Psychiatric hospitalizations, arrests, emergency room visits, and homelessness of clients with serious and persistent mental illness: findings from a randomized trial of two ACT programs vs. usual care.

OBJECTIVE: This investigation examined several adverse outcomes in clients with serious mental illness in a randomized trial of Assertive Community Treatment (ACT) versus usual care.

METHOD: 163 subjects were randomized to one of two ACT experimental conditions (staffed by consumers or non-consumers) or usual community care. Conditions were compared on psychiatric hospitalization, emergency room visit, arrest, and homelessness, within the two-year study period. Demographic, program, and client variables were examined for significant associations with outcomes.

RESULTS: Significant differences were found between ACT and usual care in time to first arrest, but not hospitalization, homelessness or ER visits. Shorter time to first hospitalization was associated with male gender, diagnoses other than schizophrenia, high psychiatric symptomatology and lower provider case load. ER visits were associated with increased client symptomatology. Shorter times to homelessness were predicted by poorer therapeutic alliance between case manager and clients. Shorter time to first arrest was predicted by client minority status and enrollment in usual care.

CONCLUSIONS: The paucity of significant main effects may have been due to a prolonged "start-up" phase of the ACT programs, poor ACT implementation, restricted availability of psychiatric hospital beds, or changes in usual care services delivered over the study period.