Purchase this article with an account.
Eva K Fenwick, Peng Guan Ong, Ryan EK Man, Charumathi Sabanayagam, Ching-Yu Cheng, Tien Yin Wong, Preeti Gupta, Ecosse Luc Lamoureux; Longitudinal impact of changes in visual acuity on vision-specific function. Invest. Ophthalmol. Vis. Sci. 2016;57(12):No Pagination Specified. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
There are limited data on the longitudinal association between visual acuity (VA) and vision-specific function (VF), especially in Asia. We examined the association in a population-based cohort of Singapore Malays.
1895 adults (862 (45.5%) males, aged 40-86 years) from the Singapore Malay Eye Study had clinical and VF data (VF-11 scale) at baseline (2004-06) and follow-up (2010-13). Best-corrected VA was measured using a LogMAR chart. Baseline and follow-up data of the VF-11 were validated using Rasch analysis and the stacking method. Mean differences and effect sizes (ES) for the composite VF-11 score and individual items were calculated for 3 categories of better eye VA change over the follow-up period (VA improved ≥2 lines; no change [-2<VA<2]; VA loss ≥2 lines). The VA loss group was stratified by incident vision impairment (VI) (baseline: VA ≥6/12 and follow-up: <6/12) and VI progression (VI worsening by ≥2 lines over time). The associations between VA and VF changes were assessed using multivariate linear regression models, with VA change analyzed continuously (per 2 line decrease) and categorically.
At follow-up, 405 (21.4%) participants had lost ≥2 lines of VA (mean VA loss 0.28 LogMAR, p<0.001), which was associated with a mean -0.81±2.07SD logit reduction in VF (moderate ES: -0.58). Greatest VF changes were for seeing stairs, watching TV and cooking (-0.95, -0.94 and -0.92 logit, respectively). Of the 405 subjects, 124 (30.6%) had incident VI and 281 (69.4%) had VI progression, which were associated with -1.44±2.43 (ES -0.96) and -0.63±1.82 (ES -0.40) logit reductions in VF, respectively. Participants with a ≥2 line gain in VA (n=31, 1.6%) and no VA change (n=1459, 77.0%) also had significant mean VF declines of -0.46±2.45 and -0.30±1.88 logits, although ES were small (-0.23). In adjusted multivariate models, a -0.31 (CI [confidence interval] -0.42, -0.21, p<0.01) logit reduction in VF was associated with a 2 line decrease in VA. Compared to those with no VA change, those with ≥2 line VA loss had 66% loss of VF (β=-0.22 CI -0.43, -0.02).
One in five adult Malays lost ≥2 lines of VA over 6 years resulting in a sizable deterioration in ability to perform vision-specific tasks. Loss of VF was markedly greater in those who developed VI compared to those whose VI worsened. Strategies to prevent and delay the progression of VI are warranted.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
This PDF is available to Subscribers Only