Influence of asthma definition on the asthma-obesity relationship

BACKGROUND: Epidemiological studies suggest an association between obesity and asthma in adults and children. Asthma diagnosis criteria are different among studies. The aim of this study was to test the influence of asthma definition on the asthma-obesity relationship. METHODS: In a cross-sectional analysis of 1922 men and women, subjects completed a translated questionnaire from the European Community Respiratory Health Survey and underwent spirometry and a bronchial challenge test. Weight, height and waist circumference were measured. Multiple logistic regression analysis was carried out to assess the association of variables related to obesity and asthma. Asthma was defined either by the presence of symptoms with bronchial hyperresponsiveness (BHR) or by a self-report of a physician-made diagnosis. The following variables were separately tested for associations with asthma: socioeconomic characteristics, schooling, physical activity, smoking status, anthropometry and spirometry. RESULTS: No association was detected between asthma confirmed by BHR and obesity indicators, odds ratio (OR) = 1.08 (95% confidence interval: 0.69 - 1.68) for obesity assessed by body mass index ≥ 30 kg/m(2); OR = 1.02 (0.74 - 1.40) for obesity assessed by abnormal waist-to-height ratio; and, OR = 0.96 (0.69 - 1.33) for abnormal waist circumference. On the contrary, a previous diagnosis of asthma was associated with obesity, OR = 1.48 (1.01 - 2.16) for body mass index ≥ 30 kg/m(2); OR = 1.48 (1.13 - 1.93) for abnormal waist-to-height ratio; and, OR = 1.32 (1.00 – 1.75) for abnormal waist circumference. Female gender, schooling ≥ 12 years and smoking were associated with BHR-confirmed asthma. Physically inactive subjects were associated with a previous diagnosis of asthma. CONCLUSIONS: Our findings indicate that the relationship between asthma and obesity in epidemiological studies depends on the definition adopted. Certain components of asthma, for instance, symptoms may be more prone to the obesity influence than other ones, like bronchial hyperresponsiveness.