Physical Limitations Contribute to Food Insecurity and the Food Insecurity–Obesity Paradox in Older Adults at Senior Centers in Georgia

Abstract

The purpose of this study was to explore the relationship of obesity and physical limitations with food insecurity among Georgians participating in the Older Americans Act (OAA) congregate meal-site program (N = 621, median age = 76 years, 83% female, 36% Black, and 64% White, convenience sample). Food insecurity was assessed using the modified 6-item US Household Food Security Survey Module; obesity was defined as Body Mass Index (BMI) or waist circumference (WC) class I or II obesity; and physical limitations (arthritis, joint pain, poor physical function, weight-related disability) were based on the Disablement Process. A series of multivariate logistic regression models found weight-related disability and obesity (WC class II) may be potential risk factors for food insecurity. Thus, obesity and weight-related disability may be risk factors to consider when assessing the risk of food insecurity and the need for food assistance in this vulnerable subgroup of older adults.

KEYWORDS:

• BMI
• food insecurity–obesity paradox
• older adults
• physical function
• physical limitations
• waist circumference
• weight-related disability

This research project was supported by Georgia Division of Aging Services, the 12 Georgia Area Agencies on Aging, the Department of Foods and Nutrition and the Georgia Agricultural Experimentation, University of Georgia (HATCH #GEO 00576), and a Doctoral Dissertation Award, University of Georgia to Dr. Dawn Brewer.

Notes

N = 621.

N = 621.

p ≤ 0.05 is considered statistically significant.

¹All models included gender and education, but these were not significantly associated with food insecurity, and thus are not shown.

²Model 1: logistic regression analyses that included age, gender, race/ethnicity, education, and obesity. ***p < 0.001, **p < 0.01, *p < 0.05.

³Model 2: logistic regression analysis that included all model 1 variables, and arthritis, joint pain, physical function, disability, and self-reported health. ***p < 0.001, **p < 0.01, *p < 0.05.

⁴Model 3: forward stepwise logistic regression analysis that included for selection all model 2 variables; only the OR for variables that met the criteria for retention in the models are shown (p ≤ 0.05).