Mid-Life Physical Activity and Late-Life Cognitive Performance among American Indians

Abstract

<b><i>Introduction:</i></b> Research on factors associated with late-life cognitive performance in diverse racial/ethnic groups is increasingly important due to the growing size and racial diversity of the elder population. <b><i>Methods:</i></b> Using data on American Indians (AIs) from the Strong Heart Study, we measured associations between mid-life physical activity (PA), assessed by a questionnaire or pedometer, and performance on tests of general cognitive function, phonemic fluency, verbal learning and memory, and processing speed. Cognitive tests were administered 7–21 years after PA measurements. To estimate associations, we used regression models with and without inverse-probability weights to account for potential attrition bias in the cohort. <b><i>Results:</i></b> Questionnaire and pedometer measures of PA were positively associated with cognitive function. Participants in the top quartile of questionnaire-based PA had Modified Mini-Mental State examination scores 3.2 (95% CI: 1.5–4.9) points higher than participants in the lowest quartile. Phonemic fluency scores also trended higher for participants in the top compared to the bottom categories for both PA measures: top questionnaire quartile = 2.7 (95% CI: 0.6–4.8) points higher and top pedometry tertile = 6.7 (95% CI: 2.7–10.7) points higher. We observed no associations between PA and tests assessing verbal learning and memory, or processing speed. Weighted model results were similar, but less precise. <b><i>Conclusions:</i></b> In this cohort of AIs with relatively low levels of PA, positive associations between mid-life PA and late-life cognitive performance were dose-dependent and of modest clinical significance.