|| Checking for direct PDF access through Ovid
This paper investigates whether self-rated health (SRH) covaries with individual hospital records. By linking the Danish Longitudinal Survey on Ageing with individual hospital records covering all hospital admissions from 1995 to 2006, I show that SRH is correlated to historical, current, and future hospital records. I use both measures separately to control for health in a regression of mortality on wealth. Using only historical and current hospitalization controls for health yields the common result that SRH is a stronger predictor of mortality than objective health measures. The addition of future hospitalizations as controls shows that the estimated gradient on wealth is similar to one in which SRH is the control. The results suggest that with a sufficiently long time series of individual records, objective health measures can predict mortality to the same extent as global self-rated measures. Copyright © 2015 John Wiley & Sons, Ltd.