Concurrent and predictive validity of indicator scales developed for the Comprehensive Assessment and Referral Evaluation interview schedule.

Evidence for the concurrent and predictive validity of indicator scales developed to assess psychiatric, social, and medical conditions of elderly adults is presented. These scales were developed from the Comprehensive Assessment and Referral Evaluation (CARE) on probability samples of 445 elderly community residents in New York City and 396 in London, England. Corroborative information was also collected from key collaterals of a random subsample of 162 of the New York elderly adults. Concurrent validity of the cognitive impairment and activity limitation scales was tested in relation to family report of inconvenience and decision to institutionalize. Predictive validity of the scales was assessed using morbidity and mortality as outcome variables. Individuals classified as having medical and psychiatric disorders at Time 1 were significantly more likely to manifest such disorders 1 year later at Time 2 than were persons not so classified. In addition, the odds that individuals with cognitive impairment, somatic complaints, activity limitation, and difficulty ambulating at Time 1 would be dead within 1 year were two to three times greater than for those without the disorder, a result that provides support for the contention that these scales are useful predictors of outcomes.