Client typology based on functioning level assessments: utility for service planning and monitoring.

Cluster analysis was used to develop a six-group typology based on level of functioning data from 2,447 clients with serious mental illness served by the Michigan public mental health system. The groups are described in terms of level of functioning in 16 domains, global functioning, diagnosis, demographic characteristics and services used. A group of older, poor functioning clients with high health needs was identified. A second group of highly dysfunctional clients with few health needs was also found. Two groups of young adult "chronic" clients were found: one with extensive substance abuse problems and one with moderate substance abuse problems and extreme levels of suicidal and aggressive behavior. Two groups of relatively good functioning clients were also identified: one with some problems in many areas and a second with functional problems in a number of life areas and pronounced symptoms of depression and demoralization. The results point to the utility of cluster analysis as a mechanism to organize and identify patterns within the rich array of information provided by multidimensional level of functioning assessments. The uses of empirically derived client types in planning and monitoring mental health services are discussed.