Table 2.
Association results in the reference group (REF) on the outcome child BMI z-score as continuous trait (N = 495).
| Determining variable | Child GRS | Maternal BMI | Child GRS × maternal BMI (interaction) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| β | 95% CI | p | β | 95% CI | p | β | 95% CI | p | |
| Child BMI-GRS | |||||||||
| Unadjusted | 0.17 | (0.09,0.26) | 4.4E−05 | 0.28 | (0.17,0.39) | 5.1E−07 | |||
| Adjusted | 0.17 | (0.09,0.25) | 4.6E−05 | 0.28 | (0.17,0.39) | 5.2E−07 | 0.03 | (0.00, 0.13) | 0.59 |
| Child obesity-GRS | |||||||||
| Unadjusted | 0.22 | (0.14,0.30) | 1.2E−07 | 0.28 | (0.17,0.39) | 5.1E−07 | |||
| Adjusted | 0.22 | (0.14,0.30) | 1.7E−07 | 0.27 | (0.17,0.38) | 7.1E−07 | 0.07 | (0.01, 0.17) | 0.24 |
Results from multiple linear regression analyses are given as estimated effects (β) in SD units (95% CI) of the various predictors on continuous child BMI z-score. We present unadjusted effects of the child GRSs and maternal BMI where child GRS and maternal BMI were used by themselves (one at a time), and adjusted effects where child GRS was adjusted for maternal BMI and maternal BMI was adjusted for child GRS from models without the interaction terms. Finally, the interaction effect is from the interaction model (child GRS × maternal BMI). The sample size was N = 495 for all reported analyses besides for the association with maternal BMI due to missing BMI information for one mother (N = 494).
GRS genetic risk score, REF group, reference group (randomly selected mothers and their children).