Objectives: To determine whether older women who maintain optimal cognitive function into old age differ from those who experience minor cognitive decline typically associated with normal aging.
Design: Prospective cohort study.
Setting: The Study of Osteoporotic Fractures.
Participants: Nine thousand seven hundred four older women.
Measurements: A modified Mini-Mental State Examination (mMMSE) was performed at baseline and Years 6, 8, 10, and 15. Random-effects regression was used to classify subjects as cognitive maintainers (slope>/=0), minor decliners (slope < 0 but > lowest tertile), or major decliners (slope <or= lowest tertile). Stepwise logistic regression was used to identify factors most predictive of being a cognitive maintainer versus a minor decliner (excluding major decliners).
Results: Women had a mean age of 72 at baseline and 85 at follow-up. Nine percent maintained optimal cognitive function, 58% experienced minor decline, and 33% experienced major decline. Most factors differed progressively over the three cognitive groups. After adjustment for key confounders, odds ratios for factors most predictive of being a cognitive maintainer as opposed to minor decliner were 1.9 (95% confidence interval (CI)=1.2-2.9) for lack of diabetes mellitus, 1.2 (95% CI=1.0-1.4) for lack of hypertension, 1.7 (95% CI=1.3-2.3) for lack of smoking, 1.2 (95% CI=1.1-1.5) for moderate alcohol consumption, 1.4 (95% CI=1.1-1.7) for lack of difficulty with instrumental activities of daily living, and 1.2 (95% CI=1.0-1.4) for lack of low social network.
Conclusion: Almost 10% of older women maintained optimal cognitive function into old age. Cognitive maintainers were less likely to have comorbid medical conditions, less likely to have difficulty with daily activities or poor social networks, and more likely to engage in healthy behaviors than minor cognitive decliners.