The 2024 Lancet Commission on dementia, in plain English — and the 14 modifiable risk factors

The 2024 Lancet Commission on dementia — a standing group of leading researchers who periodically review the global evidence — estimated that around 45% of dementia cases could, in principle, be prevented or delayed by addressing 14 modifiable risk factors across the course of life. Those factors are: less education, hearing loss, high LDL cholesterol, depression, head injury, physical inactivity, diabetes, smoking, high blood pressure, obesity, excessive alcohol, social isolation, air pollution and untreated vision loss. This is the honest version of what the report actually says — what the 45% means and what it doesn't, what changed in the 2024 update, and what each of the 14 factors looks like in practice, with one thing you can do about each.

What the Lancet Commission actually is

The Lancet is one of the oldest and most respected medical journals in the world. Its Commission on dementia prevention, intervention and care is a standing panel of leading dementia researchers, led by Professor Gill Livingston, who periodically pull together all the available evidence and publish a consensus report. They did so first in 2017, updated it in 2020, and published the latest version in July 2024.

That's worth knowing, because it's why the report carries weight. When people cite "the Lancet Commission," they aren't pointing to a single splashy study or a health headline — they mean an expert consensus that has weighed the whole body of research, argued about it, and put its name to a summary. It's about as close to a settled overview as this field gets.

It also explains why the report keeps changing. Each edition re-examines the evidence rather than restating the last one, which is how the list has grown from nine factors in 2017 to twelve in 2020 to fourteen today. A list that moves is a sign the field is still learning, not a sign it can't be trusted. It also means the list is not the whole field: some of the most-discussed findings to arrive since 2024, including the research linking shingles vaccination to lower dementia risk, sit outside the fourteen and would need a future edition to weigh them.

The headline: "~45% could be prevented or delayed"

Here is the exact claim, and it rewards careful reading. The 2024 report estimates that around 45% of dementia cases worldwide could, in principle, be prevented or delayed by eliminating 14 modifiable risk factors. Three parts of that sentence do a lot of work:

• "Could, in principle." This is a population-level modelling estimate under ideal conditions — what statisticians call a population attributable fraction. It describes what might happen if every one of the 14 factors were completely removed from the whole population. No society will ever fully achieve that, so the 45% is best read as a ceiling, not a forecast.
• "Prevented or delayed." Even where risk is reduced, that often means pushing dementia later rather than erasing it. That still matters enormously — more good years — but it is not the same as "cure."
• Modifiable, not guaranteed. Many of these links come from observational evidence. Addressing the factors is expected to lower risk across a population; it is not a guarantee for any single person.

Put plainly: a large share of dementia risk, across whole populations, runs through things people can influence. That is genuinely hopeful, and worth acting on — as long as nobody stretches it into a personal promise.

What "modifiable" actually means

A modifiable risk factor is simply one that can potentially be changed — unlike age or genetics, which cannot. That single word is doing more work than most headlines allow. It is why the Commission's list contains no mention of the two things people worry about most: getting older, and what runs in the family. Those are left off not because they are unimportant — they are the largest drivers of all — but because there is nothing to do about them. The list is deliberately a list of levers.

"Modifiable" also does not mean "easy," and it does not mean "yours alone to fix." Air pollution is modifiable at the level of a city's policy far more than at the level of one household's air filter. Years of education is modifiable for a country's next generation, not for an adult's own childhood. Several factors on this list are as much public-health targets as personal ones, which is part of why the 45% is framed as a population figure.

Two caveats before the list itself. First, having a risk factor does not mean you will develop dementia — these are shifts in odds, not diagnoses. Many people with several risk factors never develop dementia, and some with none do. Second, the factors overlap and compound; they're best thought of as a stack you chip away at over years, not a checklist you clear in a weekend.

What changed in the 2024 update

The 2020 report listed 12 factors and put the figure at about 40%. The 2024 update added two more, bringing the total to 14 factors and the estimate to about 45%:

• High LDL cholesterol in midlife. Raised "bad" cholesterol from around age 40 is now linked to higher dementia risk — added on the strength of cohort studies of more than a million people, plus a genetic (Mendelian randomisation) meta-analysis of 27 studies. It's one of the larger single contributors, with the Commission attributing roughly 7% of cases to it.
• Untreated vision loss in later life. Uncorrected vision problems — such as cataracts left untreated — are now on the list, based on two large meta-analyses, and account for about 2% of cases.

The 2024 report also reclassified several factors — depression, physical inactivity, smoking and diabetes — from later life into midlife. That's a useful nudge, and arguably the most practical change in the whole update: a lot of the action happens in your 40s, 50s and 60s, not only in old age. It also means midlife is now by far the busiest stage on the list, holding ten of the fourteen factors. (The updated risk shares were recalculated using data from Norway's large HUNT population study.)

The 14 factors, across the life course — and one thing you can do

The Commission organises the factors by when in life they tend to matter most. Below is the full list in that order, in plain English, each with one practical thing a person can do. None of these is a cure, and none guarantees an outcome — they are levers associated with lower risk. You can also explore them interactively in the interactive 14-factor explainer.

Early life

• Less education — fewer years of early education is linked to higher risk, likely through lower "cognitive reserve," the brain's capacity to absorb damage before symptoms show. You can do: keep learning at any age — a language, an instrument, a course. Novel, effortful learning is what counts, not passive consumption.

Midlife

• Hearing loss — one of the largest midlife factors, and one of the most treatable. You can do: get your hearing checked and use hearing aids if they're recommended; untreated hearing loss is the version linked to risk.
• High LDL cholesterol — raised LDL ("bad") cholesterol from around age 40 is linked to higher risk, and is one of the two factors new in 2024. You can do: know your numbers and manage them with your doctor through diet, activity and, if advised, medication.
• Depression — moved to midlife in the 2024 report; depression is associated with higher dementia risk. You can do: treat it — talk to a professional. Effective care exists, and low mood is not something to "push through" alone.
• Head injury — traumatic brain injury is associated with higher risk. You can do: wear seatbelts and helmets, take concussion seriously rather than playing on, and reduce fall hazards at home as you age.
• Physical inactivity — also moved to midlife in 2024, and one of the most consistently protective behaviours in the research. You can do: move regularly — brisk walking counts; aim for something most days rather than perfection some days.
• Diabetes — now a midlife factor; type 2 diabetes is linked to higher risk. You can do: get screened, and if you have it, manage blood sugar with clinical support.
• Smoking — reclassified to midlife in 2024. Smoking raises risk, and stopping is associated with lower risk over time. You can do: stop, with support — quitlines, nicotine replacement and coaching all improve the odds of it sticking.
• High blood pressure — a clear link between vascular health and brain health, and one of the best-evidenced levers on the list. You can do: check it and keep it in a healthy range; what protects your heart tends to protect your brain.
• Obesity — midlife obesity is associated with higher risk, partly through its overlap with blood pressure and blood sugar. You can do: work toward a healthier weight through sustainable diet and activity changes, with support if needed.
• Excessive alcohol — heavy drinking is linked to higher risk. You can do: keep alcohol within recommended limits, or lower.

Later life

• Social isolation — low social contact is associated with higher risk. You can do: protect regular connection — a standing call, a class, a group; treat it as maintenance, not a luxury.
• Air pollution — exposure to air pollution is linked to higher risk, and is the factor most shaped by policy rather than personal choice. You can do: reduce exposure where you can — ventilation, indoor air filtration, and awareness on high-pollution days.
• Untreated vision loss — the other 2024 addition; untreated vision impairment is associated with higher risk, and the word untreated is the operative one. You can do: get regular eye checks and treat correctable problems such as cataracts.

Note what the life-stage grouping is and isn't saying. It marks the period when a factor appears to matter most, not the only window in which it counts. Blood pressure still matters at 75; hearing still matters at 45. The grouping is a hint about where to look first at a given age, not an expiry date.

Which ones you actually control

Look down that list and a pattern appears: almost all of it is checkable and actionable. Age and genetics — the things people worry about most — aren't on it, precisely because they can't be modified. What is on it clusters into a few practical levers:

• Your heart and metabolic numbers — blood pressure, LDL cholesterol, blood sugar and weight. This cluster shows up four times on the list of fourteen, which is the single strongest hint the report gives about where to start.
• Your senses — hearing and vision. Both are among the most treatable factors, both are easy to ignore until checked, and both entered the list on the strength of the word "untreated."
• How you live day to day — physical activity, not smoking, moderate alcohol, continued learning and social contact.
• Your mood and your environment — treating depression, and reducing head-injury and air-pollution exposure.

Two factors sit at the edge of personal control and deserve saying out loud. Early education is fixed by the time anyone reads an article like this — the modifiable part, for an adult, is continued learning rather than schooling already had. And air pollution is largely a matter of where you live and what your city does about it; individual mitigation helps at the margins, but this one is genuinely a public-health lever more than a personal one.

The honest framing is that you influence odds, not certainties. But "influence" is not nothing — it's most of the list.

Why the combination matters more than any single factor

It's tempting to pick the one factor that feels easiest and call it done. The research points the other way: the payoff tends to come from acting on several factors together. The FINGER trial (2015) tested a two-year multidomain programme — diet, exercise, cognitive training and vascular monitoring combined — and found it helped maintain or improve cognition in at-risk older adults compared with a control group. The ACTIVE trial followed older adults for a decade and linked speed-of-processing cognitive training to lower dementia risk. And SPRINT-MIND (2019) found that intensive blood-pressure control reduced the risk of mild cognitive impairment versus standard control.

None of these studies is about a single silver bullet, and none claims prevention or a cure. Read together, they suggest the same thing the Lancet list implies: brain-health risk is spread across many factors, so the most reliable approach is to chip away at several of them steadily rather than betting everything on one.

There is a statistical reason for this too. Because the factors overlap — high blood pressure, diabetes and obesity travel together; hearing loss and social isolation feed each other — their individual percentages cannot simply be added up. That cuts both ways. It means you can't stack the numbers to build your own total, and it means addressing one factor often nudges its neighbours in the same direction.

What the 45% does — and doesn't — mean

Because this number gets stretched in both directions, it's worth being precise about its limits:

• It is not a personal figure. You can't subtract 45% from your risk by ticking boxes.
• The factors overlap and compound, so you can't simply add their percentages together.
• Age and genetics remain major drivers of dementia — they're absent from the list precisely because they can't be modified, not because they don't matter.
• "Prevented or delayed" is not "prevented" or "cured." The Commission is careful with that wording, and so should everyone quoting it be.
• It describes an ideal-conditions ceiling — the total elimination of all 14 factors from an entire population — which is a planning benchmark, not an achievable target.

None of that undercuts the core message. Across three reports and many years, the evidence keeps pointing the same way: a substantial share of dementia risk is modifiable at the population level. That's a reason for measured optimism and steady action — not fear, and not false hope.

Turning a global number into a personal plan

A worldwide statistic can't tell you what to do on a Tuesday, and fourteen factors is a lot to hold in your head — a generic "do everything" is how good intentions stall. What's actually useful is knowing where you stand across these 14 factors: which ones you've already handled, and which one or two are worth a conversation with your doctor this month. Some will already be fine. Some will need a number you don't currently have, which usually means a simple check rather than a lifestyle overhaul. That's the difference between an interesting report and an actual plan, and it's the whole idea behind mapping your own modifiable factors and watching them change over time.

How we research and review our content →

This article is educational and is not medical advice, diagnosis, or treatment. Figures and studies cited describe population-level research from the 2024 Lancet Commission and the trials named, not Solenna. Solenna does not prevent, treat, or cure Alzheimer's disease or any form of dementia; individual results vary and no outcome is guaranteed. If you are concerned about your cognitive health, consult a qualified healthcare professional.