Related Experiment Video
Updated: Feb 2, 2026
Author Spotlight: Advancing Diabetes Research with Static Exercise Training in Mice
Published on: March 29, 2024
Patient clusters based on HbA1c trajectories: A step toward individualized medicine in type 2 diabetes
Tomas Karpati1, Maya Leventer-Roberts1, Becca Feldman1
1Clalit Research Institute, Tel Aviv, Israel.
Insights
Machine learning identified three patient clusters with distinct type 2 diabetes (T2D) glycemic trajectories. These clusters reveal varying risks for complications, enabling personalized treatment targets for better health outcomes.
Area of Science:
- Endocrinology
- Data Science in Healthcare
- Diabetes Management
Background:
- Type 2 diabetes (T2D) management relies on monitoring glycated hemoglobin (HbA1c).
- Understanding longitudinal HbA1c patterns is crucial for predicting patient outcomes.
- Individualized glycemic targets may improve T2D complication prevention.
Purpose of the Study:
- To identify clinically meaningful patient clusters based on HbA1c trajectories in T2D.
- To assess the reproducibility and clinical relevance of identified HbA1c clusters.
- To inform personalized glycemic target setting for T2D management.
Main Methods:
- Retrospective cohort study of 60,423 T2D patients.
- Unsupervised machine learning clustering for HbA1c trajectory analysis.
- Random forest analysis for cluster validation and multivariable analysis for clinical relevance.
Main Results:
- Three distinct HbA1c trajectory clusters identified: stable (n=45,679), descending (n=6,084), and ascending (n=8,660).
- Clusters were highly reproducible (99.8% accuracy) via random forest.
- HbA1c thresholds for minimizing risk varied by cluster, showing a J-shape association with outcomes.
Conclusions:
- Unsupervised machine learning effectively clusters T2D patients by HbA1c trajectory.
- Identified clusters demonstrate differential risks for diabetes complications.
- These clusters provide a foundation for personalized glycemic targets in T2D care.
Aims:
To identify clinically meaningful clusters of patients with similar glycated hemoglobin (HbA1c) trajectories among patients with type 2 diabetes.
Methods:
A retrospective cohort study using unsupervised machine learning clustering methodologies to determine clusters of patients with similar longitudinal HbA1c trajectories. Stability of these clusters was assessed and supervised random forest analysis verified the clusters' reproducibility. Clinical relevance of the clusters was assessed through multivariable analysis, comparing differences in risk for a composite outcome (macrovascular and microvascular outcomes, hypoglycemic events, and all-cause mortality) at HbA1c thresholds for each cluster.
Results:
Among 60,423 patients, three clusters of HbA1c trajectories were generated: stable (n = 45,679), descending (n = 6,084), and ascending (n = 8,660) trends, which were reproduced with 99.8% accuracy using a random forest model. In the clinical relevance assessment, HbA1c levels demonstrated a J-shape association with the risk for outcomes. HbA1c level thresholds for minimizing outcomes' risk differed by cluster: 6.0-6.4% for the stable cluster, <8.0% for the descending cluster, and <9.0 for the ascending cluster.
Conclusions:
By applying unsupervised machine learning to longitudinal HbA1c trajectories, we have identified clusters of patients who have distinct risk for diabetes-related complications. These clusters can be the basis for developing individualized models to personalize glycemic targets.
More Related Videos
05:53Measuring Frailty in HIV-infected Individuals. Identification of Frail Patients is the First Step to Amelioration and Reversal of Frailty
Published on: July 24, 2013
16:26Regulatory T cells: Therapeutic Potential for Treating Transplant Rejection and Type I Diabetes
Published on: August 20, 2007
Related Concept Videos
01:22Diabetes Mellitus: Type 2 and Gestational
01:22Diabetes Mellitus: Overview and Type I Subtype
Type 1 diabetes is an autoimmune disease in which the immune system mistakenly attacks and destroys the insulin-producing beta cells in the pancreas. As a result, the body is unable to produce sufficient insulin, and individuals with...
01:20Types of Step-Growth Polymers: Polyesters
Polyesters are commonly prepared from terephthalic acid and ethylene glycol; the crude product is known as poly(ethylene terephthalate) or PET. However, polyesters are synthesized industrially by transesterification of dimethyl terephthalate with ethylene glycol at 150 °C. The two reactants and the polymer...
01:26Orthogonal Trajectories
01:30Impact of Individuals on Individuals
01:20Pathophysiology of Diabetes
Type 1 diabetes is characterized by autoimmune-mediated destruction of pancreatic β cells, with environmental factors potentially triggering this process in genetically susceptible individuals. Despite many not having a family history, certain genes increase susceptibility,...