Cataract statistics¶
Last curated: 2026-08-31. Figures are source-specific. Populations, acuity thresholds, attribution rules and follow-up differ; conflicting estimates are shown side by side and are never averaged.
Global burden¶
| Figure | Estimate (95% interval where reported) | Population/year | Method | Source |
|---|---|---|---|---|
| Cataract blindness | 15.2 million (12.7–18.0) | Adults ≥50 years, 2020 | Hierarchical models from population surveys | GBD cause analysis, PMID 33275949 |
| Cataract moderate/severe vision impairment | 78.8 million (67.2–91.4) | Adults ≥50 years, 2020 | Same model | GBD cause analysis, PMID 33275949 |
| Cataract blindness | 17.0 million; 39.6% of all blindness | All ages, 2020 | Systematic review/meta-analysis and modelling | Vision Loss Expert Group, PMID 38461217 |
| Cataract moderate/severe vision impairment | 83.5 million; 28.3% of all MSVI | All ages, 2020 | Same model | Vision Loss Expert Group, PMID 38461217 |
| Female share of cataract blindness/MSVI | 60% / 59% | All ages, 2020 | Same model | Vision Loss Expert Group, PMID 38461217 |
| Change in cataract-blind count | +29.7% | Global, 1990–2020 | Modelled trend | Vision Loss Expert Group, PMID 38461217 |
| Change in cataract-related MSVI count | +93.1% | Global, 1990–2020 | Modelled trend | Vision Loss Expert Group, PMID 38461217 |
| Change in age-standardized cataract blindness prevalence | −27.5% | Global, 1990–2020 | Modelled trend | Vision Loss Expert Group, PMID 38461217 |
| Avoidable blindness prevalence change | −15.4% (−16.8 to −14.3) | Adults ≥50, 2010–2019 | Cataract + undercorrected refractive error model | GBD cause analysis, PMID 33275949 |
| Avoidable blindness case-count change | +10.8% (8.9–12.4) | Adults ≥50, 2010–2019 | Same model | GBD cause analysis, PMID 33275949 |
| Avoidable MSVI case-count change | +31.5% (30.0–33.1) | Adults ≥50, 2010–2019 | Same model | GBD cause analysis, PMID 33275949 |
| All-cause blindness | 43.3 million (37.6–48.4) | Global, 2020 | Systematic review and hierarchical model | GBD vision analysis, PMID 33275950 |
| Projected all-cause blindness | 61.0 million (52.9–69.3) | Global, 2050 | Forecast model | GBD vision analysis, PMID 33275950 |
| Cataract MSVI (earlier VLEG model) | 52.6 million (80% UI 18.2–109.6) in 2015; 57.1 million (17.9–124.1) projected 2020 | All ages | Global Vision Database regression models | Flaxman 2017, PMID 29032195 |
| Cataract blindness (earlier VLEG model) | 12.6 million (3.4–28.7) in 2015; 13.4 million (3.3–31.6) projected 2020 | All ages | Same model | Flaxman 2017, PMID 29032195 |
Why two 2020 cataract estimates differ. The 15.2-million estimate is restricted to adults aged 50 years or older; the 17.0-million estimate covers all ages and uses a later cataract-specific synthesis. They are not competing measurements of an identical denominator (PMIDs: 33275949, 38461217).
Access, coverage and inequality¶
| Indicator | Estimate | Population/year | Method | Source |
|---|---|---|---|---|
| Effective cataract surgical coverage by education | 31.0% in illiterate participants; 59.7% after class 10 education | 31 Indian districts | District surveys | Gupta 2024, PMID 38622863 |
| Typical cataract surgical coverage | Around 50% or lower in most countries reviewed | Global evidence to review date | Review of published coverage | Hashemi 2025, PMID 39638415 |
| Economic-estimate geography | 103/155 regional estimates (66%) from high-income countries | Vision-impairment economic literature | Systematic review | Marques 2022, PMID 35340626 |
| Cataract surgical affordability index | 17%–189% in developed; 29%–133% in developing countries | Multi-country | Cost-effectiveness/affordability synthesis | Lansingh 2007, PMID 17383730 |
| CSR association with development | CSR correlated with HDI and GDP per capita | 152 countries, longitudinal | Ecological analysis | Yan 2019, PMID 30362287 |
| Global eCSC (6/18 threshold) | 48.2% (39.7–57.2) in 2025; predicted +8.4 percentage points (8.1–8.6) 2020→2030, from 43.9% to 52.3% against a 30-point target | Adults ≥50; 233 datasets, 68 countries, 2003–24 | Mixed-effects logistic regression on population-based surveys | McCormick 2026, PMID 41687671 |
| eCSC country range | 2.1% (0.9–3.4) Burundi 2024 to 77.7% (72.9–82.5) Qatar 2023 | Adults ≥50 | Same | McCormick 2026, PMID 41687671 |
| Uncorrected refractive error share of non-good postoperative outcomes | Median 26.4% per survey; correcting it estimated to raise eCSC(6/12) by a median 3.7 percentage points | Same | Same | McCormick 2026, PMID 41687671 |
| eCSC by income stratum | High income 60.5% (IQR 55.6–65.4, n=2 surveys); low income 14.8% (IQR 8.3–20.7, n=14 surveys) | 148 RAAB surveys, 55 countries | Secondary analysis | McCormick 2022, PMID 36240806 |
| eCSC sex gap | Risk difference 3.2% (95% CI 2.3–4.1); risk ratio 1.20 (1.15–1.25) favouring men | Same 148 surveys | Pooled analysis | McCormick 2022, PMID 36240806 |
| eCSC relative quality gap | 10.8% (Argentina 2013: CSC 65.7%, eCSC 58.6%) to 73.4% (Guinea-Bissau 2010: CSC 14.3%, eCSC 3.8%) | Same | Same | McCormick 2022, PMID 36240806 |
| Effective cataract surgical coverage, India | eCSC 36.7% (33.6–39.9) vs CSC 57.3% (53.3–61.2); relative quality gap 36.0% | 31 districts, RAAB pooled 2015–19, ≥50 y | District surveys | Gupta 2024, PMID 38622863 |
| Cataract surgical coverage vs outcome, Hungary | CSC (VA<3/60) 90.0%; good visual outcome in only 79.5% of operated eyes; ocular comorbidity caused 78.1% of poor outcomes | 3,523 examined, ≥50 y | National RAAB | Sándor 2020, PMID 32309181 |
Coverage is threshold-dependent: conventional cataract surgical coverage counts surgery among people judged to need it, whereas effective coverage additionally requires a good visual outcome. Neither is equivalent to raw cataract surgical rate.
Intraoperative and postoperative complication rates¶
| Outcome | Estimate | Population/follow-up | Method | Source |
|---|---|---|---|---|
| Posterior capsule rupture | 31,749/2,853,376 (1.1%); annual range 0.60–1.65%, declining (P<.001) | EUREQUO 2008–2018 | Register-based cross-sectional | Segers 2022, PMID 34074994 |
| PCR risk factors | Corneal opacities OR 3.21 (3.02–3.41); diabetic retinopathy 2.74 (2.59–2.90); poor preoperative acuity 1.98 (1.88–2.07); white cataract 1.87 (1.72–2.03) | Same | Multivariate logistic regression | Segers 2022, PMID 34074994 |
| Outcomes after PCR | CDVA 0.13 ± 0.21 vs 0.05 ± 0.16 logMAR; absolute prediction error 1.15 ± 1.60 vs 0.41 ± 0.45 D; corneal oedema aOR 2.80 (2.27–3.45); endophthalmitis aOR 4.40 (2.48–7.81); uncontrolled IOP aOR 14.58 (11.16–19.06) | 12,196 PCR cases in 1,371,743 EUREQUO surgeries | Register-based | Segers 2022, PMID 35179858 |
| PCR and acuity loss | OR 5.74 for doubling of visual angle — the only modifiable adverse risk indicator | 55,567 UK operations, 406 surgeons | Multicentre electronic audit | Sparrow 2012, PMID 22441022 |
| Dropped nucleus | 1,221/1,715,348 (0.071%), decreasing over time | EUREQUO 2008–2018 | Register-based | Lundström 2020, PMID 32126043 |
| Cataract surgery after previous vitrectomy | 19,416/1,715,348 (1.1%); CDVA ≥0.5 in 82.8% vs 95.6%; absolute prediction error 0.52 vs 0.43 D | EUREQUO, 15 countries | Register-based | Lundström 2020, PMID 32649433 |
| Anaesthesia and PCR risk vs topical | Sub-Tenon OR 0.80 (0.71–0.91); regional 0.74 (0.71–0.78); general 0.53 (0.50–0.56); intracameral 0.76 (0.64–0.90). Endophthalmitis with regional vs topical OR 0.60 (0.44–0.82) | 1,354,036 EUREQUO surgeries | Register-based, multivariate | Segers 2022, PMID 36449673 |
| Late in-the-bag IOL dislocation | Cumulative risk 0.1% at 5 and 10 y, 0.2% at 15 y, 0.7% at 20 y, 1.7% at 25 y | 14,471 extractions in 9,577 residents, Olmsted County 1980–2009 | Population-based cohort with nested case-control | Pueringer 2011, PMID 21683329 |
| Positive dysphotopsia | Up to 67% immediately; 2.2% persisting at 1 year; surgery indicated in 0.07% | Review | Narrative synthesis | Pusnik 2022, PMID 36676002 |
| Negative dysphotopsia | Up to 26% early; 0.13–3% persisting at 1 year | Same | Same | Pusnik 2022, PMID 36676002 |
PCR is the pivot: it is the commonest serious intraoperative event, the strongest modifiable predictor of acuity loss, and the multiplier for endophthalmitis, corneal oedema and uncontrolled IOP (PMIDs: 34074994, 35179858, 22441022).
Endophthalmitis and prophylaxis¶
| Outcome | Estimate | Population/follow-up | Method | Source |
|---|---|---|---|---|
| Total postoperative endophthalmitis | 29/16,603; 20 proven infective | Multicentre cataract surgery trial | Randomized 2×2 factorial | ESCRS 2007, PMID 17531690 |
| Risk without intracameral cefuroxime | OR 4.92 (1.87–12.9) | Same trial | Multivariable analysis | ESCRS 2007, PMID 17531690 |
| Risk with clear-corneal vs scleral-tunnel incision | OR 5.88 (1.34–25.9) | Same trial | Multivariable analysis | ESCRS 2007, PMID 17531690 |
| Risk with surgical complication | OR 4.95 (1.68–14.6) | Same trial | Multivariable analysis | ESCRS 2007, PMID 17531690 |
| Pooled postoperative endophthalmitis | 4,502/6,809,732 eyes (0.066%) | 51 studies | Network meta-analysis | Kato 2022, PMID 36258003 |
| Intracameral route | OR 0.19 (99.4% CI 0.12–0.30) | Same network | Route sensitivity analysis | Kato 2022, PMID 36258003 |
| Weighted incidence with cefuroxime | 0.0332% | Comparative prophylaxis studies | Meta-analysis | Bowen 2018, PMID 29326317 |
| Weighted incidence with moxifloxacin | 0.0153% | Same synthesis | Meta-analysis | Bowen 2018, PMID 29326317 |
| Weighted incidence with vancomycin | 0.0106% | Same synthesis | Meta-analysis, predominantly non-randomized | Bowen 2018, PMID 29326317 |
| Endophthalmitis with vs without postoperative topical antibiotic after intracameral prophylaxis | 0.016% vs 0.017% | Large service study | Observational comparison | Rathi 2020, PMID 33120637 |
The antibiotic rankings above are not equally certain: cefuroxime has direct randomized evidence; agent-to-agent rankings are heavily influenced by observational data and must not be read as a randomized league table (PMIDs: 17531690, 36258003).
Posterior capsule opacification and retinal/macular outcomes¶
| Outcome | Estimate | Population/follow-up | Method | Source |
|---|---|---|---|---|
| Any early PCO | 29.93% | 1,039 eyes, 3 months | Prospective imaging cohort | Gu 2022, PMID 34727350 |
| Grade 3–4 early PCO | 2.98% | Same cohort | Graded retroillumination imaging | Gu 2022, PMID 34727350 |
| PCO after prior vitrectomy | OR 2.664 | Same cohort | Multivariable model | Gu 2022, PMID 34727350 |
| PCO with <180° capsulorhexis–IOL overlap | OR 5.403 | Same cohort | Multivariable model | Gu 2022, PMID 34727350 |
| Nd:YAG at 1 year, sharp vs round edge | OR 0.30 (0.05–1.74) | 742 eyes in 6 studies | Cochrane meta-analysis | Maedel 2021, PMID 34398965 |
| Nd:YAG at 3 years, sharp vs round edge | RR 0.21 (0.11–0.41) | 538 eyes in 6 studies | Cochrane meta-analysis | Maedel 2021, PMID 34398965 |
| Nd:YAG at 5 years, sharp vs round edge | RR 0.21 (0.10–0.45) | 306 eyes in 4 studies | Cochrane meta-analysis | Maedel 2021, PMID 34398965 |
| Retinal detachment after cataract surgery | 36,886/5,480,448; 0.66 per 100 patients | Pooled surgical literature | Systematic review/meta-analysis | Alshammari 2024, PMID 39172224 |
| Clinical pseudophakic CME | 0.1%–2.35% | Modern cataract surgery literature | Narrative synthesis | Zur 2017, PMID 28351047 |
| Pooled PCO incidence | 11.8% (9.3–14.3) at 1 y; 20.7% (16.6–24.9) at 3 y; 28.4% (18.4–38.4) at 5 y | ECCE and phaco with posterior-chamber IOL, pre-sharp-edge era | Meta-analysis; significant heterogeneity | Schaumberg 1998, PMID 9663224 |
| Nd:YAG at 2 years, sharp vs round edge | RR 0.35 (0.16–0.80); 89 fewer cases per 1,000 | 703 eyes in 6 studies | Cochrane meta-analysis | Maedel 2021, PMID 34398965 |
| Sharp vs round edge (older review) | PCO score −8.65 (−10.72 to −6.59) on 0–100 scale; Nd:YAG rate 0.19 (0.11–0.35) | 66 studies | Cochrane meta-analysis | Findl 2010, PMID 20166069 |
| Nd:YAG rate by IOL material | Acrylic vs PMMA −24% (−29 to −20); silicone vs PMMA −9% (−17 to −1); hydrogel vs acrylic +19% (8–30); hydrogel vs silicone +28% (10–46); silicone vs acrylic 4% (−2 to 10) | 23 RCTs | Meta-analysis of risk differences | Cheng 2007, PMID 17224119 |
| Steroid vs NSAID monotherapy and Nd:YAG | HR 0.70 (0.52–0.88, P=.001) favouring steroid after adjustment; combination vs steroid alone HR 1.11 (0.68–1.80) | 13,368 analysed patients, mean follow-up 22.8 ± 15.7 months | Retrospective registry cohort | Hecht 2020, PMID 32061757 |
PCO presence, visually significant PCO and Nd:YAG capsulotomy are different endpoints. Retinal-detachment estimates depend strongly on age, axial length, posterior capsule status and duration of follow-up.
Technique, refractive and paediatric outcomes¶
| Outcome | Estimate | Population/follow-up | Method | Source |
|---|---|---|---|---|
| FLACS vs phaco uncorrected distance acuity | Difference −0.01 logMAR (−0.05 to 0.03) | FACT trial, 3 months | Randomized non-inferiority trial | Day 2020, PMID 32386810 |
| FLACS vs phaco composite surgical success | 41.1% (289/704 eyes) vs 43.6% (299/685 eyes); adjusted OR 0.85 (95% CI 0.64–1.12), p=0.250 | FEMCAT, 907 randomised patients (1,476 eyes), 870 analysed, 3 months | Multicentre participant-masked randomised superiority trial with sham laser | Schweitzer 2020, PMID 31954466 |
| FLACS vs phaco incremental cost-effectiveness | €10,703 saved per additional treatment success with conventional phacoemulsification | Same trial | Trial-based economic analysis | Schweitzer 2020, PMID 31954466 |
| Immediate vs delayed bilateral surgery within ±1.0 D | 97% vs 98%; difference −1% (90% CI −3 to 1) | 865 randomized participants | Multicentre non-inferiority trial | Spekreijse 2023, PMID 37201546 |
| Societal cost, immediate vs delayed bilateral surgery | €403 lower per participant | Same trial | Trial-based economic analysis | Spekreijse 2023, PMID 37201546 |
| Infant IOL vs aphakia median acuity at 10.5 years | 0.89 vs 0.86 logMAR; P=0.82 | 110/114 randomized infants assessed | Randomized follow-up | Lambert 2020, PMID 32077909 |
| Good treated-eye acuity in infant trial | 27/110 (25%) achieved ≤0.30 logMAR | Age 10.5 years | Randomized follow-up | Lambert 2020, PMID 32077909 |
| Poor treated-eye acuity in infant trial | 50/110 (44%) had ≥1.00 logMAR | Age 10.5 years | Randomized follow-up | Lambert 2020, PMID 32077909 |
| Additional intraocular surgery at 1 year, infant IOL vs aphakia | 63% vs 12%; P<0.001 | 114 infants | Randomized trial | Lambert 2010, PMID 20457949 |
| Adverse events by 5 years, infant IOL vs aphakia | 81% vs 56%; P=0.008 | 114 infants | Randomized follow-up | Plager 2014, PMID 25077835 |
| Additional intraocular surgery by 5 years, infant IOL vs aphakia | 72% vs 16%; P<0.0001 | 114 infants | Randomized follow-up | Plager 2014, PMID 25077835 |
| Short-eye formula evidence base | 15 studies; 2,395 eyes; 11 formulas | Eyes with short axial length | Systematic review/meta-analysis | Shrivastava 2022, PMID 35225507 |
| Long-eye formula evidence base | 11 studies; 4,047 eyes | Axial length >24.5 mm | Systematic review/meta-analysis | Wang 2018, PMID 29498180 |
Patient-important outcomes¶
| Outcome | Estimate | Population/follow-up | Method | Source |
|---|---|---|---|---|
| Falls before first-eye surgery | 1.17 (0.95–1.43) per person-year | Older adults awaiting surgery | Longitudinal cohort | Keay 2022, PMID 35702892 |
| Catquest-9SF validation evidence | Multiple language/population validations located | Cataract populations | Systematic review | Kabanovski 2020, PMID 31862206 |
| Patient experience after surgery | High overall satisfaction with variable negative experiences | Postoperative participants | Qualitative study | Webber 2020, PMID 32654259 |
| Falls after first-eye surgery | 0.81 (0.63–1.04) per person-year | Same cohort | Longitudinal cohort | Keay 2022, PMID 35702892 |
| Expedited vs delayed second-eye surgery, fall rate | Rate ratio 0.68 (0.39–1.19), P=0.18 — not significant | 239 women >70, 12-month follow-up | Randomised controlled trial | Foss 2006, PMID 16364936 |
| Toric vs non-toric IOL | Uncorrected distance acuity mean difference −0.07 logMAR (−0.10 to −0.04); spectacle independence RR 0.51 | 13 RCTs, 707 vs 706 eyes | Meta-analysis, GRADE high | Kessel 2016, PMID 26601819 |
| Toric IOL rotational stability | Pooled mean absolute rotation 2.36° (2.08–2.64) | 51 studies, 4,863 eyes | Single-arm meta-analysis | Li 2024, PMID 38768060 |
| Trifocal vs monofocal, uncorrected near acuity | MD −0.32 logMAR (95% CrI −0.46 to −0.19) | 27 RCTs, 2,605 patients | Bayesian network meta-analysis | Cho 2022, PMID 36136323 |
| Refractive accuracy benchmark | 71% of eyes within ±0.5 D and 92–93% within ±1.0 D of target, both arms | FACT trial, 785 patients, 3 months | Randomised trial | Day 2020, PMID 32386810 |
Known conflicts and caveats¶
- Cataract blindness depends on presenting-acuity threshold, attribution rule and whether one or both eyes define the person-level outcome.
- Age-standardized prevalence can fall while absolute case counts rise with population growth and ageing.
- Person-level and eye-level denominators cannot be interchanged.
- PCO incidence is not interchangeable with visually significant PCO or Nd:YAG capsulotomy.
- Registry complication rates reflect case mix, coding, follow-up completeness and surgeon/service structure.
- Refractive accuracy must state formula, lens constants, axial-length range and percentage within a prespecified dioptre band.
- Patient-reported improvement and acuity improvement measure different outcome domains.
- Network meta-analysis rankings (SUCRA) are frequently reported where all pairwise comparisons are non-significant; a ranking is not a difference.
- Registry associations (anaesthesia technique, surgeon grade) are confounded by case mix and are not causal estimates.
Provenance¶
Every figure on this page was re-verified against live PubMed records again on 2026-09-01. No further numerical error was found. The prior audit correction is retained for provenance: the FEMCAT surgical-success row once read "96.5% vs 96.3%"; the trial's composite primary endpoint was met by 41.1% versus 43.6% of eyes (adjusted OR 0.85, 95% CI 0.64–1.12, p=0.250) (PMID 31954466).