Dementia has long been treated as a problem of late life. The assumption has been that if your brain is going to fail, it will fail slowly, and most often after the age of 70. That assumption is being challenged. Across high-income countries, neurologists are seeing more patients in their fifties, forties, and even thirties with the kind of cognitive decline that used to be called "presenile". The umbrella term for this is young-onset dementia, sometimes called early-onset dementia, and it covers any progressive cognitive disorder whose symptoms begin before the age of 65. New research is starting to explain why the picture is changing, and what can be done about it.
The scale of the problem
The numbers are easier to describe than to pin down. The World Health Organization estimates that around 55 million people worldwide are living with dementia, with nearly 10 million new cases each year and a projected total of 139 million by 2050 [6]. Most of those numbers refer to late-onset disease, but somewhere between 5 and 10% of all dementia cases are now classified as young-onset, which still translates to millions of people under 65 [1]. In the United States, the Alzheimer's Association puts the under-65 Alzheimer's population in the hundreds of thousands, with lifetime risk at age 65 of roughly one in five for women and one in ten for men [7].
Whether the underlying incidence is genuinely rising, or whether we are simply better at recognising cases that were once missed, is one of the more contested questions in the field. A 2024 Dutch population study covering three decades found that, after accounting for diagnostic improvements, the rate of new early-onset cases in the Netherlands was broadly stable, or even slightly declining, in most age groups [3]. A separate global review published in Nature Reviews Neurology in 2024 argued the opposite, that reported incidence in young adults is climbing across many high-income countries and that the rise is real, not just an artefact of better testing [14]. Both can be partly right: a real increase in vascular and lifestyle-driven risk layered on top of steady improvements in brain imaging, genetic testing, and reduced stigma around cognitive symptoms in working-age adults.
What the latest evidence points to
What researchers do agree on is the growing list of modifiable risk factors. The 2024 update of the Lancet Commission on dementia prevention, intervention, and care expanded the original 2020 list to 14 factors that together account for an estimated 45% of dementia cases worldwide [2]. The two new additions, vision loss and high LDL cholesterol, joined the familiar roster: less education in early life, hearing loss, depression, social isolation, smoking, air pollution, head injury, physical inactivity, diabetes, hypertension, alcohol, obesity, and poor sleep. The Commission was careful to point out that this 45% is an upper estimate: removing all 14 factors would not abolish dementia, but tackling even a fraction of them could delay onset, sometimes by years, for large populations.
For young-onset cases specifically, three of those factors stand out because they are common, modifiable, and supported by strong recent evidence. The first is alcohol. A 2024 BMJ cohort study of 1.33 million adults in France found that an alcohol use disorder was associated with substantially higher dementia risk, including early-onset forms, and that risk rose steadily with consumption rather than showing the protective effect at low doses that some earlier studies had suggested [4]. The second is air pollution. A 2024 analysis in The Lancet Planetary Health found that long-term exposure to fine particulate matter (PM2.5) was linked to higher dementia risk at exposure levels below the current US annual standard, with stronger associations in people with cardiovascular vulnerability [5]. The third is traumatic brain injury. A 2024 systematic review in The Lancet Neurology reported that a single severe head injury was associated with an approximately 70% increase in long-term dementia risk, with the strongest effect when the injury occurred in younger adulthood [11].
Why it is missed for years
The other thing researchers agree on is that young-onset dementia is harder to recognise, and slower to diagnose, than late-onset disease. A 2024 BMJ Open qualitative study found that the average time from first symptoms to a formal diagnosis of young-onset dementia is around four to five years, and that many patients are first given alternative diagnoses: depression, anxiety, burnout, perimenopause, or simply stress at work [12]. Part of the problem is that memory loss, the symptom most people associate with dementia, is often not the first sign in younger patients. Frontotemporal dementia, the second most common cause of young-onset dementia after Alzheimer's, usually starts with changes in personality, behaviour, or language rather than memory, and the average age of onset is around 58, with cases well documented in the thirties and forties [8]. That profile is easily misread as a midlife crisis, a relationship problem, or a psychiatric condition, and the delay can be costly: younger patients are often in the middle of careers, mortgages, and caring for children, and the years between first symptoms and a firm diagnosis are typically the most disruptive of the whole illness.
The rarer but more clearly genetic forms of early-onset Alzheimer's are worth flagging because they are sometimes the only kind a younger patient has heard of. Autosomal dominant Alzheimer's, caused by mutations in one of three genes (PSEN1, PSEN2, or APP), accounts for a small fraction of all Alzheimer's cases but a much larger share of young-onset ones [13]. These mutations run in families, often produce symptoms in the forties or earlier, and are the reason some people with a parent who had early Alzheimer's seek genetic counselling. For the vast majority of younger adults with cognitive symptoms, the cause is not a single dangerous gene but a combination of vascular, lifestyle, and environmental factors acting over decades.
What midlife is for
What can a younger adult do with this information, given that the evidence base is still settling? The honest answer is that the same things that protect your heart and metabolism appear to protect your brain, and the protective effect seems to be largest in midlife, between 45 and 65, not in old age. A 2024 UK Biobank analysis in The Lancet Public Health found that higher cumulative vascular risk (a composite of blood pressure, cholesterol, smoking, diabetes, and body mass index) was associated with both earlier dementia onset and higher overall incidence, and that midlife risk predicted outcomes more strongly than late-life risk [10]. Sleep matters too. A 2023 meta-analysis in Sleep Medicine Reviews found that both short sleep (under six hours) and long sleep (over nine hours), as well as fragmented sleep and untreated sleep apnoea, were independently associated with higher dementia risk, with mechanisms thought to involve the brain's overnight waste-clearance system [9]. The implication is not that anyone can dementia-proof their future, but that the decades from your forties onwards are when the brain's reserve is being built or quietly eroded.
What is still uncertain
Two scientific uncertainties are worth holding open. The first is whether the apparent rise in young-onset dementia is a real biological shift or mostly a recognition effect. The second is whether moderate alcohol use is harmful or, as some older studies suggested, slightly protective. The 2024 BMJ cohort study came down firmly on the side of harm, finding no protective threshold, but the question is not fully closed. Both issues are good examples of how a single headline ("dementia is rising in young people", "a glass of wine a day keeps dementia away") can be misleading without the underlying uncertainty spelled out.
The practical takeaway for anyone in their thirties, forties, or fifties is unspectacular but worth repeating. Get blood pressure, cholesterol, blood sugar, and body composition checked and managed. Treat hearing loss, sleep apnoea, and depression rather than soldiering through them. Cut down on alcohol rather than the reverse, particularly if there is a family history of early cognitive problems. Use a helmet for cycling, contact sport, and DIY projects that risk a head injury. Avoid breathing fine particulate matter where you can, both indoors and out. None of these steps is a guarantee, and none of them replaces the need for proper neurological assessment if you or someone close to you is experiencing progressive cognitive or personality change, but together they are the closest thing the current evidence offers to a realistic, decades-long defence against a condition that is, slowly and unevenly, getting harder to ignore in working-age adults.