Background and aimExposure to phthalates and phosphorus flame retardants (PFRs) is considered to be a risk factor for asthma and allergies. However, little is known about the contribution of loss-of-function mutations in the gene encoding filaggrin (FLG) gene, which are considered to be predisposing factors for eczema and asthma, to these associations. We investigated the associations between exposure to phthalates and PFRs in dust and eczema/wheeze among Japanese children, taking into consideration loss-of-function mutations in FLG. MethodsThis study was part of the Hokkaido study on Environment and Children's Health. Seven phthalates and 11 PFRs in household dust were measured by gas chromatography–mass spectrometry. Eczema and wheeze were assessed in children aged 7 years using the International Study of Asthma and Allergies in Childhood questionnaire. Eight FLG mutations previously identified in the Japanese population were extracted from cord blood samples. Children with one or more FLG mutations were considered to be positive for FLG mutations. The study included 296 children who had complete data (birth records, FLG mutations, first trimester and 7 years questionnaires, and phthalate/PFR levels). Odds ratios (ORs) and 95% confidential intervals (CIs) of eczema and wheeze were calculated for log-transformed phthalate/PFR levels by logistic regression. We also performed stratified analyses based on FLG mutations. ResultsThe prevalence rates of eczema and wheeze were 20.6% and 13.9%, respectively. Among children without any FLG mutations, tris (1, 3-dichloro-2-propyl) phosphate (TDCIPP) increased the OR of wheeze, (OR: 1.22, CI: 1.00–1.48). Significant p values for trends were found between tris (2-butoxyethyl) phosphate (TBOEP) and eczema and di-iso-nonyl phthalate (DiNP) and eczema among children without any FLG mutations, respectively. ConclusionsDespite our limited sample size and cross-sectional study design, the effects of indoor environmental factors on childhood eczema and wheeze were clearer in children without loss-of-function mutations in FLG than in children with mutations. Children with FLG mutations might already be cared for differently in terms of medication or parental lifestyle. Further studies in larger populations are warranted so that severity of symptoms and combinations of FLG mutations can be investigated.