Macular degeneration: 21 million worldwide with sight loss by 2050

The world has a new forecast for age-related macular degeneration: 21.34 million people with sight loss by 2050, against 8.06 million in 2021. Both come from the Global Burden of Disease 2021 analysis of AMD in The Lancet Global Health, and both are modelled estimates rather than counts. What it means for macular degeneration in the UK turns on a second paper, published this month.

That study, published in Ophthalmology Retina on 18 September, re-sorted the world's prevalence studies into one classification, Beckman, and reported the stages separately. Jeong and colleagues, mostly in Seoul, with L Smith at Anglia Ruskin in Cambridge. It is paywalled, so everything here comes from its abstract.

What the two studies did

Both are burden models, not trials, and neither examined a new patient. The GBD analysis used 163 rapid assessment surveys and 107 other population studies to model 204 countries from 1990 to 2021. It counted only moderate to severe vision loss and blindness, acuity worse than 6/18, and forecast to 2050 with an ensemble of 12 submodels. The Gates Foundation funded it, among others, with no role in the work.

The Ophthalmology Retina paper is a systematic review modelled with DisMod-MR. The original studies did not use Beckman, so an expectation-maximisation algorithm split the pooled prevalence into early, intermediate, geographic atrophy and neovascular disease. The two measure different things: sight loss caused by AMD, and the disease at any stage.

What was found

Vision impairment from AMD more than doubled between 1990 and 2021, from 3.64 million people to 8.06 million, and is projected to reach 21.34 million by 2050: 9.02 million males (95% uncertainty interval 5.72 to 14.20) and 12.32 million females (8.88 to 17.08). Rates per head went the other way. Global age-standardised prevalence of that impairment fell 5.53% over the period, western Europe's from 107.40 to 85.55 per 100,000, about 20%. The totals rise with ageing and population growth, not with severity.

The stage split matters more in clinic. In 2021 the Beckman model put global prevalence of AMD at any stage at 10.03%: early 5.40%, intermediate 4.01%, geographic atrophy 0.32% and neovascular AMD 0.31%. For every person at a late stage there are about six at the intermediate stage, with large drusen, acuity still reasonable and nothing licensed to offer them. The abstract states no age range, so those are whole-population figures, not the rate in a chair.

Why it matters for macular degeneration in the UK

The UK sits inside both models as one of 204 countries, fitted from a worldwide evidence base rather than from UK data. Our own study is older. Owen and colleagues, in the British Journal of Ophthalmology in 2012, applied prevalence from 31 population studies to ONS figures and estimated 513,000 people with late AMD, 2.4% of those aged 50 and over, rising to 4.8% at 65 and 12.2% at 80, roughly 1 in 8. They put 71,200 new cases a year on it and projected 679,000 by 2020, stopping there because data beyond that point may be less reliable.

The Royal College of Ophthalmologists' commissioning guidance has already decided who monitors them:

Do not routinely monitor people with early AMD or late AMD (dry) at HES unless in clinical research.

That guidance uses the NICE classification, which has no intermediate category, so the Beckman stages do not map onto it cleanly. It says those patients continue to attend routine sight tests with their primary care optometrist, and that OCT is the most sensitive monitoring tool, naming it for community provision in patients at high risk of new neovascular AMD or monitored for stable neovascular AMD. Denmark's community OCT programme is the other half of that arithmetic.

Three things follow. When wet AMD is suspected, speed is the whole job: NICE's quality standard expects treatment within 14 days of referral to the macular service, and the Royal College asks for acuity, fundoscopy findings and the OCT images in the letter. Second, the smoking conversation is for everyone: tobacco accounted for 18.72% of male disability-adjusted life years from AMD against 3.89% of female, a gap the paper puts down to how many men smoked, not to a different macula. Third, the European Medicines Agency confirmed its refusal of pegcetacoplan on 19 September 2024 and the MHRA rejected it here too, so the first candidate treatment for geographic atrophy never reached UK patients.

The caveats

• Both papers are models. The 21.34 million adds two point estimates whose 95% uncertainty intervals run 5.72 to 14.20 million males and 8.88 to 17.08 million females.
• They count different things: GBD counts sight loss from AMD, at acuity worse than 6/18, Beckman the disease at any stage.
• The Ophthalmology Retina paper is paywalled, so only its abstract could be read here, it gives no 2050 figure, and its stages are imputed, not observed.
• GBD modelled smoking alone as a risk factor, could not separate wet from dry disease without fundus photography, and says the burden may be underestimated.
• The UK estimate is from 2012, from studies in white populations, and assumes AMD is lifelong and mortality unchanged by it.

Sources

1. 1.Global, regional, and national burden of vision impairment due to age-related macular degeneration, 1990-2021, with forecasts to 2050 · The Lancet Global Health (GBD 2021 Global AMD Collaborators), 1 July 2025
2. 2.Global burden of age-related macular degeneration, with forecast up to 2050: a modelling study · Ophthalmology Retina, 18 September 2026
3. 3.The estimated prevalence and incidence of late stage age related macular degeneration in the UK · British Journal of Ophthalmology, 1 May 2012
4. 4.Commissioning guidance on age-related macular degeneration services: executive summary · Eye (Royal College of Ophthalmologists), 28 June 2022
5. 5.Serious eye disorders, quality statement 3: treatment of late AMD (wet active) · NICE quality standard QS180, 12 February 2019
6. 6.Syfovre (pegcetacoplan): refusal of marketing authorisation · European Medicines Agency, 19 September 2024
7. 7.Syfovre FAQs and next steps for dry AMD and geographic atrophy treatment · Macular Society, 1 February 2025