The short answer
Usually not in a way that decides who gets it. A parent or sibling with Alzheimer's roughly doubles your risk — about 1.7 to 2.6 times in large family studies — and each further affected relative raises it again. Even so, most people with an affected parent never develop it. The exception is early-onset familial Alzheimer's, caused by a change in one of three genes (APP, PSEN1 or PSEN2): each child of a carrier has a 50% chance of inheriting it, and it usually starts between the 30s and early 60s, in several generations. It accounts for only about 1 in 100 cases. The common APOE ε4 variant raises risk without deciding it.
Your risk, by family history
| Family history | What it does to your risk |
|---|---|
| The average person, from age 45 | A lifetime risk of about 1 in 5 for women and 1 in 10 for men |
| One parent or sibling, diagnosed in their 70s or 80s | About 1.7 to 2.6 times the average |
| Two or more parents or siblings | About 4 times the average |
| A parent diagnosed before about 60, with others in their family diagnosed young | Possibly a single-gene form: a 50% chance of inheriting it |
To put the middle rows in plain numbers: in the MIRAGE study, about 27% of the parents, brothers, sisters and children of people with Alzheimer’s had developed dementia by 85. For the patients’ husbands and wives — the same age, the same generation, but no shared genes — it was about 10%.
So a close relative with Alzheimer’s is a real increase. It is also a long way from a prediction: on those figures, more than seven in ten of them reach 85 without dementia.
The age it started is the most important clue
Almost everything about your own risk turns on one question: was your relative diagnosed in old age, or unusually young?
Late-onset Alzheimer’s — diagnosed after 65, most often in the late 70s and 80s — is by far the most common kind. Many genes each nudge the risk a little, alongside age, blood vessels, education, hearing and more. It clusters in families, but it has no fixed inheritance pattern.
Early-onset Alzheimer’s — before 65 — is less than 1 in 10 cases, and most of those are not caused by a single gene either. But a small group of families carries a change in APP, PSEN1 or PSEN2. That is a dominant condition: each child of a carrier has a 50% chance of inheriting it, and almost everyone who inherits it develops Alzheimer’s, typically in their 40s or 50s and sometimes in their 30s.
That family usually looks unmistakable once it is written down: a parent, a grandparent and an aunt or uncle, all developing memory problems at similar and unusually young ages, generation after generation.
One more thing changes the picture: people with Down syndrome carry an extra copy of the APP gene, and around half or more develop Alzheimer’s, often in their 50s or 60s.
Why your own number is probably different
- How many relatives, and how close. In a study of Utah families traced back to the 1800s, one affected parent or sibling raised the risk by about 1.7 times and two raised it about fourfold. Affected grandparents, aunts and uncles raised it further still.
- Your own age. Alzheimer’s risk climbs steeply after 75. Being 50 and well tells you very little; being 85 and well tells you a great deal.
- The relatives who did not get it. A family where most people lived into their 90s with clear minds is evidence in its own right.
- Whether it was really Alzheimer’s. “Dementia” on a death certificate can mean Alzheimer’s, vascular dementia, Lewy body dementia or frontotemporal dementia — and they run in families differently.
What about the APOE4 gene?
APOE comes in three common versions. ε4 is the one that matters for risk:
- about 15–25% of people carry one copy, and 2–5% carry two;
- one copy raises the odds of Alzheimer’s roughly three times;
- two copies raise them much more — around 15 times in one large analysis of white patients, and higher still in Japanese patients.
But APOE4 is a risk factor, not a cause. Many people with two copies never develop Alzheimer’s, and many people with Alzheimer’s have none. That is why genetic testing for APOE is not recommended to predict whether a healthy person will get it.
If you have found out your APOE result from a consumer DNA test, it is worth talking to your doctor about it — mainly because the things that lower dementia risk are the same whoever you are.
If your relative had a different kind of dementia
Frontotemporal dementia — changes in personality, behaviour or language, often starting between 45 and 65 — is far more often inherited than Alzheimer’s. In a sizeable minority of families it follows a dominant, 50% pattern, most often through the C9orf72, GRN or MAPT genes. If that is the diagnosis in your family, genetic counselling is worth asking for even if nobody else has been affected.
Vascular dementia follows the family risk of strokes and heart disease more than any dementia gene.
What lowers the risk, whatever your family history
The 2024 Lancet Commission on dementia estimated that about 45% of dementia cases worldwide are linked to 14 things that can be changed. Among the most practical:
- treating hearing loss — hearing aids, if you need them;
- keeping blood pressure and LDL cholesterol down from middle age;
- managing diabetes, not smoking, and drinking less;
- staying physically active and socially connected;
- getting vision loss treated.
None of these cancels a family history. They do shift the odds, and they are the same steps whether you carry APOE4 or not.
When genetic testing makes sense
Testing is usually offered when someone develops symptoms young and the family history is strong — and it works best when it starts with the affected relative. If a family change in APP, PSEN1 or PSEN2 is found, healthy relatives can have a predictive test for that exact change, through a genetic counselling process much like the one for Huntington’s disease.
For ordinary late-onset Alzheimer’s, there is no genetic test that tells a healthy person whether they will get it.
Sources and further reading
- National Institute on Aging: Alzheimer’s disease genetics fact sheet
- Chêne G et al. Gender and incidence of dementia in the Framingham Heart Study from mid-adult life. Alzheimer’s & Dementia, 2015
- Green RC et al. Risk of dementia among relatives of patients with Alzheimer disease (MIRAGE). JAMA, 2002
- Cannon-Albright LA et al. Relative risk for Alzheimer disease based on complete family history. Neurology, 2019
- Farrer LA et al. APOE genotype and Alzheimer disease: a meta-analysis. JAMA, 1997
- Livingston G et al. Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission
Common questions
- My mother had Alzheimer's. Will I get it?
- Most likely not. Your risk is roughly double the average, which is a real increase from a starting point where most people never develop it. If your mother was diagnosed in her late 70s or 80s, that is the typical picture. If she was diagnosed before about 60, and her parent or siblings were too, ask for a referral to genetic counselling.
- Is Alzheimer's passed down from the mother or the father?
- Either. Some studies find a slightly stronger link through the mother and others do not, so it is not settled. In the rare single-gene forms, a father and a mother pass it on equally: a 50% chance for each child.
- Does Alzheimer's skip a generation?
- Late-onset Alzheimer's follows no fixed pattern, so it can appear in a grandparent and a grandchild with nobody in between — often because the parent died of something else first, or is not yet old enough. The single-gene early-onset forms do not really skip; a generation looks skipped when a carrier died before the age it would have started.
- A home DNA test says I have APOE4. What does that mean?
- It means a higher risk, not a diagnosis. One copy raises the odds about threefold, and two copies considerably more, but many people with APOE4 never develop Alzheimer's and many without it do. The US National Institute on Aging does not recommend APOE testing to predict who will get it. A result is a good reason to take the things you can change seriously, starting with blood pressure and hearing.
- My grandparent had dementia in their 80s. Am I at higher risk?
- Only a little. A grandparent adds less than a parent or sibling, and dementia after 85 is common in every family. It is a reason to know the risk factors, not a reason for testing.
- Is early-onset Alzheimer's always genetic?
- No. Fewer than 1 in 10 people with Alzheimer's develop symptoms before 65, and of those only about 10–15% have a change in APP, PSEN1 or PSEN2. A single relative diagnosed at 63 is much more likely to have the ordinary form of the disease.