The 45% estimate concerns the share of dementia cases in a population that might theoretically be prevented or delayed. It does not mean one person can cut their lifetime dementia risk by 45%. The estimate is a population attributable fraction, or PAF. This model combines how common several risk factors are with how strongly each is associated with dementia.
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
- What is the denominator?
- Which risk factors are included?
- Why the percentages cannot simply be added
- Is 45% the same as lifetime dementia risk?
- How should readers use the estimate?
What is the denominator?
The denominator is all dementia cases in a population. The 2024 Lancet Commission estimated that 45.3% of cases could theoretically be prevented or delayed if 14 potentially modifiable risk factors were eliminated, according to the Commission's report. That percentage does not describe the probability that an individual will develop dementia.
It also does not mean that 45% of dementia has one preventable cause. Imagine a population expected to experience 1,000 dementia cases under current conditions. A 45.3% PAF describes a modeled difference under the unrealistic scenario that all 14 exposures disappear. It does not predict that a particular person can erase 45.3 percentage points from their own risk.
Which risk factors are included?
The 14 factors span different stages of life. They include less education; hearing loss; high LDL cholesterol; depression; traumatic brain injury; physical inactivity; smoking; diabetes; hypertension; obesity; excessive alcohol; social isolation; air pollution; and untreated vision loss. The life-stage framing matters. Less education appears earlier in life.
Many cardiovascular, health, and injury-related factors are assigned to midlife, while social isolation, air pollution, and untreated vision loss appear later. High LDL cholesterol and hearing loss each received a 7% weighted PAF. Less education and social isolation each received 5%. These are modeled population contributions, not personal risk reductions available to everyone.
Why the percentages cannot simply be added
Risk factors often occur together. Someone with diabetes may also have hypertension, obesity, or inactivity, while hearing or vision problems may contribute to isolation. The Commission therefore adjusted for clustering using information from 37,000 Norwegian HUNT-study participants aged 45 or older.
Five statistical components explained 54% of the variation among the exposures, showing substantial overlap. This adjustment prevents the model from treating every factor as fully independent. Removing hearing loss and lowering LDL cholesterol, for example, should not be interpreted as a guaranteed 14% reduction for one person.
Is 45% the same as lifetime dementia risk?
No. Population-attributable estimates and lifetime-risk estimates answer different questions. A separate U.S.
study estimated that 42% of adults who were dementia-free at age 55 would develop dementia by age 95 after accounting for death as a competing risk. Estimated risks were about 45% to 60% among women, Black adults, and APOE ε4 carriers, according to the Nature Medicine report from ARIC investigators. The numerical similarity between 42% and 45.3% is coincidental. One estimate concerns the modeled preventable share of population cases; the other concerns the probability of developing dementia across the remaining lifetime of a defined group.
How should readers use the estimate?
Treat the figure as a case for broad prevention, not a personal guarantee. The evidence supports attention across education, cardiovascular health, sensory care, injury prevention, social conditions, and air quality, consistent with the WHO risk-reduction guidelines.
A practical response is to: The estimate remains uncertain. The Lancet Commission cautions that some associations may be only partly causal, risk magnitudes may be overestimated or underestimated, and global evidence comes disproportionately from high-income countries.
- Ask a clinician which cardiovascular risks are relevant to you.
- Address hearing or vision problems instead of assuming they are harmless features of aging.
- Discuss physical activity, smoking, alcohol use, diabetes, blood pressure, cholesterol, and weight in the context of your health.
- Consider injury prevention and opportunities for social connection.
- Support community measures involving education, accessible sensory care, cleaner air, and healthier living conditions.
