Table 1.

Specification of cohorts used in the different steps of our analysis

Subject	Aim	Cohort	Included Neonates (n)	Statistical Methods
GFR reference values	Determination and internal validation of mGFR reference values	(1) IPD: Individual mGFR data	367	Generalized linear mixed model to model the association between mean mGFR and postnatal age; ten-fold crossvalidation of the mean GFR reference values; sensitivity analyses to determine the robustness of our findings, by adding an interaction term between postnatal age and a covariate to the original model, for several covariates
	Verification of mGFR reference values	(2) IPD and aggregated mGFR data	881	Visual comparison between predicted values
k-Value determination	Development of new k-value	(3) Population creatinine concentrations as reported by Boer et al.1 combined with height data from CDC growth charts	Population approach	Linear regression using least-square analysis
	External validation of k-value	(4) Individual creatinine and mGFR data measured and estimated in term-born neonates as part of the HERO study	43	Determining agreement by using accuracy and precision