Development and validation of a model using routine clinical features to improve diabetes diagnosis.
4 affiliations
- 1Department of Medicine, Section of Gastroenterology, Hepatology and Nutrition, University of Chicago Medicine, Chicago, IL 60637, USA.
- 2Department of Medicine, Section of Hematology and Oncology, University of Chicago Medicine, Chicago, IL 60637, USA.
- 3Department of Medicine, Section of Endocrinology, Diabetes and Metabolism, University of Chicago Medicine, Chicago, IL 60637, USA.
- 4Department of Medicine, Section of Pulmonary and Critical Care, University of Chicago Medicine, Chicago, IL 60637, USA.
Abstract
Context
Hemoglobin A1c (A1c) and fasting plasma glucose (FPG) miss diabetes diagnosis in many individuals, but the alternative, oral glucose tolerance testing (OGTT), is difficult to obtain.Objective
This work aimed to develop and validate a model to identify diabetes missed by A1c and FPG but detected by OGTT and associate its outcomes with mortality.Methods
We trained machine-learning models using National Health and Nutrition Examination Survey (NHANES) data, applied them to corresponding National Center for Health Statistics mortality data, and externally validated the models with Diabetes Prevention Program Outcomes Study (DPPOS) data. We tested models including the A1c, vital signs, complete blood count, and comprehensive metabolic panel with and without the FPG. This retrospective analysis included NHANES study participants from 1999 to 2016 and DPPOS study participants from 1996 to 2020. Main outcome measures included the area under the receiver operating curve, sensitivity, and specificity of models to detect diabetes.Results
The A1c/FPG+ model had better sensitivity for DPPOS individuals with A1c less than 6.5% (48 mmol/mol) and FPG less than 126 mg/dL compared to a logistic regression of A1c and FPG and recalibrated FPG for individuals both on (0.69 vs 0.53 vs 0.57, respectively) and off (0.75 vs 0.65 vs 0.68, respectively) medication. NHANES participants without diabetes but with A1c/FPG+-predicted diabetes had elevated risk of mortality compared to those without A1c/FPG+-predicted diabetes (HR = 1.4; P < 2 × 10⁻¹⁶).Conclusion
Machine-learning models can improve the detection of individuals who would need OGTT assessment to diagnose diabetes using standard nonfasting laboratory values and vital signs with and without FPG.
Related
- Uncoded Clinical Features from Multilingual Electronic Health Records in Catalonia: Development and Validation Study
- Development and Internal Validation of a Clinical Prediction Model for Coronary Slow Flow Phenomenon in Type 2 Diabetes: A Retrospective Observational Study.
- Primary pulmonary NUT carcinoma: pathological features, clinical diagnosis, treatment, and prognosis — An extended review
- Development and Temporal Validation of a Three-Metabolite Cerebrospinal Fluid Model for the Diagnosis of Leptomeningeal Metastasis from Solid Tumors
- Machine learning prediction model for distant metastasis at diagnosis in pediatric neuroblastoma: development, interpretation, and temporal validation
BibTeX
@article{hutchison2026development,
title = {{Development and validation of a model using routine clinical features to improve diabetes diagnosis.}},
author = {Alan L Hutchison and Simar Narula and Brandon Faubert and Celeste C Thomas and Mary E Rinella and Raghavendra G Mirmira and William F Parker},
journal = {Journal of the Endocrine Society},
year = {2026},
month = sep,
doi = {10.1210/jendso/bvag225},
url = {https://europepmc.org/article/MED/42898234},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Development and validation of a model using routine clinical features to improve diabetes diagnosis.
AU - Alan L Hutchison
AU - Simar Narula
AU - Brandon Faubert
AU - Celeste C Thomas
AU - Mary E Rinella
AU - Raghavendra G Mirmira
AU - William F Parker
PY - 2026
DA - 2026/09/28
JO - Journal of the Endocrine Society
DO - 10.1210/jendso/bvag225
UR - https://europepmc.org/article/MED/42898234
AB - Context Hemoglobin A1c (A1c) and fasting plasma glucose (FPG) miss diabetes diagnosis in many individuals, but the alternative, oral glucose tolerance testing (OGTT), is difficult to obtain.Objective This work aimed to develop and validate a model to identify diabetes missed by A1c and FPG but detected by OGTT and associate its outcomes with mortality.Methods We trained machine-learning models using National Health and Nutrition Examination Survey (NHANES) data, applied them to corresponding National Center for Health Statistics mortality data, and externally validated the models with Diabetes Prevention Program Outcomes Study (DPPOS) data. We tested models including the A1c, vital signs, complete blood count, and comprehensive metabolic panel with and without the FPG. This retrospective analysis included NHANES study participants from 1999 to 2016 and DPPOS study participants from 1996 to 2020. Main outcome measures included the area under the receiver operating curve, sensitivity, and specificity of models to detect diabetes.Results The A1c/FPG+ model had better sensitivity for DPPOS individuals with A1c less than 6.5% (48 mmol/mol) and FPG less than 126 mg/dL compared to a logistic regression of A1c and FPG and recalibrated FPG for individuals both on (0.69 vs 0.53 vs 0.57, respectively) and off (0.75 vs 0.65 vs 0.68, respectively) medication. NHANES participants without diabetes but with A1c/FPG+-predicted diabetes had elevated risk of mortality compared to those without A1c/FPG+-predicted diabetes (HR = 1.4; P < 2 × 10⁻¹⁶).Conclusion Machine-learning models can improve the detection of individuals who would need OGTT assessment to diagnose diabetes using standard nonfasting laboratory values and vital signs with and without FPG.
SN - 2472-1972
ER -