Methods: The study draws from the State of Maryland Outcome Measurement System (OMS) data collected as part of service reauthorization. Data were collected between January 2006 and September 2009. The OMS is comprised of the BASIS, a reliable and valid measure designed to assess overall mental health functioning in community mental health clients, along with demographic, criminal justice, and housing questions. We included individuals with diagnoses of schizophrenia or other psychotic disorders whose first interview in the data was also an intake interview, resulting in a sample of 2881 consumers and 4882 surveys. We estimated random effect growth models and growth mixture models (GMM) to predict BASIS scores.
Results: The sample was 58% male and 62% African American. Average age at intake was 36 years old. The average respondent completed 2 surveys with 6 being the maximum. While there was extensive attrition, clinical severity at baseline was the best predictor of number of interviews. Growth models predicting BASIS scores suggest that individuals showed modest improvement over time. Negative correlations (r = -.61) between the random consumer intercept and time slope indicate that those with poorer mental health status at baseline had the largest improvement over time. Being male, older, or employed was significantly associated with better mental health status. Substance use or depression diagnoses, being arrested, or being homeless was significantly associated with poorer mental health status. There was a significant interaction between homelessness and time, suggesting that, compared with housed consumers, homeless consumers had more rapid improvement following intake. Similarly, the GMM found two distinct growth trajectory patterns. Trajectory class one was comprised of consumers with good mental health status at intake that were stable over time. Trajectory class two was comprised of consumers with high BASIS scores at intake, indicating poor mental health status, but who had rapid improvement over time.
Conclusions: Our findings suggest that consumers diagnosed with schizophrenia in the Medicaid financed community mental health system can be classified by one of two sets of outcome trajectories: low and stable or crisis followed by improvement. While attrition and regression to the mean are reasons to be cautious about these findings there is evidence that treatment in the community mental health system is associated with both stability and improved mental health status, especially among the most vulnerable populations such as homeless individuals.
