Disability and living arrangements among elderly American men.

The living arrangements of disabled elderly people are an important policy-related issue. This investigation focused on testing four models which posit alternative ways in which disability might affect coresidence with adult relatives. The models were tested for a pooled sample of elderly men drawn from the 1976 and 1981 rounds of the National Longitudinal Survey of Mature Men. The methodological strategies for model assessment included descriptive and logistic regression techniques. Men with multiple disabling conditions, but not a single disability, were more likely to be coresiding with adult relatives than were nondisabled men, independent of a set of selected background characteristics. These results provided general support for a model focused on assistance norms. There was some suggestion that the absence of a spouse reinforces this pattern.