Table 3.
The dose-response relationships between frequency of cosmetics use and adverse pregnancy outcomes.
| Pregnancy outcomes | Cosmetics non-users (n = 4652) | Cosmetics users (times/week) | OR (95%CI)d | OR (95%CI)e | OR (95%CI)f | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| 1–2(n = 3816) | 3–4(n = 892) | ≥5(n = 350) | Unadjusted | Adjusteda | Unadjusted | Adjusteda | Unadjusted | Adjusteda | ||
| Preterm birthb | ||||||||||
| Yes | 146(3.1) | 113(3.0) | 26(2.9) | 9(2.6) | 0.94(0.73–1.21) | 0.95(0.73–1.22) | 0.93(0.61–1.42) | 0.87(0.55–1.37) | 0.82(0.41–1.61) | 0.90(0.45–1.79) |
| No | 4506(96.9) | 3703(97.0) | 866(97.1) | 341(97.4) | ||||||
| Birth weight | ||||||||||
| Low birth weightc | 96(2.1) | 91(2.4) | 17(1.9) | 10(2.9) | 1.16(0.86–1.55) | 1.18(0.87–1.59) | 0.91(0.54–1.54) | 0.98(0.57–1.69) | 1.39(0.72–2.70) | 1.60(0.82–3.11) |
| Normal birth weight | 4247(91.3) | 3484(91.3) | 824(92.4) | 318(90.9) | ||||||
| Macrosomiac | 309(6.6) | 241(6.3) | 51(5.7) | 22(6.3) | 0.95(0.80–1.13) | 0.97(0.81–1.16) | 0.85(0.63–1.16) | 0.92(0.67–1.25) | 0.95(0.61–1.49) | 1.00(0.62–1.60) |
| Birth weight and gestational age | ||||||||||
| SGAc | 301(6.5) | 295(7.7) | 68(7.6) | 36(10.3) | 1.20(1.01–1.42)* | 1.18(0.99–1.40) | 1.19(0.90–1.57) | 1.23(0.92–1.64) | 1.68(1.16–2.41)* | 1.83(1.25–2.69)* |
| AGA | 3965(85.2) | 3246(85.1) | 753(84.4) | 283(80.9) | ||||||
| LGAc | 386(8.3) | 275(7.2) | 275(8.0) | 31(8.9) | 0.87(0.74–1.02) | 0.90(0.76–1.06) | 0.97(0.74–1.26) | 1.00(0.76–1.32) | 1.13(0.77–1.65) | 1.25(0.84–1.87) |
Results are presented as odds ratios (OR) with 95% confidence intervals (95%CIs). SGA stands for small for gestational age, AGA stands for appropriate for gestational age, LGA stands for large for gestational age. aAnalyses were adjusted for maternal age, residence, ethnicity, education level, occupation, parity, abnormal reproductive history, pre-pregnancy body mass index and pregnancy comorbidity. bThe odds ratio for preterm birth was calculated by binary logistic regression. cThe odds ratio for low birth weight/macrosomia and SGA/LGA was calculated by multinomial logistic regression. dOdds ratio of adverse pregnancy outcomes compared cosmetics users (1–2 times/week) to cosmetics non-users. eOdds ratio of adverse pregnancy outcomes compared cosmetics users (3–4 times/week) to cosmetics non-users. fOdds ratio of adverse pregnancy outcomes compared cosmetics users (≥5 times/week) to cosmetics non-users.