@Diabeticliberty, I presume that you are living in the UK? While I don't and therefore not following local developments that you may have in your regulations, I do note the standards set in this area for the UK since 2016 as described on this very website:
-----
Accuracy standards set to improve
By the end of May 2016,
new standards are being implemented to ensure that blood glucose meters meet stricter accuracy standards.
Under the new standard, meters will need to meet the accuracy guidelines 95% of the time:
- Within± 0.83 mmol/L of laboratory results at concentrations of under 5.6 mmol/L
- Within ± 15% of laboratory results at concentrations of 5.6 mmol/L or more
-----
As you can also find them described on this website:
http://www.diabetes.co.uk/blood-glucose-meters/blood-glucose-meter-accuracy.html
And please notice the regulatory wording carefully:
"...meters will need to meet the accuracy guidelines
95% of the time."
So 5% of the time they can show whatever and still be compliant !!
I used myself the Roche AccuCheck Mobile since it came out till I got the Libre, so know it as a solid trustworthy workhorse.
But I think you mistook my point in my posting above?
Me, you, everybody have to understand the fundamentals of the mechanics and biologics involved when taking a blood sample from a fingerprick to measure the bg level. Understanding these will enable you better to assess what your bg is and potential action you want to take with regards to your therapy. Even the Roche workhorse is way off one way or the other, depending on the situations you may be in and how you conducted the blood sample testing. Aka, does the blood come dripping out of a warm finger tip or do you squeeze a cold bitten finger for 1 minute to get just a micro-drop out? (the mmol/L will wary widely between these two)
Same as well with your Libre sensor, where it is well known some of its shortcomings. E.g. you lay down on the side where the sensor is under the squeezed upper arm. You will then get much lower bg readings due to the restricted transfusion in the tissue around the sensor.
In any case, regarding your sample above, you will probably decide to take some glucose.
But as a veteran T1 you will really welcome the Libre indicator that the 2.9mmol/L trend is flat-lining!
On the Roche you have no such information available. And I can assure you that 5.8mmol/L in a vertical drop is far worse than a 2.9mmol/L in stable flat. I dont have to state all the many benefits of CGM devices over the single point measure meters, as they are well described already on this website.
You are btw the only I have ever heard of using the test solutions, so hats off to that!
As diabetics we often tend to become somewhat complacent with the never ending routine of bg measures and insulin shots.