Abstract Body (Do not enter title and authors here): INTRODUCTION:
Research has suggested that 41% of US dementia cases may be prevented by eliminating several specific modifiable risk factors. Still, few studies examine risk factor variation between midlife and late-life. Previous studies have also been limited by sample diversity and size. The Dementia Risk Prediction Project (DRPP) is a consortium of longitudinal cohorts including several underrepresented racial/ethnic groups pooled at the individual level. We aimed to quantify the contribution of eight midlife (45-64 years of age) and six late-life (≥65 years of age) modifiable risk factors to dementia in the US population through calculating population attributable fractions (PAF) and potential impact fractions (PIF).
METHODS:
In a sample of 41,432 unique participants, risk factor exposure was measured in midlife for 29,983 participants (60% female, 77% white) and in late-life for 30,633 participants (56% female, 70% white). During follow-up, 5,667 unique participants (4,186 midlife; 4,893 late-life) developed dementia. We utilized cause-specific Cox proportional hazard models and nationally representative risk factor prevalences to estimate PAF and PIF.
RESULTS:
Eliminating midlife risk factors could prevent a greater proportion of dementia cases [PAF 38.2%, 95% CI 22.79%–53.66%] compared to late-life risk factors [PAF 14.62%, 95% CI -2.55%–31.78%). Midlife physical inactivity was the greatest contributor (PAF 12.7%, 95% CI 9.8%–15.6%). Given the US prevalence of dementia in 2020, we estimate that a 15% reduction in midlife physical inactivity alone may prevent 100,456 (PIF 1.91%, 95% CI 1.4%–2.5%) dementia cases.
DISCUSSION:
We found that the contribution of modifiable risk factors to dementia varies with age, highlighting the importance of appropriately-timed preventive interventions.