Postpartum contraceptive plans and fulfillment for pregnant people experiencing incarceration with substance use histories
Department of Obstetrics and Gynecology, School of Medicine, 6797University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
Department of Epidemiology, Gillings School of Global Public Health, 6797University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
Carolina Population Center, 6797University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
Lineberger Comprehensive Cancer Center, 6797University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
6797University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
Division of General Medicine and Clinical Epidemiology, School of Medicine, 6797University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
Department of Gynecology and Obstetrics, 519921Johns Hopkins School of Medicine, Baltimore, MD, USA
Andrea K. Knittel, MD, PhD, Department of Obstetrics and Gynecology, School of Medicine, University of North Carolina at Chapel Hill, 101 Manning Drive, CB#7570 Chapel Hill, NC 27599-7570, USA. Email: Andrea_Knittel@med.unc.eduAbstract
Background
Pregnant patients experiencing incarceration with a history of substance use face complex challenges accessing reproductive health services. Despite research indicating interest in these services, contraception counseling and services are often excluded from routine carceral health services.
Objectives
We sought to examine the postpartum contraceptive plans and fulfillment of pregnant patients in the North Carolina prison system with histories of substance use.
Design
This is a secondary analysis of a retrospective chart review quantifying the contraceptive plans and fulfillment of pregnant people experiencing incarceration.
Methods
We abstracted chart data from the prenatal records and scanned delivery records of patients experiencing incarceration in the North Carolina state prison system who accessed prenatal care. We categorized substance use as licit (e.g., tobacco, alcohol, and cannabis) or illicit (e.g., opioids, amphetamines). We assessed postpartum contraceptive plan documentation and contraception type provided at delivery overall and across type of pre-incarceration substance use.
Results
Of the 890 patients that met eligibility criteria, the majority (90.9%) reported the use of at least one type of substance during pre-incarceration and 72.1% reported use of multiple substances. Among patients with either licit or illicit substance use, 167 (20.6%) had a documented plan for long-acting reversible contraceptive (LARC), 119 (14.7%) for permanent contraception, 107 (13.2%) for a non-LARC method, 50 (6.2%) with a documented plan for no method or plan to be abstinent, and 366 (45.2%) had no documented contraception plan. Among the patients who delivered while incarcerated, 17.6% of patients received a non-LARC method, 9.5% received a LARC, 5.3% of patients received permanent contraception, three participants (0.6%) received no contraceptive method, and 316 (66.9%) had no documentation of receiving postpartum contraception. The proportion of patients receiving each postpartum contraceptive method type did not differ by pre-incarceration substance use type.
Conclusion
Postpartum contraceptive preferences of pregnant patients experiencing incarceration may not be adequately prioritized. Our findings indicate a potential unmet need for contraceptive services for pregnant and postpartum people with history of substance use experiencing incarceration.
Plain language summary
Why was the study done?
Pregnant patients in prisons with a history of substance use have unique challenges accessing critical reproductive health services, including contraception counseling and provision. This is particularly important for patients pregnant and in custody as they are at greater risk of poor reproductive health outcomes than pregnant patients not in custody, especially those with a history of substance use.
What did the researchers do?
The research team reviewed medical charts and abstracted data from prenatal records and delivery records from the North Carolina Department of Adult Corrections. They described the characteristics of the patients and their postpartum contraception plans and fulfillment, overall and by pre-incarceration substance use type.
What did the researchers find?
The majority (90.9%) of the 890 patients in our sample reported substance use prior to incarceration, with nearly three-quarters reporting use of more than one substance. Among patients with either licit or illicit substance use, 167 (20.6%) had a plan for long-acting reversible contraception (LARC), 119 (14.7%) for permanent contraception, 107 (13.2%) for a non-LARC method, 50 (6.2%) for no method for planned abstinence, and 366 (45.2%) had no documented contraception plan. Among the patients who delivered while incarcerated, 17.6% of patients received a non-LARC method, 9.5% received a LARC, 5.3% of patients received permanent contraception, three patients (0.6%) received no contraceptive method, and 316 (66.9%) had no documentation of postpartum contraception. The type of postpartum contraception was not different by substance use.
What do the findings mean?
This study indicates an unmet need for contraceptive services for pregnant patients in prison. Postpartum contraception preferences of pregnant patients, particularly for those with histories of substance use, may not be adequately prioritized during incarceration.
Introduction
The intersection of substance use, pregnancy, and incarceration poses significant challenges for patients accessing reproductive health services. 1 National rates of substance use among pregnant patients have risen over the past several years, with a 71% increase in the number of patients with a documented opioid use disorder (OUD) during delivery hospitalization from 2012 to 2016. 2 Due to the criminalization of substance use, particularly during pregnancy, incarceration rates have increased among patients who have used substances and are capable of giving birth. Women experiencing incarceration report higher rates of substance use disorder (SUD) (70%) compared to men (60%). 3 However, despite the fact that many sentences are substance use-related, treatment for substance use during pregnancy in carceral settings remains limited as medications for opioid use disorder (MOUD) are not consistently available, and treatment for other forms of SUD is even scarcer.4,5
Furthermore, patients experiencing incarceration – especially those with a history of substance use – face a greater risk of poor reproductive health outcomes, including unintended pregnancy, poor perinatal outcomes such as preterm birth and low birth weight, and limited access to contraception compared to their non-incarcerated counterparts. 6 Prior research indicates that most incarcerated women believe that contraception services should be available during incarceration. 7 Counseling regarding future pregnancies, safe birth spacing, contraceptive methods, and access to method initiation are routine during prenatal and postpartum care in the community but are not consistently provided to pregnant and postpartum patients in the carceral system for reasons including a lack of formal, standardized policies and budget limitations.7–9 Additional barriers to contraceptive care in carceral settings include broader mistrust of community and carceral healthcare systems, contraception-related training gaps for carceral medical providers, and delayed provision of care beyond the patient length-of-stay in the carceral setting.7,8 Incarcerated patients may remain at a higher risk for unintended pregnancy following incarceration due to no or inconsistent contraception and poor access to healthcare providers in the community, even though research shows that the majority desire to use contraception. 6
There is reason to suspect that patients with substance use histories would have different postpartum contraception experiences during incarceration than patients without, although substance use could either increase or decrease access to and quality of postpartum contraceptive counseling and contraception. Studies of non-incarcerated postpartum patients suggest that those with substance use may be less likely to receive contraception in the community.10,11 It is possible that the disparities that derive from limited access in the community would be further exacerbated in a carceral setting. Alternatively, the stigma and discrimination related to substance use during pregnancy are likely to intersect with those of incarceration and raise concerns about bias in contraceptive counseling and provision as well as the possibility that health care providers might be more motivated to provide contraception to patients who use substances.1,5,12,13
Given the complexities in access to reproductive healthcare faced by pregnant and postpartum patients experiencing incarceration with histories of substance use, we sought to examine the postpartum contraceptive plans and fulfillment within this population in the North Carolina (NC) prison system. Understanding the contraceptive needs and preferences of people experiencing incarceration is critical to improving reproductive health services and health outcomes during and after incarceration.
Methods
Eligibility criteria
The study population included pregnant patients experiencing incarceration in the NC state prison system aged 21 years or older who accessed prenatal care at the NC Correctional Institution for Women between January 1, 2016, and December 31, 2021. We included patients who returned to the community prior to the end of the pregnancy, patients who experienced miscarriage or perinatal loss, and patients who had live births during incarceration. We did not collect data on patients under age 21 due to the Medicaid prohibition on permanent contraception for that group. Although pregnant people are excluded from Medicaid during incarceration, this inclusion criterion meant that all included patients, were they in the community, could potentially have received any desired method of contraception. There was no maximum age. Not every patient was incarcerated during the full duration of their pregnancy, although if a patient returned to the community and was reincarcerated during the same pregnancy, unique identifiers allowed for both incarceration episodes to be included in the data on the single pregnancy. Data were gathered only from medical visits and hospital encounters that occurred during periods of incarceration.
Statistical analyses
This is a secondary analysis of a retrospective chart review. We abstracted data from the North Carolina Department of Adult Corrections electronic health record (EHR) system, which includes prenatal records and scanned delivery records.
For this analysis, the primary exposure was self-reported pre-incarceration substance use, which was abstracted for detailed substance use information and then divided into licit substances, illicit substances, or no reported substance use for analysis. “Licit substances” included alcohol, tobacco, and/or cannabis; cannabis was included in this category given the shifts toward legalization of cannabis outside of NC and increasing access to legal non-marijuana cannabinoids within NC during the study period. “Illicit substances” included opioids (including heroin, fentanyl, prescription opioid medications taken without a prescription, and MOUD as an indicator of OUD), amphetamines, cocaine, sedatives, hallucinogens, benzodiazepines, ecstasy/MDMA, nitrous oxide, and non-opioid prescription medications taken without a prescription.
The primary outcome for this secondary analysis was the documented postpartum contraceptive method plan. We categorized methods into three categories. Non-long-acting reversible contraception (LARC) methods included the combined oral contraceptive pill, progestin-only pill, vaginal ring, and depot medroxyprogesterone acetate (DMPA). LARC included the etonogestrel implant and the levonorgestrel and copper intrauterine devices (IUD). Permanent contraception, including any tubal sterilization procedure, was a separate category. Documented plans for abstinence from sexual intercourse in the community, abstinence due to prolonged incarceration, or no contraceptive method were grouped together.
We abstracted demographic characteristics of maternal age at delivery, parity (dichotomized into less than two/two or more), and race using the categories available in the EHR (i.e., American Indian/Alaskan Native, Another Race, Black or African American, Unknown/Not Reported, or White). We additionally abstracted legal status (i.e., housed in prison during a jail incarceration, sentenced to prison, or unknown), duration of incarceration, whether delivery occurred during incarceration, and adequacy of prenatal care. We assessed adequacy of prenatal care using Kotelchuck’s Adequacy of Prenatal Care Utilization Index SAS Macro.14,15 We modified the index to designate whether adequacy could not be established, either because the patient did not experience birth while incarcerated or was classified as inadequate care but received an indeterminate amount of care in the community before incarceration. When delivery occurred during incarceration, we additionally abstracted gestational age at delivery.
We used SAS 9.4 software for the analysis. We described key demographic, pregnancy, and criminal legal sample characteristics and detailed substance use characteristics of the entire sample using frequencies and proportions for categorical variables and median and interquartile range for continuous variables, as these were not normally distributed. We compared documented postpartum contraception plans by categories of pre-incarceration substance use, and also compared the type of contraception provided at delivery by pre-incarceration substance use category. We used Fisher’s Exact Test with a significance level of α = 0.05 to compare subsamples. The reporting of this study conforms to the STROBE guidelines [see Additional file 1]. 16
Ethics approvals and informed consent
This project was reviewed and approved by the Institutional Review Board at the University of North Carolina at Chapel Hill (IRB #22-2250) on December 14, 2022 and was reviewed by the North Carolina Department of Adult Corrections. Informed consent was waived by the IRB for secondary analysis of existing clinical data.
Results
There were 890 pregnancies among pregnant patients experiencing incarceration in our sample who met eligibility criteria. There were 34 patients contributing multiple pregnancies during incarceration over the study period. The average age of patients at delivery was 28 years, and 91.5% were multiparous. Demographic characteristics are outlined in Table 1. Substance use was highly prevalent, with 90.9% of patients reporting the use of at least one substance during pre-incarceration. Half of the patients (53.0%) included in this study delivered while incarcerated, which did not differ by substance use type. A significant proportion (72.1%) of participants in the sample reported use of more than one substance (Table 2, Figure 1). The most frequently used substances documented in the medical record were tobacco (69.9%), opioids, (56.5%), cannabis (26.5%), and amphetamines (26.4%). The most commonly reported opioids used prior to incarceration were heroin (34.7%), prescribed buprenorphine (20.4%), prescribed methadone (12.6%), and non-prescribed prescription opioids (12.8%) (Table 2, Figure 2). Among the 503 patients with reported pre-incarceration opioid use, 42.5% had opioid withdrawal and 85.5% had opioid use disorder (OUD) documented as a pregnancy complication in the prenatal record. Only 54.9% of patients with OUD received MOUD during incarceration; 56.4% of those were treated with buprenorphine, 41.1% with methadone, and 2.5% started with one MOUD and changed to the other during the course of incarceration.Characteristic All N = 890 Illicit substances N = 654 Licit substances only N = 155 No substance use N = 59 n % n % n % n %
Maternal age at delivery (years; median, Q1, Q3)
28.0 25.0, 31.0 28.0 25.0, 32.0 26.0 23.0, 30.0 26.0 24.0, 31.0
Parity less than 2 at admission
76 8.5 62 9.5 12 7.7 1 1.7 <2 >2 814 91.5 592 90.5 143 92.3 58 98.3
Legal incarceration status
225 25.3 173 26.5 34 21.9 15 25.4 Housed in prison during jail incarceration Sentenced to prison 659 74.0 476 72.8 120 77.4 44 74.6 Unknown 6 0.7 5 0.8 1 0.6 0 0.0
Adequacy of prenatal care
10 1.1 6 0.9 3 1.9 1 1.7 Adequate Adequate plus 66 7.4 57 8.7 8 5.2 0 0.0 Inadequate 264 29.7 194 29.7 45 29.0 18 30.5 Inadequate but received care prior to incarceration 127 14.3 80 12.2 31 20.0 14 23.7 Missing information 11 1.2 8 1.2 1 0.6 0 0.0 Unable to calculate (i.e., didn’t deliver) 412 46.3 309 47.2 67 43.2 26 44.1
Gestational age at delivery (weeks; median, Q1, Q3)
*
39.0 37.3, 39.4 38.9 37.0, 39.3 39.1 38.2, 39.6 38.9 38.5, 39.2
Duration of incarceration (weeks; median, Q1, Q3)
10.9 6.4, 13.4 10.6 6.1, 13.0 11.3 8.9, 12.9 11.2 5.9, 17.4
Delivered while incarcerated
472 53.0 340 52.0 87 56.1 33 55.9
Did not deliver while incarcerated
412 46.3 309 47.2 67 43.2 26 44.1
Missing
6 0.7 5 0.8 1 0.6 0 0.0
Race
†
11 1.2 8 1.2 3 1.9 0 0.0 American Indian/Alaska Native Another race 11 1.2 2 0.3 5 3.2 4 6.8 Black or African American 193 21.7 67 10.2 85 54.8 33 55.9 Unknown or not reported 12 1.3 6 0.9 1 0.6 5 8.5 White 663 74.5 571 87.3 61 39.4 17 28.8 Substance n %
Tobacco
622 69.9
Alcohol
137 15.4
Cannabis
236 26.5
Cocaine
215 24.2
Amphetamines
235 26.4
Sedatives
82 9.2
Hallucinogens
5 0.6
Opioids
503 56.5 Heroin 309 34.7 Fentanyl 16 1.8 Methadone (prescribed) 112 12.6 Buprenorphine/Subutex/Suboxone (prescribed) 182 20.4 Non-prescribed prescription opioids 114 12.8 Other opioids 35 3.9 Unknown 75 8.4
Benzodiazepines (Xanax, Librium, Ativan)
20 2.2
Molly/ecstasy/MDMA
2 0.2
Intravenous Drug Use
16 1.8
Other, specify
*
10 1.1
Unknown
22 2.5
None
59 6.6
There were 393 patients (44.2%) with documented plans for one or more types of contraceptive method, making a total of 480 plans for specific contraceptive methods. Across all documented plans, the most requested postpartum contraception method was LARC (175/480, 36.5%), followed by permanent contraception (129/480, 26.9 %), a non-LARC method (121/480, 25.2%), and 55/480 (11.5%) of plans documented no method or plans to be abstinent. Among patients with licit or illicit substance use (n=809), 167 (20.6%) requested a LARC, 119 (14.7%) requested permanent contraception, 107 (13.2%) requested a non-LARC method, 50 (6.2%) requested no method or planned to be abstinent, and 366 (45.2%) had no documented contraception plan. The proportion of patients requesting non-LARC or LARC methods did not differ by pre-incarceration substance use type (p=0.31 and 0.41, respectively) (Table 3). However, there was a significant difference in documented plans for permanent contraception by substance use type, with patients with “illicit” substance use (16.2%) requesting permanent contraception more commonly than those with “licit” substance use (8.4%) or no substance use (10.2%) (p=0.03) (Table 3).Contraception type All N = 890 Missing substance use n = 22 No substance use n = 59 Licit substances only n = 155 Illicit substances n = 654 p-value n n % n % n % n % No method 55 1 4.5 4 6.8 11 7.1 39 6.0 0.86 Non-LARC
1
121 2 9.1 12 20.3 21 13.5 86 13.1 0.31 LARC method
2
175 0 0.0 8 13.6 31 20.0 136 20.8 0.41 Permanent method
3
129 4 18.2 6 10.2 13 8.4 106 16.2 0.03 No documented plan 410 15 68.2 29 49.2 79 51.0 287 43.9 —
Among the 472 patients who delivered while incarcerated, 60.0% had at least one postpartum contraceptive method plan documented in the medical record. Of those who delivered, 83 (17.6%) received a non-LARC method, 45 (9.5%) received LARC, and 25 (5.3%) received permanent contraception. Three patients (0.6%) had documented a plan of no method use postpartum. Postpartum contraception receipt after the delivery hospitalization was not routinely documented in the medical record, with two-thirds (66.9%) of patients having no documentation of any postpartum contraception receipt. The proportion of patients receiving each contraceptive method type did not significantly differ by pre-incarceration substance use type (Table 4).Contraception type All N = 472 Missing substance use n = 12 No substance use n = 33 Licit substances only n = 87 Illicit substances n = 341 p-value n n % n % n % n %
No method
3 0 0.0 0 0.0 0 0.0 3 0.9 0.61
Non-LARC
1
83 2 16.7 6 18.2 15 17.2 60 17.6 0.96
LARC
2
45 0 0.0 2 6.1 9 10.3 34 10.0 0.78
Permanent method
3
25 0 0.0 2 6.1 3 3.4 20 5.9 0.73
No documented plan
316 10 83.3 23 69.7 60 69.0 224 65.7 —
Discussion
Our results indicate that a history of substance use is common among pregnant people experiencing incarceration, with nearly three-quarters of our sample reporting pre-incarceration use of multiple substances and over half reporting opioid use. Our findings that postpartum contraceptive counseling is frequently missing from the prenatal record and fulfillment of postpartum contraception plans is low for those with and without a history of substance use align with existing literature indicating that pregnant and postpartum patients experiencing incarceration have unmet contraceptive needs.17,18
Among those in our sample with documented plans for contraception, both LARC and permanent contraception were common. Although both of these methods are used by many women in the community, prior literature indicates that providers in community contexts may encourage patients with substance use disproportionately towards longer-acting methods like LARC.13,19 That a larger proportion of patients in our sample with illicit substance use had documented plans for permanent contraception may indicate that these biases are also present among clinicians in carceral contexts.
Despite these relatively common postpartum contraception plans, however, fulfillment rates were low among patients with pre-incarceration substance use, suggesting that contraceptive preferences of pregnant people experiencing incarceration with a history of substance use may not be adequately prioritized. This may not be so different from non-incarcerated residents of North Carolina; state-level data from the 2022 Behavioral Risk Factor Surveillance System suggest that 40% of North Carolinians in the community have unfulfilled contraceptive preferences. 19 Low rates of fulfillment of all methods of contraception could also be interpreted as the result of appropriate efforts on the part of hospital and outpatient clinicians to confirm genuine desire for a requested method of contraception prior to fulfillment. Incarcerated individuals and individuals with histories of substance use and criminal legal system involvement have long been a target of coercive LARC placement and sterilization practices that restrict their autonomy in contraceptive decision-making, including contemporary events in California and Tennessee.20–22 The lack of documentation of patients declining postpartum contraception, however, suggests that this is not the entire explanation for these findings.14,15,20,21 While only a minority of patients had any documented postpartum contraception plan, very few of those patients declined postpartum contraception altogether.
Equitable and comprehensive contraceptive counseling should be integrated into standard care for pregnant and postpartum people experiencing incarceration with a history of substance use in order to prioritize the autonomy and informed decision-making of incarcerated individuals. Policies that aim to limit contraceptive coercion for this population should not preclude postpartum access to all FDA-approved reversible contraceptive methods in custody. The lack of access to postpartum contraception places patients, including those with a history of substance use, at higher risk for short interval pregnancies when they are released from custody, which thus increases their risks of pregnancy and maternal complications.8,17,18
Our findings further echo the literature on unmet SUD management for pregnant people experiencing incarceration. Postpartum contraception is only one part of comprehensive, integrated perinatal SUD care. The rate of MOUD treatment in our study population was slightly higher than in existing literature, but may still reflect a significant gap. 23 MOUD during pregnancy is recommended by the American College of Obstetricians and Gynecologists and has the potential to reduce fetal exposure to non-prescribed opioid use, improve adherence to prenatal care and addiction treatments, increase neonatal birthweights, and lower the risk of relapse and overdose. 24 To provide the recommended perinatal care and improve maternal and infant outcomes, facilities should leverage academic and community partnerships to expand access to SUD care in custody and referrals for continuation of SUD care in the community.
Limitations
There are a few limitations to note. Our reliance on electronic health record data introduces data quality issues both in terms of missingness (e.g. contraceptive counseling, contraception utilization, substance use type, and prenatal care adequacy) as well as the inability to discern outcomes post-incarceration. Additionally, our findings may not be generalizable beyond the state prison system studied. For this secondary analysis of observational data, we also did not conduct a separate power calculation. Lastly, the study is quantitative in nature and does not account for the clinical quality and content that may influence a patient’s contraceptive plan.
Conclusion
Our study highlights critical gaps in the provision of reproductive healthcare and substance use treatment for pregnant and postpartum patients in custody with a history of substance use. Addressing the gaps in MOUD provision and contraceptive needs for pregnant people with a history of substance use during incarceration and with continuity of care in the community requires systemic reforms to ensure consistent availability and access.
Supplemental material
ORCID iDs
Grace A. Trompeter https://orcid.org/0000-0003-2539-7131
Andrea K. Knittel https://orcid.org/0000-0002-4292-3255
Ethical considerations
This project was reviewed and approved by the Institutional Review Board at the University of North Carolina at Chapel Hill (IRB #22-2250) on December 14, 2022 and was reviewed by the North Carolina Department of Adult Corrections.
Consent to participate
Written consent was waived by the IRB for this secondary analysis of existing medical record data.
Data Availability Statement
Data are available from the corresponding author on reasonable request.*