Effects of self-rated health on sick leave, disability pension, hospital admissions and mortality : A population-based longitudinal study of nearly 15,000 observations among Swedish women and men

BACKGROUND: Simple global self-ratings of health (SRH) have become increasingly used in national and international public health monitoring, and in recent decades recommended as a standard part of health surveys. Monitoring developments in population health requires identification and use of health measures, valid in relation to targets for population health. The aim of the present study was to investigate associations between SRH and sick leave, disability pension, hospital admissions, and mortality, adjusted for effects of significant covariates, in a large population-based cohort. METHODS: The analyses were based on screening data from eight population-based cohorts in southern and central Sweden, and on official register data regarding sick-leave, disability pension, hospital admissions, and death, with little or no data loss. Sampling was performed 1973--2003. The study population consisted of 11,880 women and men, age 25--99 years, providing 14,470 observations. Information on SRH, socio-demographic data, lifestyle variables and somatic and psychological symptoms were obtained from questionnaires. RESULTS: There was a significant negative association between SRH and sick leave (Beta -13.2, p<0.0001, and -9.5, p<0.01, in women and men, respectively), disability pension (Hazard ratio 0.77, p<0.0001 and 0.76, p<0.0001, in women and men, respectively), and mortality, adjusted for covariates. SRH was also significantly associated with hospital admissions in men (Hazard ratio 0.87, p<0.0001), but not in women (Hazard ratio 0.96, p0.20). Associations between SRH on the one hand, and sick leave, disability pension, hospital admission, and mortality, on the other, were robust during the follow-up period. CONCLUSIONS: SRH had strong predictive validity in relation to use of social insurance facilities and health care services, and to mortality. Associations were strong and robust during follow-up