Association between Smoking Status and Obesity in a Nationwide Survey of Japanese Adults

Abstract

Objective: A positive association between the number of cigarettes smoked per day and obesity has been reported, whereas how other smoking-related indices, such as pack-years and duration of smoking, are related with obesity has been less investigated. We analyzed the age-adjusted cross-sectional association between smoking and obesity in a general Japanese population.

Methods: We used data from a nationwide epidemiological study of Japanese adults (N = 23,106). We compared the prevalence of obesity (defined as body mass index ≥ 25kg/m2) among groups classified by smoking behavior, pack-years, number of cigarettes per day, duration of smoking, and duration and time of smoking cessation.

Results: In men, current smokers had a lower odds ratio (OR) for obesity of 0.80 (95% confidence interval (CI): 0.72-0.88) compared to non-smokers, whereas past smokers had a higher OR of 1.23 (95% CI: 1.09-1.37) compared to current smokers. In women, there were no differences in obesity between the three groups classified by smoking behavior. However, in both sexes, the prevalence of obesity tended to increase with pack-years and the number of cigarettes per day, but not with duration of smoking in current and past smokers. Further, in male smokers, the risks for obesity were markedly higher in short-term heavy smokers compared with long-term light smokers, even with the same number of pack-years. Regarding the impact of smoking cessation, female past smokers who quit smoking at an age > 55-years had an elevated OR of 1.60 (95% CI:1.05-2.38) for obesity.

Conclusions: In a general Japanese population, obesity is progressively associated with pack-years and number of cigarettes per day, but not with the duration of smoking. When investigating the association between obesity and cigarette smoking, the daily smoking burden and the duration of smoking require to be independently considered.

Fig 1. Association between age and the prevalence of obesity by smoking status in men and women.

Data were determined using the second order logistic regression model with intercept, linear, and quadratic terms: logit(P) = intercept +αAge +β(Age-mean Age)2.

Fig 2. Age-adjusted odds ratios (OR) for obesity of the male smokers by the number of cigarette per day (CPD) and the duration of smoking.

Age-adjusted ORs for obesity of the categories stratified by both CPD and duration of smoking in male current smokers and male past smokers. CPD was stratified by ≤ 10, 11–20, and >20. Duration of smoking was stratified by <10 years, 10–34 years, and ≥35 years.