What is it about?
HbA1c is the most widely used blood test to evaluate long-term glucose control in people with type 1 diabetes. It reflects average blood sugar levels over the past 2–3 months and is often used to guide treatment decisions. At the same time, many people now use continuous glucose monitors (CGM), which measure glucose levels throughout the day and night. However, HbA1c and CGM readings do not always agree. Some individuals consistently have HbA1c values that are higher or lower than what their CGM data would suggest. This mismatch can create confusion for both patients and clinicians and may lead to treatment adjustments that are not optimal. In this study, we showed that this difference between HbA1c and CGM is common - and importantly, that it is often consistent within the same individual over time. By using a person’s own historical data, we developed personalized models that better align CGM measurements with their laboratory HbA1c values. Our findings suggest that instead of relying on “one-size-fits-all” formulas, diabetes care could benefit from more individualized interpretations of glucose data. This may lead to more precise, fair, and clinically meaningful treatment decisions.
Featured Image
Photo by Towfiqu barbhuiya on Unsplash
Why is it important?
Continuous glucose monitoring (CGM) is rapidly becoming standard care in type 1 diabetes, and clinical decisions increasingly rely on CGM-derived metrics such as Time in Range and estimated A1c. At the same time, HbA1c remains the gold standard laboratory marker used for diagnosis, risk assessment, and guideline-based treatment targets. Our study addresses an important and often overlooked issue: these two measures do not always agree and this disagreement is not random. We show that HbA1c–CGM discordance is frequently consistent within individuals over time shorter time, suggesting that biological and individual-specific factors play a major role. What makes this work unique is the move from a “population-average correction” toward a personalized modeling approach. Rather than assuming that one formula fits everyone, we demonstrate that using an individual’s own historical data can substantially improve alignment between CGM-derived estimates and laboratory HbA1c values.
Perspectives
For me, this study represents an important shift in how we think about glycemic metrics. For years, I have been interested in how data-driven tools can support better, fairer, and more individualized diabetes care. Working with large-scale CGM datasets has repeatedly shown me that biological systems rarely behave according to simple, uniform rules - yet many of our clinical formulas still assume that they do. Personally, I see this publication as part of a broader movement toward precision medicine in diabetes. We now collect enormous amounts of glucose and other data. The real challenge is no longer measurement, it is interpretation. If we continue to apply one-size-fits-all models to individualized data streams, we underuse the potential of the information available to us.
Simon Cichosz
Aalborg Universitet
Read the Original
This page is a summary of: Narrowing the A1c gap: Personalized modeling of HbA1c– continuous glucose monitor discordance in type 1 diabetes, PLOS Digital Health, February 2026, PLOS,
DOI: 10.1371/journal.pdig.0001229.
You can read the full text:
Resources
Contributors
The following have contributed to this page
