Related Experiment Video
Updated: Jun 12, 2026
06:46
A High-Throughput Multiplexed Screening for Type 1 Diabetes, Celiac Diseases, and COVID-19
Published on: July 5, 2022
Prospective validation of an AI algorithm to identify adult-onset type 1 diabetes misclassification: protocol for a
Irene Brusini1, Suyin Lee2, Alvi Lai3
1Applied AI Science, IQVIA, London, UK irene.brusini@iqvia.com.
BMJ Open
|June 10, 2026
Summary
An algorithm can identify adult-onset type 1 diabetes (T1D) misclassified as type 2 diabetes (T2D). This prospective study validates its real-world precision in flagging high-risk patients for improved clinical care.
Area of Science:
- Endocrinology and Diabetes Research
- Health Informatics and Data Science
- Clinical Trial Methodology
Background:
- Adult-onset type 1 diabetes (T1D) is frequently misdiagnosed as type 2 diabetes (T2D), leading to delayed treatment and increased complication risks.
- An electronic medical record (EMR)-based algorithm was developed to identify likely misclassified adult T1D cases.
- Prospective validation is crucial to assess the algorithm's clinical utility and real-world accuracy.
Purpose of the Study:
- To prospectively validate an EMR-based algorithm for identifying adult-onset T1D misclassified as T2D.
- To assess the algorithm's real-world precision in flagging high-risk patients for clinician review.
- To evaluate the clinical utility and operational impact of the algorithm in healthcare settings.
Main Methods:
- Prospective, multicentre, non-interventional cohort study involving two healthcare organizations.
- Adult patients with T2D are scored using an algorithm analyzing 24-month historical health information exchange (HIE) data.
- Flagged patients undergo structured chart review by endocrinology or primary care providers for confirmation and guideline-concordant follow-up.
Main Results:
- The primary endpoint is the positive predictive value for confirmed adult-onset T1D among flagged patients.
- Secondary endpoints include operational cascade metrics such as attribution, provider recommendations, and diagnosis updates.
- Exploratory endpoints will examine provider adoption, interpretability, and workflow integration.
Conclusions:
- The prospective validation will determine the algorithm's reliability in identifying misclassified adult T1D.
- Successful implementation could improve diagnostic accuracy and timely treatment for adult-onset T1D.
- This research aims to enhance patient outcomes by reducing diagnostic delays and optimizing care pathways.
Related Concept Videos
01:25Type II Diabetes Mellitus III: Clinical Manifestations and Diagnosis
Type 2 diabetes mellitus develops gradually and is often asymptomatic in early stages.Clinical ManifestationsWhen symptoms appear, they include fatigue, blurred vision, pruritus, delayed wound healing, and recurrent infections, particularly candidal infections. Peripheral neuropathy may present as numbness or tingling in the extremities. Classic hyperglycemia symptoms—polyuria, polydipsia, and polyphagia—are less common. Most patients are overweight and frequently have associated hypertension...
01:15Diabetes: Symptoms, Diagnosis, and Complications
For most patients, experiencing several weeks of polyuria, polydipsia, fatigue, and significant weight loss may indicate the presence of diabetes. Furthermore, adults displaying the phenotypic appearance of type 2 diabetes (particularly those who are obese and not initially insulin-requiring), may have islet cell autoantibodies, suggesting autoimmune-mediated β cell destruction and a diagnosis of latent autoimmune diabetes of adults (LADA). The categorization of glucose homeostasis is based on...
01:12Type I Diabetes I: Introduction
Type 1 diabetes mellitus is a chronic metabolic disorder characterized by an absolute deficiency of insulin resulting from the autoimmune destruction of pancreatic β-cells. Although it can occur at any age, it is most commonly diagnosed in childhood, adolescence, or early adulthood. The loss of insulin production impairs cellular glucose uptake, resulting in persistent hyperglycemia and necessitating lifelong insulin therapy.Autoimmune Destruction of β-CellsThe hallmark of type 1 diabetes is an...