Association between data-driven diabetes subtypes and kidney, cardiovascular, and all-cause mortality outcomes in adults: a systematic review and meta-analysis.

Abstract

Data-driven clustering has identified reproducible diabetes subtypes with distinct metabolic profiles and complication risks. However, their incremental prognostic value and clinical utility beyond conventional risk assessment remain uncertain. We conducted a systematic review and meta-analysis to evaluate associations between diabetes subtypes and long-term clinical outcomes in adults. PubMed/MEDLINE, Scopus, Web of Science, and the Cochrane Library were searched from inception through February 25, 2026, with an updated targeted literature search and citation/reference screening performed through August 2026. Observational cohort studies classifying adults with diabetes using data-driven clustering and reporting longitudinal kidney, cardiovascular, or mortality outcomes were included. Random-effects meta-analyses pooled study-reported hazard ratios (HRs), and narrative synthesis was conducted when pooling was not possible due to differences in comparator subtype, outcome definitions, or clustering frameworks. Thirty-four eligible study reports were included; 12 contributed to quantitative synthesis and 22 were synthesized narratively. Because several reports drew on overlapping source cohorts, a unique review-wide participant total was not calculated. Compared with mild age-related diabetes (MARD), severe insulin-resistant diabetes (SIRD) demonstrated the strongest associations with kidney outcomes, including CKD stage ≥3 (HR 1.84; 95% CI 1.53-2.20), advanced CKD (HR 2.34; 95% CI 1.84-2.99), ESKD (HR 3.83; 95% CI 2.47-5.93), and incident macroalbuminuria (HR 2.08; 95% CI 1.39-3.12). Severe insulin-deficient diabetes (SIDD) was also associated with increased kidney risk. Cardiovascular associations were more heterogeneous, although SIDD was associated with increased cardiovascular mortality compared with mild obesity-related diabetes (MOD) (HR 1.89; 95% CI 1.37-2.62). In an exploratory two-study analysis using MOD as the reference subtype, both SIRD (HR 2.05; 95% CI 1.46-2.87) and SIDD (HR 2.04; 95% CI 1.25-3.32) were associated with increased all-cause mortality. Data-driven diabetes subtypes are associated with differences in kidney, cardiovascular, and mortality risk, with the most consistent associations observed for SIRD and kidney outcomes. Whether subtype classification improves prediction or clinical decision-making beyond conventional risk factors remains uncertain. CRD420261299731.

Abstract as supplied by the indexing source. © the authors / publisher.