Abstract
Purpose :
Geographic atrophy (GA) is a vision-threatening form of age-related macular degeneration (AMD). We examined differences in health-related quality of life (QOL) before and after GA development in Age-Related Eye Disease Study (AREDS) participants.
Methods :
The AREDS (1992-1998) clinical trial evaluated efficacy of nutritional supplements on the risk of developing late AMD. A masked reading center assessed AMD status using annual color fundus photographs. Participants completed the National Eye Institute Visual Function Questionnaire-25 (NEI VFQ-25) annually starting in 1997; the composite score (AVTOT) represents a measure of vision-related QOL. We also used the method of successive dichotomizations to estimate person measures for the overall NEI VFQ-25 (M2C) and for two derived subscales: visual functioning (M2VF) and social-emotional functioning (M2SE). We used regression spline models with random effects for time and eye-within-participant (SAS 9.4) to compare the rate of change in QOL (difference in slope for each of the four QOL measures) before and after development of GA (with separate models for GA subtypes: central GA (CGA), non-central GA (NCGA), and any GA (anyGA)). Models were adjusted for age and visual acuity. If neovascular AMD developed after a GA outcome, we censored the subsequent observations.
Results :
(see Figure) Among AREDS participants with NEI VFQ-25 data available, 358 eyes (298 participants) developed any GA. None of the QOL measures differed significantly pre- and post-CGA. M2VF and AVTOT scores declined more quickly after NCGA (difference in slope (post minus pre): -0.09 logit/yr (95%CI: -0.17, -0.01), p=.03; -0.76 points/yr (95% CI: -1.45, -0.06), p=.03, respectively) and after anyGA (-0.08 logit/yr (95% CI: -0.15, -0.01), p=.02; -0.67 points/yr (95% CI: -1.28, -0.06), p=.03, respectively).
Conclusions :
We observed a trend for worsening QOL after development of NCGA and anyGA in AREDS participants across several QOL measures. Development of CGA was not associated with any significant changes in the QOL measures, possibly due to smaller sample size and limited power. Our findings highlight the importance of examining the relationship between GA subtypes and different indices of QOL.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.