A correct determination of the glomerular filtration rate (GFR) is necessary and at the same time difficult. Using gold standard methods, such as measurement of inulin clearance, are not feasible in clinical practice raising the need for methods to estimate GFR using easy to measure endogenous biomarkers. Plasma concentrations of the filtration markers creatinine and cystatin C alone are not adequate to easily calculate kidney function. This is mainly due to a non-linear relationship between plasma concentrations and GFR and GFR-independent factors influencing the plasma concentrations. Therefore, formulae have been developed to estimate GFR using easily available variables. Currently, the most useful formulae are those developed by the modification of diet in renal disease (MDRD) study and more recently by the chronic kidney disease epidemiology (CKD-EPI) collaboration. For older individuals some specifically validated formulae were developed some years ago, among them the Berlin initiative study 1 (BIS-1) and BIS‑2 formulae. The accuracy of the estimated filtration rate (eGFR) with respect to the true GFR depends on various factors. The accuracy of the formula is especially low in the GFR range above 60 ml/min · 1.73 m, during recent or rapid changes of GFR and in the case of extreme physical traits, especially a very high or low muscle mass. In older individuals an eGFR around 60 ml/min · 1.73 m alone is not sufficient to discriminate between age-related and disease-related decline in GFR. Nonetheless dosing of medications with predominantly renal excretion should be made according to the eGFR.