Table 3.
Final adjusted logistic regression model including early life predictors of low IQ at age 6
| Predictor | OR a (95% CI) | p-value | Weights |
|---|---|---|---|
| Male | 1.5 (1.2–1.8) | p = 0.0002 | 15 |
| Skin color: both mother and father non-white | 1.9 (1.5–2.1) | p = 0.0000 | 25 |
| Father unemployed at the child’s birth | 1.6 (1.2–2.0) | p = 0.0002 | 18 |
| Mother unemployed during the child’s first 12 months of life | 1.5 (1.2–1.8) | p = 0.0003 | 15 |
| Household income at the child’s birthb | 1.3 (1.2–1.5) | p = 0.0000 | 12b |
| Maternal educationb | 1.8 (1.6–2.2) | p = 0.0000 | 23b |
| Number of siblings at the child’s birth: 3 or more | 1.8 (1.3–2.3) | p = 0.0001 | 22 |
| Number of persons per room at age 12 month: 3 or more | 1.6 (1.3–2.0) | p = 0.0000 | 18 |
| At least one smoking parent during pregnancy | 1.3 (1.1–1.6) | p = 0.0145 | 10 |
| Duration of breastfeeding | p = 0.0000 | ||
| <1 month | 2.2 (1.6–3.1) | 31 | |
| 1–11 months | 1.3 (1.0–1.6) | 10 | |
| ≥12 months | 1 | ||
| Head circumference-for-age deficit during the first year of life | 1.7 (1.2–2.4) | p = 0.0022 | 20 |
| Height-for-age deficit during the first year of life | 1.3 (1.0–1.7) | p = 0.0524 | 10 |
| Maternal perception of the child’s health status (good/fair/poor)c | 1.4 (1.2–1.8) | p = 0.0009 | 14 |
aAdjusted for child’s age, interview setting, IQ test evaluator.
bThe effect indicates increased odds of low IQ by predictor category.
cThe reference category is excellent/very good health.
CI = confidence intervals; IQ = intelligence quotient.
The 2004 Pelotas Birth Cohort (n = 3312).