Self-reported health and subsequent mortality: An analysis of 767 deaths from a large Thai cohort study

Background: Few studies have examined the link between self-reported health (SRH) and subsequent mortality in developing countries, and very few considered mortality effects of changes in SRH. We examined the relationship between SRH and subsequent all cause or cause-specific mortality in Thailand. We also noted if mortality varied after people changed their SRH. Methods: We used longitudinal data including SRH from a nationwide Thai Cohort Study (baseline 2005 - follow-up 2009) and linked to official death records (2005-2012). Cox regression examined the association between SRH in 2005 and subsequent all-cause mortality or cause-specific mortality, with results given as confounder-adjusted hazard ratios (HR). We further assessed association between changes in SRH during 2005-2009 and mortality from 2009 to 2012. Results: Poor SRH at baseline independently relates strongly with subsequent cardiovascular disease (CVD) mortality (HR = 2.8, CI: 1.3-5.9) and "other" causes of death (HR = 1.9, CI: 1.1-3.3) but moderately with cancer mortality (HR = 1.4, CI: 0.7-3.0). SRH did not exhibit a relationship with injury mortality (HR = 1.0, CI: 0.5-2.1). Worsening SRH from 2005 to 2009 associated with increased mortality in 2009-2012 for females but not for males. Conclusions: In Thailand, SRH is a good predictor of population mortality due to internal causes (e.g. CVD). SRH is holistic, simple to measure and low cost; when repeated it measures dynamic health status. In many developing countries chronic diseases are emerging and morbidity information is limited. SRH could help monitor such transitions in burdens and trends of population health.