TABLE 3.
Multivariable associations of prenatal dietary patterns during the third trimester of pregnancy and birth outcomes in MADRES (n = 465)1
| Prenatal | Continuous | Dichotomous | |||
|---|---|---|---|---|---|
| Birth weight2 | Gestational age2 | Premature,3 8.0% | SGA,3 9.0% | LGA,3 9.7% | |
| Dietary patterns | β (95% CIs)4 | β (95% CIs)4 | OR (95% CIs)4 | OR (95% CIs)4 | OR (95% CIs)4 |
| SRC | |||||
| Quartile 1 | Referent | Referent | Referent | Referent | Referent |
| Quartile 2 | 0.10 (−0.20, 0.40) | 0.09 (−0.18, 0.36) | 0.93 (0.33, 2.63) | 0.54 (0.22, 1.33) | 1.23 (0.49, 3.09) |
| Quartile 3 | 0.40 (0.10, 0.70) | 0.19 (−0.11, 0.49) | 0.73 (0.24, 2.23) | 0.23 (0.08, 0.72) | 1.46 (0.53, 4.02) |
| Quartile 4 | 0.20 (−0.10, 0.60) | 0.14 (−0.23, 0.51) | 0.82 (0.21, 3.15) | 0.27 (0.07, 0.99) | 1.52 (0.39, 5.84) |
| Ptrend | 0.045 | 0.323 | 0.673 | 0.023 | 0.448 |
| Adj. Ptrend5 | 0.056 | 0.138 | 0.253 | 0.046 | 0.176 |
| VOF | |||||
| Quartile 1 | Referent | Referent | Referent | Referent | Referent |
| Quartile 2 | 0.20 (−0.10, 0.40) | 0.01 (−0.25, 0.28) | 0.57 (0.22, 1.43) | 0.53 (0.22, 1.30) | 1.36 (0.51, 3.61) |
| Quartile 3 | 0.30 (0.00, 0.60) | 0.25 (−0.02, 0.52) | 0.53 (0.21, 1.34) | 0.71 (0.30, 1.64) | 1.93 (0.74, 4.99) |
| Quartile 4 | 0.40 (0.10, 0.70) | 0.32 (0.03, 0.61) | 0.31 (0.10, 0.95) | 0.18 (0.06, 0.58) | 2.46 (0.89, 6.79) |
| Ptrend | 0.001 | 0.013 | 0.033 | 0.006 | 0.063 |
| Adj. Ptrend5 | 0.011 | 0.036 | 0.049 | 0.028 | 0.062 |
Abbreviations: HS, high school; LGA, large for gestational age; MADRES, Maternal and Developmental Risks from Environmental and Social Stressors Study; SGA, small for gestational age; SRC, dietary pattern of solid fat, refined grain, and cheese; VOF, dietary pattern of vegetables, oils, and fruit.
Z-scores were calculated by standardizing birth weight and gestational age at birth by newborn sex.
A premature birth was defined as a birth occurring before the start of the 37th week of pregnancy; being SGA was defined as having a newborn sex-standardized birth weight <10th percentile; and being LGA was defined as having a newborn sex-standardized birth weight >90th percentile (28).
Estimates are mean differences (β; 95% CIs) and ORs (95% CIs) of continuous and dichotomous birth outcomes, respectively, by quartiles of each prenatal dietary pattern during the third trimester of pregnancy, adjusting for maternal age (years), maternal education (less than HS, HS diploma or equivalent, more than HS), prepregnancy BMI (continuous), birth order (number), and mean total energy intake (kcals/d). The Ptrend values were calculated by modeling the midpoint value of each quartile of each derived dietary pattern as a continuous measure.
Adjusted Ptrend values based on a 2-stage false discovery rate procedure (43).