Methods:

This was a population-based cohort study from three French cities (the Three-City Study). Subjects were enrolled, starting in 1999, with baseline ascertainment of sociodemographic, lifestyle, medication, anthropometric, laboratory, and neuropsychological data. Repeat cognitive testing occurred every 2-3 years to evaluate for cognitive decline and incident dementia. Ascertainment of dementia stopped in 2016. Subjects who screened positive for cognitive impairment were evaluated by a neurologist and dementia diagnoses were adjudicated by a blinded committee of neurologists using the Diagnostic and Statistical Manual of Mental Disorders criteria. Cognitive decline was defined using neuropsychological performance on tests of global cognition and memory. Cardiovascular health was categorized by poor, intermediate, or optimal adherence to the American Heart Association’s (AHA’s) Simple 7s. These metrics include healthy behaviors (physical activity and body weight, not smoking, and health eating) as well as biologic measurements (blood pressure, total cholesterol, and blood glucose). The AHA measures were adapted to fit the data collected in the study and subjects were given 0, 1, or 2 points for each metric at poor, intermediate, or optimal levels, respectively. The scores were summed to create a global cardiovascular health score, which ranged from 0-14. Analyses were adjusted for sex, education, and apolipoprotein E ε status.