Table 4. Associations of Maternal Milk Intake in the NICU With Neurodevelopmental Outcomes at School Age: Modification by Gestational Agea.
| Gestational age at birth, wk | No. | β (95% CI) | P value for interaction | β (95% CI) | P value for interaction | β (95% CI) | P value for interaction |
|---|---|---|---|---|---|---|---|
| Intelligence (WASI b ) | |||||||
| Outcome | Full scale | <.001 | Verbal | .004 | Performance | .004 | |
| 25 | 32 | 2.17 (1.00 to 3.34) | 1.63 (0.33 to 2.93) | 2.12 (1.04 to 3.20) | |||
| 26 | 26 | 1.78 (0.81 to 2.76) | 1.27 (0.18 to 2.36) | 1.81 (0.90 to 2.71) | |||
| 27 | 30 | 1.40 (0.60 to 2.20) | 0.91 (0.02 to 1.80) | 1.49 (0.75 to 2.24)) | |||
| 28 | 50 | 1.02 (0.37 to 1.66) | 0.54 (−0.18 to 1.27) | 1.18 (0.56 to 1.80 | |||
| 29 | 54 | 0.63 (0.08 to 1.19) | 0.18 (−0.44 to 0.80) | 0.86 (0.31 to 1.42) | |||
| 30 | 79 | 0.25 (−0.30 to 0.80) | −0.18 (−0.78 0.41) | 0.55 (−0.02 to 1.11) | |||
| 31 | 100 | −0.14 (−0.77 to 0.49) | −0.54 (−1.21 to 0.12) | 0.23 (−0.41 to 0.88 | |||
| 32 | 96 | −0.52 (−1.29 to 0.25) | −0.91 (−1.72 to −0.10) | −0.08 (−0.86 to 0.70) | |||
| 33 | 107 | −0.91 (−1.85 to 0.04) | −1.27 (−2.26 to −0.28) | −0.40 (−1.34 to 0.55) | |||
| Academic achievement (WRAT-4 b ) | |||||||
| Outcome | Reading standard | <.001 | Spelling | <.001 | Math computation | .02 | |
| 25 | 32 | 3.48 (1.84 to 5.13) | 2.81 (1.31 to 4.31) | 2.02 (0.76 to 3.29) | |||
| 26 | 25 | 2.98 (1.60 to 4.37 | 2.33 (1.07 to 3.60) | 1.75 (0.68 to 2.82) | |||
| 27 | 29 | 2.49 (1.34 to 3.63) | 1.86 (0.81 to 2.91) | 1.48 (0.60 to 2.37) | |||
| 28 | 50 | 1.99 (1.06 to 2.91) | 1.39 (0.52 to 2.25) | 1.21 (0.48 to 1.95) | |||
| 29 | 51 | 2.49 (0.71 to 2.26) | 0.91 (0.18 to 1.64) | 0.94 (0.31 to 1.58) | |||
| 30 | 78 | 0.99 (0.27 to 1.70) | 0.44 (−0.24 to 1.12) | 0.67 (0.06 to 1.29) | |||
| 31 | 100 | 0.49 (−0.29 to 1.26) | −0.03 (−0.77 to 0.71) | 0.41 (−0.27 to 1.08) | |||
| 32 | 96 | −0.01 (−0.94 to 0.92) | −0.51 (−1.38 to 0.37) | 0.14 (−0.67 to 0.94) | |||
| 33 | 107 | −0.51 (−1.66 to 0.63) | −0.98 (−2.04 to 0.09) | −0.13 (−1.11 to 0.84) | |||
| Behavior and executive functioning | |||||||
| Outcome | ADHD symptoms to Conners Third Edition ADHD Index T score | .04 | BRIEF global executive composite | .39 | SDQ total difficulties | .22 | |
| 25 | 32 | 0.59 (−1.29 to 2.47) | 0.06 (−1.27 to 1.39) | 0.12 (−0.51 to 0.75) | |||
| 26 | 26 | 0.22 (−1.36 to 1.81 | −0.05 (−1.17 to 1.07) | 0.05 (−0.48 to 0.58) | |||
| 27 | 31 | −0.14 (−1.45 to 1.17) | −0.16 (−1.09 to 0.77) | −0.02 (−0.47 to 0.42) | |||
| 28 | 51 | −0.50 (−1.58 to 0.57) | −0.26 (−1.04 to 0.51) | −0.10 (−0.47 to 0.27) | |||
| 29 | 54 | −0.87 (−1.78 to 0.05) | −0.37 (−1.04 to 0.30) | −0.17 (−0.50 to 0.15) | |||
| 30 | 79 | −1.23 (−2.09 to −0.37) | −0.48 (−1.14 to 0.18) | −0.25 (−0.57 to 0.08) | |||
| 31 | 98 | −1.59 (−2.53 to −0.66) | −0.59 (−1.32 to 0.14) | −0.32 (−0.68 to 0.04) | |||
| 32 | 92 | −1.96 (−3.07 to −0.84) | −0.70 (−1.57 to 0.17) | −0.39 (−0.83 to 0.04) | |||
| 33 | 104 | −2.32 (−3.68 to −0.96) | −0.80 (−1.85 to 0.25) | −0.47 (−0.99 to 0.05) | |||
Abbreviations: ADHD, attention-deficit/hyperactivity disorder; BRIEF, Behavior Rating Inventory of Executive Function; NICU, neonatal intensive care unit; SDQ, Strengths and Difficulties Questionnaire; WASI, Weschler Abbreviated Scale of Intelligence; WRAT-4, Wide Range Achievement Test, Fourth Edition.
Gestational age at birth was fitted as a continuous variable. The β estimates indicate the difference in points per 25 mL/kg daily of additional maternal milk intake in the NICU for an exact gestational age (eg, 25 weeks plus 0 days). The sample size provided for each gestational age is the number of participants in the gestational age category (eg, 25 completed weeks [25 weeks plus 0 days through 25 weeks plus 6 days]). Estimates for gestational age categories with fewer than 5 participants are not reported. P values are for the interaction of gestational age interval. All estimates are adjusted for treatment, center, maternal tertiary education, maternal occupation, number of adults and children in the home, maternal smoking and alcohol use in pregnancy, parity, race, birth weight (continuous) and antenatal steroid exposure. Generalized estimating equations were used to account for clustering for multiple gestation. Sample size may be slightly different across subscales.
Higher scores on WASI and WRAT-4 indicate more favorable outcomes, whereas lower score on the Conners Third Edition ADHD Index, BRIEF, and SDQ total difficulties scales indicate more favorable outcomes.