New South Wales cohort links incarceration to health and social outcomes across 22 years
A new population-scale cohort from New South Wales is built to answer a question that correctional and health systems have struggled to measure with precision: what happens after prison release when the same people move between custody, hospitals, mental health care, drug treatment, mortality records, and the justice system? The Prison Outcomes STudy (POST) assembles linked administrative data for every adult admitted to full-time custody in New South Wales from 2000 through 2022, creating a longitudinal platform large enough to examine post-release outcomes at the level of the whole population rather than a narrow clinic sample.
The cohort includes 200,486 adults. Women made up 15% of the group, or 30,698 people. First Nations people accounted for 27% of the cohort. Those proportions matter because incarceration does not fall evenly across the population, and the clinical consequences of custody are shaped by inequities that begin long before imprisonment and continue after release.
The study is not just descriptive. It is designed to measure treatment for drug dependence and serious mental illness, estimate health and social outcomes after custody, and project what might change if access to care were expanded in prison and after release. In practical terms, POST is positioned to do for correctional health what large registry studies have done for cardiovascular and oncology care: expose the size of the gap between need and service delivery.
A linked data platform built from 18 administrative sources
The cohort was assembled from custody records linked probabilistically to 18 health, justice, and mortality datasets. That architecture gives the study reach across settings that usually operate in silos. A person’s prison history can be connected with community mental health contact, alcohol and other drug treatment, and death records, allowing investigators to follow events before, during, and after custody rather than treating release as a clean break.
This approach is especially valuable in correctional medicine because the period after release is medically unstable. People leaving prison often face abrupt changes in medication access, housing, social support, and drug exposure. The linked design can capture those transitions in a way that single-service data never can. It also allows comparisons across subgroups, which is essential when the burden of incarceration and the burden of untreated illness are both concentrated in the same communities.
The cohort profile reports 2,282,367 person-years of follow-up. Of that time, 257,545 person-years, or 11%, were spent in custody. The rest unfolded in the community, where the consequences of interrupted care, relapse, psychiatric instability, and re-entry stress are most likely to appear.
One result stands out immediately: half of the cohort, 95,563 people, had at least one contact with community mental health services.
Mental health and substance use care dominate the post-release clinical picture
The care patterns documented in POST point to the dual burden of serious mental illness and drug dependence in people who have been incarcerated. Community mental health contact in 48% of the cohort suggests that psychiatric morbidity is not peripheral to this population; it is central to their health trajectory. Alcohol and other drug treatment had been received by 27% of the cohort, or 54,100 people, showing that substance use disorders are also deeply embedded in the service needs of this group.
Those figures do not imply that treatment needs are limited to the people already seen by services. They show that a large fraction of the cohort has already intersected with care, yet many will still cycle through custody and the community. In a setting where psychiatric symptoms, withdrawal, overdose risk, and social instability can intensify one another, the timing and continuity of treatment matter as much as the treatment itself.
The cohort structure will allow investigators to compare outcomes across people with differing service exposure, including those who received care in prison versus those whose engagement occurred only after release. That distinction is likely to be clinically important because incarceration can either interrupt treatment or create a narrow window for intervention, depending on whether services are available, sustained, and coordinated across settings.
Population-wide evidence for justice, health, and community policy
POST is built to inform policy at multiple levels. Within prisons, it can estimate the effects of treatment for drug dependence and serious mental illness. Outside prison, it can assess whether community follow-up changes the trajectory of health, mortality, and recidivism. Across both settings, it can identify subgroups with the highest burden and the weakest service access.
The modeling component adds a forward-looking dimension. Mathematical models will test the impact of expanding access to care in prison and after release on outcomes in the community. That matters because the system does not need merely better description; it needs estimates of what scale of benefit might follow from increasing coverage. For administrators and policy makers, a linked cohort with this breadth can support decisions about where to place limited resources: inside prison health services, at the prison gate, or in post-release community care.
The inclusion of justice outcomes alongside health outcomes is also important. Incarceration is often treated as a separate domain from medicine, yet recurrent justice involvement can worsen housing insecurity, employment loss, and healthcare discontinuity, all of which feed back into health. A cohort that tracks both domains can show how recidivism and illness reinforce one another over time.
Why the cohort profile matters before the outcome papers arrive
This publication is a protocol and cohort profile rather than a definitive outcomes report, so it does not yet answer whether any particular intervention reduced death, psychiatric crisis, or re-incarceration. What it does establish is the infrastructure needed to examine those questions with population coverage across more than two decades.
The scale is the story here. More than 200,000 adults, more than 2.2 million person-years of observation, and linked records spanning custody, health, and mortality give POST the capacity to detect patterns that smaller studies would miss. Because the cohort includes women, First Nations people, and adults with contacts in both mental health and alcohol and other drug services, it should be able to describe whether the burden of poor outcomes is evenly distributed or concentrated in specific groups.
The strongest clinical implication is straightforward: incarceration should be treated as a high-risk health transition, not only a legal event. POST is designed to quantify that transition in real-world data and to test whether expanded treatment before and after release can change what happens next.