gms | German Medical Science

Associations between physical activity and age-related cognitive decline are documented. There are also promising findings about positive effects of cardiovascular fitness on cognitive performance. However, cross-sectional or intervention studies with short-term treatment and selective samples predominate. In addition, physical fitness is mainly limited to cardiorespiratory fitness. Gaps remain on longitudinal studies in representative population including exercise behaviors, and in differential aspects of fitness. Therefore, the primary aim of this study is to examine whether the level and the rate of change of cognitive functions varied across older persons. The second aim is to determine whether these intercepts and trajectories differ by change of physical fitness and change of exercise behavior.

Data were collected from 216 older persons (born 1930-32; 48% women) of the Interdisciplinary Longitudinal Study of Adult Development in 1997. Follow-up was 4 years later. Cognitive performance were assessed including executive function (ZST; digit-symbol substitution test), information processing speed (ZVT; trail making), attention (d2), and memory (digits recall). Fitness was operationalized using measures of endurance, hand grip strength, balance and flexibility (as a factor score and single parameters). 4 categories of exercise behavior were considered: continuously active and inactive persons, beginners, and breakups within 4 years. To examine intraindividual change and stability trajectories of cognitive function were estimated by using individual growth modeling. The results were presented for men and women separately.

The results showed differences in all cognitive tasks. Age-related differences were only found in digits recall and d2 in men, and ZST in women (all p<.05). Exercise behavior significantly predicts level of d2 (t=-2.87; p<.01) and ZST (t=-3.36; p<.01) in women. Inactive women show the lowest level of attention and of executive function than their counterparts. The change of physical fitness is not a significant predictor, neither the factor score nor the single parameters.

Intraindividual differences in the rate of change exist in trail making in men and women and in d2 in women (all p<.001). Change in parameters of physical fitness and change of exercise behavior did not significant predict the differences in rate of change.

These findings suggest that change in sport behavior and change in physical fitness can not explain change in cognitive performance. However, there are indices that change in sport behavior (i.e. inactivity) predict lower level in some cognitive tasks.