My spouse has dementia. Does that raise my own risk?

Short answer: Somewhat, on average — but far less than the headlines suggest, and not because dementia is catching. A 2026 study of nearly a million married people in Taiwan found that spouses of someone diagnosed with dementia were about 70% more likely to be diagnosed themselves over the following years than similar people whose spouses did not have dementia. In everyday numbers, that was roughly 3 extra diagnoses for every 100 spouses over five years. The likeliest explanations are things couples share — similar backgrounds, habits and health — plus the real strain of caregiving. Several of those are things a carer can work on, and support for carers is one of the few places where trials show a measurable benefit for carers' own wellbeing.

This is educational content, not medical advice. It describes associations found in population research, not a prediction about you. If you are caring for someone and are worried about your own memory, talk to your clinician.

Dementia is not contagious

It's worth saying plainly, because "spouses face higher risk" can read like a warning about infection. None of the research here suggests dementia passes from one person to another. The studies compare couples who share a life, not people who share germs. What they find is that a household carries shared risk — the same question you'd ask about two people who have eaten the same diet and lived on the same income for forty years.

What the 2026 study found

The largest study so far was published in JAMA Network Open in July 2026 by researchers at Taiwan's National Health Research Institutes. They used the country's national health-insurance records from 1998 to 2022 and compared:

• 190,000+ people whose spouse developed dementia (118,904 women and 72,117 men), with
• up to four similar people each whose spouses did not — matched by birth year, sex, income and whether they lived in a city or a rural area. In total, 955,105 people.

The same pattern showed up for Alzheimer's disease and vascular dementia separately.

Why "74% higher" and "about 3 in 100" are both true

Look at the table again. Women had the slightly bigger relative increase (1.74 vs 1.69), but men had the bigger absolute one (3.5 vs 2.75 extra cases per 100). (The two relative figures are close; the authors describe the link as broadly similar in women and men.) That isn't a contradiction. Men in the study started from a higher baseline risk, so a slightly smaller percentage increase on top of it added more actual cases.

The study found the same thing, more strongly, by age. Among women whose spouse was diagnosed before they were 55, the relative figure was 2.21 but the absolute increase was about 0.3 extra cases per 100 over five years; among women aged 75 or over it was 1.48, but about 5 extra cases per 100. The relative increase was largest in younger spouses, while the absolute extra risk grew with age, because dementia itself becomes more common as we get older.

This is why a headline percentage, on its own, can't tell you how worried to be. For a full walk-through of this and other traps, see our guide to how to read a dementia statistic.

What could explain it

The researchers, and earlier studies, point to three overlapping explanations. None has been proven to be the main one.

1. People tend to marry people like themselves. Researchers call this assortative mating. Couples often share similar education, income and health habits — and those are linked to dementia risk on their own. The Taiwanese records had no data on education, which is one of the strongest of those shared factors, so this can't be ruled out.

2. Couples share a life. The same kitchen, the same activity level, often the same smoking and drinking habits, for decades. A Korean study of 784 older couples, which assessed both partners at one point in time, found that when one partner had mild cognitive impairment or dementia, the other had roughly 1.7 times the odds of having mild cognitive impairment or dementia too. Shared factors, including age, past head injury, and physical inactivity working through depression, appeared to account for about three-quarters of that link — though, as the authors note, a one-time snapshot can't show which came first. (In the Taiwanese study, the two groups had broadly similar rates of vascular, metabolic and psychiatric conditions, which the authors read as a sign that shared lifestyle factors alone are unlikely to explain everything.)

3. Caregiving is hard on the carer. Looking after a partner with dementia is linked with depression, broken sleep, social isolation and less time for one's own health — and each of these is, separately, associated with dementia risk. In the US Health and Retirement Study, people who had cared for a spouse with dementia showed faster cognitive decline than people caring for a spouse with other conditions.

A fourth possibility that the study did not test is worth keeping in mind too: once a family has been through a dementia diagnosis, the healthy spouse may simply be more likely to be noticed, assessed and diagnosed. Health-insurance records capture diagnoses, not the underlying disease, so some of the gap could reflect who gets checked rather than who gets ill.

Why the older "six times higher" number is not the one to use

You may see a much scarier figure: a six-fold higher risk. That comes from the Cache County study in Utah, published in 2010. It followed about 1,200 married couples and found a hazard ratio of 6.0 — but with a 95% confidence interval of 2.2 to 16.2. That width tells you the real figure was very uncertain: only 34 people whose spouse had dementia went on to develop it themselves, and with numbers that small, a study can produce a large estimate by chance.

The 2026 study is several hundred times larger, and its estimate is far narrower: around 1.7. The Taiwanese authors themselves suggest their smaller effect may reflect differences in design, how representative the sample was, and statistical precision. The Korean study's estimate, for a broader outcome, was in the same range.

What the study doesn't tell you

• It doesn't predict your personal risk. It reports averages across very large groups. Your own health, habits and family history are a large part of your individual picture, and the study can't weigh them against your spouse's diagnosis.
• It doesn't show cause and effect. It is an observational study. It can show that risk travels with a household, not why.
• It comes from one country's health system. The authors say the findings may apply best to places with similar health care systems and cultural contexts.
• Most spouses in the study did not develop dementia within five years — more than nine in ten, in both groups.

Who seemed less affected — and what that suggests

The extra risk, in absolute terms, was smaller among people with higher incomes and those with more children. The authors are careful here: the two overlap (people with fewer children were more often in the lowest income group), and income and number of children are rough stand-ins for things the records didn't measure, such as paid help, respite and family support. Still, the direction is consistent with an older observation from the US: in one Health and Retirement Study analysis, men and lower-income people caring for a spouse looked most vulnerable, and the link to day-to-day functional problems disappeared once depression was taken into account.

The practical reading: support may matter. That is a reasonable inference, not a proven one.

What a caregiving spouse can do

None of the following is proven to lower dementia risk in carers specifically. They are the same evidence- based levers that apply to everyone — which is the point. A carer's own risk factors are ordinary risk factors, and they tend to get neglected first.

1. Get support early, not when you're at breaking point. In the REACH II trial, a six-month structured programme of home and phone sessions for dementia carers (covering stress, self-care, mood and handling difficult behaviour) improved carers' quality of life, and clinical depression was less common in the group that received it (12.6% vs 22.7%). The trial measured carer wellbeing, not dementia — but depression is itself linked to dementia risk. Ask your partner's care team or a dementia charity what carer support exists where you live.
2. Protect your sleep. Night-time disturbance is common when a partner has dementia. If you are chronically short of sleep, tell your clinician — carers often put up with it for longer than they need to. See sleep and dementia risk.
3. Keep your own social life alive. Carers often lose friendships without noticing. Even short, regular contact counts. See loneliness and dementia.
4. Keep moving. Physical inactivity was one of the shared factors in the Korean couples study. Walking together, when that is possible, helps both of you. See exercise and dementia.
5. Keep your own appointments. Carers often skip their own check-ups. Blood pressure, blood sugar, cholesterol and hearing are among the best-supported modifiable factors. Discuss them with your clinician — and mention that you are a carer, because it changes what support they can offer.
6. If you notice changes in your own memory, raise them. Stress, low mood and exhaustion can all affect memory, and they're not the same as dementia. See is it anxiety or dementia?.

See where you stand

Your spouse's diagnosis is one piece of information. Your own risk also depends on factors that are about you — and many of them can be changed. The free Solenna risk profile takes about 10 minutes, covers the modifiable factors with the strongest evidence, and shows which ones are worth your attention first. It is educational, not a diagnosis, and it works entirely in your browser.

How we research and review our content →

An independent physician audited the Solenna app. We published all nine findings →

Educational content only. It does not diagnose any condition and is not a substitute for professional medical advice. If you are worried about your own memory, or about your health as a carer, talk to your clinician.