Body weight trajectories from midlife are associated with cognitive decline in advanced age

Abstract

Fluctuations in body weight may impact cognitive decline, but current evidence is inconclusive. The aim of this study is to investigate associations between body weight trajectories from midlife to later life and cognitive decline. This retrospective study analyzed harmonized data from two population-based longitudinal studies, the Progetto Veneto Anziani and the Italian Longitudinal Study of Aging, encompassing baseline and two follow-up assessments over 9 years. Weight changes were recorded from baseline to the last available follow-up or from 50 years (self-reported data) to the last available follow-up. Cognitive function was assessed using the Mini-Mental State Examination (MMSE), and cognitive decline was defined as experiencing a MMSE change from baseline to the follow-up within the lowest quartile of the change distribution in the total sample. In a sample of 3852 individuals (46% females, age 65-96 years at baseline), we investigated the impact of weight change on cognitive decline with two sets of analyses. First, using weight measurements obtained during old age, growth mixture modelling identified three weight trajectories: decreasing, stable, and increasing. None of these trajectories was significantly associated with cognitive decline. Second, we considered weight at age 50 as the baseline assessment to capture weight changes from midlife. Among the three trajectories detected (increasing, stable, and decreasing), the decreasing trajectory was significantly associated with a higher likelihood of cognitive decline in males (HR 1.44, 95% CI 1.06-1.94) and females (HR = 1.37, 95%CI 1.23-1.67), whereas the increasing trajectory was associated with cognitive decline only in females (HR = 1.33, 95%CI 1.01-1.76). These results suggest that changes in body weight from middle to older age are associated with cognitive decline in advanced age. Since body weight is influenced by multiple factors, a broader assessment of health-including metabolic, vascular, behavioral, and social dimensions-should be considered in both research and clinical settings.

Keywords: Body weight; Cognitive decline; Dementia; Pooled cohorts; Trajectories analysis.

Fig. 1

This figure describes weight trajectories identified from baseline to the follow-up, by sex. Value “0” for time (years) axes represent the baseline. Females: BIC-16601.13; probability of “decreasing group” membership (average posterior probability, as a measure of entropy) 12.7% (95% CI 11.4–14%), probability of “stable group” membership 78.8% (95% CI 77.3–80.4%), probability of “increasing group” membership 8.4% (95% CI 7.4–9.5%). Males: BIC-13369.51; probability of “decreasing group” membership (average posterior probability, as a measure of entropy) 7.5% (95% CI 7–9.1%), probability of “stable group” membership 78.5% (95% CI 77–80.1%), probability of “increasing group” membership 13.5% (95% CI 12.2–14.8%). Mean body weight values for: decreasing group (males: 76.4 kg at baseline, 67.2 kg at the first follow-up and 66.7 kg at the second; females: 68.6 kg at baseline, 59.6 kg at the first and 58.6 kg at the second follow-up); stable group (males: 74.4 kg at baseline, 74.6 kg at the second follow-up; females: 65.9 kg and 66 kg, respectively); increasing group (11.9% for males, 7.9% for females) had an increasing weight that, in particular for females, tended to stabilize after the first follow-up (for males: 68.9 kg at the baseline, 75.1 kg at the first follow-up and 78.9 kg at the second; for females, 60.4 kg at the baseline, 68 kg at the first follow-up and 68.6 kg at the second follow-up).

Fig. 2

This figure shows weight trajectory identified from age 50 to the second follow-up. Value “0” for time (years) axes represent the baseline. Females: BIC-25120.76; probability of “decreasing group” membership (average posterior probability, as a measure of entropy) 35.9% (95% CI 34.3–37.5%), probability of “stable group” membership 9.8% (95% CI 8.7–10.8%), probability of “increasing group” membership 54.3% (95% CI 52.7–55.9%). Males: BIC-19575.88; probability of “decreasing group” membership (average posterior probability, as a measure of entropy) 14.1% (95% CI 12.8–15.3%), probability of “stable group” membership 64% (95% CI 62.4–65.7%), probability of “increasing group” membership 21.9% (95% CI 20.4–23.4%).