Zheng Z, Waikar SS, Schmidt IM, et al. - Given that CKD subgroups with different risk profiles of adverse results may be identified by consensus clustering, so, researchers used unsupervised consensus clustering on 72 baseline features in a total of 2,696 participants in the prospective Chronic Renal Insufficiency Cohort (CRIC) study to determine novel CKD subgroups that best represent the data pattern. The cutoff of ± 0.3 was used for the calculation of the standardized difference of each parameter, to demonstrate subgroup characteristics. Three unique CKD subgroups that best represented patients’ baseline features were identified using the algorithm. Overall, not only the patterns of baseline clinical and laboratory measures were synthesized by consensus clustering, but it also unveiled distinct CKD subgroups, which were identified to be related to markedly different risks of important clinical results (CKD progression, cardiovascular disease, and death). Next steps toward precision medicine may be afforded by further assessment of patient subgroups as well as related biomarkers.