First community mental health contact after youth justice supervision among adolescents in New South Wales, Australia: timing, setting, and association with reincarceration in a population-based data-linkage study

Technical Summary: First Community Mental Health Contact After Youth Justice Supervision

Problem Statement
Mental disorders are disproportionately prevalent among adolescents under youth justice supervision, with clinical assessments identifying disorders in over 60% of this population, compared to approximately 13% in community samples. A critical gap exists in the continuity of care as young people transition between custody, community supervision, and release. Responsibility for care is fragmented across custodial health services, community mental health teams, and primary care, often resulting in a breakdown of relational, informational, and management continuity. While previous research has established the high prevalence of disorders and aggregated service use, it has failed to characterize the specific timing, setting, and delays of the first community mental health contact post-supervision. Furthermore, the relationship between the timeliness of this initial contact and subsequent reincarceration remains unquantified, particularly regarding whether systemic barriers produce slower contact for overrepresented groups, such as Aboriginal and Torres Strait Islander adolescents.

Methodology
This study employed a population-based retrospective cohort design using de-identified, probabilistically linked administrative records from New South Wales (NSW), Australia.

• Cohort: The study included 1,589 adolescents (aged 10–17) involved in the youth justice system, drawn from four total-population health surveys conducted between 2003 and 2015 (Young People in Custody Health Surveys and Young People on Community Orders Health Survey).
• Data Linkage: Records were linked via the Centre for Health Record Linkage (CeHReL) with an estimated false-positive rate below 0.5%. Linked datasets included the NSW Reoffending Database (justice), Admitted Patient Data Collection (APDC), Mental Health Ambulatory Data Collection (MH-AMB), Emergency Department Data Collection (EDDC), Controlled Drugs Data Collection, and the Registry of Births, Deaths and Marriages.
• Definitions:
  • Mental Health Disorder: Ascertained via ICD-10-AM codes in health records and/or self-report of prior diagnosis.
  • Community Mental Health Contact: Defined as an outpatient episode, admitted-patient episode with a mental health diagnosis, emergency presentation with a mental health diagnosis, or opioid agonist treatment record occurring in custody-free community time.
  • Follow-up: Measured exclusively during custody-free community time (24 months maximum). Time spent in custody was censored.
• Statistical Analysis:
  • Time-to-Event: Flexible parametric survival models (Royston-Parmar) were used to model the time to first community contact, accommodating non-proportional hazards identified in diagnostics. Time-varying effects were reported as adjusted hazard ratios (aHR) at specific time points.
  • Reincarceration: Landmark analyses were conducted at 7, 14, 21, and 28 days post-release to estimate the association between early contact and subsequent reincarceration. Modified Poisson regression with robust standard errors was used to calculate adjusted risk ratios (aRR) and risk differences (RD), avoiding immortal time bias.

Key Contributions and Results
The study provides granular data on the breakdown of care continuity and its association with justice outcomes:

1. Prevalence and Gaps in Contact:

   • Mental health disorders were identified in 42.5% (676/1,589) of the cohort.
   • Among those with a disorder, only 17.8% made a community mental health contact within 30 days of baseline.
   • By 24 months, 48.4% (327/676) of adolescents with a disorder had no recorded community mental health contact.
   • Of the 349 first contacts recorded, 29.2% occurred via emergency presentation, same-day hospital episode, or psychiatric admission, indicating a reliance on acute rather than preventative care.

2. Factors Associated with Delayed Contact:

   • Aboriginal and/or Torres Strait Islander Status: Contact was significantly slower for Indigenous adolescents (aHR 0.74, 95% CI 0.58–0.94).
   • Custody Status: Contact was progressively slower after release from custody compared to community supervision, with the hazard ratio dropping to 0.44 (95% CI 0.27–0.66) at 24 months.
   • Disorder Type: Contact was faster for adolescents with mood disorders (aHR 1.64) and psychosis (aHR 1.44).
   • Ascertainment Source: Adolescents whose disorder was identified only by self-report (with no prior health record diagnosis) experienced significantly slower contact compared to those with health-record diagnoses.

3. Association with Reincarceration:

   • Among adolescents released from custody with a mental health disorder, early contact was associated with a statistically significant reduction in reincarceration risk across all landmarks.
   • Adjusted risk ratios ranged from 0.82 to 0.84.
   • Adjusted risk differences indicated a reduction in reincarceration risk of 12.5 to 14.1 percentage points for those making contact within 7 to 28 days compared to those who did not.

Significance and Claims
The paper claims that the transition from youth justice supervision to community care represents a critical clinical handover point where continuity frequently fails. The study's significance lies in:

• Disaggregating Care Failures: It distinguishes between "delayed first contact" and "first contact occurring only through emergency/inpatient settings," two failures often concealed by aggregated service use data.
• Identifying Vulnerable Subgroups: It quantifies that Aboriginal and Torres Strait Islander adolescents and those with disorders identified solely by self-report face the greatest delays in accessing care.
• Linking Continuity to Justice Outcomes: It provides evidence that early linkage to community mental health care is associated with lower reincarceration rates, suggesting that continuity of care is not merely a clinical goal but a justice outcome.
• Policy Implications: The authors argue for treating the post-supervision period as a shared priority for youth justice and child/adolescent mental health services. They advocate for proactive service navigation, transitional care models adapted for adolescents, and routine linked-data reporting to monitor continuity and evaluate reforms.

The study acknowledges limitations, including the inability to observe private or interstate care, potential residual confounding in the landmark analyses, and the historical span of the data (2003–2015), but maintains that the findings robustly highlight the need for improved active linkage between justice and health systems.