TABLE 2.

Associations Between Sugar‐Containing Beverage Intake in Infancy and Liver Fat Fraction and NAFLD in School‐Aged Children

Sugar‐containing beverages (servings/day) at 1 year	MR‐Measured Liver Fat at School Age, n = 1,940
	Liver fat fraction, SD	NAFLD, yes/no
Basic model	0.037 (0.02, 0.06)	1.44 (1.27, 1.64)*
Confounder model	0.026 (−0.02, 0.07)	1.34 (1.06, 1.69) †
Mediator model	−0.004 (−0.05, 0.04)	1.34 (0.97, 1.83)

Values are regression coefficients (95% CIs) from linear regression models that reflect differences in liver fat fraction (in SD) per sugar‐containing beverage intake per day at 1 year. Values are ORs (95% CIs) that reflect the risk of NAFLD per sugar‐containing beverage intake per day at 1 year. The basic model is adjusted for children at 10 years of age, sex, and total energy intake. The confounder model is the basic model additionally adjusted for maternal prepregnancy BMI, education, net household income, child ethnicity, physical activity, and screen time. The mediator model is the confounder model additionally adjusted for sugar‐containing beverage intake at 8 years of age and BMI at 10 years of age. NAFLD was defined as “yes” when liver fat ≥5.0% and as “no” when liver fat <5.0%.

^* P value < 0.01.

^† P value < 0.05.