Abstract
Purpose: :
Previous research has documented that visual impairment (VI) is associated with reduced survival in adults. However, it is not clear is this association is a marker for frailty or other related conditions associated with the aging process.
Methods: :
The National Health Interview Survey (NHIS) is a population-based annual survey of the US non-institutionalized civilian population. Over 124,000 adults who participated in the 1997-2000 NHIS, and who provided sufficient information for record linkage with the National Death Index to 2002 were included in the analysis. Participants were asked, "Do you have any trouble seeing, even when wearing glasses or contact lenses?" (some VI), and "Are you blind or unable to see at all? (severe VI). A frailty indicator was created by summing responses to 12 questions measuring physical functioning and activities of daily living (e.g., "How difficult is it for you to walk a quarter mile?"). Item responses included: not at all difficult (0), only a little difficult (1), somewhat difficult (2), very difficult (3), and can’t do at all (4). Statistical methods included Cox regression models with adjustments for age, gender, race/ethnicity, smoking status, visual impairment status, and frailty.
Results: :
The prevalence of participants reporting no VI, some VI, and severe VI were 89.8%, 9.7%, and 0.5%, respectively. Frailty scores ranged from 0-48 with 2/3 of respondents receiving a score of 0. There were 5,840 deaths during the 3-5 year follow-up period. In the multivariable hazard model those mild/moderate (score 1-13) and severe frailty (14-48) scores at baseline were at significantly increased risk of death relative to those reporting no difficulties: hazard ratio=1.69; [95% confidence interval=1.55-1.85] and 4.21 [3.86-4.60], respectively. After controlling for covariates including frailty, both mild/moderate and severe visual impairment remained associated with a significantly increased risk of death in the same model (1.26 [1.16-1.36] and 1.31 [1.06-1.62], respectively).
Conclusions: :
Control for baseline frailty levels reduced, but did not eliminate the association between VI and survival in community-residing adults. Reduction of frailty levels and VI may lower mortality risk.
Keywords: clinical (human) or epidemiologic studies: prevalence/incidence • clinical (human) or epidemiologic studies: risk factor assessment