Global causes of blindness and distance vision impairment 1990-2020: a systematic review and meta-analysis.
Level 1 - systematic review of randomized trials
Systematic review and meta-analysis of population-based studies
PubMed 29032195 · doi:10.1016/S2214-109X(17)30393-5
What was done
MEDLINE, Embase, and WHO Library databases were systematically searched without language restrictions for population-based studies published before July 8, 2014, covering data from 1980 to 2014. From 288 studies encompassing 3,983,541 participants across 98 countries, regression models estimated the prevalence and causes of moderate or severe vision impairment (presenting visual acuity <6/18 to ≥3/60 in the better eye) and blindness (<3/60 in the better eye) by age, region, and year from 1990 to 2015, with forecasts to 2020.
What was found
In 2015, 216.6 million people (80% uncertainty interval [UI] 98.5–359.1 million) had moderate or severe vision impairment, led by uncorrected refractive error (116.3 million [80% UI 49.4–202.1 million]), cataract (52.6 million [18.2–109.6 million]), age-related macular degeneration (AMD; 8.4 million [0.9–29.5 million]), glaucoma (4.0 million [0.6–13.3 million]), and diabetic retinopathy (2.6 million [0.2–9.9 million]). Blindness affected 36.0 million people (80% UI 12.9–65.4 million) in 2015, primarily from cataract (12.6 million [3.4–28.7 million]), uncorrected refractive error (7.4 million [2.4–14.8 million]), and glaucoma (2.9 million [0.4–9.9 million]). In adults aged ≥50 years, cataract and uncorrected refractive error combined caused 77% of vision impairment and 55% of blindness. Projections for 2020 estimated 237.1 million people with vision impairment and 38.5 million people with blindness. Diabetic retinopathy (OR 2.52 [1.48–3.73]) and cataract (OR 1.21 [1.17–1.25]) were more prevalent causes among women, whereas glaucoma (OR 0.71 [0.57–0.86]) and corneal opacity (OR 0.54 [0.43–0.66]) were more common in men.
Why it matters
The vast majority of global blindness and vision impairment is avoidable with existing treatments like spectacles and cataract surgery. These estimates establish global public health priorities to scale eye care delivery for expanding and aging populations.
Limits
Data availability was uneven across countries, requiring statistical imputation for regions with sparse population-based surveys. The 80% uncertainty intervals were wide for several causes, and post-2015 numbers were model projections rather than empirical measurements. Source studies also spanned 34 years with potential heterogeneity in clinical diagnostic methods.