However, this study also has several limitations. First, although moldy conditions at home could be an important factor for asthma,54 mold counts at home were not directly measured because of the large number of subjects surveyed in this study. However, we believe it to be unlikely to change our conclusion. Our main findings are positive relation between current asthma (especially ASROH) and fungal spore counts of Aspergillus/Penicillium and basidiospores. Among those with home spore levels similar to or higher than those in the schools, such effects would not have been observed, thus, reducing the observed effects toward null. In addition, a previous study has found that households with visible mold on walls had higher indoor spore counts.55 Self-reported visible mold on walls at home was adjusted in the regression models in this study. Thus, effects caused by unmeasured spore levels at home were adjusted. Second, because fungal sensitization was not assessed by skin prick or serologic tests, we were unable to determine whether the observed effects were due to allergic or nonallergic mechanisms. Although previous reports showed low prevalence of fungal sensitization in Taiwanese schoolchildren,56,57 inappropriate fungal matrix used in current commercial kits had been considered as a potential problem for the low detection rates.33 The findings from this current study underline the importance of developing measurements of immunologic markers more specific for regional fungal allergens, which would allow for more sensitive detection of atopic reaction to fungal allergens. Third, ambient levels of pollen were not measured in this study; thus, potential confounding effects of pollen on the observed fungal effect on asthma could not be totally ruled out. However, according to a large survey on Taiwanese children with allergy, the prevalence of pollen sensitization was very low, ranging from 1% to 2%,56 mainly due to grasses.58 The peak seasons of grass pollens are summer and autumn,59,60 different from those of fungal spores (winter and spring).61 Therefore, although the confounding effects of pollens could not be completely ruled out, they should be minimal. Fourth, our fungal measurements cannot distinguish Aspergillus from Penicillium, since their spores look almost identical under light microscopy. Because the two fungal species share common indoor substrates, this will not affect the strategy for fungal remediation in classrooms. However, previous study showed Aspergillus as a common mold in outdoor air, and only 6.8% of classrooms had visible mold on walls in this current study. Methods to control outdoor spore penetration (ie, closing windows during times of high outdoor spore levels) may have to be considered as well. Further study is needed to differentiate the effects between Aspergillus and Penicillium. Fifth, because of feasibility and budget concerns, the asthma defined in this study is self-reported asthma, without further confirmation tests. To resolve such a problem, a validated questionnaire is used, as well as “physician-diagnosed asthma” with information adopted from parents. This minimizes the possibility of diagnosis misclassification.