Table 2.

Statistical uncertainty of predicted and observed effect doses for the mixture of DEHP, vinclozolin, prochloraz, and finasteride.

	Effect doses for the mixture (mg/kg/day)
	Observeda	Predicted by DA	Predicted by IA
Effect level	Mean (95% CI)	Mean (95% CI)	Mean (95% CI)
Relative AGD index
90%	11.8 (4.3–34.7)	12.2 (0.3–53.8)	12.3 (0.2–58.2)
60%	77.1 (59.4–99.8)	195.6 (147.2–262.3)*	214.3 (179.2–250.6)*
No. of nipples/areolas
3	10.6 (6.5–15.3)	8.8 (5.6–13.6)	6.7 (4.1–13.5)
6	23.3 (17.0–29.4)	25.8 (16.8–35.7)	25.1 (16.8–41.4)
Organ weights in relation to the controls
Ventral prostate (mg)
90%	44.7 (13.9–79.3)	44.6 (6.2–65.2)	24.4 (0.1–68.6)
60%	103.7 (78.9–130.5	162.1 (128.6–226.5)	148.4 (98.4–200.8)
LABC (mg)
90%	51.0 (19.3–81.9)	48.1 (3.5–78.3)	25.9 (0.02–93.1)
60%	118.9 (93.9–152.1)	179.7 (142.8–325.3)	176.2 (109.1–252.8)
Likelihood for malformation at PND16 (cleft phallus)
10%	25.5 (21.0–31.2)	96.5–133.1b*	140.0–140.1b*
50%	37.5 (32.6–44.6)	149.1–226.2b*	226.2–226.4b*

Abbreviations: DA, dose addition; IA, independent action. Predicted effect doses are based on pooled data from studies for vinclozolin, prochloraz, finasteride, and DEHP and were calculated from the respective dose–response functions given in Supplemental Material, Table 2 (doi:10.1289/ehp.0900689.S1).

^aEffect doses as calculated from the dose–response functions given in Supplemental Material, Table 2 (doi:10.1289/ehp.0900689.S1).

^bNo pups with malformations were observed for the tested doses of DEHP and prochloraz. For this reason, the calculation of expected mixture effect was based on two conjectures: For doses exceeding the tested dose range, we assumed either that effects were absent or that they reached maximal response. These two worst-case extrapolations define the only possible range of effects and were used to calculate the predicted effect doses for the mixture.

^*Statistically significant compared with observed effect doses.