Blood testing accuracy, using glucose tracker apps, reporting, picking up on lows, and what’s the point of the point?
For many of us, blood testing is a several-times-a-day activity. We do it the way other people look at their watches. All Type 1s and some Type 2s are likely to be testing daily, although some T2s may only be testing weekly, but even so it’s a big deal.
At a conference, Advanced Technologies & Treatments for Diabetes (ATTD), held in Madrid in early 2020 data was presented that showed the impact of blood glucose monitor selection on glycaemic control when using a diabetes management app. In other words, your choice of blood test meter can affect the advice you get out of some of the apps that are out there, the point being that the apps make suggestions based on the information they get, and if that information is not accurate, how can the advice be?
James Richardson, Medical lead on BGM for Ascensia, explains, “Any blood glucose app requires input of data, such as a blood test result. It then does something with that inputted data, whether from a continuous glucose monitor (CGM) or blood glucose monitor (BGM). To date, this was usually manually inputted. Now we are moving towards interoperable devices, where your choice of pump and your choice of CGM or BGM ‘talk’ to each other and your app.
“Anyone with insulin–treated diabetes would agree that the accuracy of the data is particularly important for hypos, (less than 3.8mmol). Yet the truth is that picking up blood glucose levels accurately absolutely nosedives when blood glucose goes down. Put simply, there is simply less glucose to find to test and measure, so meters are less accurate at detecting it when levels are low.
“Blood glucose meters do have to conform to an ISO standard.* Roughly speaking, they have to be within a 15% accuracy compared to a laboratory test. A margin of 15% – above or below – is considered acceptable; it translates to roughly 95% accuracy. An app is given a value, which may prompt it to advise an action based on that value. For example, it might say ‘consider taking some carbohydrates, it looks like your glucose is going down’. However, if the value was wrong then the advice from the app could be inappropriate. The app is not necessarily wrong, but if the value is not accurate then its advice might not be.
The Ascensia study looked at 5000 diabetes apps, which shows you how many there are out there) of which 370 met the criteria to be included, one of which was that the app be in English. As many as 60% of these gave an alert when the value was low, and that action was possibly required. Not all actions were within American Diabetes Association (ADA) guidelines, either the suggested action was wrong or too vague.
One of the questions in the study had been, how did the app perform? Says Richardson, “If the glucose reading is inaccurate and the app is also incorrect that means that there are anomalies. You need accurate test results and a good quality app in order to get the correct advice and outcome. Ascensia is a leader in accuracy where it matters – at the low end.”
It’s a brave soul who goes for a meter that does not conform to ISO 151197! The ISO standard for self-testing glucose monitoring systems required that manufacturers of these systems provided even greater accuracy than previously. Project leader for ISO 15197, Dr. Alan Cariski, commented at the time of the new standard (in 2013), “More accurate glucose measurements will help patients to better regulate their diabetes through more informed treatment decisions that may affect, for example, dietary intake and medication dose, especially insulin.” https://www.iso.org/home.html
A note about CGM accuracy: this is defined by the Mean Absolute Relative Difference (MARD). This is a measure that shows on average how far away the glucose sensor reading is from a blood glucose reading, irrespective of whether the difference observed is more or less than the blood glucose reading. Source: www.freestylediabetes.co.uk
READER SURVEY RESULTS
Our recent reader survey showed that the most-used blood glucose meters (in order) are:
1. Roche (the Accu-Chek brand)
2. Abbott (FreeStyle brand)
3. Ascensia (Contour Next brand),
4. GlucoRx (Nexus range of meters)
When asked, ‘how often and why do you test?’, the responses showed:
76% test between one and five times a day
30.3% test up to five times a day
29.4% more than five times a day.
7.2% said that they were testing more than six times a day in order to qualify for a FreeStyle Libre.
Responders said that they tested a lot ‘as I like to know what’s going on with my blood glucose level’ (33.4%) and ‘I don’t mind blood tests and test as much as I want to’ (31.8%).
Accuracy matters – the detail
The first-of-its-kind data highlighted the importance of meter selection for diabetes management apps by assessing accuracy at low blood glucose ranges. Ascensia presented data from a study that assessed the performance of blood glucose monitoring systems (BGMSs, or just ‘meters’) when used with mobile applications (apps) for diabetes management1. At the event, data from Ascensia’s presentation highlighted the challenges of hypo management using apps, demonstrating that not all BGMSs using apps are capable of detecting hypos with a high probability (which has the potential to make them less effective in supporting overall glycaemic control).
Apps for Type 1 and Type 2 diabetes management often rely on data from glucose monitoring devices, such as BGMSs. These apps are designed to provide support that can help to reduce the occurrence of hypos and enable more effective glycaemic control. Therefore, the performance of these apps in hypo management is dependent on the quality and accuracy of the data from the BGMSs, particularly in the low blood glucose range. The low blood glucose range (LBGR) is defined as less than 3.9 mmol/L (70 mg/dl).
Ascensia sponsored the study used a statistical model applied to real meter data to calculate the probability of achieving ±15% of a reference blood glucose (BG) value of 50 mg/dl for a variety of available BGMSs that can link to apps for diabetes management1. The blood glucose meters included were Ascensia’s Contour Next One, Accu-Chek Aviva Connect, FreeStyle Freedom Lite, GlucoMen Areo and OneTouch Verio. The study demonstrated that not all systems were capable of detecting hypos with a high probability.
These results follow two other recent studies that demonstrated the accuracy of products in the Ascensia Contour portfolio in the low blood glucose range. Together, the results of these studies highlight the importance of selecting a highly accurate meter in order to enable effective diabetes management through an app.
Sabina Furber, M.D., Chief Medical Officer at Ascensia Diabetes Care, commented, “These findings demonstrate the importance of the accuracy of blood glucose monitoring systems in the low blood glucose range where the risks to health are the highest and the impact of accuracy on the effectiveness of diabetes management apps. We want to be at the forefront of scientific excellence in assessing the quality of diabetes devices and digital solutions. These data further validate the accuracy of Ascensia’s world-renowned Contour product portfolio, which many people depend upon to manage their diabetes.”
- Richardson J. et al., Challenges ofHypoglycemiamanagement using mobile applications. Poster presented at International Conference on Advanced Technologies & Treatments for Diabetes, Madrid, Spain, 2020.
- Stuhr A. et al., Accuracy of CONTOUR®NEXT ONE blood glucose monitoring system in low blood glucose range using probability methodology. Poster presented at International Conference on Advanced Technologies & Treatments for Diabetes, Madrid, Spain, 2020.
- Shaginian R. et al., Blood glucose monitoring systems’ performance in low blood glucose range and its clinical implications. Poster presented at International Conference on Advanced Technologies & Treatments for Diabetes, Madrid, Spain, 2020.