- What doubles is new cases a year: an estimated 11.4 million people developed dementia in 2023, projected to rise to 24.7 million a year by 2050, in a JAMA Neurology meta-analysis of 38,418 people in 22 cohorts across 18 countries and territories
- The rise comes mainly from ageing populations. The projection assumes the rate of dementia at each age stays constant, so the extra cases come from more, and older, people
- Women's risk was higher even after adjusting for age: 10.6 new cases per 1,000 person-years against 8.6 in men. The study doesn't explain why
- The 2024 Lancet Commission linked about 45% of dementia to 14 modifiable risk factors, led by hearing loss and high LDL cholesterol (7% each). These are population estimates from mostly observational data, not a promise for any individual
- Trial evidence is thinner: intensive blood pressure control cut mild cognitive impairment, lifestyle programmes nudged cognition up, and hearing aids helped only a higher-risk subgroup
This week's headlines were blunt: dementia is set to double worldwide by 2050, and women are more at risk. Both points come from a real study, published in JAMA Neurology on 12 October 2026. But the headline hides two things that matter more to anyone in their 40s, 50s or 60s.
First, the doubling is mostly about a larger and older population, not about dementia getting more common at any given age. Second, a large share of dementia is linked to things that can be changed, several of them in midlife. Here's what the evidence says, including where it's weaker than it sounds. It isn't medical advice: talk to your doctor about your own risks.
What the New Study Actually Found
Evidence: Strong for rates — projection rests on an assumptionThe study comes from COSMIC (Cohort Studies of Memory in an International Consortium), led by the Centre for Healthy Brain Ageing at UNSW Sydney. Lead author Ashleigh Vella and colleagues pooled individual-level data from 22 population-based studies that follow people over time, covering 38,418 participants in 18 countries and territories on six continents, 6 of them low- or middle-income and 12 high-income. Over 287,632 person-years of follow-up, 3,001 people developed dementia (Vella, JAMA Neurol 2026; UNSW).
The headline rate: after standardising for age and sex, 9.7 people per 1,000 developed dementia each year (95% CI 7.89 to 11.97). Rates were higher in low- and middle-income countries (13.6 per 1,000) than in high-income countries (8.3). UNSW reports that an estimated 11.4 million people developed dementia in 2023. Applying the age- and sex-specific rates to the projected 2050 population gives 24.7 million new cases a year, "assuming constant incidence over time", as the abstract puts it.
That assumption is the key. The rate of dementia at each age is held fixed, so the doubling in this projection comes from the world's population growing and getting older, not from rising rates at any given age. An earlier Global Burden of Disease analysis reached the same conclusion about the number of people living with dementia. It projected a rise from 57.4 million in 2019 to 152.8 million in 2050, while the age-standardised prevalence "remained stable". Its authors said the increase "could largely be attributed to population growth and population ageing" (GBD 2019 Dementia Forecasting Collaborators, Lancet Public Health 2022).
So when you read "dementia will double", the more precise version is: about twice as many people will develop dementia each year, mainly because many more people will reach their 80s and 90s. For you, what matters is your own risk, and that you can influence.
Why women face more risk
More women than men live with dementia. The Global Burden of Disease analysis put the female-to-male ratio of people living with it at 1.69 in 2019, and 1.67 projected for 2050. Raw counts like these partly reflect who lives longest. But the new study's rates were already adjusted for age, and women still had higher incidence: 10.6 new cases per 1,000 person-years (95% CI 8.61 to 13.11) against 8.6 in men (6.94 to 10.72).
UNSW's summary says the study "does not establish why" incidence was higher among women or in lower-income countries, and that the differences may reflect "a complex combination of health, social, economic, educational and environmental factors" (UNSW).
New dementia cases per 1,000 person-years, by group
Vella et al., JAMA Neurology 2026. Individual-participant meta-analysis of 22 population-based cohorts (38,418 people, 18 countries or territories). Rates by sex are age-standardised; the global rate is age- and sex-standardised; income-group rates are standardised as reported in the abstract.
The 14 Risk Factors You Can Change
Evidence: Moderate — mostly observational, some trialsThe more hopeful half of the story comes from the Lancet Commission on dementia prevention, intervention and care, a panel of 27 dementia experts led by Gill Livingston of University College London. Its 2024 report, published on 31 July 2024, concluded that "nearly half of dementias could theoretically be prevented" by eliminating 14 risk factors (Livingston, Lancet 2024; UCL). The figure usually quoted is 45%.
Twelve of the factors were already in the Commission's 2020 report, where together they were linked to about 40% of cases: less education, hearing loss, high blood pressure, smoking, obesity, depression, physical inactivity, diabetes, excessive alcohol, traumatic brain injury, air pollution and social isolation. The 2024 update added two:
- High LDL cholesterol in midlife, linked to 7% of cases. The Commission cites a pooled analysis of three UK cohorts in which each 1 mmol/L increase in LDL was associated with an 8% higher rate of dementia.
- Untreated vision loss in later life, linked to 2% of cases.
The factors are grouped by life stage. The Commission defines midlife broadly, as roughly 18 to 65, with later life above 65. Less education in early life accounts for 5%. Midlife factors add up to 30%, led by hearing loss and high LDL (7% each). Later-life factors add 10%, led by social isolation (5%).
Share of dementia linked to each modifiable risk factor
Livingston et al., Lancet 2024, as presented in the Commission's summary graphic (via NIHR Dementia Researcher) and by Diabetes on the Net and the Goodfellow Unit. Bars are population attributable fractions: the estimated share of dementia cases that would not occur if the risk factor were removed, assuming the link is causal. Gold bars were added in 2024. Total: 45%.
What "45%" does and doesn't mean
A population attributable fraction combines how strongly a factor is linked to dementia with how common it is. Hearing loss scores high partly because so many people have it. It describes a population, not you.
- It assumes cause and effect. The Lancet's press release states that "the prevention estimates assume there is a causal relationship between risk factors and dementia" (EurekAlert). Most of the underlying evidence comes from observational studies.
- Some links may run backwards. UCL notes that "some associations may only be partly causal", and that depression in late life "may be caused by dementia". The Commission now counts depression as a midlife risk factor, where the risk is clearer.
- It's mostly rich-country data. UCL says "nearly all the evidence for dementia still comes from high-income countries", yet the new study found the highest rates outside them.
- No guarantee. Risk reduction "affects the population, and does not guarantee that any individual will avoid dementia."
What Happens When You Test It
Evidence: Mixed — modest effects, often on secondary outcomesDementia develops over decades, so trials usually measure cognitive scores or mild cognitive impairment (MCI) over a few years, not dementia itself.
Hearing aids: ACHIEVE
ACHIEVE, published in The Lancet in July 2023, randomised 977 adults aged 70 to 84 with untreated hearing loss at four US sites. Half got audiological counselling and hearing aids; half got a health-education programme on healthy ageing. Over three years, cognition declined almost identically in both groups: −0.200 vs −0.202 standard deviations, a difference of 0.002 (p=0.96) (Lin, Lancet 2023).
The trial drew from two groups. In 238 people recruited from an existing long-term study (ARIC), a group the authors describe as at increased risk of cognitive decline, the hearing intervention was linked to a 48% reduction in 3-year cognitive change (difference 0.191 SD, 95% CI 0.022 to 0.360). In the other 739, there was no significant difference. The authors' summary: hearing intervention "did not reduce 3-year cognitive decline" overall but may help older adults at increased risk. That's a subgroup finding, encouraging but not proof.
Blood pressure: SPRINT MIND
SPRINT tested aiming for a systolic blood pressure below 120 mm Hg against below 140 in 9,361 adults aged 50 and over with high blood pressure and high cardiovascular risk, but no diabetes or stroke (mean age 67.9; 35.6% women). The main trial stopped early in 2015 because intensive treatment cut cardiovascular events and deaths (Williamson, JAMA 2019; NIH).
Over a median 5.11 years of follow-up, probable dementia was lower with intensive control (7.2 vs 8.6 cases per 1,000 person-years), but the difference wasn't statistically significant (HR 0.83, 95% CI 0.67 to 1.04), and the authors note the trial may have been underpowered for it. Mild cognitive impairment was significantly reduced: 14.6 vs 18.3 per 1,000 person-years, about 19% lower risk (HR 0.81, 95% CI 0.69 to 0.95).
Lifestyle bundles: FINGER and US POINTER
The Finnish FINGER trial randomised 1,260 people aged 60 to 77 with a raised dementia risk score (CAIDE score of 6 or more) to two years of diet advice, exercise, cognitive training and vascular risk monitoring, or to general health advice. Cognitive scores improved in both groups, slightly more with the programme: the between-group difference was 0.022 z-score units per year on a standardised test battery (95% CI 0.002 to 0.042, p=0.030). Adverse events, mostly musculoskeletal pain, occurred in 7% of the intervention group vs 1% of controls (Ngandu, Lancet 2015).
The US POINTER trial, published in JAMA in 2025, compared a structured, higher-intensity lifestyle programme with a self-guided one in 2,111 sedentary adults aged 60 to 79 with a less-than-ideal diet and other risk factors. Over two years, global cognition rose in both groups, by 0.243 vs 0.213 standard deviations a year, a small but significant advantage of 0.029 SD a year (95% CI 0.008 to 0.050, p=.008) for the structured programme (Baker, JAMA 2025). There was no no-treatment group, and dementia itself wasn't an outcome.
What the trials tested, and what they found
| Trial | Who | Intervention | Result |
|---|---|---|---|
| ACHIEVE Lancet 2023 | 977 adults 70–84 with untreated hearing loss | Hearing aids and counselling vs health education, 3 years | No overall effect on cognitive decline 48% less decline in 238 higher-risk people |
| SPRINT MIND JAMA 2019 | 9,361 adults 50+ with high BP and high heart risk | Systolic target <120 vs <140 mm Hg | MCI risk 19% lower dementia: HR 0.83, not significant |
| FINGER Lancet 2015 | 1,260 Finns 60–77 at raised risk | Diet, exercise, brain training, vascular checks vs health advice, 2 years | Small cognitive benefit difference 0.022 z-score units a year |
| US POINTER JAMA 2025 | 2,111 sedentary adults 60–79 at raised risk | Structured vs self-guided lifestyle programme, 2 years | Small extra benefit for structured 0.029 SD a year; both groups improved |
Lin et al., Lancet 2023; Williamson et al., JAMA 2019; Ngandu et al., Lancet 2015; Baker et al., JAMA 2025. MCI = mild cognitive impairment. None of these trials was large or long enough to show a clear reduction in dementia diagnoses.
The strongest trial signals are for blood pressure and bundles of healthy habits, with modest effects. For LDL, vision and social isolation, the evidence is still mainly observational.
What to Do at 40 to 65, in Order
Lead author Gill Livingston's message was that it's never too early or too late to act. Midlife factors make up the biggest slice of the Commission's estimate, and most of them also matter for your heart. A sensible order, weighing the size of each link, the trial evidence and how easy it is to act on:
- Know your blood pressure. It has the best trial evidence. The Commission recommends keeping systolic pressure at 130 mm Hg or less from age 40. If yours is higher, talk to your doctor about lifestyle changes and whether you need medication.
- Get your hearing tested. Hearing loss is the joint-largest factor at 7%. The Commission recommends hearing aids for everyone who needs them, plus less exposure to harmful noise. ACHIEVE didn't prove hearing aids protect the brain, but treating hearing loss has few downsides. Ask your doctor or an audiologist.
- Check your LDL cholesterol. The Commission recommends detecting and treating high LDL "in midlife from around age 40 years". Ask your doctor what your number means for you. Our guide to cholesterol numbers explains the basics.
- Move, don't smoke, and keep alcohol moderate. Exercise was a core part of both the FINGER and US POINTER programmes, alongside diet and brain training; smoking and alcohol are on the Commission's list of 14. Our steps guide is a practical place to start.
- Get a baseline eye exam at 40. The American Academy of Ophthalmology recommends one for adults without symptoms or risk factors (AAO). If your vision is getting worse, get it corrected; that's the "untreated" in untreated vision loss.
- Stay connected and treat depression. Social isolation is the biggest later-life factor (5%). If your mood has been low for weeks, see your doctor; the Commission recommends treating depression effectively. See our piece on loneliness and healthy years.
- Protect your head and manage diabetes and weight. The Commission recommends helmets in contact sports and on bikes. If you have diabetes, keep it well managed with your doctor.
Frequently Asked Questions
More cases ahead, but your rate isn't fixed.
The doubling in this week's headlines is mostly arithmetic: more people will live into their 80s and 90s, and the projection holds each age's dementia rate constant. Women's higher rate held even after adjusting for age, and nobody yet knows exactly why.
About 45% of dementia is linked to 14 changeable factors, a population estimate from mostly observational data; trials show smaller gains. The to-do list is familiar: check your blood pressure, hearing, LDL and eyes, stay active and connected, and talk to your doctor.
Sources
- Vella AS, Visontay R, Lipnicki DM, et al.; Cohort Studies of Memory in an International Consortium (COSMIC). Global Incidence of All-Cause Dementia: A Meta-Analysis. JAMA Neurology, published online 12 October 2026 (doi:10.1001/jamaneurol.2026.3540; journal page). Abstract read via Crossref.
- UNSW Sydney. New yearly dementia cases to double by 2050: study. 13 October 2026. See also Global dementia study reveals 11.4 million new cases in 2023.
- GBD 2019 Dementia Forecasting Collaborators. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the Global Burden of Disease Study 2019. Lancet Public Health 2022;7(2):e105–e125 (doi:10.1016/S2468-2667(21)00249-8).
- Livingston G, Huntley J, Liu KY, et al. Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. Lancet 2024;404:572–628, published online 31 July 2024 (doi:10.1016/S0140-6736(24)01296-0).
- University College London. Nearly half of dementia cases could be prevented or delayed by tackling 14 risk factors. 31 July 2024.
- The Lancet. Nearly half of dementia cases could be prevented or delayed by tackling 14 risk factors starting in childhood, including two new risks—high cholesterol and vision loss. EurekAlert, 31 July 2024.
- Per-factor attributable fractions from the 2024 Commission as presented in its summary graphic (NIHR Dementia Researcher), by Diabetes on the Net and by the Goodfellow Unit, University of Auckland.
- Lin FR, Pike JR, Albert MS, et al.; ACHIEVE Collaborative Research Group. Hearing intervention versus health education control to reduce cognitive decline in older adults with hearing loss in the USA (ACHIEVE): a multicentre, randomised controlled trial. Lancet 2023;402(10404):786–797 (doi:10.1016/S0140-6736(23)01406-X).
- SPRINT MIND Investigators for the SPRINT Research Group; Williamson JD, et al. Effect of intensive vs standard blood pressure control on probable dementia: a randomized clinical trial. JAMA 2019;321(6):553–561 (doi:10.1001/jama.2018.21442). See also NIH news release, 28 January 2019.
- Ngandu T, Lehtisalo J, Solomon A, et al. A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER): a randomised controlled trial. Lancet 2015;385(9984):2255–2263.
- Baker LD, Espeland MA, Whitmer RA, et al. Structured vs self-guided multidomain lifestyle interventions for global cognitive function: the US POINTER randomized clinical trial. JAMA, published online 2025 (PMID 40720610).
- American Academy of Ophthalmology. Get an Eye Disease Screening at 40.