Popular glucose monitors used to take regular blood sugar readings could be driving poor diets and food restrictions because of inaccurate measurements, according to a recent study.
Continuous Glucose Monitors (CGMs) take blood sugar readings every five minutes and were originally designed for people with diabetes to assess how their body responds to different foods.
However, a spiral in popularity in recent years has led to them increasingly being used by the health-conscious to track their diet and avoid glucose spikes.
Now, research by University of Bath scientists in the American Journal of Clinical Nutrition, suggest they may not be as accurate as once thought.
The study warns that monitors may be overestimating blood sugar levels in healthy adults and prompting them to make unnecessary changes to their diet, reports The Independent.
Carried out in healthy, non-diabetic volunteers, the research compared results from a CGM to the gold standard finger-prick test for blood sugar levels.
Scientists found that the CGMs consistently reported higher levels than the finger-prick test. The monitors overestimated the time spent above the Diabetes UK’s recommended blood sugar level threshold by nearly 400%, causing unnecessary concern for people whose blood sugar was actually well-controlled.
Professor Javier Gonzalez, from the university’s Department of Health, warned people should stick with the finger-prick test if they are looking for accurate readings.
“Continuous glucose monitors (CGMs) are fantastic tools for people with diabetes, because even if a measurement isn’t perfectly accurate, it’s still better than not having a measurement at all,” he said.
“However, for someone with good glucose control, they can be misleading based on their current performance.
“For healthy individuals, relying on CGMs could lead to unnecessary food restrictions or poor dietary choices.”
After consuming a smoothie, one monitor showed a GI (Glycaemic index) of 69, which is medium, while the finger-prick test showed a GI of 53, which is low.
Whole fruits, classified as low-GI by finger prick tests, were classified as medium or high-GI foods by CGMs.
Gonzalez said CGMs may be inaccurate because they measure glucose in the fluid surrounding your cells, not directly in the blood. “This can lead to discrepancies due to factors like time delays, blood flow, and how glucose moves between different parts of the body,” he added.
Study details
Continuous glucose monitor overestimates glycaemia, with the magnitude of bias varying by postprandial test and individual – a randomised crossover trial.
Katie Hutchins, James Betts, Dylan Thompson, Aaron Hengist, Javier Gonzalez.
Published in The Journal of Clinical Nutrition on 26 February 2025
Abstract
Background
Continuous glucose monitors (CGM) are used to characterise postprandial glycaemia, yet no study has directly tested how different test foods/beverages alter CGM accuracy.
Objective
Assess glycaemic responses to test foods/drinks using CGM versus capillary sampling (criterion).
Methods
Fifteen healthy females(n=9) and males(n=6) completed 7 laboratory visits in a randomised crossover design with ≥48h washout between visits. During each visit, participants consumed an oral carbohydrate challenge comprising either 50g glucose (CONTROL), or equivalent carbohydrate as whole-fruits (WHOLE), blended-fruits (BLEND), commercial fruit smoothie (PRODUCT), commercial smoothie ingested over 30±4min (SLOW), commercial smoothie with 5g inulin (FIBRE), commercial smoothie providing 30g carbohydrate (DOSE). Glycaemia was recorded from both CGM and capillary samples every 15min for 120min and expressed as incremental areas under the curve (iAUC). Glycaemic index (GI) was calculated relative to CONTROL where appropriate. Exploratory analyses examined 1) inter-individual heterogeneity of CGM bias versus criterion; and 2) whether CGM bias could be improved with adjustment for baseline differences.
Results
CGM-estimated fasting and postprandial glucose concentrations were (mean±SD) 0.9±0.6 and 0.9±0.5 mmol/L higher than capillary estimates, respectively (both, p<0.001). CGM bias varied by postprandial test such that GI for PRODUCT was higher with CGM (69, 95%CI: 48, 99) versus capillary (53, 95%CI: 40,69; P=0.05). Furthermore, differences in CGM versus capillary fasting glucose concentrations varied by participant(p=0.001). Unadjusted, CGM overestimated time >7.8mmol/L by ∼4-fold, and adjustment for baseline differences reduced this overestimate to ∼2-fold (both p<0.01).
Conclusions
CGM overestimated glycaemic responses in numerous contexts. At times this can mischaracterise the GI. In addition, there is inter-individual heterogeneity of the accuracy of CGM to estimate fasting glucose concentrations. Correction for this difference reduces, but does not eliminate, postprandial overestimate of glycaemia by CGM. Caution should be applied when inferring absolute or relative glycaemic responses to foods using CGM, and capillary sampling should be prioritised for accurate quantification of glycaemic response
See more from MedicalBrief archives:
Continuous glucose monitoring improves glycaemic control versus finger-prick
Wearable sensor provides minimally invasive glucose monitoring
Genetic link between blood sugar levels, headaches – Australian study
Continuous glucose monitoring improves glycaemic control versus finger-prick