|
nif:isString
|
-
A representative random sample of mothers from maternity hospitals throughout France was established. The target population consisted of every one in 50 children born in 2011 in French maternity units [22]. After a written informed consent was obtained, medical records of the mother were consulted. The mother then answered a general questionnaire about the health of the baby and mental, social, family, and environmental exposures at age one. Follow-up is planned until age 18. The children were selected to represent all seasons of the year, having been born between April 1–4, June 27—July 4, September 27—October 4, and November 28—December 5, 2011. The exclusion criteria were: prematurity, childbirth under 18 years old, and no response to the questionnaire after two months. Mothers with multiple pregnancies with more than two children and those unable to give informed consent in one of the languages in which the information and consent forms were translated were also excluded. Before the study began in 2011, a pilot study started in 2006–2007 in a small cohort (n = 470) to test the different parts of the study, and the questionnaires. Questions showing poor completion rates or that were not well understood by the participants were removed or corrected. Before the study, a training course was organized for the investigators (questionnaire, filters inside the questionnaire). Investigators traveled to maternity units and undertook the study in person with the parents. Initially, 18 322 children including 289 pairs of twins were enrolled in the study at birth. Among the 18,322 selected families who gave their consent, 281 eventually refused to participate, leaving 18,041 families in the study. The children were followed up at two months of age and then again at one year. At two months, trained investigators interviewed the parents in a face to face interview at home. When the child reached one year old, parents answered a standardized questionnaire through a telephone conversation conducted by specially trained professionals (phone-platforms currently involved in health studies in France, by the French institute of statistics (INSEE)). By the time of the phone interview, 4270 participants had been lost during the follow-up, leaving a total of 14059 children included in the present analysis (Fig 1).
Figure data removed from full text. Figure identifier and caption: 10.1371/journal.pone.0196711.g001 Flow chart. Four groups were defined: non-wheezers (no current wheezing symptoms at both two months and one year old), intermittent (current wheezing symptoms at two months but no wheezing at one year), persistent (current wheezing at two months and at one year) and incident (no current wheezing at two months but wheezing at one year). Wheezing at two months was defined by a positive answer yes to the question “Does your child currently have wheezing or whistling in the chest? “[23]. Wheezing at one year was defined by a positive answer to the question “Does your child ever have wheezing or whistling in the chest? “ The questionnaire assessed the sociodemographic characteristics (age, gender, region, number of rooms and number of persons living in the house, level of education for both parents), type of delivery, Apgar scores at one and five minutes respectively, mother’s previous pregnancies, and family history of asthma in parents and siblings of the child and other potential risk factors of asthma (heating system used inside the house, dampness inside the house, work done during pregnancy in the house, presence of pets in the house, child’s history of recurrent otitis, rash history of the child, bronchiolitis, bronchitis episodes, and breastfeeding). Information concerning parental allergies was recorded by questions about family history of eczema, asthma, eczema, and rhinitis (hay fever) for both parents and siblings. Finally, environmental exposure to toxic substances was considered: smoking during pregnancy, passive smoking, occupational and domestic pesticide exposure, and the use of cleaning products. Respiratory symptoms at two months were taken into account. The presence of coughing was ascertained by an affirmative answer to the question: “Has your child had a dry cough at night, apart from a cough associated with a cold or chest infection?” To know whether the child had respiratory problems, the following questions were used: “Has your child had any sign of respiratory distress?” and “At present does your child have any sign of respiratory distress?”. To investigate the presence of excess bronchial secretions, the following questions were used: “Has your child had an excess of bronchial secretion or phlegm?” and “At present are you noticing that your child has an excess of bronchial secretion or phlegm?”. A question was further asking if the child had ever been diagnosed with a bronchitis, and whether a doctor had prescribed medications, and a list of medications by class was proposed, to the parents.
Each parent gave written informed consent at the maternity unit. Every level of the project and each procedure was monitored by the National Council for Statistical Information (CNIS) and approved by the French Data Protection Authority (CNIL), in conjunction with the Consultative Committee for Data Processing in Health Research (CCTIRS), which assesses the scientific relevance of health projects, properly conducted from start to finish. Each parent retains the right to withdraw his/her child at any point during the study.
Data analysis was performed by SPSS software, version 23. Two-sided statistical tests were used to test the association of each respiratory symptom at two months across all four wheezing groups; the Chi-2 test for dichotomous or multinomial qualitative variables, with Fisher’s exact test whenever appropriate. Regarding multivariate analysis, backward logistic regressions were performed by taking into account the variables in the bivariate analysis that showed a p-value <0.10 [24, 25];which included recommended covariates (such as maternal asthma, smoking, and educational level of the mother used as a proxy for socio-economical status). Educational level of the mother was categorized into 3 categories (low, intermediate and high level). Number of siblings was categorized into 0 (reference), 1, 2 and “3 or more” siblings. Several logistic regressions were conducted taking the dependent variables as follows: (1) intermittent vs persistent wheezing, (2) non-wheezers vs persistent wheezers and (3) non-wheezers vs incident wheezers.
|
![[This page as RDF]](http://data.gesis.org/somesci/static/rdf-icon.gif){kind=icon}