Diabetes clusters and risk of diabetes-associated diseases

We read with interest the study by Zaharia and colleagues,1Zaharia OP Strassburger K Strom A et al.Risk of diabetes-associated diseases in subgroups of patients with recent-onset diabetes: a 5-year follow-up study.Lancet Diabetes Endocrinol. 2019; 7: 684-694Summary Full Text Full Text PDF PubMed Scopus (229) Google Scholar who adopted the diabetes five-cluster algorithm proposed in 2018 by Ahlqvist and colleagues2Ahlqvist E Storm P Käräjämäki A et al.Novel subgroups of adult-onset diabetes and their association with outcomes: a data-driven cluster analysis of six variables.Lancet Diabetes Endocrinol. 2018; 6: 361-369Summary Full Text Full Text PDF PubMed Scopus (984) Google Scholar and characterised a cohort of patients with different degrees of whole-body and adipose-tissue insulin resistance. The authors1Zaharia OP Strassburger K Strom A et al.Risk of diabetes-associated diseases in subgroups of patients with recent-onset diabetes: a 5-year follow-up study.Lancet Diabetes Endocrinol. 2019; 7: 684-694Summary Full Text Full Text PDF PubMed Scopus (229) Google Scholar reported distinct metabolic alterations and specific risk patterns for the development of diabetes-related comorbidities and complications after 5 years of follow-up. Here, we comment on two methodological aspects of this study. First, application of the diabetes five-cluster algorithm2Ahlqvist E Storm P Käräjämäki A et al.Novel subgroups of adult-onset diabetes and their association with outcomes: a data-driven cluster analysis of six variables.Lancet Diabetes Endocrinol. 2018; 6: 361-369Summary Full Text Full Text PDF PubMed Scopus (984) Google Scholar in the ongoing German Diabetes Study by Zaharia and colleagues1Zaharia OP Strassburger K Strom A et al.Risk of diabetes-associated diseases in subgroups of patients with recent-onset diabetes: a 5-year follow-up study.Lancet Diabetes Endocrinol. 2019; 7: 684-694Summary Full Text Full Text PDF PubMed Scopus (229) Google Scholar yielded a reproducibility of only 77% after following up patients with newly diagnosed diabetes for 5 years. In particular, the clustering pattern of the severe insulin-deficient diabetes (SIDD) group at baseline was almost reversed by the end of the 5 years of follow-up, which suggests that the diabetes five-cluster algorithm2Ahlqvist E Storm P Käräjämäki A et al.Novel subgroups of adult-onset diabetes and their association with outcomes: a data-driven cluster analysis of six variables.Lancet Diabetes Endocrinol. 2018; 6: 361-369Summary Full Text Full Text PDF PubMed Scopus (984) Google Scholar might not be suitable for this German diabetes cohort and the results should thus be interpreted cautiously. To validate this claim, assessment tools for cluster analysis, such as Jaccard coefficient,3Hennig C Cluster-wise assessment of cluster stability.Comput Stat Data Anal. 2007; 52: 258-271Crossref Scopus (376) Google Scholar a similarity measure between sets, are recommended to judge the effectiveness and stability of a proposed clustering algorithm in other independent groups.4Newby PK Tucker KL Empirically derived eating patterns using factor or cluster analysis: a review.Nutr Rev. 2004; 62: 177-203Crossref PubMed Google Scholar, 5Olsen SF Martuzzi M Elliott P Cluster analysis and disease mapping—why, when, and how? a step by step guide.BMJ. 1996; 313: 863-866Crossref PubMed Scopus (90) Google Scholar Second, the current German diabetes cohort involved 1105 patients at baseline, but only 33·2% of these patients were assessed during the 5 years of follow-up. Zaharia and colleagues interpreted this high loss rate as the result of ongoing cohort monitoring. To facilitate extrapolation of results, additional analyses should be done to see whether baseline characteristics are comparable between patients who were eligible but lost to follow-up at 5 years. We declare no competing interests. Diabetes clusters and risk of diabetes-associated diseases – Authors' replyWe thank Shufang Liu and Wenquan Niu for their interest in our Article,1 which applied the clustering algorithm proposed by Ahlqvist and colleagues2 to patients with diabetes in the German Diabetes Study (GDS).3 We would like to respond to their comments on the methodological aspects of our study. Full-Text PDF Risk of diabetes-associated diseases in subgroups of patients with recent-onset diabetes: a 5-year follow-up studyCluster analysis can characterise cohorts with different degrees of whole-body and adipose-tissue insulin resistance. Specific diabetes clusters show different prevalence of diabetes complications at early stages of non-alcoholic fatty liver disease and diabetic neuropathy. These findings could help improve targeted prevention and treatment and enable precision medicine for diabetes and its comorbidities. Full-Text PDF