Association between maternal and perinatal characteristics and body mass index (BMI) from birth to 12 years of age in the study population, children born between 1994 and 1996 in Stockholm, using conditional mixed model.

Bold numbers are statistically significant
1Age at first antenatal clinic visit, almost around week 12−14 of gestation
2Smoking at least one cigarette per day, at any time during pregnancy
3Diabetes, preeclampsia and eclampsia, hypertension, anemia, renal disease, liver disease and placental disorders which might affect fetal growth
4Total breast feeding: number of months that child has been breastfed, both exclusive and partial
5AGA =  appropriate for gestational age
6SGA =  small for gestational age
7LGA =  large for gestational age
8Time variable in this model is ‘Age’ (year) at weight and height measurement in children. Since the changes in BMI at different ages is not linear or a straight line, different polynomial functions of ‘Age’ has been used to take into account non-linear changes in BMI in different ages. Then, ‘Age’ accounts for linear changes, ‘Age*Age’ accounts for unidirectional curvilinear changes, ‘Age*Age*Age’ accounts for bi-directional curvilinear changes and ‘Age-log’ account for sharp changes, either peak or bottom.
Association between maternal and perinatal characteristics and body mass index (BMI) from birth to 12 years of age in the study population, children born between 1994 and 1996 in Stockholm, using conditional mixed model.