Accelerometer-Assessed Physical Activity and Cognitive Performance among European Adults Aged 50+: The Mediating Effects of Social Contacts and Depressive Symptoms

Abstract

Background: Cognitive decline is a major public health concern worldwide and it is vital to identify and better understand effective population-based means to improve cognitive performance in old age. The current study set out to examine the links between accelerometer-based physical activity with cognitive performance in later life, as well the indirect pathways through one's social network contacts and depression.

Method: We used data from 855 participants aged 50 and above who took part in a cross-sectional accelerometer study as part of the Survey of Ageing, Retirement and Health (SHARE). Cognitive function was measured as an average score of fluency, immediate and delayed recall tests, social contacts were the average contact frequency with members of the social support network, and depression was the Euro-D summary score of depressive symptoms. A multiple mediation analysis was conducted to test the direct and indirect associations between total physical activity (intensity gradient) and cognitive function, as well as the mediation of this association by social contacts and depressive symptoms.

Results: Intensity of physical activity was directly related to better cognitive performance (B = 0.170, p = 0.007). The association was partially mediated by social contacts (B = 0.022, 95% CI 0.005, 0.046) and depressive symptoms (B = 0.009, 95% CI 0.009, 0.025), such that total physical activity was linked to cognitive health via more frequent contacts with network members and low depressive symptoms.

Conclusions: Practitioners might consider encouraging a physically active lifestyle that involves social interactions to support better cognitive aging and mental health.

Keywords: accelerometer-based physical activity; adults; cognitive health; depression; intensity gradient; social contacts.

Figure 1

The main study model—a multiple mediator model depicting the direct effect without mediators (weight C) and with mediators (weight C′); and the indirect effects (sum of all a × b weights) of the intensity gradient via the two mediators (contact with the social network and depressive symptoms), controlling for the covariates.