A lower relative dementia risk does not automatically mean a large personal benefit. Relative risk compares groups, while absolute risk describes a person's chance of developing dementia over a specified period. The same relative reduction can produce different absolute results when the starting risk changes. Age, follow-up period, and the population studied all affect how meaningful a reported percentage is.
Medical information disclaimer: This article is for general educational purposes only and does not provide medical advice, diagnosis, or treatment. Always consult a physician or other qualified health professional about symptoms, medications, tests, or treatment decisions.
Table of Contents
- Relative risk versus absolute risk
- How the starting risk changes the result
- What population estimates can—and cannot—tell you
- Why an association is not proof of prevention
- How to use broad prevention claims responsibly
Relative risk versus absolute risk
Relative risk is a ratio between two groups. A report saying that one group has a 20% lower relative risk means its risk is 80% of the comparison group's risk. Absolute risk tells you how many people experience the outcome.
NIH News in Health explains that relative risk alone can make an effect sound larger than its actual difference in cases. NIH News in Health explains the difference between relative and absolute risk For example, a 20% relative reduction does not mean that 20 out of every 100 people avoid dementia. It means that the reduction equals 20% of the original risk.
How the starting risk changes the result
A U.S. ARIC-based analysis estimated cumulative dementia risk at 4% by age 75 and 20% by age 85. These figures show why the same relative change can create different absolute differences at different ages, according to the National Institute on Aging. The National Institute on Aging reports these age-specific dementia risks Applying a hypothetical 20% relative reduction to a 4% risk by age 75 lowers the risk to 3.2%.
That is an absolute reduction of 0.8 percentage points, or about 8 fewer cases per 1,000 people. Applying the same hypothetical reduction to a 20% risk by age 85 lowers the risk to 16%. That is an absolute reduction of 4 percentage points, or about 40 fewer cases per 1,000 people. The relative claim stays the same, but the number of potentially avoided cases is larger when the starting risk is higher. These are illustrations, not predictions for a particular person.
What population estimates can—and cannot—tell you
The ARIC analysis estimated a 42% lifetime dementia risk after age 55. Most of that risk accrued after age 85, and the estimated lifetime risk was 48% for women and 35% for men, largely because women tend to survive to older ages.
A lifetime estimate describes a population, not an individual's guaranteed outcome. It also does not mean that a person has the same risk at every age or that a particular intervention will produce a matching reduction. When reading a dementia-risk claim, check whether it states: Without those details, a percentage may be technically accurate but difficult to apply to a real decision.
- The starting or baseline risk
- The follow-up period or age range
- The people included in the study
- The relative measure used
- The absolute difference in cases
Why an association is not proof of prevention
A 2025 study of adults aged 51 to 71 found that dementia occurred in 8.9% of people after hearing-disorder onset, compared with 4.11% of people without the disorder. Its adjusted hazard ratio was 1.91, meaning the relative association remained after statistical adjustment. The American Journal of Preventive Medicine study record reports both the event frequencies and hazard ratio Those numbers illustrate why relative and absolute results should be read together.
They also do not prove that hearing loss causes dementia or that treating hearing loss will produce the same difference. The study reported evidence of moderate reverse causation in some analyses. That means early, undetected changes related to dementia may have influenced hearing-disorder timing, rather than hearing loss alone explaining the association.
How to use broad prevention claims responsibly
The 2024 Lancet Commission estimated that 14 potentially modifiable factors accounted for a 45.3% population attributable fraction. That measure describes a modeled share of cases linked to specified factors under causal assumptions; it is not a 45% personal guarantee of prevention. The 2024 Lancet Commission describes the model and its limits A practical response is to focus on manageable health decisions while avoiding promises about personal outcomes.
When evaluating any claim, ask what risk it started with, how long people were followed, and whether the evidence shows an association or a tested preventive effect. A lower relative risk may still matter, especially when the baseline risk is high. But the clearest measure of likely impact is the absolute difference: how many fewer cases occurred among a comparable number of people over a stated period.
