Abstract
A substantial proportion of dementia risk may be attributable to modifiable factors, yet these are often examined in isolation despite their interrelated nature and tendency to co-occur. It remains unclear whether the relationship between modifiable factors and dementia risk is influenced by individual characteristics such as sex and genetic susceptibility. We investigated longitudinal associations between the Lifestyle for Brain health (LIBRA) score and risk of dementia, cognitive performance, and brain structure, and whether relationships differed by sex and APOE ɛ4 carrier status.Participants were aged > 50 years, dementia-free at baseline, 50% female and predominantly (97%) white/Caucasian. The LIBRA score included 11 modifiable factors (e.g., hypertension, obesity, physical inactivity). Magnetic resonance imaging estimated brain volume, domain-specific cognitive composite scores were calculated, and dementia diagnoses were determined based on self-reported and linked healthcare data.Across a mean follow-up of 10.2 years, a higher LIBRA score was associated with greater odds of developing dementia (OR = 1.20, 95% CI 1.18-1.22). This association was stronger in APOE ɛ4 non-carriers compared to ɛ4 carriers. Cross-sectionally, higher LIBRA scores related to poorer cognition, smaller whole-brain gray and white matter volumes, and increased ventricular cerebrospinal fluid (CSF), however, only the association with increased ventricular CSF persisted longitudinally (mean follow-up 3.4 years).Each one-point increase on the LIBRA score was associated with 20% increased odds of developing dementia. These results reinforce the need to target modifiable dementia risk factors and to tailor dementia prevention strategies to individual risk profiles to maximize the impact on brain health.