Aim To examine the association between weight change and baseline body mass index (BMI) over 8 years in a cohort of continuing and quitting smokers.
Design Prospective cohort.
Setting Oxfordshire general practices nicotine patch/placebo trial with 8-year follow-up.
Participants Eighty-five participants were biochemically proven abstinent at 3, 6, 12 months and 8 years (abstainers). A total of 613 smoked throughout the 8 years (smokers), 26 quit for a whole year but were smoking again by 8 years (relapsed); 116 smoked for the first year but were abstinent at 8 years (late abstainers).
Measurements Weight and BMI was measured at baseline and at 8 years. Regression models were used to examine weight gain by smoking status and the association of BMI at the time of quitting.
Findings Abstainers gained 8.79 kg [standard deviation (SD) 6.36; 95% confidence interval (CI) 7.42, 10.17]. Smokers gained 2.24 kg (SD 6.65; 95% CI 1.7, 2.77). Relapsed smokers gained 3.28 kg (SD 7.16; 95% CI 0.328, 6.24). Late abstainers gained 8.33 kg (SD 8.04; 95% CI 6.85, 9.81). The association between baseline BMI and weight change was modified by smoking status. In smokers there was a negative linear association of BMI, while in abstainers a J-shaped curve fitted best. These models estimated weight change over 8 years in abstainers of +9.8 kg, +7.8 kg, +10.2 kg, +19.4 kg and in smokers of +3.9 kg, +2.6 kg, 1.0 kg and −0.8 kg, where BMI was 18, 23, 29 and 36, respectively.
Conclusion Obese smokers gain most weight on quitting smoking, while obese continuing smokers are likely to remain stable or lose weight. Obese quitters have the greatest need for interventions to ameliorate weight gain.