Background Diabetic nephropathy is a major end stage kidney disease and its timely diagnosis helps in combating the morbidity and mortality associated with it, thereby improving the quality of the patients suffering from the condition. Though identification of microalbuminuria has long been used to diagnose the nephropathy among diabetics, there has always a need for sensitive and specific biomarker that could predict the susceptibility of the diabetic patients to develop nephropathy much ahead of the onset of microalbuminuria. Methods A prospective study was conducted among 125 diabetic patients at various levels of nephropathy like the normoalbuminuric, microalbuminuric and macroalbuminuric patients and 45 healthy controls based on the Joint Committee of Diabetic Nephropathy Classification. A detailed history, clinical characteristics, duration of illness and other baseline investigations along with urinary podocin, urine nephrin, serum carboxy-methyl lysine levels, glycated albumin, glycaemic control, lipid profile were analyzed among the various nephropathy stages and compared with the controls. Results The sensitivity, specificity, positive and negative predictive value of the various studied biomarkers were compared between the controls and diabetic patients with progressing levels of albuminuria. According to the AUC values the studied biomarkers were categorised into five groups as follows : i. AUC 1.0-0.9 : excellent diagnostic accuracy : podocin, nephrin, CML, microalbumin ii. AUC 0.9-0.8 : very good diagnostic accuracy : glycated albumin, HbA1c iii. AUC 0.8-0.7 : good diagnostic accuracy : HOMA-IR iv. AUC 0.7-0.6 : sufficient diagnostic accuracy : none among the studied biomarkers v. AUC 0.6-0.5 : no diagnostic accuracy : serum albumin, bilirubin Conclusion Urinary Podocin, Nephrin, n-Carboxy methyl lysine, Glycated Albumin seem to have very good to excellent diagnostic value in accordance to microalbumin and HbA1c followed by HOMA-IR in the diagnosis of onset and progression of diabetic nephropathy.