Subjective well-being and mortality in Chinese oldest old

The present study investigates the relationship between subjective well-being (SWB) and mortality risk, using a large sample (N=7852) from the Chinese Longitudinal Healthy Longevity Study (age range 80-105) conducted in 2000 and 2002. Initially, we intended to contribute to the understanding of system relations between SWB, mortality risk, and unobserved heterogeneity by treating SWB as an endogenous variable, using a multi-process model. However, failure to identify unobserved heterogeneity in the mortality equation prevents us from employing this model. Given this limitation, the study examines three issues. First, we argue that the mortality model with duration dependency on the age of the study subjects is specified and that the model with duration dependency on time since the interview is misspecified. Second, we address problems associated with the identification of unobserved heterogeneity in the mortality equation. Third, we examine the association between SWB and mortality risk in the Chinese oldest old as well as the risk pattern by gender, without considering unobserved heterogeneity. We find that SWB is not a significant predictor of mortality risk when we control for socio-demographic characteristics and health status. Health plays a very important role in the relationship between SWB and mortality risk in the oldest old. Gender differences in the predictive pattern of SWB on this risk are negligible in the sample.