Association for Public Policy Analysis & Management

Individuals involved with state and county correctional systems, both prisons and local jails, have high rates of infectious disease and chronic health conditions. Mental health and substance use disorders are also quite prevalent and are linked to high levels of re-incarceration. Despite these significant health needs, many lack insurance coverage and rely on emergency departments for care.  Due to policy silos between health care and criminal justice systems, limited information is available to examine the intersection between health delivery systems and justice-involved populations resulting in inefficiencies and higher costs.

In this study, we examine the overlap between frequent users of hospital emergency departments (ED) and correctional systems in California. The non-public hospital discharge data we use contain detailed information on the universe of ED encounters in California hospitals between 2009 and 2011 including patient demographics and clinical information for each visit. We link all ED records (outpatient and inpatient) using a unique linkage number to create a patient-level dataset. We identify patients who have direct contact with correctional systems based on whether the patient was admitted from or discharged to jail or prison.  We perform bi-variate and multi-variate analysis to profile frequent ED users with correctional system contact and compare them with other frequent users with no criminal justice involvement.

We find that nearly 22,000 non-elderly adults made at least one emergency department visit in 2011 and had contact with a correctional institution recorded on at least one hospital visit during the year. These patients made a total of 81,775 ED visits with the majority (57%) of visits occurring while individuals were not in custody. About one-quarter of patients with some correctional contact were classified as frequent ED users -- defined as four or more ED visits in the year -- compared to only 8% of all California ED patients. Although patients with correctional contact represented only 0.6% of all ED patients, they accounted for nearly 2% of frequent ED users and more than 4% of heavy ED users (more than 10 annual ED visits) in the state. They also have higher rates of serial frequent use, making more than 4 ED visits across multiple years, compared to other heavy ED users with no observed criminal justice involvement. Very few had private sources of insurance coverage with half of patients uninsured for all or some ED visits, and another quarter with Medicaid coverage. Finally, the presence of behavioral health issues was high; more than half of patients with correctional contact had a coded mental health condition and over one-third an alcohol or substance use condition. 

Our findings provide strong evidence of a cohort of adults whose use of healthcare and correctional systems significantly overlaps. This suggests contact with either system could potentially be put to better use. Targeting individuals for policy interventions, including extensive case management, health coverage, and connections to coordinated physical and behavioral health services, could contribute to reductions in both high ED use and re-arrest rates.