TABLE 2

Regression Model of Weight Gain and Child IQ, Adjusted for Potential Confounders

Weight GainBirth to 8 wk (N = 4730 [82.0%])a8 wk to 9 mo ( N = 4867 [84.3%])bBirth to 9 mo (N = 4642 [80.4%])b
ParameterSEPParameterSEPParameterSEP
Continuous variable0.840.23.0002−0.230.20.250.190.21.36
Dichotomous (lowest 5%)−1.621.00.110.141.01.89−2.711.03.008
Categorical
    >1.5 SD1.491.02.14−0.190.73.130.020.80.98
    >0.5 to 1.5 SD0.620.53.240.080.53.87−0.790.54.14
Reference group
    Less than −0.5 to −1.5 SD−0.920.53.080.820.56.13−0.590.57.30
Less than −1.5 SD−1.580.93.090.540.93.56−1.860.95.051
  • a Adjusted for (1) variability in who tested the child, age at test, and month of test; (2) potential confounders for IQ: poor maternal education, low paternal social class, insecure tenure, no transport, larger families, low birth-weight, short duration or no breastfeeding, recent illness before the IQ test, and long-term hospitalized treatment over the 8 years; and (3) potential confounders for infant growth: time between weight measurements, short and tall maternal height, young and old maternal age, infant health before weight measurement, and feeding difficulties (general and poor sucking) in the first few weeks of life.

  • b Adjusted for (1) variability in who tested the child, age at test, and month of test; (2) potential confounders for IQ: poor maternal education, low paternal social class, insecure tenure, no transport, larger families, low birth weight, short duration or no breastfeeding, recent illness before the IQ test, and long-term hospitalized treatment over the 8 years; and (3) potential confounders for infant growth: short and tall maternal height, small maternal BMI (birth to 9 months only), and feeding difficulties (refusing non-breast milk and taking too small a quantity) in the first months of life.