Table 1.
Single-pollutant associations between average phthalate exposures and gestational age
| Phthalate Metabolite | Logistic* | Cox | AFT | |||
|---|---|---|---|---|---|---|
| OR | 95% CI | HR | 95% CI | % Change | 95% CI | |
| MEHP | 1.50 | (1.10, 2.07) | 1.07 | (0.98, 1.17) | −0.79% | (−1.67, 0.09%) |
| MEHHP | 1.05 | (0.75, 1.47) | 1.04 | (0.94, 1.15) | −0.51% | (−1.01, 0.32%) |
| MEOHP | 1.23 | (0.87, 1.75) | 1.09 | (0.98, 1.21) | −0.76% | (−1.74, 0.23%) |
| MECPP | 1.66 | (1.20, 2.30) | 1.21 | (1.09, 1.33) | −1.19% | (−2.11, −0.26%) |
| Σ DEHP | 1.47 | (1.06, 2.03) | 1.14 | (1.04, 1.26) | −1.03% | (−1.95, −0.01%) |
| MBzP | 1.27 | (0.79, 1.65) | 1.15 | (1.03, 1.27) | −0.60% | (−1.63, 0.43%) |
| MBP | 1.35 | (0.99, 1.87) | 1.17 | (1.05, 1.29) | −0.74% | (−1.75, 0.28%) |
| MiBP | 0.98 | (0.66, 1.45) | 1.00 | (0.90, 1.12) | 0.30% | (−0.82, 1.44%) |
| MEP | 1.22 | (0.87, 1.71) | 1.01 | (0.91, 1.12) | −0.11% | (−1.06, 0.86%) |
| MCPP | 1.27 | (0.93, 1.72) | 1.10 | (1.01, 1.20) | −0.28% | (−1.18, 0.62%) |
Phthalates were averaged across the first three visits and log-transformed. All models were adjusted for average specific gravity, maternal age at first visit, race, and education. The single-pollutant models for MBzP, MBP, MiBP, MEP, and MCPP are also adjusted for health insurance provider. Bolded cells indicate significant (p < 0.05) odds ratios (OR), hazard ratios (HR), and percent changes (% Change). Odds ratios, hazard ratios and percent changes are all calculated on IQR (Interquartile Range) scale. Abbreviations: CI, Confidence Interval; Logistic, Logistic Regression; Cox, Cox Proportional Hazards Model; AFT, Accelerated Failure Time Model; MEHP, Mono-(2-ethyl)-hexyl phthalate; MEHHP, Mono-(2-ethyl-5-hydroxyhexyl) phthalate; MEOHP, Mono-(2-ethyl-5-oxohexyl) phthalate; MECPP, Mono-(2-ethyl-5-carboxypentyl) phthalate; Σ DEHP, Summed DEHP metabolites; MBzP, Mono-benzyl phthalate; MBP, Mono-n-butyl phthalate; MiBP, Mono-isobutyl phthalate; MEP, Mono-ethyl phthalate; MCPP, Mono-(3-carboxypropyl) phthalate. *Results are similar to Ferguson and colleagues (2014) [14]