Abstract
Purpose: :
Although previous studies suggest visual impairment is associated with increased mortality, it is unclear whether visual impairment is an independent, biological marker of ageing, or whether this association is linked to other predictors of mortality. We aimed to investigate whether the presence of specific mortality markers in older persons with visual impairment explained this association.
Methods: :
At baseline 3654 baseline participants aged 49+ were examined in the Blue Mountains Eye Study (1992-4). Mortality and causes of death were obtained by data linkage with the Australian National Death Index and age-standardised mortality rates calculated. Cox regression models were used to estimate risk ratios (RR) and 95% confidence intervals (CI).
Results: :
As of 31/12/2005, there were 1273 deaths. Compared to persons without visual impairment, age- and sex- adjusted all-cause mortality was higher among visually impaired participants, defined either using presenting all ages (RR 1.49, 95% CI 1.29-1.73) or best corrected visual acuity (RR 1.43, 95% CI 1.15-1.78). This association was somewhat stronger in persons aged <75 years (RR 1.69, 95% CI 1.32-2.17 and RR 2.48; 95% CI 1.36-4.50, p for age interaction 0.09 and 0.03 respectively). Visually impaired participants were more likely to be hypertensive, diabetic, overweight or obese, and to report a past history of myocardial infarction or stroke, but were less likely to be smokers. They were also more likely to have walking disability, a history of falls or cancer and low self rated health at baseline. However, after adjusting for age, sex and these associated factors, presenting visual impairment remained a significant predictor of increased mortality (RR 1.32, 95% CI 1.13- 1.55), and was again most evident in participants aged <75 years when visual impairment was defined by best-corrected visual acuity (RR 2.08; 95% CI 1.10-3.96). Analysis of persons with none of these mortality markers confirmed that visually impaired participants had a higher mortality risk (RR 1.34; 95% CI 1.15-1.55).
Conclusions: :
These population-based data indicate that visual impairment independently predicts mortality, particularly among persons <75 years. Presence of co-morbid cardiovascular risk factors in the older population does not explain the association between visual impairment and premature mortality.
Keywords: clinical (human) or epidemiologic studies: risk factor assessment • aging