| Univariable effects(adj R2 0.085),β coefficient (95% CI) | Multivariable effects† | ||
| Model 1(adj R2 0.251),β coefficient (95% CI) | Final model(adj R2 0.313),β coefficient (95% CI) | ||
| pBMI (kg/m2) | 29 (19 to 39)*** | 19 (10 to 28)*** | 21 (11 to 30)*** |
| <18.5 | −358 (−550 to −167)*** | −205 (−380 to −30) | −266 (−434 to −98)* |
| 18.5–19.9 | −208 (−333 to −83)** | −129 (−243 to −16)* | −140 (−250 to −32)* |
| 20–24.9 | Reference | Reference | Reference |
| 25–29.9 | 133 (18 to 248)* | 101 (−3 to 205) | 70 (−32 to 173) |
| ⩾30 | 221 (52 to 390)* | 127 (−26 to 281) | 147 (−4 to 298) |
| Birth weight (kg) | 862 (783 to 940)*** | 857 (773 to 944)*** | 791 (705 to 876)*** |
| Gestational age (weeks) | 51 (18 to 84)** | −82 (−116 to −49)*** | −62 (−95 to −30)*** |
| Female gender | −609 (−688 to −531)*** | −518 (−595 to −442) | −525 (−595 to −452)*** |
| Duration of breastfeeding | |||
| None | Reference | Reference | |
| <1 month | 138 (6 to 271)* | 65 (−56 to 186) | |
| 1–3 months | 83 (−53 to 218) | −28 (−152 to 95) | |
| 4–6 months | −125 (−261 to 11) | −223 (−348 to −98)*** | |
| >6 months | −115 (−256 to 27) | −267 (−397 to −136)*** | |
| Maternal age (years) | −12 (−20 to −4)** | −18 (−25 to −10)*** | |
| Maternal height (cm) | 25 (19 to 30)*** | 24 (18 to 30)*** | |
| Paternal height (cm) | 19 (14 to 24)*** | 21 (15 to 26)*** | |
| Maternal education | |||
| <5 years | 196 (80 to 311)** | – | |
| 5–10 years | −15 (−105 to 75) | ||
| >10 years | Reference | ||
| Ethnicity | |||
| Dutch | Reference | Reference | |
| Surinamese | −187 (−377 to 2) | 84 (−98 to 267) | |
| Moroccan | 322 (173 to 472)*** | 478 (324 to 633)*** | |
| Turkish | 512 (323 to 701)*** | 749 (561 to 938)*** | |
| Other non-Western country | 5 (−125 to 135) | 298 (165 to 430)*** | |
| Other Western country | −164 (−318 to −10)* | 73 (−71 to 213) | |
†Multivariable effects are results with pBMI continuous in the model.
*p<0.05; **p<0.01; ***p<0.001.
All adjusted for age of the child at measurement weight 14 months. The residuals from all models were normally distributed and variances were homogeneous. Parity, pre-existing hypertension, diabetes, alcohol use, smoking and cohabitant status were not significantly associated with child weight at age 14 months in any model.