Predictors of Case Management for Persons with Alzheimer's Disease

Abstract

This study examines case management service allocations to a clinical, nonrandom sample of persons with Alzheimer's disease (N = 822). Service allocation rules are estimated using logistic regression. The latent index of the likelihood of providing case management services is regressed on patients' enabling, need, and predisposing characteristics and organization affiliation. Significant predictors include patient unmanageability, memory problems, gender, previous case management use, no other service use, family problems, and center affiliation. Less functional patients may receive case management services when more functional patients may not, thus indicating a temporal dimension to case management allocation because people later in their disease trajectory are more likely to be directed toward case management services. This suggests that the efficacy of case management services may depend on the timing of the help in relation to the client's disease severity.