Service-planning issues necessary for successful implementation of an indirect needs-assessment model to estimate geographical differences in the prevalence of needs for alcohol, drug abuse, and mental health services are discussed. The issues include selection of need categories as target groups, estimation model selection, and understanding…

A summary is provided of a large-scale needs assessment research project in Colorado that tested the validity of several statistical procedures or models for indirectly estimating the need for alcohol, drug abuse, and mental health services across geographic subregions using a Colorado Social Health Survey sample of 4,745 adults. (SLD)

A stratified probability sample of 4,745 adults in Colorado was surveyed to determine the need for alcohol, drug abuse, and mental health services. Findings that variation in need prevalence across 48 subareas of the state substantially challenge the assumptions of a uniform rate of need in a state. (SLD)

Three methods of designating everyday-dysfunction cases in the Colorado Social Health Survey are compared for a sample of 4,745 adults. The clinical judgment approach was chosen as the primary method of identification, with the statistical deviation criterion a backup. The patient criterion group approach was less useful. (SLD)

Six existing statistical procedures (models) for indirectly assessing the prevalence of need for alcohol, drug abuse, and mental health services across subareas of a state are described and critiqued in terms of epidemiological and statistical characteristics. Specific advantages and disadvantages are listed for each model. (SLD)