Trends in the Burden of Vision Impairment in Saudi Arabia, 1990-2023: A Global Burden of Disease Study Analysis.

Abstract

Background: Vision impairment (VI) in Saudi Arabia is common, largely preventable or treatable, and increasingly linked to chronic disease rather than solely to the causes that have historically dominated blindness. Recent Saudi-specific evidence places adult VI prevalence at approximately 17% and blindness at under 2%, with diabetic retinopathy affecting roughly one in three Saudi adults with diabetes; however, no study has yet examined how this burden has trended nationally over time using the full 1990-2023 window made available by the Global Burden of Disease (GBD) 2023 round. This study quantified trends in the burden of VI in Saudi Arabia by severity, cause, sex, and age, and characterised the contribution of population ageing to the observed pattern. Methods: This repeated cross-sectional trend analysis used GBD 2023 national-level modelled estimates for Saudi Arabia (1990-2023). The estimated annual percentage change (EAPC) was calculated via log-linear regression on age-standardised prevalence and years-lived-with-disability (YLD) rates. A Durbin-Watson check confirmed autocorrelation in every annual series, and because GBD point estimates carry their own reported uncertainty, EAPC confidence intervals were derived two further ways alongside the naive ordinary-least-squares (OLS) interval: a Newey-West heteroskedasticity- and autocorrelation-consistent (HAC) interval, and a Monte Carlo interval that resampled each annual estimate within its GBD-reported 95% uncertainty interval. A trend is reported here as confirmed only where all three methods agreed on direction, and as suggestive where they did not. Segmented-regression sensitivity checks, cause decomposition, and comparison of age-standardised against crude rates were used to characterise breakpoints and the contribution of ageing. Results: Age-standardised blindness and severe VI declined substantially and robustly (blindness prevalence EAPC = -2.60%/year, confirmed decreasing in females and males separately, both p < 0.05 on a formal sex-difference test), while moderate VI rose robustly overall and among males specifically (EAPC = +0.61%/year); the female-specific rise in moderate VI, and the decline in presbyopia, were directionally consistent but not robust to the choice of uncertainty method. Diabetes-attributable rates increased faster than any other tracked cause and did so robustly under every uncertainty specification (crude prevalence EAPC ≈ +4.47%/year), and males carried a disproportionate, formally sex-different share of diabetes-attributable blindness relative to females; these tracked causes covered only 44-51% of the total VI envelope, so cause-specific conclusions are restricted to the causes examined. The largest proportional increases in case counts were concentrated in working-age adults (30-59 years) rather than older adults, and the gap between crude and age-standardised trends is consistent with an important contribution from Saudi Arabia's changing age structure, although the study design does not support assigning a specific numerical fraction to that contribution. Segmented-regression analysis further indicated that the national decline in blindness and the rise in moderate VI were concentrated in 1990-2019 and 1990-2008 respectively, with the post-breakpoint segments no longer showing a statistically distinguishable trend. Conclusions: Saudi Arabia's VI burden is shifting in composition rather than simply declining. Diabetes-related causes, sex differences, and working-age adults emerge as priorities for vision-health policy, and the post-breakpoint plateaus identified here warrant continued monitoring in future GBD rounds.