Background: Evidence from the literature suggests that the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation more accurately estimates glomerular filtration rate (GFR) than the Modification of Diet in Renal Disease (MDRD) Study equation using the same variables, especially at higher GFR. The Bland-Altman graphical method has been used to compare the two equations, but it is biased because the assumption that the mean and the standard deviation of the difference between the two equations are constant across all GFR values does not hold. This study is aimed to suggest an alternative method, based on polynomial fractional regression, to compute the confidence interval of the difference between the two equations. Methods: The data used are drawn from the Italian PIRP (Prevention of Progressive Renal Insufficiency) registry, that collects prospectively demographic and clinical information on 10,687 outpatients with CKD in the Emilia-Romagna region since 2004. Results: Evaluation of the agreement between these formulas showed that there was a good correlation between CKD- EPI and MDRD for values of eGFR between 40 and 10 mL/min/1.73m2. In this range the two formulas may be interchangeable. Conclusions: In the patients with CKD in its early stages (CKD2, CKD3a), we suggest to monitor the time course of renal function with the same estimation formula, because the deviation between the values obtained is quite large. Clinical implications of this method in terms of accurate prediction of risk of renal failure will be discussed.