Abstract
Objectives:
For women with opioid use disorder (OUD), transitioning from pregnancy to postpartum and from incarceration to the community are times of increased risk for opioid overdose. This prospective project evaluated the extent of Jenna’s Project success in preventing overdoses and improving recovery outcomes by coordinating post-release care in incarcerated OUD pregnant and postpartum women.
Methods:
Participants (N=132) were pregnant or postpartum incarcerated women with OUD who self-referred for post-release services. From March 2020 to October 2021 participants could receive up to 6-months of care coordination services. Outcomes included overdose (non-fatal or fatal), reincarceration, active Medicaid, receipt of medication to treat opioid use disorder (MOUD), presence of children living with them, open child protective services (CPS) cases and number of referrals for services.
Results:
There were 0% non-fatal and 0% fatal overdoses at both 1 and 6 months post-release, and 2% returned to incarceration. Significantly fewer participants had Medicaid at release and after 6-month post-release than before prison (p<0.001 for all three pairwise comparisons). At 6-months post-release, significantly more participants reported both MOUD receipt compared to baseline (p <0.001). There was no significant change in the number of open CPS cases and referrals for childcare or parenting services were the most common type of referral provided.
Conclusion:
Immediate post-release care coordination for pregnant and post-partum women with OUD was highly feasible and effective in preventing overdose (fatal or non-fatal), reincarceration, and promoting recovery outcomes. Care coordination immediately post-release is a promising, pragmatic intervention for saving lives and improving outcomes.
Keywords: Opioid use disorder, Pregnancy, Addiction, Opioid medication treatment, Prison, Jail, Incarceration, Release, post-release, pregnant people
Introduction
Approximately 55,000 pregnant people are incarcerated yearly1,2, and 26%−50% are diagnosed with opioid use disorder (OUD)3. Transitioning from pregnancy to postpartum and from incarceration to the community increases opioid overdose risks. A person returning to their community within two weeks of state prison release was 129 times more likely to die from an overdose compared to the general public4. In North Carolina (NC), formerly incarcerated people were 40 times more likely to die of an opioid overdose two weeks post-release from state prison than those without an OUD5. Such overdose vulnerability is partly due to reduced opioid tolerance developed and limited access to OUD medication during incarceration6. To date, treating OUD with medications and providing naloxone reduces overdose mortality after prison release7. Further, OUD accounts for 10% of postpartum mortality, with an increased risk of fatal overdose within the first year post-delivery8,9,10.
This project was borne from the tragic overdose death of a former UNC Horizons patient who was released from prison, received opioids from a significant other upon her release, and died before receiving help. This pilot project aimed to examine how to prevent such deaths from happening to others by immediately connecting women leaving a carceral setting by providing care coordination and treatment resources for 6 months following release.
Materials and Methods
Participants and Setting.
This prospective evaluation enrolled 132 unduplicated pregnant and postpartum women from March, 2020-October, 2021. They were first identified in various NC incarcerated settings encompassing jails and prisons. Due to NC law and the COVID 19 pandemic emergency response, pregnant women were often transferred among different incarcerated settings for medical care. The UNC Institutional Review Board exempted the evaluation from review. Eligibility included: age 18 or older, current OUD diagnosis, and pregnant or postpartum at referral. Clients were self-referred for post-release services and were connected to Horizons’ staff members who identified treatment needs and conducted baseline clinical assessments upon release.
Care Coordination.
Horizons’ staff members met women at carceral facilities immediately upon release to provide care coordination services (e.g., regular communication, transportation, and/or help arranging emergency housing, treatment, etc.) for 6 months post-release. Treatment services were consistent with the participant’s American Society of Addiction Medicine risk rating and Diagnostic and Statistical Manual-5 scores11. Participants requesting medication to treat OUD (MOUD) received an appointment for medication assessment and initiation. Participants already receiving MOUD, had their medication facilitated post-release. All participants received Narcan kits and overdose prevention training.
Data Collection.
Data were self-reported responses to interview questions at both release (retrospectively asking about pre-incarceration for certain measures) and 6 months post-release. Domains in the initial assessment and subsequent follow-up included: Current substance use, treatment needs, employment, housing needs, children, Child Protective Service involvement, parenting support, child care needs, social and community support and relationships, transportation, physical, dental, and mental health needs and legal issues (see Table 2).
Table 2.
Overdose, changes in health insurance status, medication to treat opioid use disorder, children living with the participant, and open child protective services case from baseline to follow-up, and for referrals for services at 6-month follow-up (N=132)
| Outcomes | % | ||
|---|---|---|---|
| Overdose not fatal | 0% | ||
| Overdose fatal one month | 0% | ||
| Overdose fatal six month | 0% | ||
| Re-incarcerated since leaving incarceration at 6-months | 2% | ||
| Before incarceration | At Release | 6 month follow-up after release | |
| Least Squares Means (Standard Errors) | |||
| Health Insurance | .87 (.03) | .36 (.04) | .60 (.04) |
| Prescribed Medication to treat Opioid Use Disorder | .39 (.04) | .51 (.04) | |
| Reported their children living with them | .07 (.02) | .32(.04) | |
| CPS Case Open | .11 (.03) | .11 (.03) | |
| Number of referrals per participant made during their UNC Horizons treatment?? for: | |||
| Child care or parenting services | 7.6 | ||
| Medical care | 7.0 | ||
| Mental health | 7.1 | ||
| SUD treatment | 7.0 | ||
| Housing | 6.7 | ||
| Medication for OUD | 6.1 | ||
| Peer services | 5.3 | ||
| Employment | 5.0 | ||
| Dental care | 5.0 | ||
| Education services | 4.8 | ||
| Legal help | 2.5 | ||
| Transportation | 1.7 | ||
Notes:
Outcome data were collected from132 unduplicated participants from 29 different urban and rural counties in North Carolina with an average duration of incarceration was 3.3 months (range 3–36 months). All had an OUD diagnosis (83% pregnant at baseline) who received services from UNC Horizons for the 6-month post release period.
Outcome data were collected by project staff either face-to-face or by phone and by review of public records. Outcome variables of interest included: Overdose (self-report and review of available medical records indicating receipt of Narcan, overdose requiring hospitalization etc.) and deaths from overdose (examined by searching death records for every person out of contact to determine occurrence of any overdose death). Substance use was determined by self-report, treatment records, and urine drug testing. New legal system involvement was determined by both self-report and a search in public databases for new charges, arrests, nights incarcerated. Other service-delivery variables (women receiving various services and number of services) were captured through an electronic records system. Contact information was collected, reviewed, and updated at each clinical contact.
Active health insurance (0=no, 1 = yes) was recorded for prior to and upon incarceration release, and at 6-month follow-up. Baseline and 6-month follow-up assessments included: taking medication to treat opioid use disorder (0=none, 1 = yes), children living with the participant (0=no 1=yes), and having an open child protective services case (0=no, 1 =yes). Reasons for incarceration were largely due to 90-day substance-related probation violations (74%), while another 19% had a probation violation for other reasons (e.g., inability to pay a fee), and 7% had another reason for incarceration.
Least squares means are the estimated likelihood of occurrence of event, and so can be considered the proportion of the answer=1 for each outcome.
Statistical methods.
A generalized estimating equations (SAS 9.4 GENMOD) model examined change in the four outcomes (Table 2) with a single fixed effect for Time point of assessment, using an exchangeable correlation structure. For health insurance, pairwise comparisons of the levels were used following detection of a significant effect for Time.
Results
Participant characteristics are shown in Table 1.
Table 1.
Participant Demographic Characteristics and Current Substance Use at Baseline (N=132)
| Demographics | |
|---|---|
| Mean (SD) | |
| Age in Years (n=131) | 30.3 (4.7) |
| (%) | |
| Race: | |
| Black | 14% |
| White | 82% |
| Multiple Race or other | 4% |
| Hispanic | 2% |
| Unemployed | 99% |
| Not Married | 94% |
| Homeless | 48% |
| Current Substance Use | |
| Opioid use disorder, severe | 73% |
| Opioid use disorder, mild or moderate | 27% |
| Tobacco | 81% |
| Amphetamine | 43% |
| Cocaine | 25% |
| Cannabis | 20% |
| Alcohol | 14% |
| Benzodiazepine | 11% |
Notes. SD = Standard Deviation. Racial category was based on self-report.
Overdose and re-incarceration.
Table 2 shows that no participant had either a non-fatal or fatal overdose at 6-months post-release, and only 3 (2%) participants experienced re-incarceration.
Insurance and MOUD outcomes.
The Time effect was significant for active health insurance, prescription of MOUD, and children living with them (Table 2). Open CPS cases did not differ between time points. For active health insurance, post-hoc testing indicated that all time points were significantly different. The proportion of participants having active health insurance before incarceration was significantly higher than the proportion at 6-month follow-up. Both of these proportions were significantly higher than the proportion upon release from incarceration. Medicaid was the only health insurance reported. In contrast, the proportion of participants receiving MOUD was significantly higher at 6-months post-release than at baseline, and that the proportion of women with children living with them was higher at 6-months follow-up than at baseline.
Service referrals.
Table 2 shows the mean number of referrals each participant received for different services. These varied from 1.7 for transportation to 7.6 for childcare or parenting services.
Discussion
The most powerful finding in this evaluation was that no pregnant or postpartum woman with OUD who received services at prison release had experienced a non-fatal or fatal overdose by 6-months post-release. These data are congruent with a other studies showing significant improvements in outcomes at six months post-release with patient navigation intervention compared to a comparison condition.12 These data may be the first to show the potentially protective power of immediate and consistent care coordination post-prison release to prevent overdose among pregnant or post-partum women with OUD. These data are especially important given that among former NC inmates, females, like males, have an all opioid overdose mortality risk 40.5 times higher within 2 weeks and 10.6 times higher within 1 year post-release than in the general NC population.5
Remarkably, only 3 (2%) participants were re-incarcerated by 6-months post-release follow-up. This finding is lower than the 28% re-incarceration rate in the aforementioned.12 This low re-incarceration rate may be due to immediate engagement in comprehensive services.12 This result and the finding that more participants reported having their children live with them 6 months post-release compared to before incarceration (32% versus 7%, respectively) are likely attributable to the receipt of mother-child dyad services. The typical duration and legal complexity of CPS cases may explain the lack of change in open CPS cases during the project.
These findings add perinatal specific data to the growing evidence that treating OUD with medications and providing care coordination upon release holds promise for preventing overdose mortality post-release7. Given the observed increase in MOUD receipt at 6-months post-release, there remains a substantial role for carceral facilities in scaling up MOUD provision during incarceration and ensuring seamless continuation of medication (including naloxone training and kits) post-release13.
Related to the receipt of MOUD is access to active health insurance. Even at 6 months post-release the proportion of participants with Medicaid remained lower than pre-incarceration. This highlights an opportunity for NC to consider following California in making a 1115 request from the Centers for Medicare and Medicaid Services (CMS) to provide coverage for reentry services 90 days prior to release.
Study strengths include being prospective, verification of possible deaths using public records, and high retention during follow-up (100% at the 6-month follow-up). This rate is higher than both the 72% follow-rate at 6 months12 and similar to the 1-month follow-up rate for those released from prison and randomized to receive methadone (96%) 14.
Findings are limited by several issues. Self-reported data may have recall bias; however, self-reported data has been validated in similar cohorts. The current sample may not be generalizable to other incarcerated populations. Insights into the post-release experience of participants are limited to 6 months. While harms post-release (e.g., overdose) are typically most acute immediately after prison release, poor outcomes are seen over 24 months15.
In conclusion, immediate post-release care coordination is feasible and yields outcomes that are promising for pregnant and post-partum women and people with OUD.
Acknowledgments
Sources of support/funding for the work: Foundation for Opioid Response Efforts
Footnotes
Conflicts of interest: None
Statement of adherence to preprint policy: Our manuscript has not been posted on a preprint server.
References
- 1.Sufrin C, Beal L, Clarke J, Jones R, & Mosher WD (2019). Pregnancy Outcomes in US Prisons, 2016–2017. American journal of public health, 109(5), 799–805. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Sufrin C, Jones RK, Mosher WD, & Beal L. (2020). Pregnancy Prevalence and Outcomes in U.S. Jails. Obstetrics and gynecology, 135(5), 1177–1183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Knittel AK, Swartzwelder RA, Zarnick S, Tsujimoto THM, Horne T, Lin FC, Edwards J, Amos E, Alexander J, Thorp J, & Jones HE (2022). Medications for opioid use disorder during pregnancy: Access and continuity in a state women’s prison facility, 2016–2019. Drug and alcohol dependence, 232, 109308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Binswanger IA, Stern MF, Deyo RA, Heagerty PJ, Cheadle A, Elmore JG, & Koepsell TD (2007). Release from prison--a high risk of death for former inmates. The New England journal of medicine, 356(2), 157–165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ranapurwala SI, Shanahan ME, Alexandridis AA, Proescholdbell SK, Naumann RB, Jr DE, & Marshall SW (2018). Opioid Overdose Mortality Among Former North Carolina Inmates: 2000–2015. American Journal of Public Health, 108(9), 1207–1213. doi: 10.2105/ajph.2018.304514 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.National Academies of Sciences, Engineering, and Medicine; Health and Medicine Division. (2019). Board on Health Sciences Policy; Committee on Medication-Assisted Treatment for Opioid Use Disorder. Medications for Opioid Use Disorder Save Lives. (Mancher M, Leshner AI, eds.). National Academies Press (US); Accessed September 15, 2023. http://www.ncbi.nlm.nih.gov/books/NBK538936/ [PubMed] [Google Scholar]
- 7.Green TC, Clarke J, Brinkley-Rubinstein L, Marshall BDL, Alexander-Scott N, Boss R, & Rich JD (2018). Postincarceration Fatal Overdoses After Implementing Medications for Addiction Treatment in a Statewide Correctional System. JAMA psychiatry, 75(4), 405–407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gemmill A, Kiang MV, & Alexander MJ (2019). Trends in pregnancy-associated mortality involving opioids in the United States, 2007–2016. American journal of obstetrics and gynecology, 220(1), 115–116. [DOI] [PubMed] [Google Scholar]
- 9.Schiff DM, Nielsen T, Terplan M, Hood M, Bernson D, Diop H, Bharel M, Wilens TE, LaRochelle M, Walley AY, & Land T. (2018). Fatal and Nonfatal Overdose Among Pregnant and Postpartum Women in Massachusetts. Obstetrics and gynecology, 132(2), 466–474. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Smid MC, Stone NM, Baksh L, Debbink MP, Einerson BD, Varner MW, Gordon AJ, & Clark EAS (2019). Pregnancy-Associated Death in Utah: Contribution of Drug-Induced Deaths. Obstetrics and gynecology, 133(6), 1131–1140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.American Psychiatric Association. (2013). Neurodevelopmental disorders. In Diagnostic and statistical manual of mental disorders (5th ed., text rev.). [Google Scholar]
- 12.Binswanger IA, Whitley E, Haffey PR, Mueller SR, & Min SJ (2015). A patient navigation intervention for drug-involved former prison inmates. Substance abuse, 36(1), 34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Knittel AK, Swartzwelder RA, Zarnick S, et al. Medications for opioid use disorder during pregnancy: Access and continuity in a state women’s prison facility, 2016–2019. Drug Alcohol Depend. 2022;232:109308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Rich JD, McKenzie M, Larney S, Wong JB, Tran L, Clarke J, Noska A, Reddy M, & Zaller N. (2015). Methadone continuation versus forced withdrawal on incarceration in a combined US prison and jail: a randomised, open-label trial. Lancet (London, England), 386(9991), 350–359. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Stewart AC, Cossar R, Walker S, Wilkinson AL, Quinn B, Dietze P, Winter R, Kirwan A, Curtis M, Ogloff JRP, Kinner S, Aitken C, Butler T, Woods E, & Stoové M. (2021). Strategies to maximise study retention and limit attrition bias in a prospective cohort study of men reporting a history of injecting drug use released from prison: the prison and transition health study. BMC medical research methodology, 21(1), 185. [DOI] [PMC free article] [PubMed] [Google Scholar]