Smoking and dementia risk: how much quitting actually helps

Short answer: Smoking is one of the 14 modifiable risk factors the Lancet Commission links to dementia, and the useful news is about quitting. In a study of 32,802 US adults followed for up to 25 years, people who quit had a 16% lower rate of dementia than those who kept smoking — and the longer they stayed smoke-free, the more their risk fell, levelling off at roughly the same rate as people who had never smoked after about seven years. A separate analysis across 12 countries found that cognitive decline measurably slowed in the years after quitting: verbal fluency declined about half as fast, and memory about 20% more slowly, compared with people who continued. Both studies are observational, so they show a pattern rather than proving cause. But the direction is consistent, it holds for people who quit in their fifties and sixties, and the honest headline is the encouraging one: it does not appear to be too late.

This is educational, not medical advice. If you want to stop smoking, your doctor, pharmacist or a national quitline can offer treatments that roughly double or triple your chances compared with willpower alone. That conversation is worth having regardless of anything on this page.

How strong is the link between smoking and dementia?

Smoking sits on the 2024 Lancet Commission's list of 14 modifiable risk factors that together account for around 45% of dementia cases worldwide. In the 2024 update it was reclassified from a late-life factor to a midlife one — meaning the exposure that matters most is happening in your forties and fifties, decades before symptoms.

Its population attributable fraction — the share of dementia cases at a population level statistically attributable to it — is about 2%.

That number surprises people, and it's worth explaining, because it's easy to misread in both directions.

A population attributable fraction is not your personal risk. It reflects two things multiplied together: how much a factor raises risk and how common it is. Smoking's PAF is modest largely because smoking rates have fallen dramatically. If you personally smoke, the relevant figure isn't the 2% — it's the per-person risk elevation, which is meaningful, and which sits on top of smoking's much larger and better established effects on your heart, lungs and blood vessels.

There's also a mechanism that makes the link biologically plausible rather than a statistical curiosity. Smoking damages small blood vessels, and vascular damage is one of the clearest routes to cognitive decline — the same pathway that makes blood pressure, blood sugar and cholesterol matter. We covered how much that vascular burden adds up to in what midlife vascular risk costs in dementia-free years, where smoking is one of the three factors measured.

What happens when you quit

This is the part with the real news in it, and there are two recent studies worth knowing.

The dementia study

Researchers used the US Health and Retirement Study to follow 32,802 adults who were dementia-free at the start, average age 60.5, checking their smoking status and weight every two years between 1995 and 2020. Over a median of about 10 years, 5,868 developed dementia.

Compared with people who kept smoking:

Read down that column: people who quit end up close to people who never smoked. And the study was able to look at how quickly that happens — risk fell gradually with time since quitting and plateaued at around seven years, at approximately the never-smoker level.

(Chen H et al., Neurology, May 2026)

The cognition study

A second analysis, from UCL, took a different and rather clever approach. Rather than comparing smokers with non-smokers, it matched about 4,700 people who quit against about 4,700 who continued, drawn from three long-running studies across 12 countries — 9,436 people in total, average age 58 — and compared their cognitive test scores for six years before and six years after the quit date.

Before quitting, the two groups were declining at essentially the same rate. Afterwards they separated:

• Verbal fluency decline roughly halved — about six months less decline per year
• Memory decline slowed by about 20% — about three to four months less per year

That "before" comparison matters more than it might look. It's evidence that the two groups weren't already different — which is one of the main ways a study like this could mislead you.

(Bloomberg M et al., The Lancet Healthy Longevity, 14 October 2025)

The researchers were careful about one thing, and so should we be: this study measured cognitive test scores, not dementia diagnoses. In their words, "further research will be needed that specifically examines dementia to confirm this." Slower decline on a memory test is encouraging. It isn't the same as fewer dementia cases.

Is it too late to quit?

On the evidence above, no — and this appears to be true well into later life.

Both studies were done in middle-aged and older adults. The dementia study's average participant was 60. The cognition study included people up to 89, and the researchers' summary was direct: it is never too late to quit.

That's worth sitting with if you're 55 or 65 and have assumed the damage is done. The seven-year convergence in the dementia data means a 60-year-old who stops now is looking at a risk profile approaching that of a never-smoker by their late sixties — during precisely the years when dementia risk starts climbing.

Two honest caveats. Both studies are observational: people weren't randomly assigned to quit, so some of the difference could reflect other things about people who successfully stop. The Neurology authors explicitly flag "potential residual confounding and measurement error." And neither study can tell you what will happen to you personally. What they show is a consistent pattern across large populations, pointing the same way.

About weight gain after quitting

The Neurology study found something that needs handling carefully, because it is easy to misread into a genuinely harmful conclusion.

The apparent benefit of quitting was largest in people who gained little or no weight in the two years afterwards (5 kg / about 11 lb or less). In people who gained more than 10 kg (about 22 lb), the association with lower dementia risk was not statistically significant — hazard ratio 1.33, with a confidence interval of 0.87 to 1.82.

Look at that interval, because it's the whole point. It runs from a meaningful benefit to a meaningful harm and includes 1.0 — "no difference." That is a finding of uncertainty, not a finding that quitting with weight gain is bad for you. The honest reading is that substantial weight gain may blunt some of the brain benefit, and that we don't yet know by how much, or whether it does at all.

What it is definitely not is a reason to keep smoking. Smoking's effects on cardiovascular disease, cancer and mortality dwarf anything in this analysis, and none of those are in doubt. The authors' own conclusion is about support, not deterrence: it highlights "the need for weight management in cessation programs."

So the practical version: quit, and if you can, plan for the weight side at the same time — which usually means physical activity and food quality, both of which are on the dementia-risk list in their own right. Ask your doctor or quitline to help you with both together.

What about vaping?

The honest answer is that we don't know yet, and anyone telling you otherwise is ahead of the evidence.

E-cigarettes haven't been in widespread use long enough to produce the kind of study you'd need — decades of follow-up to a dementia diagnosis. There is no good long-term human evidence that vaping raises dementia risk, and there is no good evidence that it doesn't.

What can be said: the vascular pathway that plausibly connects smoking to dementia is a reason for caution rather than reassurance, and "no evidence of harm" is not the same as "evidence of no harm." If you're using vaping as a route away from cigarettes, that's a conversation to have with a clinician or quitline, who can talk about the full range of options.

Where smoking sits among the other levers

If you smoke, stopping is likely the single highest-value change on the whole modifiable risk factor list — not primarily because of the dementia numbers above, but because it's also the biggest lever you have on heart disease, stroke, cancer and lung disease, and because vascular health and brain health run together.

If you don't smoke, or you quit years ago, the attention is better spent on the factors that are still open: blood pressure, hearing, blood sugar, physical activity, sleep, alcohol and social connection.

Smoking is also one of the interventions the WHO's 2026 guidelines specifically recommend addressing to reduce dementia risk — one of the relatively few things on that list where the guidance is unambiguous.

Find out which factors apply to you

Quitting is the clearest single lever if you smoke. It isn't the only one, and it may not be the biggest one for you.

More context: can you reduce your dementia risk? and check your dementia risk.

How we research and review our content →

This article is educational and is not medical advice, diagnosis, or treatment. The studies cited describe findings in general populations, not any individual. Solenna is not a medical device and does not diagnose, prevent, treat, or cure Alzheimer's disease or any other condition. Risk reduction means lowering probability, not eliminating it; individual results vary and no outcome is guaranteed. If you want to stop smoking, your doctor, pharmacist or a national quitline can offer treatments that substantially improve your chances. Decisions about medication or cessation aids should be made with a qualified clinician.