All analyses were performed using R software package version R.4.4.2 Codes showing how BMI slope was created can be found in this study (reference below) Burrows K, Heiskala A, Bradfield JP, Balkhiyarova Z, Ning L, Boissel M, et al. A framework for conducting GWAS using repeated measures data with an application to childhood BMI. Nature Communications . 2024 Dec 1;15(1). doi:10.1038/s41467-024-53687-3 The values of each BMI slope were standardized with a mean of 0 and a standard deviation of 1. Logistic regression analyses were performed to calculate odds ratios (OR) and their 95% confidence intervals for the association between BMI slopes and the incidence of MetS. Three models were tested in our analysis; unadjusted, adjusted for sex, and fully adjusted for covariates. We further evaluated sex and cohort specific path analyses using lavaan in R package (34)to estimate the linear relationships among time specific BMI slopes and the development of MetS. The pathways were modelled in two ways: path a) estimating the effects of infancy BMI on MetS, mediated through late childhood BMI, path b) estimating the effects of early childhood BMI on MetS, mediated through late childhood BMI. The model accounted for data missingness through maximum likelihood estimation. Mavis Fosuaa Boateng 13.9.2026