Article: Silver, I. A., Newsome, J., & Cohen, T. (2025). “Health Inequalities: Is Adolescent Involvement in the Legal System Associated with Reduced Health Care Access and Usage during Adulthood?” Preventive Medicine, 200, 108413 https://doi.org/10.1016/j.ypmed.2025.108413
Preprint available at: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5086255
The ability to access and use health care is critical for preventing health conditions and addressing illnesses throughout an individual’s life. While many Americans maintain health insurance and receive the preventive or acute health care needed, approximately 30 million Americans remain uninsured in the United States. Prior research has highlighted that socioeconomic hardship, limited employment opportunities, geography, distrust in medical professionals, and costs of healthcare reduce health insurance enrollment and regular health care. Given these risk factors, historically disadvantaged populations often possess the lowest rates of health care access and usage.
Among the most vulnerable groups are individuals who were involved in the criminal legal system (CLS) during adolescence (before the age of 18). Existing research documents that former CLS involved adolescents are often uninsured and face challenges maintaining insurance coverage. These individuals often rely on public health insurance programs (e.g., Medicaid), as they come from impoverish families and have limited opportunity to obtain the financial or employment requirements associated with private health insurance. Additionally, youth incarcerated in juvenile or adult facilities may receive inconsistent or inadequate health care while confined and typically receive little reentry support focused on health literacy, insurance navigation, or care coordination. As such, CLS contact during adolescence may disrupt health-related socialization and access to preventative care, potentially creating lifelong disparities in health care coverage and usage.
However, research to date has yet to fully explore the association between CLS involvement before the age of 18 and health insurance coverage and health care usage. This study addresses a key gap by assessing whether adolescent CLS involvement is associated with an increased or decreased number of years with health insurance coverage, preventive care doctor visits, and visits to doctors when sick during adulthood.
2. Summary of Findings
Using 24 years of data from the National Longitudinal Survey of Youth-1997 (NLSY97), the study followed 8,961 respondents from 1997-2021 (had to be under 18 years of age in 1997). Three key outcomes were analyzed: (1) number of years respondents reported having health insurance (2003-2021), (2) number of years they reported receiving routine check-ups (2003-2021), and (3) number of years they visited a doctor when sick (2003-2008).
Descriptive Results: Individuals with no adolescent CLS contact averaged nine years of health insurance coverage and seven years of routine check-ups, compared to roughly five years for those incarcerated in juvenile or adult facilities (statistically significant differences existed between the groups).
Multivariate Results: After adjusting for 25 covariates and using gradient-boosted inverse probability weighting (IPW) and a doubly robust design, the study found (see Figure 1):
- Health Insurance Coverage: Arrest before 18 predicted a 9% reduction in the number of years with health insurance coverage (IRR = 0.91), while incarceration in juvenile and adult facilities before 18 was associated with 30% (IRR = 0.71) and 36% (IRR = 0.64) reductions in the number of years with health insurance coverage, respectively.
- Routine Check-Ups: Arrest before 18 was associated with a 3% reduction (IRR = 0.97) in the number of years reporting routine check-ups, whereas incarceration in juvenile and adult facilities before 18 predicted 16% (IRR = 0.84) and 15% (IRR = 0.85) reductions in the number of years reporting routine check-ups, respectively.
- Doctor Visits When Sick: Arrest and incarceration in an adult facility before 18 were linked to 4% and 7% increases in the number of years with a doctor visit when sick, respectively. Incarceration in a juvenile facility before 18, however, was associated with a 21% reduction (IRR = 0.79) in the number of years with a doctor visit when sick.
The consistency of results across multiple model specifications (including sensitivity analyses with and without IPWs) strengthened confidence in these findings.
3. Implications
This study underscores a persistent health care disparity for individuals involved in the CLS before age 18. The results suggest that adolescent CLS contact is associated with reduced health insurance coverage and routine health care activities into adulthood.
Policy Implications:
- Continuity of Coverage: Findings highlight the importance of maintaining Medicaid coverage during and after incarceration. While the Consolidated Appropriations Act, 2023 automatically provides Medicaid coverage 30 days prior to release for confined juveniles, expanding this window to 90 days post-release could better stabilize coverage and improve continuity of care.
- Health Care Navigation: Implementing reentry programs that include health care navigators or coordinators could facilitate enrollment in health insurance coverage and encourage engagement in preventative health care. Programs such as the Health Care Coordinators (HCC) offered by Partners for Kids in Franklin County illustrates the value of this program for youth reentering the community from juvenile detention facilities.
- Education and Health Literacy: Correctional facilities should expand access to health education and preventative services to mitigate the developmental disruptions caused by incarceration. Targeted health literacy interventions can empower youth to navigate health systems effectively after release.
- Employment and Economic Stability: Policies and treatment that improve employment opportunities for formerly incarcerated youth could increase health insurance access and usage, as employers remain the primary source of private health insurance in the United States.
Notes: Robust (sandwich) standard errors were used when estimating the models. Pre-weighting and post-weighting balancing statistics across the treatment groups are provided in Appendix A for the ATE weights.
4. Data and Methods
Sample: The NLSY97 is a nationally representative longitudinal survey of individuals born between 1980 and 1984. The analytic sample (N=8,961) included all respondents under 18 in 1997 to preserve temporal ordering of CLS exposure and covariates. Respondents were followed through 2021, when they were ages 36–41. More than 75% of participants completed each wave, with most completing over 15 interviews.
Measures:
- Dependent Variables: Years of health insurance coverage (0–14 years), years reporting routine check-ups (0–14 years), and years visiting a doctor when sick (0–6 years).
- Independent Variables: Four dichotomous indicators captured CLS involvement before 18: 1) no contact (reference group), 2) arrested, 3) incarcerated in a juvenile facility, and 4) incarcerated in an adult facility.
- Covariates: Twenty-five variables were included to adjust for potential confounding, including demographic characteristics, parental education and incarceration, household wealth, adverse experiences, peer deviance, school engagement, and baseline health insurance in 1997.
Analytical Strategy: Missing data were addressed with multiple imputation (MICE, 25 iterations). Gradient boosted multi-group IPWs were estimated using the twang package in R, followed by doubly robust Poisson regressions to estimate incident rate ratios (IRRs). Supplemental models (no imputation, no weights, ATT weights) confirmed robustness. All analyses were conducted in R/RStudio, with replication materials available at the project GitHub repository (https://github.com/ianasilver/JIF-and-Health-Care-Access) .
5. Conclusion
Involvement in the CLS before 18 – particularly incarceration in a juvenile or adult facility – has significant and lasting association with health insurance coverage and preventive health care use. These findings have important implications for health equity, juvenile reentry policy, and public health planning. Expanding Medicaid continuity, improving health literacy, and addressing structural barriers to employment and care could reduce disparities for this vulnerable population.
Disclosure: This research brief was prepared by ChatGPT and reviewed/edited by Ian A. Silver.
ianasilver.com 