A machine learning algorithm for the prediction of complications incorporated in electronic medical records improves type 2 diabetes care
Authors' conclusion
Embedding an AI-based prediction tool in routine clinical practice improved several quality indicators and therapeutic decisions. Its real-world application shows promise in overcoming clinical inertia and promoting personalized diabetes management.
Study context
- Population
- Patients with T2D in 38 Italian diabetes clinics
- Participants
- 138,558 eligible patients
- Comparison
- Patients with ML-based complication risk score generated vs eligible patients not assessed
- Testn = 20,314
- Outcome
- Quality-of-care indicators (change)
- Follow-up
- 12 months
Methodological detailsmulticentre, 38 centers, EMR-integrated machine learning prediction tool
- Multicentre
- 38 centers
- EMR-integrated machine learning prediction tool
Main finding
As reported in the abstract
HbA1c ≤7.0 %: Test vs Control
+9.0 % vs. +4.5 %
LDL-C <70 mg/dL: Test vs Control
+27.9 % vs. +20.7 %
BMI <25 kg/m²: Test vs Control
+16.5 % vs. +11.0 %
HbA1c >8.0 %: Test vs Control
−18.4 % vs. −10.1 %
Further findings
Therapy initiation
- Test group more frequently initiated antihypertensive, lipid-lowering and cardio-renal protective therapies
Original publication
A machine learning algorithm for the prediction of complications incorporated in electronic medical records improves type 2 diabetes care. Diabetes Res Clin Pract. Published 13 September 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →