• Research Review
• Journal Review

Authors: SE Saarni, K Pietiläinen, S Kantonen, A Rissanen, J Kaprio

Reference: Am J Public Health 2009; 99: 348-354, http://www.ajph.org/cgi/content/abstract/99/2/348

Reviewer: Joaquin Barnoya, MD, MPH, ProCor contributing editor, Director, Research and Education, Unidad de Cirugia Cardiovascular de Guatemala, Guatemala

Purpose of study: To examine the independent effects of smoking on weight during late adolescence and early adulthood.

Methods: The FinnTwin16 is a population-based study of 5 birth cohorts of Finish twins born in 1975 to 1979. Baseline data was collected through mailed questionnaires within 60 days of the twins' 16th birthdays. Follow-ups were at ages 17 (T2) and 18.5 (T3) and in young adulthood (mean 24.4 years, T4). Response rates ranged between 80% and 90%.  Smoking status was categorized as
never smokers, former smokers, occasional smokers, daily smokers smoking fewer than 10 cigarettes per day, and daily smokers smoking 10 or more cigarettes per day.  Smoking status at age 18 was used for the analysis.  Waist circumference was self-reported by the participants using a tape measure supplied by mail in the last follow-up. Height and weight to calculate the BMI were also self-reported. Respondents were categorized as normal weight, overweight, or obese. Analysis included known potential confounders (e.g. dietary behavior, physical activity). Self-reported measures were highly correlated when validated with a sub-sample of 566 twins. 

Results: About 50% of men and women had never smoked and about 12% were former smokers in adolescence. Heavy smoking was more prevalent among men than among women. Eighteen percent and 12% of men and women were classified as exercisers; respectively. Prevalence of overweight rose steeply after ages 16 to 18.5 years. One quarter of respondents were overweight at T4.  The percentage of obese men and women rose from 0.7% at T1 to 4% at T4. The adjusted risk of overweight for girls who smoked at least 10 cigarettes per day was 1.74 (95% CI 1.06, 2.88). No
effect was seen among men.  Adolescent smoking predicted abdominal obesity in men and women. Heavy smoking was a significant predictor of an increased risk of subsequent abdominal obesity after adjustment for potential confounders (OR 1.44, 95% CI 1.11-1.88). 

Comment: As opposed to common knowledge, this study shows that smoking actually increases the risk of abdominal obesity, especially among women.  Even though this study only provides epidemiological evidence, there is also evidence of increased insulin resistance among smokers, which gives these results biological plausibility.  Results from this study should be discussed with adolescents and others to clarify the relationship between smoking and weight.