Open Access Open Badges Research

The impact of waist circumference on function and physical activity in older adults: longitudinal observational data from the osteoarthritis initiative

John A Batsis123*, Alicia J Zbehlik2345, Laura K Barre356, Todd A Mackenzie235 and Stephen J Bartels235

Author Affiliations

1 Section of General Internal Medicine, Dartmouth-Hitchcock Medical Center, 1 Medical Center Drive, Lebanon, NH 03756, USA

2 Centers for Health and Aging, 46 Centerra Parkway, Lebanon, NH 03766, USA

3 Geisel School of Medicine at Dartmouth, 1 Rope Ferry Road, Hanover, NH 03755, USA

4 Section of Rheumatology, Dartmouth-Hitchcock Medical Center, 1 Medical Center Drive, Lebanon, NH 03756, USA

5 The Dartmouth Institute, Dartmouth College, 30 Lafayette St, Lebanon NH 03766, USA

6 Currently address at Division of Nutritional Sciences, School of Human Ecology, Cornell University, Ithaca, NY 14850, USA

For all author emails, please log on.

Nutrition Journal 2014, 13:81  doi:10.1186/1475-2891-13-81

Published: 9 August 2014



We previously demonstrated that BMI is associated with functional decline and reduced quality of life. While BMI in older adults is fraught with challenges, waist circumference (WC) is a marker of visceral adiposity that can also predict mortality. However, its association with function and quality of life in older adults is not well understood and hence we sought to examine the impact of WC on six-year outcomes.


We identified adults aged ≥60 years from the longitudinal Osteoarthritis Initiative and stratified the cohort into quartiles based on WC. Our primary outcome measures of function at six year follow-up included: self-reported quality of life [Short Form-12 (SF-12)], physical function [Physical Activity Scale for the Elderly (PASE)] and disability [Late-life Disability Index (LLDI)]. Linear regression analyses predicted 6-year outcomes based on WC quartile category (lowest = referent), adjusted for age, sex, race, education, knee pain, smoking status, a modified Charlson co-morbidity index and baseline scores, where available.


We identified 2,182 subjects meeting our inclusion criteria and stratified the study cohort by quartiles of WC. Mean age ranged from 67.5-68.7 years, 60-71% were female and 80-86% were white. The highest WC quartile compared to 50-75th, 25-50th or lowest quartile, was associated with a greater number of medications (4.3, 4.0, 3.6 and 3.4 [p < 0.001]), lower gait speeds (1.23, 1.27, 1.32, and 1.34 m/s[p < 0.001]), higher rates of knee osteoarthritis (70.2, 62.2, 60.2, 48.6;p < 0.001), higher Charlson co-morbidity scores and greater knee pain (WOMAC scores) (all p < 0.001). At follow-up, adjusted SF-12 physical function subscale and PASE scores, were lowest in the highest WC quartile as compared to the 50-75%, 25-50%, and lowest quartiles [(SF-12 scores: 45.5, 46.7, 47.6, and 47.9), and (PASE scores: 109.6, 128.7, 126.6, and 131.0). The LLDI limitation subscale for disability demonstrated lower scores in the high WC quartile as opposed to the referent group.


Elevated WC is associated with lower quality of life, a decline in physical function, and a slightly higher risk of disability over time. Intervention studies are needed to prevent functional decline in this high-risk population.

Obesity; Waist circumference; Osteoarthritis; Disability; Physical activity; Quality of life