271P
Behavioral Health Service Use Patterns Among Individuals with Serious Mental Illness and Co-Occurring Substance Use Disorders after Release from Incarceration
Returning prisoners face a substantially elevated risk of death in the year after their release. Mental health services have been found to positively impact the community reentry process for people with mental illness leaving jails and prison. But much is yet to be learned about how people with serious mental illness engage behavioral health services post release and what factors influence these service usage patterns. This study addresses this gap by examining underlying patterns of service use among persons with SMI within a year after release from jail and the impact that substance use had on these patterns.
Method:
Jail records and behavioral health claims were merged in order to identify a cohort of 922 people with serious mental illness released from a large urban jail system between 2003-04. Individuals’ diagnosis (schizophrenia spectrum and major affective disorders) and service utilization patterns were established using the behavioral health claims records. Behavioral health service utilization were categorized as follows: (a) inpatient mental health services, (b) outpatient mental health services, (c) case management services, (d) drug/alcohol services, (e) residential services, and (f) other services. LCA was conducted to classify subtypes of service use patterns for the cohort after release. Model fit was evaluated by G2 statistics, AIC, BIC, and entropy. The study also conducted a multinomial logistic regression to examine whether the presence of a co-occurring substance use disorder impacted the service use patterns identified by the LCA after controlling for demographic characteristics (e.g., gender, age, race).
Results:
Three distinctive service use patterns were emerged among individuals with SMI: (a) no service use (68.6%), (b) drug/alcohol services and other service use (22.3%), and (c) mental health services (9.1%; inpatient, outpatient, or case management). A multinomial logistic regression found that individuals with co-occurring disorders were associated with increased odds of belonging to the drug/alcohol service use and other group as compared to the no service use group (OR=3.01) after controlling for demographic characteristics. However, having dual disorders was not related to belong to the mental health service use group as compared to the no service use group (OR=1.71).
Implication:
While this study showed three distinct service utilization patterns were present it this sample, it also found that the majority of persons (68.6%) with SMI did not receive any behavioral treatment in the year after their release. This raises questions about what is getting in the way of people with serious mental illness receiving treatment after release which requires further study. Additionally these analyses showed that individuals with dual disorders were more likely to receive substance use service only. This finding highlights again the need for more integrated services to person with SMI.