Table 4. Subgroup analyses of the efficacy of prenatal micronutrient-fortified BEP supplementation on maternal and newborn FMI1.
| Subgroup factor | Control | Intervention | Unadjusted difference (95% CI) | P | Adjusted difference (95% CI) | P | ||
|---|---|---|---|---|---|---|---|---|
| n | Mean ± SD | n | Mean ± SD | |||||
| Maternal | ||||||||
| Household food security | 0.720 | 0.892 | ||||||
| Food secure | 76 | 5.39 ± 2.45 | 81 | 5.13 ± 2.37 | −0.29 (−1.21, 0.63) | 0.536 | −0.12 (−0.87, 0.64) | 0.758 |
| Food insecure | 109 | 4.96 ± 2.07 | 105 | 4.88 ± 2.03 | −0.27 (−0.86, 0.31) | 0.354 | −0.30 (−0.83, 0.23) | 0.272 |
| Maternal BMI | 0.277 | 0.112 | ||||||
| ≥21.0 kg/m2 | 110 | 5.91 ± 2.40 | 116 | 5.53 ± 2.44 | −0.58 (−1.28, 0.13) | 0.107 | −0.59 (−1.21, 0.03) | 0.063 |
| <21.0 kg/m2 | 75 | 4.00 ± 1.31 | 70 | 4.08 ± 1.20 | 0.14 (−0.34, 0.62) | 0.570 | 0.11 (−0.38, 0.61) | 0.652 |
| Maternal MUAC | 0.981 | 0.726 | ||||||
| ≥23 cm | 82 | 6.11 ± 2.69 | 76 | 6.05 ± 2.54 | −0.46 (−1.48, 0.56) | 0.371 | −0.61 (−1.54, 0.32) | 0.198 |
| <23 cm | 103 | 4.36 ± 1.39 | 110 | 4.25 ± 1.51 | −0.10 (−0.51, 0.32) | 0.649 | −0.11 (−0.46, 0.25) | 0.552 |
| Newborn | ||||||||
| Household food security | 0.643 | 0.783 | ||||||
| Food secure | 149 | 0.95 ± 1.31 | 146 | 0.77 ± 1.28 | −0.19 (−0.51, 0.13) | 0.249 | −0.12 (−0.44, 0.20) | 0.457 |
| Food insecure | 221 | 0.90 ± 1.44 | 205 | 0.90 ± 1.38 | −0.11 (−0.38, 0.17) | 0.438 | −0.12 (−0.39, 0.16) | 0.406 |
| Maternal BMI | 0.581 | 0.906 | ||||||
| ≥21.0 kg/m2 | 229 | 0.95 ± 1.42 | 221 | 0.87 ± 1.31 | −0.03 (−0.29, 0.22) | 0.805 | −0.04 (−0.30, 0.21) | 0.742 |
| <21.0 kg/m2 | 141 | 0.88 ± 1.34 | 131 | 0.81 ± 1.38 | −0.11 (−0.46, 0.22) | 0.509 | −0.05 (−0.40, 0.29) | 0.760 |
| Maternal MUAC | 0.951 | 0.753 | ||||||
| ≥23 cm | 169 | 1.06 ± 1.40 | 143 | 0.99 ± 1.39 | −0.09 (−0.42, 0.23) | 0.568 | −0.09 (−0.42, 0.24) | 0.590 |
| <23 cm | 201 | 0.81 ± 1.37 | 209 | 0.75 ± 1.29 | −0.08 (−0.34, 0.19) | 0.572 | −0.05 (−0.31, 0.22) | 0.730 |
1Linear regression models were fitted to test interaction between study arm and a subgroup factor at P < 0.10. All models contained the health center and randomization block to account for clustering by the study design. Adjusted models contained additional covariates such as maternal age, height, arm fat index, hemoglobin concentration and parity at study enrollment, household size, wealth index and food security status, and number of postpartum days before the body composition measurement.
BEP, balanced energy-protein
BMI, body mass index
CI, confidence interval
FMI, fat-mass index
MUAC, mid-upper arm circumference
SD, standard deviation.