Table 5.

Sensitivity analyses for associations between third trimester maternal urinary phthalate metabolite concentrations and percent fat mass among children 4–9 years of age in the Mount Sinai Children’s Environmental Health Study.

Metabolite	Primary analysisa	Single metabolite modelsb	Loss to follow-upc
MEP	0.12 (–1.34, 1.58)	0.08 (–1.14, 1.30)	–0.12 (–1.62, 1.36)
MnBP	–0.86 (–3.07, 1.36)	–0.07 (–1.41, 1.27)	–0.93 (–3.15, 1.30)
MiBP	0.34 (–1.54, 2.20)	0.23 (–1.12, 1.52)	1.03 (–0.91, 2.97)
MCPP	0.63 (–1.55, 2.82)	0.30 (–1.05, 1.66)	0.40 (–1.81, 2.60)
MBzP	0.67 (–1.31, 2.65)	0.40 (–0.93, 1.74)	0.48 (–1.47, 2.47)
∑DEHP	–0.89 (–2.24, 0.47)	–0.57 (–1.79, 0.57)	–0.62 (–2.01, 0.81)
Beta coefficients (95% credible intervals) per SD increase in natural log phthalate metabolite concentrations, adjusted for urine dilution and collection date; maternal race/ethnicity, age, education, work status, and smoking during pregnancy; maternal height and prepregnancy BMI; adequacy of gestational weight gain; breastfeeding; months of age and physical activity at follow-up; and child’s sex. aAssociations among children with at least one follow-up visit (n = 180) estimated in a multiple metabolite linear mixed-effects regression model. bAssociations among children with at least one follow-up visit (n = 180) estimated in a separate linear mixed-effects regression model for each metabolite. cAssociations among all children in the birth cohort with measured phthalate metabolite concentrations (n = 380) estimated in a multiple metabolite linear mixed-effects regression model using a selection model for potentially nonignorable missing outcome data.