Associations of smoking status and leisure-time physical activity with waist circumference change-10-year follow-up among twin adults.

Abstract

This cohort study investigated the associations of smoking status and leisure-time physical activity (LTPA) with weight circumference (WC) change. In the FinnTwin16 cohort, 3322 twins (46% men) reported smoking status, LTPA, and WC in early adulthood and 10 years later providing information on essential covariates at baseline. The effects of smoking status and LTPA (metabolic equivalent tasks [MET]-h/week) on WC change (cm) were estimated by modeling WC value at the end of follow-up and adjusted for baseline WC in linear regression models. Within-pair associations were analyzed using linear fixed-effect regressions among 660 dizygotic and 390 monozygotic twin pairs. During the 10-year follow-up, 36.4% (n = 273) of baseline daily smokers quit smoking. Among those who quit daily smoking, the mean WC increase was 8.4 cm (SD 8.1). Quitters who smoked daily at baseline increased WC by about 2 cm more than continuing smokers (adjusted β 2.04; 95% CI 0.94, 3.14). This association was not robust after shared familial influences were controlled for. In general, the participants decreased LTPA during follow-up, except the quitters with the mean LTPA increase of 5.0 MET-h/week (SD 35.0). Independently of smoking status, each additional MET-h/week was associated with 0.06 cm less WC increase (adjusted β -0.06; 95% CI -0.07, -0.05). This association was replicated in the within-pair analyses. Smoking cessation seems to be associated with WC increase, but familial confounding may be involved in this process. LTPA appears to mitigate increase in WC independently of smoking status and familial influences.