摘要：Objectives. We assessed the association between mortality and disability and quantified the effect of disability-associated risk factors. Methods. We linked data from cross-sectional health surveys in the Netherlands to the population registry to create a large data set comprising baseline covariates and an indicator of death. We used Cox regression models to estimate the hazard ratio of disability on mortality. Results. Among men, the unadjusted hazard ratio for activities of daily living, mobility, or mild disability defined by the Organization for Economic Co-operation and Development at age 55 years was 7.85 (95% confidence interval [CI] = 4.36, 14.13), 5.21 (95% CI = 3.19, 8.51), and 1.87 (95% CI = 1.58, 2.22), respectively. People with disability in activities of daily living and mobility had a 10-year shorter life expectancy than nondisabled people had, of which 6 years could be explained by differences in lifestyle, sociodemographics, and major chronic diseases. Conclusions. Disabled people face a higher mortality risk than nondisabled people do. Although the difference can be explained by diseases and other risk factors for those with mild disability, we cannot rule out that more severe disabilities have an independent effect on mortality. Population aging is associated with an increase in the number of people who are disabled. This increase presents a challenge for society because elderly persons disabled in 1 or more domains of life are hospitalized more often, 1 need more medical and long-term care, 2 – 5 and face a higher risk of death than nondisabled persons do. 6 – 13 Disablement refers to the impact that chronic and acute conditions have on people's ability to perform tasks necessary for daily living and normal social functioning. 14 In a broader context, the disablement process is described as a causal chain in which the progression of disease leads to functional limitations, loss of mobility, and eventually to inability to perform activities of daily living (ADLs). 14 – 17 Empirical studies have found numerous risk factors associated with disablement. These factors are usually seen as risks that increase the chance of developing a disability. The major underlying causes are (acute and progressive) chronic diseases, 18 but other risk factors including sociodemographic factors (e.g., age, gender, 19 socioeconomic status 20 ), behavioral factors (e.g., smoking), 21 nutrition, 22 physical activity, 23 comorbidity, 18 self-rated health, 24 and cognitive impairment 25 are also associated with incident disability. Disability is most often assessed in cross-sectional studies without information on mortality. The few longitudinal studies that have been conducted tend to emphasize incident disability rather than the trajectory of disability following onset because of lack of statistical power. 26 Thus, although the onset of disability has been extensively researched, there has been far less investigation into the mortality risk associated with disability. In previous studies, the study populations were often limited to specific disease groups 9 , 12 or based on small sample sizes with few control variables. 6 – 8 , 10 , 11 , 13 Moreover, the focus was often on other determinants of mortality rather than on disability. Nonetheless, disability has been found to be an independent predictor of death after adjustment for heart disease, 9 depressive symptoms, 10 physical activity, 11 socioeconomic status, 13 or health status. 10 However, no study has assessed the extent to which the relationship between disability and mortality can be explained by risk factors known to be associated with disablement. Assessment of this relationship may enhance understanding of the public health aspects of aging. If disability is found to be independently associated with mortality, developing strategies to prevent disability would not only increase disability-free life expectancy but also total life expectancy. We assessed the association between mortality and 3 disability measures reflecting different levels of disability severity. The linking of cross-sectional health surveys to municipal health registries in the Netherlands permitted the compilation of a large time-to-event data set with covariates measured at baseline. 27 We quantified the magnitude of the association between disability and mortality, unadjusted and adjusted for groups of risk factors. These risk factors included distal and proximal risk factors that may influence the speed of disablement. 28 – 31 We used hazard ratios (HRs) and life expectancy to summarize the association between disability and mortality.
