Abstract
RATIONALE: The literature on the effect of obesity and weight gain on respiratory outcomes in smokers is contradictory. OBJECTIVE: To examine the cross-sectional effect of body mass index (BMI) and the longitudinal effect of change in BMI upon spirometry and health status among smokers at risk for and with milder chronic obstructive pulmonary disease (COPD). METHODS: Participants from the Lovelace Smokers' Cohort were followed for a median period of 6 years, 75% of whom were at risk and 25% of whom had COPD at baseline examination. BMI and gain in BMI were examined as continuous independent variables overall and after stratification into three categories (normal-weight, overweight, and obese) determined on the basis of baseline weight. Spirometry and health status (as assessed by St. George Respiratory Questionnaire total and subscale scores) were dependent variables. Covariates included age, sex, ethnicity, pack-years of smoking, and current smoking status. Cross-sectional analysis used linear and logistic regression; longitudinal analysis used a mixed model approach. MEASUREMENTS AND MAIN RESULTS: In cross-sectional analyses, higher BMI was associated with worse health status among obese smokers but with better health status among normal-weight smokers. In longitudinal analyses, weight gain was associated with a decrease in FEV1 and health status among obese smokers and with an increase in these outcomes among normal-weight smokers. CONCLUSIONS: Weight gain affects respiratory outcomes differently between obese and normal-weight smokers. Whereas FEV1 and health status decrease with weight gain among obese smokers, they improve among normal-weight smokers. The nonlinear relationship between weight gain and respiratory outcomes suggests that this effect of excess weight is unlikely to be mechanical alone.