The association between criminal legal attitudes and healthcare utilization among adolescents: differences by gender and race

The study examines whether adolescents’ negative criminal legal (CL) attitudes—encompassing legal cynicism and perceptions of police legitimacy—are associated with reduced healthcare utilization, reflecting the sociological concept of system avoidance. Prior work in adults shows that criminal legal contact is linked to foregoing needed medical care, avoiding formal healthcare settings, and increased distrust of healthcare providers. Given adolescents’ frequent police contact and the developmental process of legal socialization, the authors hypothesize that youth with more negative CL attitudes will be less likely to utilize preventive healthcare services.

Data source and sample: Secondary analysis of the Future of Families and Child Wellbeing Study (FFCWS), a cohort of ~4,700 families representative of non-marital births in large U.S. cities, with six waves from birth (1998–2000) through age 15 (2014–2017). The analytic sample included adolescents who completed Wave 6 at age ~15 and had complete data (N=3,444 of 4,663 eligible). IRB approval obtained; data publicly available.

Measures: (1) Healthcare utilization: Primary caregiver report of whether the teen had a regular check-up or well visit in the past 12 months (yes/no). (2) Criminal legal attitudes: Adolescents completed six items reflecting legal cynicism and attitudes toward police (4-point scale, 0–3; higher scores indicate more negative attitudes; α=0.67). Items included respect for police (reverse-coded), permissive attitudes toward rule-breaking, and the view that police create more problems than they solve. The mean of items formed the CL attitude score. (3) Health insurance: Primary caregiver report of current coverage by public (e.g., Medicaid) and/or private insurance; coded yes/no. (4) Delinquency: Adolescents reported engagement in 13 delinquent acts in past 12 months (ADD Health items). Each item recoded to 0/1 and summed to a variety score (range 0–12). (5) Sociodemographics and psychosocial covariates: Gender (male/female), race/ethnicity (non-white vs. non-Hispanic white), age (14–19), household income (0=<$5,000 to 8=>$60,000), primary caregiver education (0=<HS to 3=college/grad), primary caregiver marital status (married vs. not), teen depression (CES-D 5-item; α=0.76), and teen anxiety (modified BSI-18 6-item; α=0.76).

Statistical analysis: Logistic regression models estimated the association between CL attitudes and preventive healthcare utilization, controlling for covariates. Models were stratified by gender and race (non-white boys, non-white girls, white boys, white girls). Analyses were conducted in Stata 17. Sensitivity analyses additionally controlled for youth ever being stopped by police.

• Descriptive statistics: 88.66% of adolescents had a preventive healthcare visit in the prior 12 months; 96.22% had current health insurance. Mean criminal legal attitudes score M=0.76 (SD=0.57). Mean delinquency M=1.09 (SD=1.66). Sample was 47.80% female, 81.93% non-white; mean age 15.60.
• Primary association: Only among non-white girls were higher (more negative) criminal legal attitudes significantly associated with reduced likelihood of having a regular check-up in the past year (aOR=0.55, 95% CI: 0.35–0.85; p≤0.01).
• Among non-white girls, delinquency was associated with an increased likelihood of having a regular check-up (aOR=1.28, 95% CI: 1.06–1.55).
• Health insurance status was significantly associated with increased likelihood of preventive healthcare utilization among non-white boys, non-white girls, and white girls, with adjusted odds ratios reported as 3.33 (95% CI: 1.60–6.93), 3.91 (95% CI: 1.77–8.62), and 5.46 (95% CI: 1.13–26.31), respectively.
• Household income increased the likelihood of healthcare utilization among non-white boys (aOR=1.10, 95% CI: 1.02–1.19).
• Sensitivity analyses: Adjusting for ever being stopped by police did not change the significant findings. Among non-white girls, delinquency became marginally significant (aOR=1.20, 95% CI: 0.99–1.47), and ever being stopped by police was marginally associated with increased likelihood of healthcare utilization (aOR=2.08, 95% CI: 1.00–4.34).

Findings support the system avoidance framework: adolescent non-white girls with more negative criminal legal attitudes were less likely to seek preventive healthcare, suggesting that mistrust of legal authorities may generalize to other surveilling institutions like healthcare. Despite expectations that delinquency would align with system avoidance, delinquency among non-white girls was associated with greater healthcare utilization, potentially reflecting health needs arising from delinquent behaviors (e.g., injuries, substance use, unintended pregnancy) that prompt care use regardless of institutional mistrust. These patterns underscore the importance of understanding how legal socialization processes intersect with race and gender to shape adolescents’ engagement with healthcare.

This study contributes evidence that negative criminal legal attitudes among non-white adolescent girls are associated with reduced preventive healthcare utilization, consistent with system avoidance across institutions. Policymakers should consider how criminal surveillance and policing practices, particularly in minority communities, may shape youths’ institutional trust with downstream effects on healthcare engagement. Future research should directly test whether mistrust in the criminal legal system fosters mistrust in healthcare, identify mechanisms linking legal attitudes to healthcare use over time, and clarify when and how adolescents with delinquent behavior engage with health services.

• Cross-sectional design precludes causal inference.
• Healthcare utilization measured crudely as any preventive check-up in the past 12 months; emergency, urgent care, and specialty visits were not assessed.
• The study did not directly measure nor model the pathway from police contact to legal cynicism in this analysis, although sensitivity analyses adjusted for ever being stopped by police.
• Potential measurement limitations include moderate internal consistency for the CL attitudes scale (α=0.67).