] guideline, and include the completed [STROBE or other] checklist as Supporting Information. When completing the checklist, please use section and paragraph numbers, rather than page numbers. Please add the following statement, or similar, to the Methods: "This study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (S1 Checklist)."

Please report your study according to the relevant guideline, which can be found here: http://www.equator-network.org/

Comments from the reviewers:

Reviewer #1: Statistical review

This paper reports a repeated cross-sectional study that investigates the changes in treatment given to pregnant women with opioid dependency. In particular the authors look at how this changes after Medicaid was expanded in a group of states compared to other states that did not expand it.

I have some comments on the statistical methods and reporting, which I have summarised below:

1. From what I understand there was a big political difference in the two types of states (expansion vs non-expansion). From what I understand from natural experiments, to get robust causal conclusions you'd want the two groups to be comparable at the time the change is made (perhaps after an adjustment through matching). In this case the prescription behaviour seems rather different both before and after the Medicaid expansion. Is there strong potential for confounding in this? Would matching by other characteristics of states be possible? Having said all that, the results do look fairly convincing in terms of the trajectory before and after expansion.

2. Abstract - would be useful to have effect sizes + CIs for the difference between the referral paths rather than just CIs for each arm separately.

3. Abstract - number of expansion and non-expansion states?

4. Line 83 - is it possible to link admissions of the same individual? If so, are there many individuals that are seen multiple times? If so, did the authors consider using a method that would allow for correlation within admissions from the same individual?

5. Line 144/145 - I understand the rationale for the washout period, but did this make a large difference to the results? Would be interesting to include 2014 to see (unless the pre-specified analysis plan excluded 2014).

6. Statistical analysis - was there any missing demographic data, or were individuals only included with complete data? Any issues of representativeness in that case?

7. Line 156 - p-values are mentioned but no p-values are given, so I don't think this sentence is needed.

8. Would be interesting to see whether same results for expansion vs non-expansion would be the case for the other referral paths too - why just looking at prison?

James Wason

Reviewer #2: This is an analysis of 1992-2017 data on 131,838 pregnant women in the US who entered treatment for opioid use.

The authors found that a large proportion of women did not receive standard of care treatment with medications, and furthermore, that those referred from the criminal justice system received even less appropriate treatment with medications. The authors also found that, not surprisingly, receipt of medications was higher in Medicaid expansion states.

The authors did not include Florida- would it be possible to look at the Florida data and see, at least if the data that is available is consistent with the rest of the country? If not, how so? The state is a large one and it would be re-assuring if the data there were at least consistent.

Reviewer #3: Thank you for the opportunity to review this manuscript. This is a novel study on an important and overlooked topic that, at its core, seems to suggest discrimination against pregnant people referred for treatment from criminal justice agencies, that they receive worse treatment than those without. The statistical approach is a compelling one to highlight this problem. It's a promising paper with an innovative approach with a variable that receives little attention when it comes to pregnant people, criminal justice system involvement. This paper would be greatly strengthened by providing more detail and clarity on a number of methodological/definitional fronts to better orient the reader and contextualize the meaning of and limits of the data we can know from this data source; specific suggestions on where to do this are outlined below.

Title, abstract, and throughout uses the term "the criminal justice system." This makes it sound like a monolithic entity, which it is not, as the authors are well aware. In calling it that, it's confusing to know whether referrals came from jails, prisons, drug courts, probation, parole. This monolithic descriptor makes it seem like they are all the same, when they all have very different processes. This should be clarified and defined up front in the abstract and the intro. It takes getting deep into the methods section to realize that this actually does not include prisons or jails, which is important to explain up front.

Perhaps saying criminal justice system agencies is better (as in line 57), and then being very explicit in the introduction and discussion that these are community based CJS agencies, or some other way of saying that this does not include jails and prisons.

Abstract-Lines 26-27-could use some more clarity in the first sentence. "data of

pregnant women in the United States who reported opioids as their primary reason for treatment

(N=131,838)." It's confusing not knowing the source of the data and opioids as primary reason for treatment from what kind of provider or agency. . . While the manuscript will get into more detail of course, this first sentence confused me as to how various data sources were defined. It's also confusing without that context what these women would be referred for if not for M-OUD. Are they just being referred for prenatal care? Or is it drug treatment referrals you are talking about and they just get behavioral treatment? Just mentioning that the data come from a national survey of admissions to substance treatment facilities would be orienting for the reader.

Intro/Lines 60-66: This part of the intro needs more explanation as to what you mean by 'pregnant women referred by the criminal justice system.' Does this mean people who are not in custody at the time of the referral? Which kinds of agencies? How do CJS agency referrals for MOUD even happen? Are they court mandated referrals? The "referral" aspect is the crux of the study, but I think it could use a little more specificity as to how/what this really means. Relatedly, it would also be helpful to contextualize what "other sources" means. It likewise sounds monolithic but likely is not, and would be useful to elaborate more on what you mean by this, in contradistinction to criminal justice agencies. While the methods section gets into this more, the intro needs some of this overall context (though not as much specificity as in the methods)—also because once you read the methods you see that "the criminal justice system" as available in the database and therefore as defined in this study actually does not include jails or prisons.

Intro would also benefit from some information on trends in women's involvement in criminal justice system agencies that are being described—pregnant women data would be ideal, but I suspect this is not available. This would just be helpful so that readers have some sense of the trends in female CJS involvement over time.

Methods-

78-85: Clarify that this is a survey done annually (if indeed it is). Perhaps I missed this mention, but it's worth saying here in the methods that it is an annual survey.

87-91- Please explain how this survey asks about/records/verifies pregnancy status, so we know how the sample population is ascertained.

94-95: Could use a brief few words that 'referral for treatment' could be just for behavioral treatment—to underscore that medication treatment is something separate that is not necessarily a given with a referral.

118-123: How are these variables assessed by the person reporting? Do we know? Is it just "reported race/ethnicity" etc? Also, how does this survey distinguish for-profit facilities that do not accept insurance? The limitations section mentions something about this, but in explaining the database to the readers in the methods, some of this information is helpful up front.

Discussion

This section too needs a reminder that this is talking about people in the community who are referred by criminal justice system agencies-- that this is not about incarcerated people.

Somewhere in the discussion it's also worth noting, even as the authors note that the database does not distinguish methadone vs. buprenorphine, that buprenorphine did not become as widely used in pregnancy until after the results of the MOTHER trial were published in 2012. Is there some way that this might come into play?

The discussion section should acknowledge that there may be other reasons pregnant people are not on M-OUD—including that they decline medication treatment. This study could not assess for people who voluntarily do not want to me on medication. And people referred by CJS agencies are different than those from other sources, eg. may have experienced more coercion, mistrust etc, and this may account for some of the difference (maybe CJS referred people were offered medications but were more likely to decline) which this study cannot measure.

We cannot assume that people who had referrals by self or other did not have CJS involvement. They certainly might, it's just that a CJS agency wasn't the one who referred them to treatment. So can we really isolate this as CJS effect? Maybe people who were told by a CJS agency that they had to be in treatment didn't want to accept medication treatment offered to them, whereas women who referred themselves, even if they might have also had CJS involvement, would be more likely to be open to medication. So it's not just that as stated in lines 315-316 this study can't account for people with CJS involvement who were not referred, it's also that people who self referred may also have CJS involvement, it's just not mandated/part of the referral.

Any attachments provided with reviews can be seen via the following link:

[LINK]


24 Feb 2020

Submitted filename: 2020.02.24.Response to Reviewers.pdf

2 Apr 2020

Dear Dr. Winkelman,

Thank you very much for re-submitting your manuscript "Medications for opioid use disorder among pregnant women referred by criminal justice agencies before and after Medicaid expansion: A retrospective study of admissions to treatment centers in the United States" (PMEDICINE-D-19-03648R1) for review by PLOS Medicine.

I have discussed the paper with my colleagues and the academic editor and it was also seen again by reviewers. I am pleased to say that provided the remaining editorial and production issues are dealt with we are planning to accept the paper for publication in the journal.

The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:

[LINK]

Our publications team (plosmedicine@plos.org) will be in touch shortly about the production requirements for your paper, and the link and deadline for resubmission. DO NOT RESUBMIT BEFORE YOU'VE RECEIVED THE PRODUCTION REQUIREMENTS.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.

Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. If you haven't already, we ask that you provide a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract.

We expect to receive your revised manuscript within 1 week. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

If you have any questions in the meantime, please contact me or the journal staff on plosmedicine@plos.org.

We look forward to receiving the revised manuscript by Apr 09 2020 11:59PM.

Sincerely,

Caitlin Moyer, Ph.D.

Associate Editor

PLOS Medicine

plosmedicine.org

------------------------------------------------------------

Requests from Editors:

Abstract – please add summary demographic information, including mean age and also is it possible to say which cities or is it truly nationwide (this would also be helpful in the main text?; Can you please add a sentence of limitations as the final sentence of the ‘Methods and Findings section’, or at least be more explicit by starting that sentence with ‘Limitations of our study are……’

Where p values are for example .2, please write as 0.2

STROBE checklist – Please amend your checklist as it currently does not appear to have sections and paragraphs. The STROBE guideline can be found here: http://www.equator-network.org/reporting-guidelines/strobe/ When completing the checklist, please use section and paragraph numbers, rather than page numbers.

Comments from Reviewers:

Reviewer #1: Thank you to the author for addressing my previous comments well. I have no further issues to raise.

Reviewer #3: The authors have an outstanding job revising this manuscript and responding to my initial review. The article provides important insights and I recommend this article for publication.

I have two extremely minor suggestions:

Line 325- consider adding a citation to this new publication, which describes use of medications for OUD for pregnant and postpartum women in jails and prisons.

https://www.ncbi.nlm.nih.gov/pubmed/32141128

Line 353- add "in" between early and pregnancy

Thank you for the opportunity to review this revised manuscript.

Any attachments provided with reviews can be seen via the following link:

[LINK]


16 Apr 2020

Submitted filename: 2020.04.03.Response to Reviewers.docx

22 Apr 2020

Dear Dr. Winkelman,

On behalf of my colleagues and the academic editor, Dr. Zirui Song, I am delighted to inform you that your manuscript entitled "Medications for opioid use disorder among pregnant women referred by criminal justice agencies before and after Medicaid expansion: A retrospective study of admissions to treatment centers in the United States" (PMEDICINE-D-19-03648R2) has been accepted for publication in PLOS Medicine.

PRODUCTION PROCESS

Before publication you will see the copyedited word document (in around 1-2 weeks from now) and a PDF galley proof shortly after that. The copyeditor will be in touch shortly before sending you the copyedited Word document. We will make some revisions at the copyediting stage to conform to our general style, and for clarification. When you receive this version you should check and revise it very carefully, including figures, tables, references, and supporting information, because corrections at the next stage (proofs) will be strictly limited to (1) errors in author names or affiliations, (2) errors of scientific fact that would cause misunderstandings to readers, and (3) printer's (introduced) errors.

If you are likely to be away when either this document or the proof is sent, please ensure we have contact information of a second person, as we will need you to respond quickly at each point.

PRESS

A selection of our articles each week are press released by the journal. You will be contacted nearer the time if we are press releasing your article in order to approve the content and check the contact information for journalists is correct. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact.

PROFILE INFORMATION

Now that your manuscript has been accepted, please log into EM and update your profile. Go to https://www.editorialmanager.com/pmedicine, log in, and click on the "Update My Information" link at the top of the page. Please update your user information to ensure an efficient production and billing process.

Thank you again for submitting the manuscript to PLOS Medicine. We look forward to publishing it.

Best wishes,

Caitlin Moyer, Ph.D.

Associate Editor

PLOS Medicine

plosmedicine.org

https://www.researchpad.co/tools/openurl?pubtype=article&doi=10.1371/journal.pmed.1003119&title=Medications for opioid use disorder among pregnant women referred by criminal justice agencies before and after Medicaid expansion: A retrospective study of admissions to treatment centers in the United States&author=Tyler N. A. Winkelman,Becky R. Ford,Rebecca J. Shlafer,Anna McWilliams,Lindsay K. Admon,Stephen W. Patrick,Zirui Song,Caitlin Moyer,Clare Stone,Clare Stone,&keyword=&subject=Research Article,Medicine and Health Sciences,Women's Health,Obstetrics and Gynecology,Medicine and Health Sciences,Women's Health,Maternal Health,Pregnancy,Medicine and Health Sciences,Women's Health,Obstetrics and Gynecology,Pregnancy,Medicine and Health Sciences,Pharmaceutics,Drug Therapy,Medicine and Health Sciences,Pharmacology,Drugs,Analgesics,Opioids,Medicine and Health Sciences,Pain Management,Analgesics,Opioids,Medicine and Health Sciences,Pharmacology,Drugs,Opioids,Social Sciences,Law and Legal Sciences,Criminal Justice System,Medicine and Health Sciences,Women's Health,Maternal Health,Pregnancy,Management of High-Risk Pregnancies,Medicine and Health Sciences,Women's Health,Obstetrics and Gynecology,Pregnancy,Management of High-Risk Pregnancies,Medicine and Health Sciences,Health Care,Health Statistics,Morbidity,Research and Analysis Methods,Research Design,Survey Research,Census,