Online dementia risk quizzes are only fair, so don’t take the score to heart
☕ Includes a question for the family ↓
Alzheimer's and dementiaPart of a story we are following · 4 articles · See every update
If you ever took an online quiz that gave you a dementia risk score, you don’t have to hold on to that number. A new review found these scores are only fair at telling who will develop dementia years later. That’s useful to know. It means one score doesn’t decide anything about you.
The quizzes ask about things like age, education, blood pressure, cholesterol and physical activity. The best of them sorted people reasonably well. None of them was close to a crystal ball.
How the study worked
Researchers pooled results from earlier studies that followed middle-aged and older adults who started out free of dementia. In all, 29 studies met their standards and 13 were combined in the final math. They were published as a systematic review and meta-analysis in GeroScience on Oct. 26, 2026. Only three of the ten scores had enough comparable data to be combined, and the authors say more testing in different groups of people is still needed.
What they found
The researchers rated each score on a scale where 0.5 is a coin flip and 1 is always right. The closer to 1, the better.
The score called CAIDE did best, at 0.74 to 0.76. It worked best for people who were middle-aged when first assessed and who were followed for longer.
Here is how the three scores compared:
| Score | Rating (0.5 = coin flip, 1 = perfect) |
|---|---|
| CAIDE | 0.74 to 0.76 |
| ANU-ADRI | about 0.65 to 0.71 |
| LIBRA, modified version | 0.65 to 0.67, and 0.70 at medium follow-up |
| LIBRA, original version | 0.51 to 0.60 |
The original LIBRA was close to a coin flip. The updated version did better. The authors say these scores may help doctors sort people by risk, which can guide earlier steps to lower it.
Why a number is not your future
A large share of dementia risk is linked to things that can change. Think blood pressure, physical activity, smoking and hearing. That is the useful part of these quizzes. Your score may be off, but the habits behind it are real and yours to work on.
So if a quiz gave you a high number, it doesn’t mean you will get dementia. If it gave you a low one, it doesn’t mean you’re in the clear. Treat it as a prompt, not a verdict.
What you can do today
Skip worrying about the score and jot down three numbers you can find this week: your last blood pressure reading, how many minutes you move in a typical week, and whether you’ve had a hearing check. Keep them handy for your next visit with your doctor or at your pharmacy’s blood pressure machine. Those three are things you can actually change.
Talk about it at home
Scientists wanted to know how well some online quizzes can guess who will have memory problems much later in life. They gathered results from many earlier studies and checked how often each quiz guessed right. It is a bit like a weather forecast: it can give a hint about the sky, but it cannot tell you exactly what tomorrow will bring.
Question for the table
If a scientist gave a quiz a score for how well it guessed, and a coin flip was the lowest possible score, why would they want to compare it against a coin flip at all?
Keep exploring
Copies this article and its questions so you can paste them into ChatGPT, Claude, Gemini or the AI you use, and keep the conversation going.
Questions and answers
- Which dementia risk score worked best?
- CAIDE did best in this review, with a rating of 0.74 to 0.76. It worked best for people who were middle-aged at the start and followed for longer.
- Can an online quiz tell me if I'll get dementia?
- No. Even the best score is only fair at predicting it. Think of it as a conversation starter, not a diagnosis.
- What do the numbers like 0.74 mean?
- They show how well a score separates people who will and won't develop dementia. 0.5 is a coin flip and 1 is always right.
- Which risk factors can I work on?
- The review links a large share of dementia risk to changeable factors such as blood pressure, physical activity, smoking and hearing.
