Table 5.

Interaction between alpha- and gamma-tocopherol on 2-year wheezing

Univariate analysis: Distribution of alpha-tocopherol concentrations by 2-year wheezing status in subjects with low, medium, and high gamma-tocopherol concentrations (tertiles)
		Gamma-tocopherol concentration (tertiles, ng/mg creatinine)
		Low	Medium	High
Alpha-tocopherol concentration (ng/mg creatinine) [IQR]	No wheezing	49 [36,74]	73 [55,98]	106 [83,123]
Alpha-tocopherol concentration (ng/mg creatinine) [IQR]	Wheezing	42 [26,53]	67 [45,82]	107 [87,121]
P value		0.004	0.059	0.42
Multivariable regression analysis: The protective association of alpha-tocopherol with 2-year wheezing outcome differs by gamma-tocopherol level
		Gamma-tocopherol concentration (tertiles, ng/mg creatinine)
		Low	Medium	High
	Unadjusted ORs*	0.51 (0.32,0.80)	0.54 (0.28,1.05)	1.64 (0.71,3.81)
Alpha-tocopherol concentration (ng/mg creatinine) IQR difference	aORs*	0.49 (0.30,0.80)	0.59 (0.29,1.18)	1.65 (0.67,4.01)

^*ORs were estimated for the association of alpha tocopherol with 2-year wheezing in low, medium, and high tertiles of gamma-tocopherol. Multivariable logistic regression was used for the adjusted association including infant sex, race, birth weight, breastfeeding status, maternal smoking during pregnancy, maternal asthma, and insurance status. P interaction (likelihood ratio test) cross-product alpha (continuous)—and gamma-tocopherol tertiles = 0.05. Alpha-tocopherol was natural log transformed.