Arigo D, Ainsworth MC, Pasko K, Brown MM, Travers L. Predictors of change in BMI over 10 years among midlife and older adults: Associations with gender, CVD risk status, depressive symptoms, and social support.
Soc Sci Med 2021;
279:113995. [PMID:
33993009 PMCID:
PMC8393364 DOI:
10.1016/j.socscimed.2021.113995]
[Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Revised: 02/10/2021] [Accepted: 04/30/2021] [Indexed: 10/21/2022]
Abstract
RATIONALE
Change in BMI is recognized as a key health indicator among midlife and older adults, though predictors of BMI change in this group have received little attention.
OBJECTIVE
The aim of this study was to examine relations between hypothesized predictors (i.e., gender, cardiovascular disease [CVD] risk status, depressive symptoms, social support) and BMI change over 10 years, among midlife and older adults.
METHODS
Participants were adults ages 50-74 at baseline (N = 5,688, 64% women) who completed four assessments over 10 years. Gender, CVD risk status (i.e., diagnosis of hypertension, type 2 diabetes, or both), depressive symptoms, and perceived social support were assessed at baseline, and BMI was calculated from height and weight reports at all assessments. Multilevel models tested for concurrent and prospective relations between predictors and BMI change (effect size estimates as semipartial correlation coefficients, sr), as well as whether observed relations were further moderated by baseline BMI category (underweight, healthy weight, overweight, or obese).
RESULTS
Baseline BMI was higher among those with (vs. without) CVD risk, higher (vs. lower) depressive symptoms, and lower (vs. higher) social support; all of these relations were moderated by gender (ps < 0.05, srs 0.03-0.32). Moreover, BMI showed significant change over 10 years, and BMI variability during this time was higher among women (vs. men) and those with (vs. without) CVD risk (ps < 0.0001). BMI change also differed by CVD risk status, and this relation was moderated by gender, baseline depressive symptoms, and baseline BMI category (ps < 0.05, srs 0.03-0.08).
CONCLUSIONS
Although the predictors of interest were not associated with steady BMI decreases (which are associated with long term health risks for older adults), findings reveal unique patterns of change in BMI among subgroups of midlife and older adults, and may allow for early identification of those with noteworthy BMI changes after age 50.
Collapse