Association of current and former smoking with body mass index: A study of smoking discordant twin pairs from 21 twin cohorts.

Background: Smokers tend to weigh less than never smokers, while successful quitting leads to an increase in body weight. Because smokers and non-smokers may differ in genetic and environmental family background, we analysed data from twin pairs in which the co-twins differed by their smoking behaviour to evaluate if the association between smoking and body mass index (BMI) remains after controlling for family background.
Methods and Findings: The international CODATwins database includes information on smoking and BMI measured between 1960 and 2012 from 156,593 twin individuals 18-69 years of age. Individual-based data (230,378 measurements) and data of smoking discordant twin pairs (altogether 30,014 pairwise measurements, 36% from monozygotic [MZ] pairs) were analysed with linear fixed-effects regression models by 10-year periods. In MZ pairs, the smoking co-twin had, on average, 0.57 kg/m2 lower BMI in men (95% confidence interval (CI): 0.49, 0.70) and 0.65 kg/m2 lower BMI in women (95% CI: 0.52, 0.79) than the never smoking co-twin. Former smokers had 0.70 kg/m2 higher BMI among men (95% CI: 0.63, 0.78) and 0.62 kg/m2 higher BMI among women (95% CI: 0.51, 0.73) than their currently smoking MZ co-twins. Little difference in BMI was observed when comparing former smoking co-twins with their never smoking MZ co-twins (0.13 kg/m2, 95% CI 0.04, 0.23 among men; -0.04 kg/m2, 95% CI -0.16, 0.09 among women). The associations were similar within dizygotic pairs and when analysing twins as individuals. The observed series of cross-sectional associations were independent of sex, age, and measurement decade.
Conclusions: Smoking is associated with lower BMI and smoking cessation with higher BMI. However, the net effect of smoking and subsequent cessation on weight development appears to be minimal, i.e. never more than an average of 0.7 kg/m2.
Competing Interests: TK reports personal consultation fees from Pfizer Finland, during the conduct of the study but outside of this study. JK reports grants from the Academy of Finland and the Sigrid Juselius Foundation, during the conduct of the study, and personal fees from Pfizer Inc., outside the submitted study. AB reports working in the HealthTwiSt GmbH but the company did not have any impact on this study. FI reports grants from JSPS KAKENHI JP (23593419,16K15978), during the conduct of the study. MG reports grants from the National Institutes of Health, during the conduct of the study, and other grants from National Institutes of Health, outside the submitted work period. All other authors have nothing to declare. None of the funders played a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials. None of the commercial affiliations affects our adherence to PLOS ONE policies on sharing data and materials. The specific roles of authors with commercial affiliations (AB, JK, TK) are articulated in the 'author contributions' section. The corresponding author had full access to all of the data in the study and had final responsibility for the decision to submit the paper for publication.