February 26, 2025

expert reaction to study looking at continuous glucose monitors and measuring blood sugar levels in people without diabetes

A study published in the American Journal of Clinical Nutrition looks at continuous glucose monitors in people without diabetes. 

Dr Sufyan Hussain, Honorary Reader and Consultant Diabetes Physician, King’s College London, said:

“Although CGMs are valuable for people living with diabetes, there is recognition that in healthy individuals, they can sometimes yield less reliable data – particularly if the devices are not designed or calibrated for non-diabetes ranges.  This is a small study, limited by comparison to blood glucose meters rather than more accurate laboratory glucose assessments.  Nevertheless, it’s an important reminder that continuous glucose monitors – now a growing trend among healthy individuals – may not always provide an accurate picture in certain situations.  Therefore, readings must be interpreted with expert guidance to ensure safe and informed decisions.”

Dr Adam Collins, Associate Professor of Nutrition, University of Surrey, said:

“The appeal of CGMs lies in their non-invasiveness, coupled with their ability to automatically measure glucose every 5 minutes for up to 2 weeks at a time.  This provides the user with a mass of real time data which can be informative for dietary and lifestyle changes, particularly useful on those who are diabetic.  However, this study by Prof Gonzalez and his team is highly relevant and informative given the increasing use of CGMS by healthy, non-diabetic individuals.

“This study nicely illustrates potential issues of accuracy with CGMs.  The robustness of the study lies in its measurement of postprandial glucose excursions on multiple occasions, under standardised conditions, in a well-controlled crossover design.  By design, CGMs sample interstitial fluid, which we assume is directly reflective of blood glucose, but yet could be influenced by other factors that impact glucose movement into tissues.  In addition, there will inevitably be some time lag between blood glucose and interstitial fluid changes.  This lag could be as short as a few minutes or as long as half an hour.  Such a lag can be problematic when detecting rapid changes in blood glucose.

“We are in the middle of a similar validation study investigating the accuracy and reliability of CGMS and have observed that discrepancies can be as high as 1-1.5mmol/L.  Such a discrepancy can be significant in those who are healthy and have normal blood glucose control, as their normal range of blood glucose is typically far smaller.  This can lead to misinterpretation and “misdiagnosis” of apparent blood glucose excursions that still remain within the boundaries of normal.  Indeed, often individuals are basing dietary and lifestyle changes on what are essentially variations of normal any way, compounded by a magnitude of error in the data itself.

“To add to this, in our current study, we are examining an agreement between two CGM devices fitted to participants at the same time, and have observed discrepancies between devices on the same person.  For example, differences between having the device fitted on your dominant or non-dominant arm.  I would also suggest that using handheld devices for measuring capillary blood, as has been the traditional practice (and used in this AJCN study), can also be subject to error.  Similarly, we have observed discrepancies between two handheld devices measuring the same individual’s blood.

“With all this said, I very much agree with the lead authors’ conclusions.  The use and interpretation of CGM devices in healthy individuals should be undertaken with caution.  They certainly shouldn’t be used as the sole basis for dietary changes, restrictions, or extreme lifestyle changes.”

Dr Nicola Guess, academic dietitian and researcher specialising in the dietary prevention and management of type 2 diabetes, University of Oxford, said:

“This was a well conducted randomised controlled trial which compared the glucose response to standard meals using two methods: a CGM (continuous glucose monitor) and capillary blood testing (“finger prick” testing) in 15 healthy males and females.  The investigators found that CGMs overestimate glucose to a clinically-relevant degree in healthy people without prediabetes or diabetes.  This means that people who have normal glucose may be led to believe they have prediabetes.

“This is one of a growing number of studies which demonstrate how cautious we need to be when interpreting CGM data from a person without diabetes.  In this study, the CGM estimated the “time out of range” as 4-fold higher compared to the capillary value.  Other studies have found CGMs can also underestimate glucose (for example during exercise) and are particularly inaccurate when glucose concentration changes rapidly.

“Taken together, I would advise people without diabetes using CGMs to interpret the output with a pinch of salt.”

‘Continuous glucose monitor overestimates glycemia, with the magnitude of bias varying by postprandial test and individual – A randomized crossover trial’ by Katie M. Hutchins et al. was published in the American Journal of Clinical Nutrition at 08:00 UK time on Wednesday 26 February 2025. 

DOI: 10.1016/j.ajcnut.2025.02.024

Declared interests

Dr Sufyan Hussain: “SH has served on the advisory board and undertaken non-promotional educational work for Abbott UK.”

Dr Adam Collins: “No conflict of interest.”

Dr Nicola Guess: “No conflicts with any CMG company.  Nutrition adviser to Beyond Meat (I advise on fortification questions, ingredients etc., and have written blogs for them).  For MyFitnessPal I provide input into their behavioural programmes and have written blogs for them.”