Mediation analysis for the path (being a chimney sweeps → PAH exposure → DNA hypomethylation), where being a chimney sweep is a binary variable (chimney sweep vs control), PAH exposure is a continuous variable (PAH metabolites) and DNA hypomethylation is a continuous variable (methylation of F2RL3 and AHRR)
| Base model† | F2RL3_CpG1 | F2RL3_cg03636183* | AHRR_CpG1 | AHRR_CpG2 | AHRR_cg05575921 | |||||
| B (95% CI) | Mediation% | B (95% CI) | Mediation% | B (95% CI) | Mediation% | B (95% CI) | Mediation% | B (95% CI) | Mediation% | |
| −0.98 (−2.2 to 0.22) | 0.02 (−1.02 to 1.07) | – | −1.5 (−3.6 to 0.54) | −2.6 (−4.4 to −0.69) | −2.0 (−4.2 to 0.063) | |||||
| 1-OH-Phe (µg/g crea.)‡ | −0.91 (−2.2 to 0.34) | 7 | – | – | −1.7 (−3.8 to 0.49) | -8 | −2.7 (−4.6 to −0.76) | -6 | −2.2 (−4.3 to 0.038) | -6 |
| ∑2-,3-OH-Phe (µg/g crea‡ | −0.73 (−2.0 to 0.56) | 26 | – | – | −1.7 (−4.0 to 0.52) | −12 | −2.7 (−4.7 to −0.72) | -7 | −2.1 (−4.4 to 0.18) | -3 |
| 4-OH-Phe (ng/g crea.)‡ | −0.79 (−2.0 to 0.46) | 20 | – | – | −1.7 (−3.8 to 0.49) | -8 | −2.6 (−4.5 to −0.64) | -1 | −2.0 (−4.2 to 0.19) | 1 |
| ∑OH-Phe (µg/g crea.)‡ | −0.77 (−2.05 to 0.51) | 21 | – | – | −1.7 (−3.9 to 0.52) | −10 | −2.7 (−4.7 to −0.72) | -6 | −2.1 (−4.4 to 0.15) | -3 |
| ∑OH-Flu (µg/g crea.)‡ | −0.30 (−1.4 to 0.84) | 69 | – | – | −0.40 (−2.4 to 1.6) | 74 | −1.5 (−3.3 to 0.31) | 42 | −0.64 (−2.6 to 1.3) | 69 |
*Mediation analysis was not performed for F2RL3_cg03636183 because there was no association with exposure group (p=0.96).
†Outcome (DNA methylation)=intercept + B.exposure group (chimney sweep vs control)+B1.age+B2.smoking (three categories).
‡Outcome (DNA methylation)=intercept + B.exposure group (chimney sweep vs control)+B1.age+B2.smoking (three categories)+B3.(PAH metabolite).
B, unstandardised beta estimate; ∑OH-Flu, sum of 2-hydroxyfluorene and 3-hydroxyfluorene; ∑OH-Phe, sum of 1-hydroxyphenanthrene, 2-hydroxyphenanthrene, 3-hydroxyphenanthrene, and 4-hydroxyphenanthrene; ∑2-,3-OH-Phe, sum of 2-hydroxyphenanthrene and 3-hydroxyphenanthrene.