. . "Adjustment for other potential confounders including BMI, socioeconomic status, marital status, maternal height, birth weight, medical complications such as diabetes or hypertension and race/ethnicity varied between the studies." . . . .