Longitudinal body weight dynamics in relation to cognitive decline over two decades: A prospective cohort study

The aim of this study was to investigate the associations of body weight change (BWC) and body weight variability (BWV) with changes in cognitive function.

In 10,340 Health and Retirement Study participants (mean age: 68.0 years), body weight was reported biennially from 1993/1994 to 2016, and cognitive function was measured biennially from 1998 to 2016. We calculated BWC and BWV as the slope and root-mean-square error by regressing body weight on time for each individual. BWC was categorized by quintiles (Q): stable weight (Q2 to Q4), weight loss (Q1), and weight gain (Q5). BWV was categorized by tertiles. We used linear mixed regression models to assess associations with cognitive change.

Compared with stable weight (median: 0 kg/y), weight loss (median: -1.3 kg/y) predicted faster cognitive decline as demonstrated by mean difference of -0.023 (95% CI: -0.027 to -0.019) in cognitive change z score per year, whereas weight gain (median: 1 kg/y) was related to slower cognitive decline (β = 0.006; 95% CI: 0.003 to 0.009). Larger BWV was also associated with faster cognitive decline (β comparing the top with bottom tertile = -0.003; 95% CI: -0.006 to -0.0002). Similar associations were observed for episodic and working memory.

Weight loss and large BWV over a long time independently predicted faster cognitive decline in middle-aged and older adults, underscoring the importance of long-term dynamic body weight monitoring.