Maternal and infant prediction of the child BMI trajectories:studies across two generations of Northern Finland birth cohorts

Abstract Background/objective: Children BMI is a longitudinal phenotype, developing through interplays between genetic and environmental factors. Whilst childhood obesity is escalating, we require a better understanding of its early origins and variation across generations to prevent it. Subjects/me...

Full description

Bibliographic Details
Main Authors: Nedelec, R. (Rozenn), Miettunen, J. (Jouko), Männikkö, M. (Minna), Järvelin, M.-R. (Marjo-Riitta), Sebert, S. (Sylvain)
Format: Article in Journal/Newspaper
Language:English
Published: Springer Nature 2020
Subjects:
Online Access:http://urn.fi/urn:nbn:fi-fe2020101484111
Description
Summary:Abstract Background/objective: Children BMI is a longitudinal phenotype, developing through interplays between genetic and environmental factors. Whilst childhood obesity is escalating, we require a better understanding of its early origins and variation across generations to prevent it. Subjects/methods: We designed a cross-cohort study including 12,040 Finnish children from the Northern Finland Birth Cohorts 1966 and 1986 (NFBC1966 and NFBC1986) born before or at the start of the obesity epidemic. We used group-based trajectory modelling to identify BMI trajectories from 2 to 20 years. We subsequently tested their associations with early determinants (mother and child) and the possible difference between generations, adjusted for relevant biological and socioeconomic confounders. Results: We identified four BMI trajectories, ‘stable-low’ (34.8%), ‘normal’ (44.0%), ‘stable-high’ (17.5%) and ‘early-increase’ (3.7%). The ‘early-increase’ trajectory represented the highest risk for obesity. We analysed a dose-response association of maternal pre-pregnancy BMI and smoking with BMI trajectories. The directions of effect were consistent across generations and the effect sizes tended to increase from earlier generation to later. Respectively for NFBC1966 and NFBC1986, the adjusted risk ratios of being in the early-increase group were 1.08 (1.06–1.10) and 1.12 (1.09–1.15) per unit of pre-pregnancy BMI and 1.44 (1.05–1.96) and 1.48 (1.17–1.87) in offspring of smoking mothers compared to non-smokers. We observed similar relations with infant factors including birthweight for gestational age and peak weight velocity. In contrast, the age at adiposity peak in infancy was associated with the BMI trajectories in NFBC1966 but did not replicate in NFBC1986. Conclusions: Exposures to adverse maternal predictors were associated with a higher risk obesity trajectory and were consistent across generations. However, we found a discordant association for the timing of adiposity peak over a 20-year period. This suggests the role of ...