Purchase this article with an account.
L.A. Lott, M.E. Schneck, G. Haegerstrom–Portnoy, J.A. Brabyn; Predictors of Mortality in the Smith–Kettlewell Institute (SKI) Vision Study . Invest. Ophthalmol. Vis. Sci. 2006;47(13):3484.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Previous researchers have reported that visual impairment is a predictor of mortality in older populations. The purpose of this study was to determine which vision measures, if any, predict death in the SKI Vision Study sample.
A battery of vision function tests was administered on three occasions to the SKI Vision Study participants (mean age at baseline [BL]=78.4 years, range 58–102 years). The test battery included acuity under conditions of high and low contrast, reduced luminance, and disability glare, as well as contrast sensitivity, glare recovery, visual fields with and without an attentional task, stereopsis, and color. All testing was binocular with habitual correction. Visual impairment was defined as high contrast acuity 20/70 or worse, with identical log unit losses for other measures.
Of the 900 original participants, 27% died prior to the third test phase (mean time to death=3.8 years, range 0.08–8.81 years). Those who died were significantly older, had fewer years of education, more self–reported medical conditions, higher depression ratings at BL, and were more likely to be male. Those who died had significantly worse performance at BL than survivors on all vision measures (mean differences ranged from 0.16 logMAR for high contrast acuity to 0.34 logMAR for low contrast acuity in glare). Logistic regression analysis was used to determine which vision measures were significant independent predictors of death when potential confounders (age, sex, education, number of medical conditions, depressive symptoms) were taken into account. In this multivariate analysis, low contrast, low luminance acuity (impairment defined as >0.84 logMAR (Snellen equivalent 20/145 or worse) and >50% loss in attentional visual field were significant independent predictors of death. Of the spatial vision measures tested, low contrast, low luminance acuity had the highest independent association with mortality, but models substituting other spatial vision measures were also statistically significant.
In agreement with previous researchers, older people with visual impairment are more likely to die than those without impairment. Impairments in spatial vision and attentional field performance are predictors of mortality.
This PDF is available to Subscribers Only