Social Support, Gender and Central Obesity: A Population-Based Secondary Analysis Study of 28,779 Adults From the Canadian Longitudinal Study on Aging

Abstract Objectives The well-established link between social isolation and mortality could occur through metabolic effects of poor social support, however research rarely considers different types of social support or effect modification by gender, which we address in this study. Methods Baseline Canadian Longitudinal Study on Aging (CLSA, 2012–2015) Comprehensive cohort data on 28,779 adults (45–85y) provided multiple measures of social support and anthropometric variables. Multiple regression and stratification assessed the association between four support types (informational, emotional, tangible, belonging) and waist circumference (WC) or odds of central obesity (females, WC ≥ 88 cm; males, WC ≥ 102 cm), adjusting for age, age2, education, smoking, and province in all models. Results Multivariable analyses indicated that average levels of WC were higher among women with lowest, compared to highest, levels of informational (1.89 cm (1.29, 2.49), P < 0.001), belonging (1.54 cm (0.95, 2.14), P < 0.001), emotional (1.47 cm (0.86, 2.07), P < 0.001), or tangible (1.18 cm (0.6, 1.76), P < 0.001) support. Mean WC was only higher among men reporting lowest levels of informational support (+0.74 (0.19, 1.29), P < 0.01), compared to highest. Adjusting for confounders, higher odds of central obesity was associated with a lack of informational, emotional, and tangible support among women (OR range: 1.10–1.23) and men (OR range: 1.10–1.22); central obesity was inversely linked to belonging support in women only (OR 1.20 (1.09, 1.31), P < 0.001). Conclusions Efforts to lower WC among older Canadians should include access to multiple forms of social support, as different supports may not benefit women and men equally. Funding Sources Our secondary analysis study used the CLSA Baseline Comprehensive Dataset version 4.0, Baseline, under Application Number 19CA003. We acknowledge our Canadian Institute of Health Research grant. We also acknowledge the CLSA cohort study (PIs: Drs. Parminder Raina, Christina Wolfson and Susan Kirkland) that is supported by the Government of Canada through the Canadian Institutes of Health Research under grant reference: LSA 94473 and the Canada Foundation for Innovation.