Abstract
Existing studies on the association between unemployment and health indices suffer reciprocal causality bias. Existing studies do not demonstrate that unemployment results in poor health rather than vice versa. This study avoids the reciprocal causality bias by measuring disability as an incidence rate and using measures of unemployment prior to the onset of the disability. Evidence from a large national longitudinal data set is presented that suggests that an individual's unemployment is useful in predicting subsequent disability. Aggregate countywide measures of unemployment, on the other hand, do not help predict an individual's probability of becoming disabled.

This publication has 16 references indexed in Scilit: