Lung health and questionnaire data
Health questionnaires were used to determine presence and frequency of wheeze and its triggers, and physician diagnosed asthma among the children. These questionnaires were administered at 3 and 6 months postpartum, repeated every 6 months from ages 1 year to 3 years, then annually to 5 years. Recurrent wheeze was defined as two or more episodes of wheeze symptoms in one year between 2-5 years of age. Asthma diagnosis by age 5 was determined by a pediatric consultant with expertise in asthma at clinic visits. This diagnosis resulted in three classifications: definite, possible or no asthma. Children who showed consistent symptoms and objective evidence of bronchodilator response (greater than 12% enhancement in FEV1) were classified as definite asthma. Those who had consistent symptoms but did not show response to bronchodilator or had no symptoms but showed response to bronchodilator were classified as possible asthma. In order to maximize statistical power, we combined definite and possible asthma cases into a single case group and those with no asthma symptoms as the control group.
In addition to recurrent wheeze and physician diagnosed asthma, questionnaires completed by parents at 3, 6, 12, 18, 24, 30 months of age documented exposures to pre- and postnatal cigarette smoking, traffic air pollution measured by nitrogen dioxide (NO2), pet ownership, and breastfeeding duration. Exposure to NO2 was measured using city-specific land use regression (LUR) models, while accounting for residential mobility and temporal variance in ambient concentrations.(22) Estimates of exposure were measured for each residence reported by participants from address at conception to the address where the participating child resided at age 1 year.(22) All measures were temporally adjusted based on local ambient monitoring data per biweekly basis.(22) NO2 exposure were grouped into three categories: high, moderate, and low exposure. High was defined as 1 standard deviation above the mean (z score > 1), low as 1 standard deviation below the mean (z score < -1) and moderate exposure fell between high and low groups ( -1 < z score < 1).