Predicting Self-Related Health among the Aged

Abstract

Self-rated health among a sample of older persons is investigated via a simple path analysis model. The strongest predictor of self-rated health is a health index measuring reported conditions and confinement at home in bed. Income and education have small but significant direct effects on self-rated health suggesting that elderly of lower SES tend to evaluate their health as poorer than those of higher SES who are of similar health status (measured by the health index). Age, ethnicity, and sex affect self-ratings only indirectly through other variables.