Variation in Body Weight Increases Risk for Dementia in Adulthood

older man thinking
asian elder man wory something and look through window
High body weight variability may be associated with increased risk for all-cause dementia and Alzheimer dementia in older patients.

High body weight variability, measured as the repeated loss and regain of weight within a specific period, may be associated with increased risk for all-cause dementia and Alzheimer dementia in older patients, according to study results published in Frontiers in Endocrinology.

Variability in body weight was previously found to be associated with increased mortality risk. The goal of the current study was to explore the association between visit-to-visit body weight variability and risk for dementia in elderly patients.

Visit-to-visit body weight variability was determined from ≥3 body weight measurements for each individual and intra-individual variability was measured using 3 indices: variability independent of the mean, coefficient of variation, and standard deviation.

The researchers used data from the National Health Insurance Service (NHIS), a mandatory social health insurance program that enrolls approximately 98% of Koreans. Of 107,367 individuals who underwent health examination in 2009, the study included 19,987 eligible patients (mean age, 73.1 years).

During a median follow-up period of 6.5 years, 1592 patients (7.97%) were diagnosed with all-cause dementia, including 1217 (6.09%) with Alzheimer dementia and 304 (1.52%) with vascular dementia.

The risk for all-cause dementia was significantly higher in patients with the highest body weight variability. Patients in the highest quartile of body weight variability independent of the mean had a 39% increased risk for all-cause dementia compared with those in the lowest quartile (adjusted hazard ratio [HR], 1.390; 95% CI, 1.206-1.603). As for dementia subtypes, the association between body weight variability with the risk for Alzheimer disease was more significant (adjusted HR, 1.462; 95% CI, 1.240-1.724), whereas the risk for vascular dementia was not significant (HR, 1.204; 95% CI, 0.880-1.647).

Interestingly, the risk for dementia was approximately 2-fold higher among underweight patients in the highest quartile of body weight variability independent of the mean (HR, 1.994; 95% CI, 1.302-3.054), whereas in patients with obesity, lower body weight variability was associated with decreased risk for dementia compared with individuals with normal body mass index in the lowest weight variability quartile (body weight variability independent of the mean quartile 1: HR, 0.664 [95% CI, 0.505-0.872]; quartile 2: HR, 0.648 [95% CI, 0.493-0.852]).

Subgroup analyses showed a stronger association between body weight variability and dementia in people aged <75 years compared with older patients.

The researchers acknowledged several study limitations, including missing data on educational level, dietary factors, or the apolipoprotein E e4 allele; establishing diagnosis of dementia solely on disease codes and not confirming them with brain images; and the relatively low incidence of vascular dementia.

Related Articles

“We for the first time demonstrated that [body weight] variability in the elderly was significantly associated with the increased risk for dementia after adjusting other covariates, suggesting that the inability to maintain one’s weight might be an indication for poorer homeostatic control and general condition,” noted the researchers, adding that “…the close relation between [body weight] variability in late-life and dementia was independent from preclinical weight loss of dementia or age.”

“Considering that [body weight] fluctuations in older persons are not [a] benign condition, monitoring the weight of an older person for fluctuations as well as episodes of weight loss might be an important aspect of geriatric care,” they concluded.

Follow @EndoAdvisor


Roh E, Hwang SY, Kim JA, et al. Body weight variability increases dementia risk among older adults: a nationwide population-based cohort study [published online May 12, 2020]. Front Endocrinol (Lausanne). doi:10.3389/fendo.2020.00291