TABLE 3.
Model estimates of the initial testing of the total effect (c1) of maternal BMI on human milk leptin, insulin and adiponectin.
| Log-HM leptin (n = 221) |
Log-HM insulin (n = 221) |
Log-HM adiponectin (n = 219) |
||||
| Estimate | 95% CI | Estimate | 95% CI | Estimate | 95% CI | |
| (a) Maternal BMI | ||||||
| Maternal BMI (kg/m2) | −0.37* | [–0.71 to (–0.02)] | – | – | –0.00 | [–0.03 to 0.02] |
| Maternal BMI (kg/m2) for mothers of males α | – | – | 0.0079 | [–0.01 to 0.03] | – | – |
| Maternal BMI (kg/m2) for mothers of females α | – | – | 0.038*** | [0.02 to 0.06] | – | – |
| Infant sex (females vs. males) | 0.28* | [0.00 to 0.5] | −0.71* | [–1.36 to (–0.06)] | –0.04 | [–0.18 to 0.10] |
| Log-plasma leptin | −0.95* | [–1.80 to (–0.10)] | – | – | – | – |
| Maternal BMI*log-plasmaβ | 0.060** | [0.02 to 0.10] | – | – | – | |
αEstimates of the influence of maternal BMI on HM insulin are shown separately for each infant sex instead of the interaction term between maternal BMI and infant sex.
βExposure-mediator interaction term between maternal BMI and maternal plasma concentration of each hormone.
All models were adjusted for visit and mean-centered age.
BMI, body mass index; CI, confidence interval; HM, human milk.
p < ***0.001; p < **0.01; p < *0.05.