Premium
Phenotypes of childhood wheeze in Japanese children: A group‐based trajectory analysis
Author(s) -
Yang Limin,
Narita Masami,
YamamotoHanada Kiwako,
Sakamoto Naoko,
Saito Hirohisa,
Ohya Yukihiro
Publication year - 2018
Publication title -
pediatric allergy and immunology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.269
H-Index - 89
eISSN - 1399-3038
pISSN - 0905-6157
DOI - 10.1111/pai.12917
Subject(s) - wheeze , medicine , pediatrics , asthma , odds ratio , respiratory sounds , odds , etiology , multinomial logistic regression , demography , logistic regression , statistics , mathematics , sociology
Background Exploring patterns of childhood wheeze may help to clarify the etiology and prognosis of respiratory diseases. The purpose of this study was to classify phenotypes of wheezing in children up to 9 years of age in Japan and to evaluate the individual and environmental risk factors for these phenotypes. Methods Wheeze was evaluated at approximately 1‐year intervals based on the mothers’ recollection of their child’s wheezing or whistling in the chest during the preceding 12 months. The children were aged 1‐9 years. In total, 1116 children who had at least five measures of wheezing at all nine time points were used for identifying trajectories. Trajectories were identified with group‐based trajectory analysis. A multinomial logit model was built to evaluate the relationships between phenotypes and risk factors. Results Five typical trajectories were identified. The probability of group membership was 43.7%, 32.2%, 6.2%, 8.6%, and 9.2% for the never/infrequent wheeze, transient early wheeze, school‐age‐onset wheeze, early‐childhood‐onset remitting wheeze, and persistent wheeze trajectories, respectively. Infant tobacco exposure increased the odds of membership in the transient early wheeze trajectory compared to the never/infrequent wheeze trajectory. Conclusions Using the group‐based trajectory modeling approach, we identified five trajectories of childhood wheeze development in a Japanese population. The trajectories shown here are based on formal statistical modeling rather than on subjective classification, and an assessment of its precision suggested that the model has high assignment accuracy.