Background Social connections are known to influence morbidity and mortality risk, and have also been associated with adiposity outcomes, including waist circumference, in older adults. However, longitudinal evidence on changes in social ties and changes in waist circumference are lacking in Canada, and potential gender differences are also unknown.

Methods We analyzed data from the Canadian Longitudinal Study on Aging (CLSA) collected at three time points: baseline (2012-15), follow-up 1 (2015-18), and follow-up 2 (2018-21). We assessed changes in marital status, living arrangements, and social participation in relation to changes in waist circumference (WC) by gender. Sample sizes vary by exposure with 13,955 participants for marital transition and living arrangement transition, and 13,936 participants for changes in social participation. Mixed models with random coefficients for province adjusted for age, age squared, follow-up time, baseline waist circumference, BMI, education, household income, home ownership, ethnicity, smoking, and chronic conditions.

Results Our final analytic sample comprised about 47% men. Preliminary results showed gender differences in the direction of association between social tie changes and WC. Compared to remaining partnered, women and gender minorities who experienced divorce had reductions in WC (-0.83 cm [CI95: -2.28, 0.63]), whereas men who divorced had increased WC (0.87 cm [CI95: -0.56, 2.30]). Widowed women and gender minorities had increased WC (0.60 cm [CI95: -0.58, 1.78]), whereas widowed men had decreased WC (-1.05 cm [CI95: -2.61, 0.51]), compared to counterparts remaining partnered. Women and gender minorities who became co-living or lone-living had decreased WC (-0.08 cm [CI95: -1.04, 0.88] and -0.16 cm [CI95: -0.89, 0.56], respectively), but men with changed accommodation had increased WC (1.10 cm [CI95: -0.16, 2.36] and 0.38 cm [CI95: -0.51, 1.28], respectively), compared to counterparts remaining co-living. Reductions in social participation were associated with significant increases in WC for women and gender minorities (0.49 cm [CI95: 0.10, 0.87]) that were double the increases seen in men’s WC (0.22 cm [CI95: -0.17, 0.60]). No gender differences were observed for increases in WC from remaining non-partnered or becoming married, for remaining lone-living, and for increases in social participation, compared to reference groups for both genders.

Conclusion Initial findings indicate that different social tie transitions may be associated with changes in WC among Canadian aging population, and that associations are gendered. Results suggest potential implications for public health interventions and social policies aimed at addressing adiposity-related health outcomes and reducing gender-specific health disparities.