Kicking the habit may add centimeters to your waist, but this twin study reveals that physical activity is the key to staying trim after quitting. Discover how genes, lifestyle, and movement shape your post-smoking health.
Study: Associations of smoking status and leisure-time physical activity with waist circumference change—10-year follow-up among twin adults. Image Credit: New Africa / Shutterstock
In a recent study published in the International Journal of Obesity, a group of researchers investigated how smoking status and higher leisure-time physical activity (LTPA) predict ten-year changes in waist circumference (WC) after accounting for shared genes and environment.
Background
Can a single habit trade one health risk for another? Worldwide, more than one billion adults smoke, and over 30% intend to quit this year despite worries about post-cessation weight gain, even as WC has surpassed body mass index (BMI) as the stronger predictor of cardiometabolic risk. Evidence suggests nicotine suppresses appetite and raises metabolic rate, but withdrawal reverses both effects, promoting visceral fat.
Regular LTPA counteracts overall weight gain, yet its influence on long-term central adiposity after quitting remains unclear. Because family genetics and upbringing blur the causal links, further research is needed.
About the study
The analysis drew on the FinnTwin16 population cohort, which surveyed Finnish twins aged 24 years in 2000-2003 and ten years later. After excluding pregnancies and missing data, 3,322 individuals (46% men) provided information on smoking behavior, LTPA, WC, and covariates.
Smoking status at each wave was classified as daily or occasional, former, or never, then merged into categories: persistent smokers, quitters, and non-current smokers. An additional 16.6% of participants with miscellaneous smoking patterns were classified as “others” and were not included in the primary interpretations due to the group’s diversity.
LTPA was quantified by multiplying reported session frequency, duration, and metabolic equivalent tasks (MET) intensity to yield metabolic equivalent task-hours per week (MET-h/week), including active commuting. WC was self-measured with standardized pictorial instructions and modeled as the follow-up value adjusted for baseline WC. BMI, alcohol intake, diet quality, sleep problems, General Health Questionnaire-12 (GHQ-12) distress score, and education served as covariates.
Linear regression with clustering by family estimated associations between smoking or LTPA and WC change, sequentially adjusting for confounders. To probe familial confounding, fixed-effects models were used to compare 660 dizygotic (DZ) and 390 monozygotic (MZ) twin pairs discordant for exposure. Significance was set at P < 0.05; analyses were performed with Stata 18 statistical software using standard two-tailed tests, and Institutional Review Board (IRB) approval and informed consent were secured.
Study results
At baseline, most participants were non-current smokers, while 19% smoked persistently, 8% reported daily smoking but succeeded in quitting during follow-up, and another 16.6% fell into the “others” group due to inconsistent smoking patterns. Overall, WC rose by a mean of 6.5 cm over ten years, confirming a trend toward increasing abdominal obesity in adulthood. Non-current smokers gained 6.1 cm, persistent smokers gained 6.6 cm, and quitters gained 8.4 cm, indicating that those who stopped daily smoking experienced the largest expansion.
Regression models adjusted for sex, age, baseline WC, and BMI showed that quitting daily smoking predicted an additional 2.0 cm WC increase compared with continuing smokers (β = 2.04 cm; 95 % confidence interval [CI] 0.94-3.14). The estimate strengthened after controlling for diet, LTPA, alcohol consumption, sleep, distress, socioeconomic background, and health, suggesting that lifestyle clustering alone could not fully explain the association. No significant difference emerged between persistent smokers and non-current smokers, underscoring cessation as the critical inflection point for abdominal fat gain.
Physical activity painted a protective picture, as each extra MET-h/week at baseline corresponded to 0.05 cm less WC growth, and every unit increase in LTPA during follow-up reduced gain by 0.06 cm, independent of smoking group. While the impact appears modest, the dose-response relationship was linear; accumulating approximately 34 MET-h/week (roughly equivalent to seven one-hour walks) would be needed to offset the excess expansion associated with quitting. Quitters were the only group to increase LTPA on average (+5 MET-h/week), implying that activity upticks cushioned their metabolic load.
Twin analyses clarified whether genes or the environment were responsible for these patterns. Among DZ pairs discordant for smoking cessation, the quitting twin showed 2.6 cm more WC growth than the smoking co-twin, mirroring individual estimates. Yet the signal disappeared in MZ pairs, where genetically identical siblings serve as controls (β = -2.6 cm; CI -5.9 to 0.6). This divergence suggests that familial factors, such as shared genes or upbringing, may moderate the weight rebound after quitting.
However, the authors stress that this finding should be interpreted with caution, as it could also reflect insufficient statistical power. The analysis was based on only 11 identical twin pairs who differed in their quitting behavior, a number the paper notes is too small to draw firm conclusions.
Sensitivity tests showed no interaction between sex and smoking or between LTPA and smoking on WC change, allowing pooling of men and women. Diagnostics satisfied normality and variance assumptions, and results were accurate across alternative covariate sets.
The authors also note that this analysis reflects a 10-year change, and other evidence suggests that over a more extended period, the weight of former smokers may eventually return to the same trajectory as that of never-smokers.
Conclusions
To summarize, smoking cessation appeared to present a public-health paradox: the cardiovascular advantages of quitting were accompanied by a measurable expansion of central adiposity. In this twin cohort, daily smokers who stopped added roughly two centimeters more to their waistlines than their peers who continued. Yet, this association was not statistically significant once genetic matching was complete.
The paper suggests that this may indicate that predisposition, rather than quitting itself, drives long-term fat redistribution, but also strongly cautions that the small sample of twins in this part of the analysis means that a lack of statistical power is another potential explanation for the result. Across all genetic strata, higher LTPA reliably tempered waist growth.
These findings suggest that embedding exercise guidance into cessation programs could protect metabolic health without undermining quit success. The study highlights the importance of such integrated programs, while also underscoring the need for future research with larger twin samples and objective measures of activity and smoking to confirm these findings. Policymakers and clinicians should prioritize such integrated interventions.
Journal reference:
- Piirtola, M., Filippone, EL., Ranjit, A. et al. Associations of smoking status and leisure-time physical activity with waist circumference change--10-year follow-up among twin adults. Int J Obes (2025). DOI: 0.1038/s41366-025-01820-7, https://www.nature.com/articles/s41366-025-01820-7