Duration of Medication Use for Opioid Use Disorder and Risk of Postpartum Overdose and Death

Ahra Kim; Andrew D Wiese; Sarah Osmundson; Chad You; Carlos G Grijalva; Margaret A Adgent; Alexandra C Sundermann; Andrew J Spieker; Amelie Pham; Stephen W Patrick; Ashley A Leech

doi:10.1001/jamanetworkopen.2026.37214

. 2026 Oct 2;9(10):e2637214. doi: 10.1001/jamanetworkopen.2026.37214

Duration of Medication Use for Opioid Use Disorder and Risk of Postpartum Overdose and Death

Ahra Kim ¹, Andrew D Wiese ¹, Sarah Osmundson ², Chad You ¹, Carlos G Grijalva ^1,³, Margaret A Adgent ¹, Alexandra C Sundermann ², Andrew J Spieker ⁴, Amelie Pham ², Stephen W Patrick ^5,^6,⁷, Ashley A Leech ^1,^✉

¹Department of Health Policy, Vanderbilt University School of Medicine, Nashville, Tennessee

²Department of Obstetrics and Gynecology, Vanderbilt University Medical Center, Nashville, Tennessee

³Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee

⁴Department of Biostatistics, Vanderbilt University Medical Center, Nashville, Tennessee

⁵Department of Health Policy and Management, Rollins School of Public Health, Emory University, Atlanta, Georgia

⁶Department of Pediatrics, Emory University School of Medicine, Emory University, Atlanta, Georgia

⁷Health Services Research Center, Emory University School of Medicine, Emory University, Atlanta, Georgia

Accepted for Publication: August 10, 2026.

Published: October 2, 2026. doi:10.1001/jamanetworkopen.2026.37214

Open Access: This is an open access article distributed under the terms of the CC-BY-NC-ND License, which does not permit alteration or commercial use, including those for text and data mining, AI training, and similar technologies. © 2026 Kim A et al. JAMA Network Open.

^✉

Corresponding Author: Ashley A. Leech, PhD, MS, Department of Health Policy, Vanderbilt University School of Medicine, 2525 West End Ave, Nashville, TN 37203 (ashley.leech@vanderbilt.edu).

Author Contributions: Mrs Kim and Dr Leech had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Wiese, Osmundson, Spieker, Leech.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Kim, Sundermann, Leech.

Critical review of the manuscript for important intellectual content: All authors.

Statistical analysis: Kim, Wiese, Spieker, Pham.

Obtained funding: Patrick.

Administrative, technical, or material support: Osmundson, You, Adgent, Patrick.

Supervision: Wiese, Pham, Patrick, Leech.

Conflict of Interest Disclosures: Dr Osmundson reported receiving grants from the National Institutes of Health (NIH) during the conduct of the study. Dr Grijalva reported receiving grants from the NIH during the conduct of the study; personal fees from Merck and GlaxoSmithKline; a contract from the Centers for Disease Control and Prevention and SyneosHealth; and grants from NIH and the Agency for Healthcare Research and Quality outside the submitted work. Dr Pham reported receiving grants from Novocuff and personal fees from Vanderbilt University Medical Center for serving on a data safety monitoring board for an NIH-funded study outside the submitted work. Dr Patrick reported receiving grants from NICHD during the conduct of the study, grants from National Institute on Drug Abuse, and personal fees as an expert consultant outside the submitted work. No other disclosures were reported.

Funding/Support: This study was funded by grant P50HD106446 from the National Institutes of Health, National Institutes of Child Health and Human Development to Drs Wiese, Patrick, and Leech.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Data Sharing Statement: See Supplement 2.

^✉

Corresponding author.

PMCID: PMC13633532 PMID: 42826020

Key Points

Question

Among pregnant and postpartum individuals with opioid use disorder (OUD) receiving medications for OUD (MOUD), is the duration of MOUD associated with lower 1-year risks of overdose or death?

Findings

This cohort study including 9360 pregnancies among 7776 individuals found that individuals with a greater days’ supply of MOUD receipt during pregnancy and early post partum were less likely to experience nonfatal overdose and death.

Meaning

Longer duration of MOUD during pregnancy and early post partum was associated with a reduced risk of death and overdose during the first year post partum, underscoring the importance of improving treatment continuity from pregnancy through the postpartum period.

This cohort study examines whether longer duration of medications for opioid use disorder dispensed during pregnancy and early post partum is associated with reduced overdose and death through 365 days post partum.

Abstract

Importance

Overdose is the leading cause of maternal mortality in the US, yet evidence on how medications for opioid use disorder (MOUD) affect postpartum overdose and mortality risk is limited. It is unclear whether cumulative MOUD exposure during this period reduces overdose or death beyond 90 days post partum.

Objective

To examine whether longer duration of MOUD dispensed during pregnancy and early post partum is associated with reduced overdose or death through 365 days post partum.

Design, Setting, and Participants

This is a retrospective cohort study of individuals delivering between 2007 and 2020, assessed through 1 year post partum or until the first outcome, through 2021. The study used Tennessee Medicaid (TennCare) data linked to vital records (birth and death certificates) and Tennessee Hospital Discharge Data. Participants were individuals aged 15 to 44 years with at least 1 day of dispensed MOUD during the exposure window (90 days before to 41 days after delivery), a delivery at least 20 weeks’ gestation, and continuous TennCare enrollment during the exposure period. Data were analyzed from July 2024 to July 2026.

Exposure

Total unique, nonoverlapping days of dispensed MOUD (buprenorphine and/or naltrexone) from 90 days before delivery through 41 days after delivery, with overlapping fills deduplicated so each calendar day was counted only once.

Main Outcomes and Measures

The primary outcome was the first occurrence of all-cause death or overdose (opioid or nonopioid) from inpatient, outpatient, and emergency department claims 42 to 365 days post partum. Multivariable Cox proportional hazards models were used to estimate the association between MOUD days’ supply and risk of death or overdose, adjusting for maternal and clinical characteristics.

Results

The cohort included 9360 pregnancies among 7776 individuals (median [IQR] age at delivery, 27 [24-31] years). The 1-year postpartum risk of death or overdose was 1.58% (148 individuals), with a median (IQR) time to event of 186.5 (92.5-258.8) days. The median (IQR) MOUD duration during the 132-day exposure period was 88 (43-111) days. Longer MOUD duration was associated with lower risk of death or overdose, with the lowest hazard at 132 days’ supply (adjusted hazard ratio for 132 vs 1 day, 0.52; 95% CI, 0.31-0.88).

Conclusions and Relevance

Longer duration of MOUD receipt during pregnancy and early post partum was associated with lower risk of overdose or death in the first year post partum, underscoring the importance of supporting continuous MOUD treatment during this period.

Introduction

Unintentional drug overdose is the leading cause of maternal death in the US.¹ Opioid use disorder (OUD) during pregnancy has increased more than 7-fold over the past 2 decades.^2,3,4 Without treatment, OUD in pregnancy has been associated with many adverse maternal and infant outcomes, such as maternal overdose and mortality, as well as preterm birth and low birth weight.^2,3,5,6,7 Medications for opioid use disorder (MOUD), including methadone, buprenorphine, and naltrexone, are Food and Drug Administration–approved for OUD treatment; however, methadone and buprenorphine are the preferred medications in pregnancy, with naltrexone used less frequently.^{2,5,6,7,8,9,10} Prior studies have shown that MOUD treatment is associated with lower risks of adverse maternal outcomes, including overdose and severe maternal morbidity,^7,11,12,13 as well as adverse neonatal outcomes, such as preterm birth and neonatal intensive care unit admissions.^7,11

Nationally, fewer than 1 in 4 individuals with OUD receive and sustain treatment for at least a year,^14,15 a gap strongly associated with opioid-related adverse events.¹⁶ Prior studies have examined MOUD use during pregnancy and early post partum but have not evaluated outcomes in the late postpartum period when pregnancy-associated deaths related to drug overdose commonly occur.¹⁷ Prior studies have shown that MOUD treatment during pregnancy is associated with improved maternal and infant outcomes⁷ and lower overdose rates among women receiving treatment,¹³ and that longer duration of MOUD treatment is associated with lower risk of nonfatal overdose during pregnancy and the early postpartum period.^11,12,13 However, the association between continuous MOUD treatment exposure and pregnancy-associated death or overdose in late post partum (42-365 days) has not been examined.

This gap is particularly important in high-burden settings such as Tennessee, where opioid-related diagnoses at delivery were nearly double the national average at 16.1 per 1000 delivery hospitalizations in 2017.³ Access to treatment has remained limited throughout and beyond the study period; as of 2022, only 22 of the state’s opioid treatment programs were located in 15 of 95 counties.^3,18 More recently, during 2021 to 2023, mental health conditions (most commonly substance use disorders) were the leading cause of pregnancy-related deaths in Tennesse, accounting for more than one-half of deaths occurring 43 days to 1 year post partum.¹⁹ Medicaid is a critical source of insurance for pregnant individuals, as it provides coverage for 41% of all births in the US,²⁰ and particularly among those with substance use disorders and OUD,^3,21,22 making it the largest payer for this population. In 2022, Tennessee Medicaid (TennCare) covered 84% of births where an infant was exposed to opioids.²³

To our knowledge, the association between cumulative MOUD exposure during pregnancy and early post partum and subsequent pregnancy-associated death or nonfatal overdose during the late postpartum period has not been evaluated. Recognizing the need for additional studies of MOUD in these high-risk populations, our study focused on pregnant and postpartum individuals enrolled in TennCare to examine MOUD treatment patterns and associated maternal outcomes. We hypothesized that a greater cumulative duration of MOUD would be associated with reduced risks of death or overdose among pregnant individuals receiving MOUD.

Methods

Study Design and Data Sources

We conducted an observational, retrospective cohort study among pregnant patients enrolled in TennCare. We obtained data from a previously established linked database that integrates TennCare enrollment, administrative claims, and pharmacy files, vital records (including birth and death certificates with causes of deaths), and data from the Tennessee Hospital Discharge Data System (HDDS) through probabilistic record linkage. By linking HDDS data, we captured overdoses and deaths beyond those recorded in claims alone, particularly during periods without TennCare coverage. Details on the linked database have been described previously.²⁴

This study was approved by the institutional review boards at Vanderbilt University Medical Center and the Tennessee Department of Health, and by the Division of TennCare. The study was approved with a waiver of consent. This study follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines for cohort studies.

Population

We included pregnant and postpartum individuals aged 15 to 44 years with at least 1 day of MOUD receipt (buprenorphine or naltrexone) (eTable 1 in Supplement 1) during the exposure window (90 days before delivery through 41 days post partum; baseline) who delivered at 20 weeks’ gestation or later between 2007 and 2020, with postpartum follow-up extending into 2021. We restricted our study to live births because birth certificate data are only available for live-born infants. Early pregnancy losses were not included since they may not result in a medical encounter and are, therefore, undercounted in claims data. We did not include methadone for OUD because TennCare did not begin covering it for this purpose until 2020, the final year of the delivery cohort; we did not observe any TennCare-covered methadone claims for OUD during the exposure window. Eligible participants also had continuous TennCare enrollment during the exposure window, permitting gaps of 5 days or fewer (Figure 1 and Figure 2). We excluded individuals with evidence of sickle cell disease during baseline because it is associated with high opioid prescribing for pain management (Figure 1).

Flowchart of pregnancy cohort selection with exclusions and final groups. Light blue and white vertical flowchart with rectangular nodes connected by downward arrows and several rightward side boxes for exclusions. Top center large box begins with bold 17934 followed by text listing enrollment criteria: pregnancies meeting enrollment criteria; aged 15 to 44 y; evidence of O U D diagnosis and or medication for O U D, minus 90 to 41 d post partum; delivered an infant greater than or equal to 20 wk’ gestation, 2007 to 2020; continuous TennCare enrollment, minus 90 to 41 d post partum; gaps in enrollment less than or equal to 5 d; no sickle cell disease. A downward arrow leads to a long horizontal box reading 17233 after excluding duplicate records from multiple gestation deliveries, twins triplets. From this level, a rightward arrow points to a separate box reading 701 excluded: duplicate records from multiple gestation deliveries. A downward arrow continues to a long box reading 17221 after excluding overdose in exposure window, with a rightward arrow to a side box reading 12 excluded: overdose in exposure window. A downward arrow leads to a long box reading 9420 greater than or equal to 1 d M O U D receipt, with a rightward arrow to a side box reading 7801 excluded: no M O U D receipt. A downward arrow leads to a long box reading 9360 final analytic sample. From this level, a rightward arrow points to a side box reading 60 excluded: naltrexone pregnancies with no clear O U D evidence, 5 A U D dx only; 55 neither O U D nor A U D diagnosis. From the final analytic sample box, the main flow continues downward and splits into two bottom boxes: left box reading 9085 buprenorphine only, and right box reading 275 naltrexone only, with O U D diagnosis, or with buprenorphine. — The outcome follow-up period extended through 2021. The final analytic sample included 9360 pregnancies with at least 1 day of MOUD receipt during the exposure window (90 days before delivery to 41 days after delivery), further stratified by MOUD type: buprenorphine only (9085 pregnancies) and naltrexone only (with OUD diagnosis) or naltrexone with buprenorphine (275 pregnancies). AUD indicates alcohol use disorder.

Flowchart timeline of baseline and follow-up periods around delivery. Five light gray rectangular nodes arranged left to right across the upper half of the figure, connected by right-pointing arrows. The leftmost node, in the upper left, contains two lines of text: 90 d Before delivery on the first line and (T −132) on the second line. A right-pointing arrow leads to the second node, positioned left of center, labeled Delivery date on the first line and (T −42) on the second line. Another right-pointing arrow leads to the third node near the center, labeled 41 d Post partum on the first line and (T −1) on the second line. A right-pointing arrow leads to the fourth node right of center, labeled 42 d Post partum on the first line and (T 0) on the second line. A final right-pointing arrow leads to the fifth node in the upper right, labeled 365 d Post partum on the first line and (T 323) on the second line. Beneath the first three nodes, a horizontal bracket line spans from below the first node to below the third node, with short vertical connector lines rising to the first and third nodes; centered under this bracket is the label Baseline period. Beneath the fourth and fifth nodes, a second horizontal bracket line spans from below the fourth node to below the fifth node, with short vertical connector lines rising to the fourth and fifth nodes; centered under this bracket is the label Follow-up period. No axes, scales, legends, or additional symbols are present. — The baseline period includes 90 days before delivery (time [T] −132) through 41 days post partum (T −1), anchored to the delivery date (T −42). The follow-up period begins at 42 days post partum (T 0) and extends through 365 days post partum (T 323). Time points are expressed relative to the start of the follow-up period.

Outcome

Follow-up started on postpartum day 42 and continued though postpartum day 365, or until overdose or death, whichever occurred first. Mortality outcomes reflected pregnancy-associated deaths occurring up to 1 year post partum, regardless of the cause, and does not include pregnancy-related deaths that occur before 42 days post partum.^19,25 Individuals who did not experience an event were censored at 365 days post partum. We identified overdose outcomes using International Classification of Diseases, Ninth Revision and/or International Statistical Classification of Diseases and Related Health Problems, Tenth Revision diagnosis code algorithms (eTable 2 in Supplement 1), using hospitalization and emergency department claims in TennCare during enrollment and HDDS records for events that occurred outside of TennCare enrollment. We identified deaths using vital statistics death certificate data. In all analyses, we counted only the first overdose per person. Because HDDS data were only available through 2019, we conducted a sensitivity analysis restricting follow-up through 2019.

Exposure

We used pharmacy claims to calculate each individual’s total days’ supply of nonoverlapping MOUD (buprenorphine or naltrexone) from 90 days before delivery through 41 days post partum, with possible values ranging from 1 to 132 days’ supply of MOUD. A numeric example illustrating this method is provided in eFigure 1 in Supplement 1. For example, if an individual received a 15-day supply of MOUD followed by another 15-day supply the following day, we counted 16 distinct days of coverage. Using the standard shift-forward approach would count 30 days of coverage in this example (sensitivity analysis; eFigure 1 in Supplement 1). For our primary analysis, we counted nonoverlapping days to avoid inflating the total days’ supply and to maintain a conservative estimate.

Covariates

We identified demographic and clinical information during the baseline period (90 days before through 41 days after delivery), including race and ethnicity, calendar year of delivery, age at delivery, gestational weeks at delivery, route of delivery (cesarean or vaginal), number of previous pregnancies, presence of a pain condition (fibromyalgia, back pain, dental pain, evidence of nonsteroidal anti-inflammatory drug use, and anticonvulsant use), use of opioids for pain, anxiety or depression, alcohol or other substance use disorder (cannabis, cocaine, sedative, stimulant, hallucinogen, and other psychoactive substance related disorders), sedative or relaxant prescriptions (benzodiazepines or muscle relaxants), tobacco use, Social Vulnerability Index geocoded to the mother’s residential address, and severe maternal morbidity defined by the Centers for Disease Control and Prevention.²⁶ Data used for race and ethnicity come from a third-party data processor that creates 5 mutually exclusive categories of race and ethnicity. Race and ethnicity data were included in the analysis to account for potential confounding related to differential uptake and effectiveness of MOUD by race and ethnicity. We obtained diagnoses using International Classification of Diseases, Ninth Revision and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes and evidence from prescription fills from pharmacy claims (eTable 3 in Supplement 1). We described the average daily dose of MOUD over the entire exposure period and separately for the antepartum and postpartum periods.

Statistical Analysis

Data were analyzed from July 2024 to July 2026. We computed descriptive statistics using frequencies (proportions) and medians (IQRs) as appropriate. The primary analysis examined the association between the number of days’ supply of MOUD and the risk of death or overdose in late post partum (42-365 days) for deliveries occurring between 2007 and 2020, with follow-up through 2021. We included a secondary analysis examining the timing of MOUD receipt (during pregnancy, post partum, or both periods) (eTable 4 in Supplement 1). We applied sequential multivariate imputation (50 iterations) to account for missing information for race and ethnicity (0.3%), delivery type (0.3%), number of previous pregnancies (2.1%), and missing Social Vulnerability Index data (6.8%). For both analyses, we first used a Kaplan-Meier estimator to compute the unadjusted survival probability of the cohort. We then used Cox proportional hazards models, upon checking assumptions, to obtain univariate and multivariable hazard ratios (HRs) with 95% CIs to estimate the crude and adjusted risks of the outcomes after controlling for covariates, respectively. A 95% CI excluding the null value of 1 indicated statistical significance. We used a linear term for number of days’ supply in the model and report adjusted HRs (aHRs) at specific values and risk differences of days’ supply (30, 60, 90, and 132 days’ supply [ie, the full exposure period] compared with 1-day supply) with all covariates set to their reference values. We fit robust SEs clustered at the individual level to account for multiple pregnancies within the same individual over time.

We performed sensitivity analyses to assess model robustness. First, we included cubic spline terms in the model to account for the potential of a nonlinear association between days’ supply and the outcomes (eTable 5 and eFigure 2 in Supplement 1). In addition, we used inverse probability of treatment weighting to adjust for confounding, given the small number of outcomes, and conducted a sensitivity analysis restricting MOUD exposure days to buprenorphine only, given that most of the population received buprenorphine and naltrexone contributed minimal days’ supply overall. In this analysis, we included individuals who received both medications, but counted only their buprenorphine days toward exposure, and excluded 172 naltrexone-only recipients. We also conducted a sensitivity analysis limiting to deliveries between 2007 to 2018, with follow-up through 2019, the period for which HDDS data were available, allowing us to capture overdoses regardless of TennCare enrollment. We also recalculated total MOUD days’ supply using a shift-forward approach (moving overlapping supply forward to subsequent uncovered days). An example illustrating this method is shown in eFigure 1 in Supplement, and results of the prespecified sensitivity analyses are shown in eTable 6 in Supplement 1. We conducted all analyses using R statistical software version 4.5.0 (R Project for Statistical Computing) and Stata statistical software version 19 (StataCorp).

Results

Study Population

Our analysis included a cohort of 9360 pregnancies among 7776 unique individuals, with a median (IQR) age of 27 (24-31) years at delivery, and a median (IQR) gestation of 39 (37-39) weeks at delivery. The median (IQR) calendar year of delivery was 2016 (2014-2018), and 64.96% of the sample (6080 pregnancies) had a vaginal delivery. Among the sample, we identified 39 deaths (0.42%) and 109 patients (1.16%) with at least 1 overdose postpartum. Individuals who experienced an adverse postpartum outcome had a median (IQR) time to event of 186.5 (92.5-258.8) days; the median (IQR) was 156.0 (57.5-238.0) days for deaths and 208.0 (111.0-262.0) days for overdoses (opioid or nonopioid). Most individuals in our sample received buprenorphine only (9085 individuals [97.06%]).

Overall, the median (IQR) number of days’ supply of MOUD was 88 (43-111). The days’ supply of MOUD ranged from 1 to 132 days in both the combined buprenorphine and naltrexone cohort and buprenorphine-only group. Table 1 describes additional characteristics of the study cohort.

Table 1. Descriptive Characteristics of Study Cohort.

Characteristic	Pregnancies, No. (%) (N = 9360)
Demographic and clinical
Race and ethnicity^a
Hispanic	79 (0.84)
Non-Hispanic Asian	11 (0.12)
Non-Hispanic Black	247 (2.64)
Non-Hispanic White	8927 (95.37)
Non-Hispanic other	62 (0.66)
Missing	34 (0.36)
Delivery type
Vaginal	6080 (64.96)
Cesarean	3246 (34.68)
Missing	34 (0.36)
Year of delivery, median (IQR)	2016 (2014-2018)
Age at delivery, median (IQR), y	27 (24-31)
Gestational age at delivery, median (IQR), wk	39 (37-39)
Previous pregnancies (n = 9160)
Median (IQR)	2 (1-3)
Missing	200 (2.14)
Social Vulnerability Index of residence (n = 8724)
Median (IQR)	58.0 (37.0-78.4)
Missing	636 (6.79)
Chronic pain conditions	7168 (76.58)
Back pain	3198 (34.17)
Dental pain	348 (3.72)
Fibromyalgia	456 (4.87)
Nonsteroidal anti-inflammatory drug use	5697 (60.87)
Anticonvulsant use	495 (5.29)
Receiving other opioids for pain	3334 (35.62)
Evidence of tobacco use	8017 (85.65)
Alcohol or other substance use	3988 (42.61)
Sedative or relaxant prescriptions (benzodiazepines or muscle relaxants)	1276 (13.63)
Severe maternal morbidity^b	638 (6.82)
Anxiety and depression combined	4062 (43.40)
Personality, bipolar, or schizophrenia disorders	1317 (14.62)
MOUD type
Buprenorphine only	9085 (97.06)
Naltrexone only	172 (1.84)
Both	103 (1.10)
Primary buprenorphine formulation^c
Buccal film	127 (1.38)
Sublingual tablet	8603 (93.63)
Sublingual film	455 (4.95)
MOUD information, median (IQR)
Total days’ supply MOUD exposure^d	88 (43-111)
Total days’ supply MOUD before delivery	62 (28-77)
Total days’ supply MOUD after delivery	30 (12-40)
Total days’ supply buprenorphine exposure	89 (45-112)
Total days’ supply buprenorphine before delivery	62 (29-78)
Total days’ supply buprenorphine after delivery	30 (12-40)
Average daily dose of buprenorphine, mg^e	14.00 (9.14-16.00)
Average daily dose of buprenorphine before delivery, mg^e	13.86 (8.00-16.00)
Average daily dose of buprenorphine after delivery, mg^e	9.86 (4.00-16.00)
Average daily dose: sublingual tablet	14.40 (9.91-16.00)
Average daily dose: sublingual film	14.04 (8.00-16.00)
Average daily dose: buccal film	6.25 (4.20-8.40)
Outcomes
Death or overdose (combined)	148 (1.58)
Any death	39 (0.42)
First occurrence of any overdose (nonfatal)	109 (1.16)
Censoring
Loss of enrollment	1057 (11.29)
Censored (day 365)	8155 (87.13)
Time to event among those with outcome, median (IQR), d
Either death or overdose	186.5 (92.5-258.8)
Any death	156.0 (57.5-238.0)
First occurrence of any overdose (nonfatal)	208.0 (111.0-262.0)

Open in a new tab

Abbreviation: MOUD, medications for opioid use disorder.

^{^a}

Data used for race and ethnicity come from a third-party data processor that creates 5 mutually exclusive categories of race and ethnicity. This grouping may not reflect other sources of Medicaid data in Tennessee (like T-MSIS). Disparities by racial and ethnic group may be underestimated because of these differences.

^{^b}

Severe maternal morbidity reflects the Centers for Disease Control and Prevention definition.²⁶

^{^c}

Pregnancy-level counts reflect the most frequently dispensed formulation during the exposure window. Pregnancies with naltrexone-only MOUD receipt (n = 172) are excluded from this row. We further excluded extended-release injectable and other sublingual formulations because of cell sizes less than 11.

^{^d}

The primary method for accumulating days of MOUD supply (varied in sensitivity analysis) was as follows: Calculated as the number of unique, nonoverlapping calendar days covered by prescriptions during the 90 days before delivery to 41 days after delivery exposure window. Overlapping fills are deduplicated so each calendar day is counted only once. Predelivery days exceeding 90 carry over into the postdelivery period (capped at 41 days). Possible range of values include 1 to 132 days.

^{^e}

Average dose reported for sublingual film, sublingual tablet, and buccal film only. Excluded from dose summary are extended-release injectable buprenorphine (n = 10 fills and 8 pregnancies), compounded powder formulation (n = 2 fills), fills with 0 days’ supply (n = 1), and naltrexone (not a buprenorphine formulation). For each prescription, we calculated average dose per day as the product of the quantity dispensed and the strength per unit divided by the total days’ supply, reported in table as median (IQR) across pregnancies.

Risk of Death or Overdose

The risk of death or first overdose at 1 year post partum was 1.58% (148 individuals). Greater MOUD duration was associated with lower risk of overdose or death, with lower estimated HRs for each additional 30 days’ supply of MOUD and with the lowest estimated HR observed for 132 days’ supply (aHR, 0.52; 95% CI, 0.31-0.88) compared with a 1-day supply (Table 2).

Table 2. Unadjusted and Adjusted HRs for First Occurrence of Overdose or Death for Selected Days’ Supply vs 1-Day Supply of MOUD Among Pregnant Individuals With Opioid Use Disorder and Any Evidence of Buprenorphine Treatment During Pregnancy and Post Partum, Tennessee Medicaid (2010-2021).

No. of days’ supply	HR (95% CI)		Absolute risk difference, % (95% CI)^c
No. of days’ supply	Unadjusted^a	Adjusted^b	Absolute risk difference, % (95% CI)^c
30	0.83 (0.73 to 0.93)	0.86 (0.76 to 0.97)	−0.15 (−0.32 to 0.03)
60	0.69 (0.53 to 0.86)	0.74 (0.58 to 0.94)	−0.28 (−0.60 to 0.04)
90	0.58 (0.37 to 0.78)	0.64 (0.45 to 0.92)	−0.39 (−0.82 to 0.03)
132	0.45 (0.21 to 0.68)	0.52 (0.31 to 0.88)	−0.53 (−1.07 to 0.01)

Open in a new tab

Abbreviations: HR, hazard ratio; MOUD, medications for opioid use disorder.

^{^a}

We calculated unadjusted HRs with a Cox proportional hazards model with a linear variable for MOUD days’ supply only.

^{^b}

We calculated adjusted HRs with a Cox proportional hazards model with a linear variable for MOUD days’ supply accounting for delivery year, mother’s age at delivery, gestational weeks at delivery, race and ethnicity, evidence of comorbid pain, anxiety or depression, personality disorder, tobacco use, alcohol or other substance use, concomitant use of other medications (opioids for pain, benzodiazepines, muscle relaxants, nonsteroidal anti-inflammatory drug, or anticonvulsants), severe maternal morbidity, delivery type, number of previous pregnancies, and Social Vulnerability Index geocoded to the mother’s residential address. We used sequential multivariate imputation (50 imputed datasets) to account for missing covariate information for a small proportion with missing race and ethnicity, delivery type, number of previous pregnancies, and Social Vulnerability Index. We calculated exponentiated linear estimates with all covariates set to their reference values: absence of comorbidities, integer values to their median (previous pregnancies, gestational weeks, maternal age, and delivery year) and continuous values to their mean (Social Vulnerability Index).

^{^c}

We calculated absolute risk differences at 365 days post partum and expressed as a percentage. The 95% CIs were pooled after bootstrapping (1000 replications) each imputed dataset.

Secondary Analysis on the Timing of MOUD Receipt

The 1-year risk of death or overdose was 1.41% (95% CI, 1.14%-1.68%) among individuals who received MOUD both before and after delivery, compared with 1.89% (95% CI, 1.12%-2.65%) among those who received MOUD only before delivery and 2.58% (95% CI, 1.51%-3.63%) among those who received MOUD only post partum. These probabilities are statistically different. The multivariable model showed that compared with individuals who received MOUD during both antepartum and postpartum periods, those receiving MOUD only antepartum had an elevated, albeit not significant, hazard of the outcome (aHR, 1.26; 95% CI, 0.80-1.99) (eTable 4 in Supplement 1). However, those receiving MOUD only post partum had a significantly higher hazard of the outcome compared with those receiving MOUD in both periods (HR, 1.71; 95% CI, 1.07-2.72). This association was attenuated after adjusting for the total days’ supply.

Sensitivity Analyses

Results were consistent across sensitivity analyses. The association between days’ supply and the outcome were similar when using cubic spline terms (eTable 5 and eFigure 2 in Supplement 1) and when using inverse probability of treatment weighting (aHR for 132-day supply compared with 1-day supply, 0.53; 95% CI, 0.24-0.83) (eTable 6 in Supplement 1). Restricting MOUD exposure to buprenorphine days only (excluding 172 naltrexone-only recipients) did not meaningfully change the findings (aHR for 132-day supply compared with 1-day supply, 0.44; 95% CI, 0.26-0.75), suggesting that the observed associations were not driven by naltrexone. Results were also consistent when limiting to deliveries between 2007 and 2018 with follow-up through 2019 (aHR for 132-day supply compared with 1-day supply, 0.38; 95% CI, 0.19-0.76). Using the shift-forward approach to calculate days’ supply resulted in little difference from the primary exposure definition (aHR for 132-day supply compared with 1-day supply, 0.62; 95% CI, 0.37-1.02) (eTable 6 in Supplement 1).

Discussion

In this cohort study of pregnant and postpartum individuals with MOUD receipt, longer duration of MOUD use was associated with lower risk of death or nonfatal overdose. This finding was consistent when we restricted the cohort to individuals receiving buprenorphine only, suggesting that the observed association was not driven by the inclusion of naltrexone. Even among individuals receiving MOUD in both pregnancy and postpartum, the most protected group, the 1-year risk of death or nonfatal overdose was 1.41% (more than 1 in 100), which is an exceptionally high risk of mortality and overdose even with optimal treatment.

These findings are consistent with prior research on buprenorphine’s protective effects on reducing adverse outcomes in pregnancy and postpartum.^{7,11,12,13,27} For instance, Jarlenski et al¹³ found that the risk of nonfatal overdose during pregnancy decreased as the duration of MOUD use increased among commercially insured individuals. Using Pennsylvania Medicaid data, Krans et al¹¹ found that nonfatal overdose during pregnancy and through 12 weeks post partum was more frequent among those with shorter durations of MOUD use, whereas another study by Jarlenski et al¹² found that nonfatal overdose through 90 days post partum was more frequent among those who discontinued buprenorphine or received lower doses during pregnancy. A longer duration of buprenorphine use during pregnancy was also associated with improved buprenorphine continuation through 90 days post partum.¹² Comparing buprenorphine treatment with no treatment, Krishnapura et al⁷ found that buprenorphine use during pregnancy was associated with lower rates of severe maternal morbidity and maternal death through 6 weeks post partum among TennCare enrollees. Similarly, Schiff et al²⁷ compared overdose rates with and without pharmacotherapy receipt in each month across the year before delivery and each 3-month postpartum interval through 1 year post partum. Overall, the study found a lower overdose rate among those receiving pharmacotherapy. However, that study examined monthly pharmacotherapy receipt rather than cumulative MOUD exposure during pregnancy and early post partum.²⁷

Our study extends this work by examining the association between cumulative MOUD exposure during pregnancy and early post partum and overdose or death in late post partum (42-365 days) among TennCare enrollees, a period and population not previously studied. We found that longer duration of MOUD during pregnancy and early post partum was linked with a lower risk of overdose or death in late post partum. We reported dose descriptively in our study since dose may influence duration of MOUD receipt. Future research examining dose-response relationships on maternal outcomes would complement these findings. Given that drug overdose is the leading cause of pregnancy-associated death in the state of Tennessee and the US more broadly,^1,28,29 our findings underscore the importance of supporting MOUD continuity among pregnant and postpartum individuals with OUD, particularly within Medicaid, a population at elevated risk. Patrick et al¹⁴ found that pregnant individuals in the US were less likely than nonpregnant individuals to obtain an appointment with a buprenorphine prescriber, and that approximately one-quarter of buprenorphine prescribers granted appointments only when patients agreed to pay cash. Initiatives that integrate MOUD into prenatal care settings with social work and care coordination services, such as the CMS Maternal Opioid Misuse Model, may reduce barriers and improve treatment retention through the postpartum period.

Strengths and Limitations

Our study leveraged the unique integration of multiple data sources to identify a large retrospective cohort of pregnant and postpartum patients with MOUD receipt, enabling measurement of MOUD use, baseline characteristics, and outcomes. A strength of this data infrastructure is the linkage to HDDS and death certificate data, allowing us to capture overdoses and deaths, even among individuals who lost TennCare eligibility during follow-up. Despite this, however, we were limited by the small number of observed outcomes, which restricted the degrees of freedom available for covariate adjustment and introduced the possibility of residual confounding. In addition, outcomes may have been incompletely captured for individuals who moved out of state during follow-up because events occurring outside of Tennessee would not be available.

Furthermore, we used prescription fill data to characterize the days’ supply of MOUD for which prior work has shown moderate to strong concordance between pharmacy prescription fills and self-reported use in pregnant individuals.^30,31,32 We did not capture medication fills outside of pharmacy claims, and treatment duration may have been underestimated during prolonged hospitalizations when outpatient prescriptions were not filled. Our study was limited to pregnant individuals who received at least 1 day of MOUD, and our analyses did not compare those receiving treatment vs not receiving treatment. Longer duration of MOUD receipt may also be a marker of treatment engagement, continuity of care, and other patient or health factors associated with overdose risk or death.

Our study covered a period of evolving OUD treatment, postpartum care guidelines, and Medicaid policy changes. We adjusted for calendar year of delivery in our models; however, the small number of outcomes limited our ability to stratify analyses by calendar year to examine temporal variation in the association between MOUD duration and the outcome. Although severe maternal morbidity could lie in the pathway between MOUD exposure and outcomes,⁷ the risk is highest in the immediate postpartum period^33,34 and was, therefore, treated as a baseline covariate given our outcome follow-up began at 42 days post partum.

Conclusions

In this observational cohort study among pregnant individuals enrolled in TennCare with at least 1 day of MOUD receipt, we found that longer duration of MOUD during pregnancy and early post partum was associated with a lower risk of overdose or pregnancy-associated death in late post partum (42-365 days). These findings highlight the importance of supporting sustained treatment in this population.

Supplement 1.

eTable 1. List of names used to identify buprenorphine or naltrexone for opioid use disorder

eTable 2. List of ICD-9 and ICD-10 codes used to identify outcomes

eFigure 1. Graphical example to calculate total days’ supply of MOUD

eTable 3. List of ICD-9 and ICD-10 codes used to identify covariates

eFigure 2. Predicted adjusted hazard ratios for overdose and death at each observed value of MOUD days’ supply

eTable 4. Secondary analysis: unadjusted and adjusted hazard ratios for overdose and death associated with timing of MOUD receipt (pregnancy only, postpartum only, or both), with and without days’ supply in the model, Tennessee Medicaid (2010-2021)

eTable 5. Cubic spline model: unadjusted and adjusted hazard ratios for overdose and death at observed values of days’ supply of medications for opioid use disorder (MOUD) during pregnancy and postpartum, Tennessee Medicaid (2010-2021)

eTable 6. Results of pre-specified sensitivity analyses

jamanetwopen-e2637214-s001.pdf^{(368.9KB, pdf)}

Supplement 2.

Data Sharing Statement

jamanetwopen-e2637214-s002.pdf^{(14KB, pdf)}

References

1.Azad HA, Goin D, Nathan LM, et al. Overdose, homicide, and suicide as causes of maternal death in the United States. N Engl J Med. 2026;394(7):722-723. doi: 10.1056/NEJMc2512078 [DOI] [PubMed] [Google Scholar]
2.Haight SC, Ko JY, Tong VT, Bohm MK, Callaghan WM. Opioid use disorder documented at delivery hospitalization—United States, 1999-2014. MMWR Morb Mortal Wkly Rep. 2018;67(31):845-849. doi: 10.15585/mmwr.mm6731a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Hirai AH, Ko JY, Owens PL, Stocks C, Patrick SW. Neonatal abstinence syndrome and maternal opioid-related diagnoses in the US, 2010-2017. JAMA. 2021;325(2):146-155. doi: 10.1001/jama.2020.24991 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Zipursky J, Juurlink DN. Opioid use in pregnancy: an emerging health crisis. Obstet Med. 2021;14(4):211-219. doi: 10.1177/1753495X20971163 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Centers for Disease Control and Prevention . Opioid use during pregnancy. May 7, 2025. Accessed August 31, 2026. https://www.cdc.gov/opioid-use-during-pregnancy/about/index.html
6.Krans EE, Patrick SW. Opioid use disorder in pregnancy: health policy and practice in the midst of an epidemic. Obstet Gynecol. 2016;128(1):4-10. doi: 10.1097/AOG.0000000000001446 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Krishnapura SR, McNeer E, Loch SF, et al. Buprenorphine treatment in pregnancy and maternal-infant outcomes. JAMA Health Forum. 2025;6(4.11):e251814. doi: 10.1001/jamahealthforum.2025.1814 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.National Academies of Sciences, Engineering, and Medicine . Medications for Opioid Use Disorder Save Lives. The National Academies Press; 2019. doi: 10.17226/25310 [DOI] [PubMed] [Google Scholar]
9.Substance Abuse and Mental Health Services Administration . Clinical guidance for treating pregnant and parenting women with opioid use disorder and their infants: HHS publication 18-5054. 2018. Accessed August 31, 2026. https://library.samhsa.gov/sites/default/files/sma18-5054.pdf
10.Leech AA, Garbett S, Yu HA, et al. Cost-effectiveness of treatment for opioid use disorder in pregnancy and its impact on birth outcomes. JAMA Pediatr. 2025;179(11):1203-1216. doi: 10.1001/jamapediatrics.2025.3067 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Krans EE, Kim JY, Chen Q, et al. Outcomes associated with the use of medications for opioid use disorder during pregnancy. Addiction. 2021;116(12):3504-3514. doi: 10.1111/add.15582 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Jarlenski M, LoCiganic WH, Chen Q, et al. Association between buprenorphine dose and outcomes among pregnant persons with opioid use disorder. Am J Obstet Gynecol. 2025;233(1):59.e1-59.e15. doi: 10.1016/j.ajog.2024.12.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Jarlenski M, Chen Q, Gao A, Rothenberger SD, Krans EE. Association of duration of methadone or buprenorphine use during pregnancy with risk of nonfatal drug overdose among pregnant persons with opioid use disorder in the US. JAMA Netw Open. 2022;5(4):e227964. doi: 10.1001/jamanetworkopen.2022.7964 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Patrick SW, Richards MR, Dupont WD, et al. Association of pregnancy and insurance status with treatment access for opioid use disorder. JAMA Netw Open. 2020;3(8):e2013456. doi: 10.1001/jamanetworkopen.2020.13456 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Leech AA, McNeer E, Roberts AW, et al. Buprenorphine out-of-pocket costs and discontinuation in privately insured adults with opioid use disorder. JAMA Intern Med. 2023;183(9):1023-1026. doi: 10.1001/jamainternmed.2023.2826 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Samples H, Williams AR, Crystal S, Olfson M. Impact of long-term buprenorphine treatment on adverse health care outcomes in Medicaid. Health Aff (Millwood). 2020;39(5):747-755. doi: 10.1377/hlthaff.2019.01085 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Margerison CE, Roberts MH, Gemmill A, Goldman-Mellor S. Pregnancy-associated deaths due to drugs, suicide, and homicide in the United States, 2010-2019. Obstet Gynecol. 2022;139(2):172-180. doi: 10.1097/AOG.0000000000004649 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Risby J, Schlesinger E, Geminn W, Cernasev A. Methadone treatment gap in Tennessee and how medication units could bridge the gap: a review. Pharmacy (Basel). 2023;11(5):131. doi: 10.3390/pharmacy11050131 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Tennessee Department of Health, Division of Family Health & Wellness . Maternal mortality in Tennessee: annual report 2025. December 10, 2025. Accessed August 31, 2026. https://www.tn.gov/content/dam/tn/health/2025-notcom/Maternal%20Mortality%20Review%20Report%202025.pdf
20.Kaiser Family Foundation. Births financed by Medicaid by metropolitan status. Accessed August 31, 2026. https://www.kff.org/medicaid/state-indicator/births-financed-by-medicaid/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D
21.Jarlenski M, Krans EE, Chen Q, et al. Substance use disorders and risk of severe maternal morbidity in the United States. Drug Alcohol Depend. 2020;216:108236. doi: 10.1016/j.drugalcdep.2020.108236 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Jarlenski M, Kim JY, Ahrens KA, et al. Healthcare patterns of pregnant women and children affected by OUD in 9 state Medicaid populations. J Addict Med. 2021;15(5):406-413. doi: 10.1097/ADM.0000000000000780 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Tennessee Department of Health . Neonatal abstinence syndrome in Tennessee 2022: 2024 NAS surveillance annual report. July 2024. Accessed August 31, 2026. https://www.tn.gov/content/dam/tn/health/program-areas/NAS-Annual-Report.pdf
24.Whitmore CC, Hawley RE, Min JY, et al. Building a data linkage foundation for mother-child pharmacoepidemiology research. Pharmaceut Med. 2021;35(1):39-47. doi: 10.1007/s40290-020-00371-8 [DOI] [PubMed] [Google Scholar]
25.Hoyert DL. Maternal mortality rates in the United States, 2024. Centers for Disease Control and Prevention. March 4, 2026. Accessed August 31, 2026. https://stacks.cdc.gov/view/cdc/174651
26.Centers for Disease Control and Prevention . Identifying severe maternal morbidity (SMM). May 15, 2024. Accessed August 31, 2026. https://www.cdc.gov/maternal-infant-health/php/severe-maternal-morbidity/icd.html
27.Schiff DM, Nielsen T, Terplan M, et al. Fatal and nonfatal overdose among pregnant and postpartum women in Massachusetts. Obstet Gynecol. 2018;132(2):466-474. doi: 10.1097/AOG.0000000000002734 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Wallace ME, Jahn JL. Pregnancy-associated mortality due to homicide, suicide, and drug overdose. JAMA Netw Open. 2025;8(2):e2459342. doi: 10.1001/jamanetworkopen.2024.59342 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Centers for Disease Control and Prevention . Circumstances contributing to pregnancy-related deaths: data from maternal mortality review committees in 38 U.S. States, 2020. April 20, 2026. Accessed August 31, 2026. https://archive.cdc.gov/#/details?url=https://www.cdc.gov/maternal-mortality/php/report/index.html
30.Olstad EW, Nordeng H, Bjørk MH, Selmer KK, Gervin K. Agreement between parental self-reported antiseizure medication use and dispensed prescription records from a national prescription database. Pharmacoepidemiol Drug Saf. 2025;34(4):e70139. doi: 10.1002/pds.70139 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Cheung K, El Marroun H, Elfrink ME, Jaddoe VWV, Visser LE, Stricker BHC. The concordance between self-reported medication use and pharmacy records in pregnant women. Pharmacoepidemiol Drug Saf. 2017;26(9):1119-1125. doi: 10.1002/pds.4264 [DOI] [PubMed] [Google Scholar]
32.de Korte BAC, Smeets NJL, Colbers A, van den Bemt BJF, van Gelder MMHJ. Adherence to prescription medication during pregnancy: do pregnant women use pharmacological treatment as prescribed? Br J Clin Pharmacol. 2023;89(5):1521-1531. doi: 10.1111/bcp.15609 [DOI] [PubMed] [Google Scholar]
33.Ukah UV, Dayan N, Potter BJ, Ayoub A, Auger N. Severe maternal morbidity and risk of mortality beyond the postpartum period. Obstet Gynecol. 2021;137(2):277-284. doi: 10.1097/AOG.0000000000004223 [DOI] [PubMed] [Google Scholar]
34.Callaghan WM, Creanga AA, Kuklina EV. Severe maternal morbidity among delivery and postpartum hospitalizations in the United States. Obstet Gynecol. 2012;120(5):1029-1036. doi: 10.1097/AOG.0b013e31826d60c5 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

eTable 1. List of names used to identify buprenorphine or naltrexone for opioid use disorder

eTable 2. List of ICD-9 and ICD-10 codes used to identify outcomes

eFigure 1. Graphical example to calculate total days’ supply of MOUD

eTable 3. List of ICD-9 and ICD-10 codes used to identify covariates

eFigure 2. Predicted adjusted hazard ratios for overdose and death at each observed value of MOUD days’ supply

eTable 6. Results of pre-specified sensitivity analyses

jamanetwopen-e2637214-s001.pdf^{(368.9KB, pdf)}

Supplement 2.

Data Sharing Statement

jamanetwopen-e2637214-s002.pdf^{(14KB, pdf)}

[zoi261003r1] 1.Azad HA, Goin D, Nathan LM, et al. Overdose, homicide, and suicide as causes of maternal death in the United States. N Engl J Med. 2026;394(7):722-723. doi: 10.1056/NEJMc2512078 [DOI] [PubMed] [Google Scholar]

[zoi261003r2] 2.Haight SC, Ko JY, Tong VT, Bohm MK, Callaghan WM. Opioid use disorder documented at delivery hospitalization—United States, 1999-2014. MMWR Morb Mortal Wkly Rep. 2018;67(31):845-849. doi: 10.15585/mmwr.mm6731a1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r3] 3.Hirai AH, Ko JY, Owens PL, Stocks C, Patrick SW. Neonatal abstinence syndrome and maternal opioid-related diagnoses in the US, 2010-2017. JAMA. 2021;325(2):146-155. doi: 10.1001/jama.2020.24991 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r4] 4.Zipursky J, Juurlink DN. Opioid use in pregnancy: an emerging health crisis. Obstet Med. 2021;14(4):211-219. doi: 10.1177/1753495X20971163 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r5] 5.Centers for Disease Control and Prevention . Opioid use during pregnancy. May 7, 2025. Accessed August 31, 2026. https://www.cdc.gov/opioid-use-during-pregnancy/about/index.html

[zoi261003r6] 6.Krans EE, Patrick SW. Opioid use disorder in pregnancy: health policy and practice in the midst of an epidemic. Obstet Gynecol. 2016;128(1):4-10. doi: 10.1097/AOG.0000000000001446 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r7] 7.Krishnapura SR, McNeer E, Loch SF, et al. Buprenorphine treatment in pregnancy and maternal-infant outcomes. JAMA Health Forum. 2025;6(4.11):e251814. doi: 10.1001/jamahealthforum.2025.1814 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r8] 8.National Academies of Sciences, Engineering, and Medicine . Medications for Opioid Use Disorder Save Lives. The National Academies Press; 2019. doi: 10.17226/25310 [DOI] [PubMed] [Google Scholar]

[zoi261003r9] 9.Substance Abuse and Mental Health Services Administration . Clinical guidance for treating pregnant and parenting women with opioid use disorder and their infants: HHS publication 18-5054. 2018. Accessed August 31, 2026. https://library.samhsa.gov/sites/default/files/sma18-5054.pdf

[zoi261003r10] 10.Leech AA, Garbett S, Yu HA, et al. Cost-effectiveness of treatment for opioid use disorder in pregnancy and its impact on birth outcomes. JAMA Pediatr. 2025;179(11):1203-1216. doi: 10.1001/jamapediatrics.2025.3067 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r11] 11.Krans EE, Kim JY, Chen Q, et al. Outcomes associated with the use of medications for opioid use disorder during pregnancy. Addiction. 2021;116(12):3504-3514. doi: 10.1111/add.15582 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r12] 12.Jarlenski M, LoCiganic WH, Chen Q, et al. Association between buprenorphine dose and outcomes among pregnant persons with opioid use disorder. Am J Obstet Gynecol. 2025;233(1):59.e1-59.e15. doi: 10.1016/j.ajog.2024.12.001 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r13] 13.Jarlenski M, Chen Q, Gao A, Rothenberger SD, Krans EE. Association of duration of methadone or buprenorphine use during pregnancy with risk of nonfatal drug overdose among pregnant persons with opioid use disorder in the US. JAMA Netw Open. 2022;5(4):e227964. doi: 10.1001/jamanetworkopen.2022.7964 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r14] 14.Patrick SW, Richards MR, Dupont WD, et al. Association of pregnancy and insurance status with treatment access for opioid use disorder. JAMA Netw Open. 2020;3(8):e2013456. doi: 10.1001/jamanetworkopen.2020.13456 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r15] 15.Leech AA, McNeer E, Roberts AW, et al. Buprenorphine out-of-pocket costs and discontinuation in privately insured adults with opioid use disorder. JAMA Intern Med. 2023;183(9):1023-1026. doi: 10.1001/jamainternmed.2023.2826 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r16] 16.Samples H, Williams AR, Crystal S, Olfson M. Impact of long-term buprenorphine treatment on adverse health care outcomes in Medicaid. Health Aff (Millwood). 2020;39(5):747-755. doi: 10.1377/hlthaff.2019.01085 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r17] 17.Margerison CE, Roberts MH, Gemmill A, Goldman-Mellor S. Pregnancy-associated deaths due to drugs, suicide, and homicide in the United States, 2010-2019. Obstet Gynecol. 2022;139(2):172-180. doi: 10.1097/AOG.0000000000004649 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r18] 18.Risby J, Schlesinger E, Geminn W, Cernasev A. Methadone treatment gap in Tennessee and how medication units could bridge the gap: a review. Pharmacy (Basel). 2023;11(5):131. doi: 10.3390/pharmacy11050131 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r19] 19.Tennessee Department of Health, Division of Family Health & Wellness . Maternal mortality in Tennessee: annual report 2025. December 10, 2025. Accessed August 31, 2026. https://www.tn.gov/content/dam/tn/health/2025-notcom/Maternal%20Mortality%20Review%20Report%202025.pdf

[zoi261003r20] 20.Kaiser Family Foundation. Births financed by Medicaid by metropolitan status. Accessed August 31, 2026. https://www.kff.org/medicaid/state-indicator/births-financed-by-medicaid/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D

[zoi261003r21] 21.Jarlenski M, Krans EE, Chen Q, et al. Substance use disorders and risk of severe maternal morbidity in the United States. Drug Alcohol Depend. 2020;216:108236. doi: 10.1016/j.drugalcdep.2020.108236 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r22] 22.Jarlenski M, Kim JY, Ahrens KA, et al. Healthcare patterns of pregnant women and children affected by OUD in 9 state Medicaid populations. J Addict Med. 2021;15(5):406-413. doi: 10.1097/ADM.0000000000000780 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r23] 23.Tennessee Department of Health . Neonatal abstinence syndrome in Tennessee 2022: 2024 NAS surveillance annual report. July 2024. Accessed August 31, 2026. https://www.tn.gov/content/dam/tn/health/program-areas/NAS-Annual-Report.pdf

[zoi261003r24] 24.Whitmore CC, Hawley RE, Min JY, et al. Building a data linkage foundation for mother-child pharmacoepidemiology research. Pharmaceut Med. 2021;35(1):39-47. doi: 10.1007/s40290-020-00371-8 [DOI] [PubMed] [Google Scholar]

[zoi261003r25] 25.Hoyert DL. Maternal mortality rates in the United States, 2024. Centers for Disease Control and Prevention. March 4, 2026. Accessed August 31, 2026. https://stacks.cdc.gov/view/cdc/174651

[zoi261003r26] 26.Centers for Disease Control and Prevention . Identifying severe maternal morbidity (SMM). May 15, 2024. Accessed August 31, 2026. https://www.cdc.gov/maternal-infant-health/php/severe-maternal-morbidity/icd.html

[zoi261003r27] 27.Schiff DM, Nielsen T, Terplan M, et al. Fatal and nonfatal overdose among pregnant and postpartum women in Massachusetts. Obstet Gynecol. 2018;132(2):466-474. doi: 10.1097/AOG.0000000000002734 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r28] 28.Wallace ME, Jahn JL. Pregnancy-associated mortality due to homicide, suicide, and drug overdose. JAMA Netw Open. 2025;8(2):e2459342. doi: 10.1001/jamanetworkopen.2024.59342 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r29] 29.Centers for Disease Control and Prevention . Circumstances contributing to pregnancy-related deaths: data from maternal mortality review committees in 38 U.S. States, 2020. April 20, 2026. Accessed August 31, 2026. https://archive.cdc.gov/#/details?url=https://www.cdc.gov/maternal-mortality/php/report/index.html

[zoi261003r30] 30.Olstad EW, Nordeng H, Bjørk MH, Selmer KK, Gervin K. Agreement between parental self-reported antiseizure medication use and dispensed prescription records from a national prescription database. Pharmacoepidemiol Drug Saf. 2025;34(4):e70139. doi: 10.1002/pds.70139 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261003r31] 31.Cheung K, El Marroun H, Elfrink ME, Jaddoe VWV, Visser LE, Stricker BHC. The concordance between self-reported medication use and pharmacy records in pregnant women. Pharmacoepidemiol Drug Saf. 2017;26(9):1119-1125. doi: 10.1002/pds.4264 [DOI] [PubMed] [Google Scholar]

[zoi261003r32] 32.de Korte BAC, Smeets NJL, Colbers A, van den Bemt BJF, van Gelder MMHJ. Adherence to prescription medication during pregnancy: do pregnant women use pharmacological treatment as prescribed? Br J Clin Pharmacol. 2023;89(5):1521-1531. doi: 10.1111/bcp.15609 [DOI] [PubMed] [Google Scholar]

[zoi261003r33] 33.Ukah UV, Dayan N, Potter BJ, Ayoub A, Auger N. Severe maternal morbidity and risk of mortality beyond the postpartum period. Obstet Gynecol. 2021;137(2):277-284. doi: 10.1097/AOG.0000000000004223 [DOI] [PubMed] [Google Scholar]

[zoi261003r34] 34.Callaghan WM, Creanga AA, Kuklina EV. Severe maternal morbidity among delivery and postpartum hospitalizations in the United States. Obstet Gynecol. 2012;120(5):1029-1036. doi: 10.1097/AOG.0b013e31826d60c5 [DOI] [PubMed] [Google Scholar]

PERMALINK

Duration of Medication Use for Opioid Use Disorder and Risk of Postpartum Overdose and Death

Ahra Kim, MPH

Andrew D Wiese, MPH, PhD

Sarah Osmundson, MD

Chad You, MD, MA

Carlos G Grijalva, MPH, MD

Margaret A Adgent, PhD

Alexandra C Sundermann, MD, PhD

Andrew J Spieker, PhD

Amelie Pham, MD

Stephen W Patrick, MD, MPH

Ashley A Leech, PhD, MS

Key Points

Question

Findings

Meaning

Abstract

Importance

Objective

Design, Setting, and Participants

Exposure

Main Outcomes and Measures

Results

Conclusions and Relevance

Introduction

Methods

Study Design and Data Sources

Population

Figure 1. Flowchart of Patient Enrollment for Pregnancies With Opioid Use Disorder (OUD) Receiving Medications for OUD (MOUD) From Tennessee Medicaid (TennCare), 2007-2020 .

Figure 2. Flowchart of Timeline of Baseline and Follow-Up Periods.

Outcome

Exposure

Covariates

Statistical Analysis

Results

Study Population

Table 1. Descriptive Characteristics of Study Cohort.

Risk of Death or Overdose

Table 2. Unadjusted and Adjusted HRs for First Occurrence of Overdose or Death for Selected Days’ Supply vs 1-Day Supply of MOUD Among Pregnant Individuals With Opioid Use Disorder and Any Evidence of Buprenorphine Treatment During Pregnancy and Post Partum, Tennessee Medicaid (2010-2021).

Secondary Analysis on the Timing of MOUD Receipt

Sensitivity Analyses

Discussion

Strengths and Limitations

Conclusions

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases