The Modifiable Share of Dementia Risk
Commission estimates put a substantial fraction of cases in reach of prevention. The estimate is a ceiling, not a promise.
Dr. Elena Vance
Chief of Research
Sep 19, 2026 • 11 min

What the data says
~40%
Share of dementia cases attributed to modifiable risk factors in Lancet Commission population estimates.
Source: Lancet Commission on dementia prevention
The commission's figure is a population attributable fraction: the share of cases that would not occur if the listed exposures were entirely removed. Nobody removes them entirely, so treat it as an upper bound.
The listed factors are unglamorous and overlap heavily with cardiovascular risk: hearing loss, hypertension, smoking, physical inactivity, diabetes, alcohol, depression, social isolation, air pollution, and education in early life.
Hearing loss is the one most often missed. Treating it is cheap relative to its attributed share, and the trial evidence, while mixed, is more encouraging in higher-risk groups.
No supplement appears on that list, which is itself informative about where the evidence sits.
Editorial note: this article summarises published research for general information. It is not medical advice. Discuss any intervention with a qualified clinician.


