Fig. 4

Bland Altman analysis comparing measured SCr in community children versus different methods to estimate bSCr. Graphs show the difference in estimated baseline SCr (bSCr) and measured (mSCr) compared to the average. The top row represents approaches starting with an assumed GFR and back-calculating bSCr using either the Bedside Schwartz equation (a, c) or the Pottel height-independent equation (b). The bottom row represents approaches to estimate bSCr directly using the upper limit of normal (d) or height (e) and age (f) based estimates from the community children. The bias represents the difference between the mSCr and estimated bSCr where the significance of bias was assessed using a one sample t-test comparing the mean bias to 0. The precision is represented as one standard deviation of the mean difference. Proportional bias was evaluated by testing if the slope of the linear regression model of the differences between estimated and measured creatinine against the average of estimated and measured creatinine differed from zero. Proportional bias represents the slope (B1) + SE (standard error) and the asterisks indicate whether the slope is statistically different from zero. Not significant (NS, where p > 0.05), *p < 0.05, **p < 0.001, ***p < 0.0001