Abstract
Child protection systems often intervene following substance exposed deliveries but are not designed to address the postpartum needs of the delivering parent. In this retrospective cohort study, we aimed to estimate the association between early child protection system removal and high-risk subsequent birth trajectories among a cohort of mothers with substance-exposed deliveries in California. Of 6,893 births in 2015 with documented prenatal drug and alcohol exposure, 20.4% of mothers experienced child protection removal in the first month of their infant’s life. First-month child protection system removal was associated with short interval birth (aHR 1.61, 95% CI 1.09-2.36) and short interval birth with documentation of substance exposure (aHR 3.17, 95% CI 1.65-6.08). We found that child separation was associated with an increase and not a reduction in subsequent substance-exposed births. These findings indicate the need for focused public health and supportive services to address the treatment, healthcare, family-building, and psychological needs of parents with substance use during pregnancy.
PRECIS:
Child protection removal of an infant in the setting of prenatal substance exposure was associated with short interpregnancy intervals and repeat substance-exposed birth for the mother.
Introduction
Interpregnancy intervals less than 6 months are associated with adverse outcomes, including preterm birth, low birth weight, early neonatal death, and maternal mortality and severe morbidity.1–3 Prenatal substance (drug and alcohol) use compared with no prenatal use, is associated with higher rates of unintentional pregnancies4 and lower contraceptive use,5 both risk factors for short interpregnancy intervals.6, 7
Families experiencing removal by the child protection system following a substance-exposed delivery are more likely to have repeat births and substance-exposed repeat births than those retaining physical custody.8 No publications, however, have examined short interval births by custody status. The objective of this study was to document birth trajectories following a substance exposed birth, comparing those who experienced a child protection removal in the first month of their child’s life with those who did not.
Methods
This retrospective cohort study linked vital records, maternal hospital discharge records, and child protection system records in order to identify a cohort of women who experienced a live birth event in 2015 and who also had an International Classification of Diseases Clinical Modification (ICD-CM) code indicating prenatal substance use (Appendix 1, available online at http://links.lww.com/xxx). For the primary exposure, we coded each birth into one of two mutually exclusive groups based on child protection system removals: (1) removal, which was measured according to whether a child protection system placement episode began within 1 month of the child’s date of birth, regardless of placement type or duration; (2) no removal.
The primary outcome was a subsequent birth that occurred between 200 and 463 days after the index birth. For models estimating risk of repeat substance-exposed births, we included only those that experienced short interval birth and examined whether child removal was associated with documented substance use in the subsequent birth.
To examine the risk of short interval birth, we specified a Cox proportional hazard regression model to produce an adjusted hazard ratio (aHR) with 95% confidence intervals determined through robust standard error adjustments. Next, we used a Fine-Gray subdistribution hazard model to analyze the cause-specific hazard of a substance exposed repeat birth, censoring repeat birth without documented substance exposure as potentially informative competing events.9 We then modeled cumulative hazard and cumulative incidence curves to illustrate each outcome between the removal and non-removal groups.10 All analyses were conducted using Stata v.17 (StataCorp, College Station, TX). Data are available to the Children’s Data Network through data sharing agreements and this project falls under approvals through both state and university institutional review boards (USC #UP-13-00455 and CA State CPHS #13-10-1366). Study data are subject to strict confidentiality and non-rerelease agreements.
Results
Overall there were 460,001 births in 2015, and 1.5% (6,893) had documented prenatal substance use. Approximately one fifth of these mothers (20.4%) experienced child protection removal in the first month of life. Women with removal were more likely to experience a short interval repeat birth (3.3% vs. 2.1%) and to have documentation of substance exposure on the subsequent birth record (47% vs. 24.1%) than those without removal. See Appendix 2, available online at http://links.lww.com/xxx, for descriptive statistics of our population across outcomes.
In a fully adjusted Cox proportional hazard model, child protection removal was significantly associated with short interval repeat births when compared with those without removal (aHR 1.61, 95% CI 1.09-2.36). Among those with short interval births, there was an association between child protection removal and documentation of substance exposure in the subsequent birth (aHR 3.17, 95% CI 1.65-6.08) (Table 1). The cumulative hazard and incidence curves showed significantly higher cumulative probabilities of short interval repeat birth and substance exposed short interval repeat birth in the group with removal compared with those without removal (Figure 1).
Table 1.
Incidence, relative risk, and adjusted hazard ratio of rapid repeat birth and substance exposed rapid repeat birth by child protection system removal
| Subsequent births | No first month removal N = 5,488 |
First month removal N = 1,405 |
|---|---|---|
| Repeat short interval births (within 200-463 days) | ||
| Follow up person months | 84,397.73 | 21,558.33 |
| Event | 116 | 47 |
| Incidence Rate* | 1.37 | 2.18 |
| Crude Relative Risk (95% CI) | Reference | 1.58 (1.33-2.21) |
| Adjusted Hazard Ratio (95% CI)† | Reference | 1.61 (1.09-2.36) |
|
| ||
| Substance exposed repeat short interval births | ||
| Follow up person months | 84,397.73 | 21,558.33 |
| Event | 28 | 22 |
| Incidence Rate* | 0.33 | 1.02 |
| Crude Relative Risk (95% CI) | Reference | 3.06 (1.76-5.35) |
| Adjusted Hazard Ratio (95% CI)† | Reference | 3.17 (1.65-6.08) |
Per 1,000 person months;
Model adjusted for maternal race and ethnicity, maternal age, maternal education, insurance type, prior births, and documentation of paternity.
Figure 1.
The fully adjusted cumulative incidence of repeat birth outcomes between those with and without first-month custody removal of a 2015 substance-exposed birth. A. The cumulative hazard of short interval repeat birth. B. The cumulative incidence of substance-exposed, short interval repeat birth.
Discussion
We found an association between infant removal by child protection and short interval repeat birth and short interval repeat birth with prenatal substance use. The observed associations may indicate a behavioral response to child removal or a level of disorganization related to substance use that affects one’s ability to adopt effective family planning strategies.11, 12 It also may indicate a broader failure of systems to engage postpartum individuals in substance use treatment or supportive parenting programs, particularly those who experience child protection system removal.13 Targeted research and services are needed to respond to the psychological sequelae and resultant needs of custody loss, with specific attention to guarding against coercive pregnancy prevention tactics and discrimination against parents who use substances.14
A study strength is the use of linked administrative records from a large, diverse state, although we could not ascertain any repeat births that may have occurred outside of California. Other limitations include reliance on hospital diagnostic codes, a failure to account for postnatal fertility, not adjusting for specific substance exposures, and a binary definition of child protection removal. Overall, our results underscore the importance of longstanding questions about how child protection and other systems respond to prenatal substance exposure, including the healthcare, family-building, and psychological needs of parents. Comprehensive public health and clinical interventions should be designed to focus on the postpartum health of birth parents with substance use and substance use disorders, particularly following child protection system involvement.
Supplementary Material
Acknowledgments
This research was supported by an Innovative Research Grant from the Center for Innovation in Child Maltreatment Policy, Research, and Training at the Washington University in St. Louis, through a grant from the National Institutes of Health National Institute of Child Health and Human Development (NICHD P50HD096719).
The authors thank Huy T. Nghiem, John Prindle, and Stephanie Cuccaro-Alamin, who assisted with the preparation of data files.
Although the findings reported and conclusions drawn from these data are solely those of the authors and should not be considered to reflect those of any agency of the California government, this analysis would not be possible without the partnership of the California Department of Social Services, the California Department of Public Health, and the California Department of Healthcare Access and Information.
Financial Disclosure
Davida Schiff reports that money was paid to their institution from the NIDA (K23DA048169). They have received funding for presentations from the AAP, UMass, and Darmouth Hitchcock. The other authors did not report any potential conflicts of interest.
Each author has confirmed compliance with the journal’s requirements for authorship.
Footnotes
Presented at the American Public Health Association Annual Meeting, Monday, November 7, 2022, in Boston, MA.
Contributor Information
Julia Reddy, University of North Carolina Chapel Hill Gillings School of Global Public Health, Chapel Hill, NC.
Davida Schiff, MassGeneral Hospital for Children, Harvard Medical School, Boston, MA.
Mishka Terplan, Friends Research Institute, Baltimore, MD.
Hendree Jones, University of North Carolina Chapel Hill Department of Obstetrics & Gynecology, Chapel Hill, NC.
Emily Putnam-Hornstein, University of North Carolina Chapel Hill School of Social Work, Chapel Hill, NC.
References
- 1.Ni W, Gao X, Su X, Cai J, Zhang S, Zheng L, et al. Birth spacing and risk of adverse pregnancy and birth outcomes: A systematic review and dose-response meta-analysis. Acta Obstet Gynecol Scand. 2023. doi: 10.1111/aogs.14648 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Schummers L, Hutcheon JA, Hernandez-Diaz S, Williams PL, Hacker MR, VanderWeele TJ, et al. Association ofshort interpregnancy interval with pregnancy outcomes according to maternal age. JAMA Internal Medicine. 2018;178(12):1661–70. doi: 10.1001/jamainternmed.2018.4696 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.De Silva DA, Thoma ME. The association between interpregnancy interval and severe maternal morbidities using revised national birth certificate data: A probabilistic bias analysis. Paediatr Perinat Epidemiol. 2020;34(4):469–80. doi: 10.1111/ppe.12560 [DOI] [PubMed] [Google Scholar]
- 4.Loree AM, Gariepy A, Ruger JP, Yonkers KA. Postpartum contraceptive use and rapid repeat pregnancy among women who use substances. Substance Use & Misuse. 2018;53(1):162–9. doi: 10.1080/10826084.2017.1327976 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Terplan M, Hand DJ, Hutchinson M, Salisbury-Afshar E, Heil SH. Contraceptive use and method choice among women with opioid and other substance use disorders: A systematic review. Prev Med. 2015;80:23–31. doi: 10.1016/j.ypmed.2015.04.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Thiel de Bocanegra H, Chang R, Howell M, Darney P. Interpregnancy intervals: impact of postpartum contraceptive effectiveness and coverage. American Journal of Obstetrics and Gynecology. 2014;210(4):311.e1–.e8. doi: 10.1016/j.ajog.2013.12.020 [DOI] [PubMed] [Google Scholar]
- 7.Gemmill A, Lindberg LD. Short interpregnancy intervals in the United States. Obstet Gynecol 2013;122(1):64–71. doi: 10.1097/AOG.0b013e3182955e58 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Grant T, Christopher Graham J, Ernst CC, Michelle Peavy K, Brown NN. Improving pregnancy outcomes among high-risk mothers who abuse alcohol and drugs: Factors associated with subsequent exposed births. Children and Youth Services Review. 2014;46:11–8. doi: 10.1016/j.childyouth.2014.07.014 [DOI] [Google Scholar]
- 9.Austin PC, Fine JP. Practical recommendations for reporting Fine-Gray model analyses for competing risk data. Stat Med. 2017;36(27):4391–400. doi: 10.1002/sim.7501 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Austin PC, Lee DS, Fine JP. Introduction to the analysis of survival data in the presence of competing risks. Circulation. 2016;133(6):601–9. doi: 10.1161/CIRCULATIONAHA.115.017719 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Canfield M, Radcliffe P, Marlow S, Boreham M, Gilchrist G. Maternal substance use and child protection: a rapid evidence assessment of factors associated with loss of child care. Child Abuse & Neglect. 2017;70:11–27. doi: 10.1016/j.chiabu.2017.05.005 [DOI] [PubMed] [Google Scholar]
- 12.MacAfee LK, Harfmann RF, Cannon LM, Minadeo L, Kolenic G, Kusunoki Y, et al. Substance use treatment patient and provider perspectives on accessing sexual and reproductive health services: Barriers, facilitators, and the need for integration of care. Substance Use & Misuse. 2020;55(1):95–107. doi: 10.1080/10826084.2019.1656255 [DOI] [PubMed] [Google Scholar]
- 13.Peacock-Chambers E, Buckley D, Lowell A, Clark MC, Friedmann PD, Byatt N, et al. Relationship-based home visiting services for families affected by substance use disorders: A qualitative study. Journal of Child and Family Studies. 2022;31(8):2121–33. doi: 10.1007/s10826-022-02313-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Abbass N, Malhotra T, Bullington W, Arora KS. Ethical issues in providing and promoting contraception to women with opioid use disorder. J Clin Ethics. 2022;33(2):112–23. doi: 10.1086/jce2022332112 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.