Prognostic models for diabetic kidney disease outcomes: A systematic review
Authors' conclusion
Current DKD prognostic models demonstrate acceptable discrimination but lack adequate calibration and external validation, highlighting the need for more rigorous study designs to enhance reliability and clinical applicability.
Study context
- Population
- Studies of prognostic models for diabetic kidney disease (DKD) outcomes; PubMed, Web of Science, Cochrane Library, Ovid Embase, China National Knowledge Infrastructure (CNKI), Wanfang, Vip and SinoMed from inception to 30 April 202627 studies · 80 models
- Outcome
- Model characteristics, performance and risk of bias
Methodological detailsCHARMS checklist, PROBAST risk of bias
- CHARMS checklist
- PROBAST risk of bias
Pooled result
As reported in the abstract
- Studies included
- 27
- Prediction models
- 80
- Prospective / retrospective studies
- 2 / 25
- Traditional regression / machine learning / both
- 22 / 5 / 4
- Studies with external validation
- 8
- AUC/C-statistic range (most models acceptable discrimination)
- 0.70 to 0.90
- Studies at high risk of bias
- All
- High concerns regarding applicability (mainly single-centre design, restrictive inclusion criteria)
- 21
Parts of this Brief are withheld: the automatic checks could not confirm them against the abstract.
Original publication
Prognostic models for diabetic kidney disease outcomes: A systematic review. Diabet Med. Published 30 June 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →