LAUSR

To the Editor, Growing urbanization imposes a huge pressure on environment and physical landscape. With hasty urbanization, there is an increasing interest in understanding how urban settings and environment affect children's health.1 The number of studies on the effect of specific characteristics of child's school and residential neighborhood on asthma and allergy is increasing.2– 4 Yet, given that children spend large periods of time at home and at school, whether and how different landuse characteristics of both neighborhoods are related to respiratory health, including asthma, in schoolaged children, is unclear. Thus, we aimed to identify the landuse characteristics in school and residential surrounding associated with asthma prevalence in schoolaged children using a decision tree learning approach. Data on demographic and health outcomes from 858 children (aged 7– 12 years) were obtained from a crosssectional study performed in Porto, Portugal (supporting information and Table S1). We did a crosssectional study with an ecological research design. Landuse characteristics within a 500 m buffer centered in each child's school (n = 20) and residential (n = 174) address were assessed based on the 2015 Portuguese official maps of land cover using a Geographical Information System (ArcMap 10.7.2) (Table S2). For residences without information on landuse characteristics, data from each child's school were imputed. The mean imputation method was used for residences without information on landuse characteristics, and the missing values are replaced with the mean of the same landuse type for each child's school. The analysis of landuse characteristics was aggregated at school level. Regression tree models were used to identity the landuse characteristics in school and residential neighborhoods associated with the prevalence of asthma among children. Linear regression models were fitted to estimate the association between each branch and asthma prevalence using the split values. The Ethical Committee of the University of Porto approved the study, and informed consent was obtained from children's legal guardians. The results demonstrated that three landuse characteristics around schools and residences had a significant effect in the prevalence of asthma in schoolchildren (Figure 1). Schools with children living in residential neighborhoods surrounded by an area of agriculture, natural, and seminatural spaces smaller than 0.77 ha, and those surrounded by an area of continuous urban fabric smaller than 65.3 ha had a higher prevalence of asthma (11%) compared to schools with children living in residential neighborhoods surrounded by an area of agriculture, natural, and seminatural spaces larger than or equal to 0.77 ha and those surrounded by an area of cultural facilities and larger than or equal to 0.67 ha (4.7%). Compared to the branch with larger areas of agriculture, natural, and seminatural spaces in residential's neighborhood and cultural facilities and historic areas around schools, the branch with smaller areas of agriculture, natural, and seminatural spaces in residential's neighborhood and continuous urban fabric around schools was associated with a higher prevalence of asthma [β (95% CI) = 0.068 (0.024; 0.112)] (Table 1). Our findings suggested that some characteristics of land use around schools and residences may be determinants of a higher prevalence of asthma in schoolaged children. Smaller areas of agriculture, natural, and seminatural spaces around residences, and the smaller presence of continuous urban fabric in school's neighborhood were significantly associated with a higher prevalence of asthma compared with larger areas of agriculture, natural, and seminatural spaces in residential's neighborhood and cultural facilities and historic areas around schools. Additionally, children living in neighborhoods surrounded by larger areas of agriculture, (semi)natural, and cultural and historic facilities had a lower prevalence of asthma. The presence of natural landuse areas may enhance environmental biodiversity in children's neighborhood, affecting the composition of their microbiota, which in turn may decrease the risk of asthma.5 In addition, the presence of natural spaces may also provide a protection from anthropogenic air pollution through absorption, providing physical barriers against emission sources, or by limiting the overall area available to sources of pollution such as traffic and industry.6 Previous studies have also evaluated the health impacts of specific land cover types, as agricultural areas and forests,5,7 corroborating our results and highlighting the influence of land cover use on respiratory health in children. Additionally, the presence of cultural and historic facilities and continuous urban fabric areas in our study was related to a lower presence of transit roads in school's neighborhood (data not shown), and possibly to a lower exposure to trafficrelated air pollution, decreasing the risk of asthma development among children.8,9 Growing evidence supports the link between air pollution and the trafficrelated pollution exposure from roadways near schools and the risk of asthma in children.10– 12 In 2020, Zeng et al.13 also suggested a protective