Background <p>Substance use disorders (SUDs) are more prevalent among youth involved in the legal system (YILS) compared to non-system-involved peers. Legal system involvement can serve as a gateway to health services, presenting an opportunity to identify YILS in need of SUD intervention. Here, aims included: (1) describe YILS patterns of substance use based on drug testing (e.g., urine) versus a self-report risk screener, (2) characterize how screening practices and results differ based on YILS demographic characteristics, and (3) compare how positive screening informs referrals to behavioral health treatment by screening type.</p> Methods <p>Administrative records were collected for all youth, ages 11–17, arrested in one of eight Indiana counties during 2019–2023 (<i>N</i> = 1,197 youth). Variables included youth demographics, most severe alleged charge at arrest, drug test results, self-report substance use screen results, and youth referrals to behavioral health services by probation officers. Prevalence of screening receipt by type, positive screens, and substance detected in drug tests are reported. Chi-square tests were conducted between demographic variables and drug test and self-report screener receipt, and positive screen per screening type. Logistic regression models were used to determine the association between positive screen and referral to services; models were compared using the <i>c</i>-statistic.</p> Results <p>Of sampled youth, 26.7% received a drug test, and 58.0% a self-report screener; of those, 54.4% and 42.0% screened positive, respectively. Cannabinoids (84.5%) and alcohol (28.2%) were detected most often. Youth whose most severe alleged charge was drug-related were more likely to receive a drug test, to receive a self-report screener, and screen positive. Positive drug test (OR = 8.48 95% CI = 6.06–11.87) and self-report screen (OR = 11.55, 95% CI = 8.33–16.01) were associated with behavioral health referral; the <i>c</i>-statistic was slightly higher for self-report screener (0.77) than for drug test (0.74).</p> Conclusions <p>Probation officers may be utilizing self-report screener results to make referral recommendations differently than drug test results. For less resourced counties that do not administer drug tests, self-report screeners may be particularly helpful in triaging youth to other services that routinely monitor substance use. Future research is needed to identify the optimal frequency of substance use screening to inform service referrals.</p> Trial registration <p>The study was registered through ClinicalTrials.gov on 7/30/2020 (ID NCT04499079).</p>