2.1 Study participants and ethics statement
Shanghai Children Allergy Study (SCAS) explores the epidemic characteristics of childhood allergies, aiming to provide a basis for formulating the intervention strategy for children allergy in Shanghai. The study was sponsored by Shanghai Education Commission. We collaborated with Shanghai Children's Medical Center to carry out the project research. This cross-sectional baseline study was conducted during April-June 2019, using a multi-stage and multi-strata sampling approach. Among the total 9 urban areas and 8 suburban/rural areas in Shanghai, 4 urban areas (Xuhui, Putuo, Yangpu, and Pudong New area) and 4 suburban/rural areas (Minhang, Jinshan, Qingpu, and Chongming) were randomly sampled. In all sampled areas, 31 kindergartens were randomly selected. The investigation purpose was explained ahead to the principals and teachers. To obtain parental permission, the purpose was conveyed to parents through parents' meetings, in which voluntary and anonymous participation was stressed. Informed consent form was then obtained from parents. Each child’s mother completed the questionnaire on behalf of herself and her child.
Overall, 6389 children were recruited to participate in this baseline survey. Among them, 6237 (97.62%) completed the questionnaires. After the exclusion of 1 missing data on allergic diseases, the valid sample of this study consisted of 6236 mother-children pairs (defined as All Population Group).
The ethical application and the consent procedure of this study were approved by the Ethics Committee of Shanghai Jiao Tong University School of Medicine.
2.2 Maternal behavioral variables
A brief Maternal Life-style Questionnaire (MLQ) was utilized to evaluate maternal life styles during pregnancy based on the literature reviewed.16-18 Daily behavioral choice during pregnancy included the duration of sleep per day, the frequency of physical activity (never/occasionally, 1-3hrs./week, 3-6hrs./week, ≥1hrs./day), and the frequency of phone/computer/television exposure (<2hrs./day, 2-4hrs./day, 4-8hrs./day, ≥8hrs./day) on average. The optimal duration of sleep for pregnant woman is not yet known, the cutoff for short sleep duration was defined as less than 8 hours/day in this study with the reference of most studies.18,19 Meanwhile, 8 hours/day was close to the 25th percentile of the distribution of total sleep duration during pregnancy in our dataset. The answers for daily physical activity were classified into two groups: <1hr./day vs. ≥1hrs./day; and screen exposure were classified into two groups: ≥2hrs./day vs. <2hrs./day.
2.3 Ascertainment of allergic diseases
The International Study of Asthma and Allergies in Childhood (ISAAC) questionnaire was applied to detect respiratory allergies.20,21 The Cronbach's alpha coefficient of the ISAAC allergic questionnaire in our sampled children was 0.91. The intra-class correlation coefficient of retest reliability at intervals of 2 weeks was 0.94. Validity presented by Kaiser-Meyer-Olkin (KMO) was 0.94 and the high validity among preschool children has been confirmed.22 All questions were answered by yes or no. For the assessment of asthma, the question was asked: “Has your child ever been diagnosed with asthma?”. We further assessed the current wheezing by questions: “Has your child had any wheezing, or breathing difficulties in the last 12 months?” or “Has your child ever experienced wheezing or whistling in the chest in the last 12 months?”. Two questions were utilized for allergic rhinitis diagnosis: “Has your child ever been diagnosed with allergic rhinitis?” and “Has your child ever sneezing, runny nose, stuffy, itchy or itchy eyes in the last 12 months not due to having a cold or flu?”. Children were screened out one of above three types of respiratory allergies were considered to be Screened Positive (SP).
In addition to using the ISAAC questionnaire to screen respiratory allergies, we further collected information on allergen tests. 4438 participants in total ever took allergen test, including skin prick test (SPT), immunoglobulin E (IgE) and the others (defined as Allergen Test Subgroup). We further asked the children who had been tested for allergens: “Did the allergen test is positive, if so, which of the following allergens?” Common allergens were listed, including dust mite, mold, milk/dairy products, eggs, fish and shrimp, peanut, nuts other than peanuts, wheat, beans or soy products, fruit, vegetables, pollen, mugwort, ragweed, cat/dog hair, and so forth. Among children with screened respiratory allergies positive, those who tested positive for any of allergens were further regarded as Screened and Test Positive (STP).
2.4 Confounding variables
2.4.1 Demographic characteristics
Demographic characteristics included age and childhood overweight/obesity (yes/no), child’s sleep duration on weekdays (<10hrs./day, ≥10hrs./day), child’s sleep duration on weekends (<10hrs./day, ≥10hrs./day), child’s exercise frequency (<1hr./day, ≥1hrs./day), child’s screen exposure frequency (<5 times./week, ≥5 times./week). BMI (body mass index) for age and sex-specific percentile over 85% was implemented to define the overweight/obesity.23 Family background contained household incomes (<4000, 4000-8000, >8000), family structures (single parent family, nuclear family, extended family), both mother and father’s educational levels (primary education, secondary education, college and above), and first-degree relative with allergies (yes/no).
2.4.2 Obstetric characteristics and health status of parents
Obstetric characteristics included gestational weeks (premature delivery, post-delivery and term delivery), delivery modes (cesarean delivery, vaginal delivery/midwifery), and full breastfeeding over 6 months (yes/no). Health status of parents contained maternal smoking exposure status (yes/no), maternal drinking habits (yes/no), gestational hypertension (yes/no), gestational diabetes (yes/no), gestational anemia (yes/no), maternal pre-pregnancy overweight/obesity (yes/no), mother's age at delivery (≥30, 25-29, ≤24). Those pre-pregnancy maternal body mass index above 25 (including 25) was defined as maternal pre-pregnancy overweight/obesity.24
2.4.3 Environmental exposure
Four questions about the state of the environment were about whether there were farms or other sources of pollution within a five-minute walk from home, including farmlands and orchards (yes/no), chemical emission sources (yes/no), smog-emitting factories (yes/no), large garbage dumps (yes/no).
2.5 Statistical analysis
Statistical descriptions were made utilizing the percentage for categorical variables. The univariate logistic regression was applied to compare differences between groups.
Univariate logistic regression was implemented to calculate the unadjusted odds ratios (OR) and 95% confidence intervals (CI), thus evaluating the relationship between maternal behavioral characteristics and allergic diseases. In All Population Group, '1' for children with screened allergies positive and '0' for children screened allergies negative. In Allergen Test Subgroup, children with both positive in allergen test and screened respiratory allergies were defined as '1', and who had neither been screened with any allergy nor found any allergen were defined as '0'. Adjustments of confounding factors were made by the multivariate regression models followed by a two-step procedure: Model I only adjusted for demographic characteristics; in Model II, obstetric characteristics and health status of parents were further adjusted.
All other analyses were performed with the Statistical Package for the Social Sciences (SPSS) (IBM-SPSS Statistics version 23.0, Inc., Chicago, IL). Statistical significance level was set at p value <.05 (two sided).