Meter readings confusion

magicaxeman58 said:

Hi, My name is Ian, I have poorly regulated type 2 diabetes due to the fact that I'm bed bound 24/7.

I have two test meters, both obtained earlier this year

An Accu-chek mobile and a one touch ultra easy.

The confusion arises as the meters give completely different readings, for instance a few mins ago I tested my blood on both machines, the accu-chek gave a reading of 17.3 mmol, the one touch gave a reading of 12.2 mmol.

This is a large difference and I really don't know which meter to believe at present.

Any thoughts or ideas?

Click to expand...

Although ISO standards are changing, current meters only have to be +\-20% accurate with high glucose levels.. So you likely have 1 reading low and the other reading high.

If your true value was say, 15, your error could be 3 (20%) either way. Hope that makes sense.

Choose the meter that you feel most comfortable with and stick with it.

It is worth ordering some calibration fluid. and checking your meters against it for your own peace of mind.
From testing a few different meters I found they rarely agree and the readings have a wider spread in the higher end of the scale.

Sirmione said:

It is worth ordering some calibration fluid. and checking your meters against it for your own peace of mind.
From testing a few different meters I found they rarely agree and the readings have a wider spread in the higher end of the scale.

Click to expand...

Accu check send you the calibration fluid for free
Once one of my machines was off so they sent me a new one and... Some free batteries!

always good to change your battery, 12 mths or sooner if you are getting funny readings. but both readings aren't that good

Being bed bound may not mean that you can't control your diabetes. I would recomend you to read about the low carb high/full-fat way of eating LCHF you may even need to see your Dr about your meds

a long page and a few good video’s
http://www.dietdoctor.com/lchf

what to expect the first week
http://lowcarbdiets.about.com/od/lowcarb101/a/firstweek.htm

From the @Southport GP study, who posts here
http://www.practicaldiabetes.com/SpringboardWebApp/userfiles/espdi/file/March 2014/PP Unwin final proofs revised.pdf

urbanracer said:

Although ISO standards are changing, current meters only have to be +\-20% accurate with high glucose levels.. So you likely have 1 reading low and the other reading high.

If your true value was say, 15, your error could be 3 (20%) either way. Hope that makes sense.

Choose the meter that you feel most comfortable with and stick with it.

Click to expand...

@Administrator Could I suggest that someone adds a Sticky about the ISO accuracies, and include ISO (2016) due soon. Perhaps a table showing how the percentages convert to +/- mmol/L readings. Also, it appears that some meters are calibrated for Plasma, and others are Whole Blood, which is I believe a factor of 1.2 A list of meters in each category might help here. And lastly, a section showing how to convert USA readings (mg/dl) to mmol/L may be of interest to those of us reading research papers on the net

Oldvatr said:

@Administrator Could I suggest that someone adds a Sticky about the ISO accuracies, and include ISO (2016) due soon. Perhaps a table showing how the percentages convert to +/- mmol/L readings. Also, it appears that some meters are calibrated for Plasma, and others are Whole Blood, which is I believe a factor of 1.2 A list of meters in each category might help here. And lastly, a section showing how to convert USA readings (mg/dl) to mmol/L may be of interest to those of us reading research papers on the net

Click to expand...

@Oldvatr
You can get % to mmol conversion on the main website......
http://www.diabetes.co.uk/what-is-hba1c.html

When I posted my previous response I was not aware that this page existed .......
http://www.diabetes.co.uk/diabetes_care/blood_glucose_monitor_guide.html

Look at the meter reviews on that page for more detailed information. Not sure if they all include Plasma or Whole Blood designations.

There is a converter on this page for mmol / mgdl here........
http://www.diabetes.co.uk/blood-sugar-converter.html

All I was suggesting is that as this subthread covers blood monitoring, then it was the first place I would tend to look, but I don't want to wade through pages of posts to find this basic info. I suggested it as a sticky so it is easy to find, and yes, maybe it is just a set of links as you have provided here. The subject of whole blood vs plasma has been raised in several threads, and not just by me, and there seems to be a lack of knowledge in the responses. I think that it becomes a whole blood calibration issue where a meter is designed to allow alternate site testing, but it is not made clear, and also the BGL range data limit info published for Newbies here on this forum does not acknowledge that some meters will read high by a factorof 1.2 I suspect my SD Codefree is whole blood since it consistently reads high, andI think my old XCEED meter suffered the same problem (both my XCEEDs read about 3 mmol high.) There is nothing in the literature that mentions whole blood, but I think all the ones I have mentioned can do alternate site tests. When I compare my SD against my Hba1c, then on both occasions it read around 1.4 mmol higher than the plasma equivalent on two occasions now.

Oldvatr said:

All I was suggesting is that as this subthread covers blood monitoring, then it was the first place I would tend to look, but I don't want to wade through pages of posts to find this basic info. I suggested it as a sticky so it is easy to find, and yes, maybe it is just a set of links as you have provided here. The subject of whole blood vs plasma has been raised in several threads, and not just by me, and there seems to be a lack of knowledge in the responses. I think that it becomes a whole blood calibration issue where a meter is designed to allow alternate site testing, but it is not made clear, and also the BGL range data limit info published for Newbies here on this forum does not acknowledge that some meters will read high by a factorof 1.2 I suspect my SD Codefree is whole blood since it consistently reads high, andI think my old XCEED meter suffered the same problem (both my XCEEDs read about 3 mmol high.) There is nothing in the literature that mentions whole blood, but I think all the ones I have mentioned can do alternate site tests. When I compare my SD against my Hba1c, then on both occasions it read around 1.4 mmol higher than the plasma equivalent on two occasions now.

Click to expand...

PS the % to mmol conversion I suggested is for showing how a +/= 20% error looks like at different readings, i.e. at 3, 5, 10, 15 and 20 mmol/L actual bgl. It is not so bad at 20%, but soon it will be 15% which needs a calculator. I think many people on this forum are unaware of how far the readings can be out and still be within spec. This is probably more important for those on insulin/

Some interesting points in this article.
http://www.healthline.com/diabetesmine/why-meters-cant-tell-us-our-blood-sugar-levels#1

Nice article and quite succinct. Thank you for sharing it

Oldvatr said:

@Administrator Could I suggest that someone adds a Sticky about the ISO accuracies, and include ISO (2016) due soon. Perhaps a table showing how the percentages convert to +/- mmol/L readings. Also, it appears that some meters are calibrated for Plasma, and others are Whole Blood, which is I believe a factor of 1.2 A list of meters in each category might help here. And lastly, a section showing how to convert USA readings (mg/dl) to mmol/L may be of interest to those of us reading research papers on the net

Click to expand...

I have finally found the relevant section where there is a Table showing the 2012 ISO accuracies vs readings as I suggested above. The link is http://www.diabetes.co.uk/blood-glucose-meters/blood-glucose-meter-accuracy.html
It took me a while to find it, so I repeat my request for a sticky to collate these links. Note: ISO 2016 is coming.....

Oldvatr said:

@Administrator Could I suggest that someone adds a Sticky about the ISO accuracies, and include ISO (2016) due soon. Perhaps a table showing how the percentages convert to +/- mmol/L readings. Also, it appears that some meters are calibrated for Plasma, and others are Whole Blood, which is I believe a factor of 1.2 A list of meters in each category might help here. And lastly, a section showing how to convert USA readings (mg/dl) to mmol/L may be of interest to those of us reading research papers on the net

Click to expand...

I finally found the ISO (2013) Table showing Error vs readings in mmol/l as requested above at http://www.diabetes.co.uk/blood-glucose-meters/blood-glucose-meter-accuracy.html
http://www.diabetes.co.uk/blood-glucose-meters/blood-glucose-meter-accuracy.html
It may need updating soon for ISO (2016)

I come back to a point I made elsewhere on this topic though. Diabetes is not a condition of exactitudes. So one meter reads differently from another? So what? As a T1, you had to use urine and test tubes and then visual colours to guesstimate what was going on. Now you get an electrochemical reading that, due to reading to 0.1 of a mmol, people believe to be "accurate".

Given that timing, absorption rate, test locations and measurements are all variable and not at all "accurate" then the numbers on blood glucose meters can only ever be a guide. The best you can hope for is consistency. Getting hung up on the accuracy of the numbers or even the accuracy of the numbers isn't going to dramatically change your life. If you target 5mmol/l on a blood glucose meter, then, based on the current and future ISO requirements, you will be measuring within the range you want. with it is 4, 5 or 6 really doesn't matter a vast amount. For a type 1, what's more important at that point is the direction of travel.

tim2000s said:

I come back to a point I made elsewhere on this topic though. Diabetes is not a condition of exactitudes. So one meter reads differently from another? So what? As a T1, you had to use urine and test tubes and then visual colours to guesstimate what was going on. Now you get an electrochemical reading that, due to reading to 0.1 of a mmol, people believe to be "accurate".

Given that timing, absorption rate, test locations and measurements are all variable and not at all "accurate" then the numbers on blood glucose meters can only ever be a guide. The best you can hope for is consistency. Getting hung up on the accuracy of the numbers or even the accuracy of the numbers isn't going to dramatically change your life. If you target 5mmol/l on a blood glucose meter, then, based on the current and future ISO requirements, you will be measuring within the range you want. with it is 4, 5 or 6 really doesn't matter a vast amount. For a type 1, what's more important at that point is the direction of travel.

Click to expand...

Although I completely agree with this the fact that certain meters read consistently high or low does make a difference if on insulin and are a driver. We are all aware of the DVLA guidelines and how strict they are ( as they should be ) so a meter that read lower than it should can make a difference.
I too look for trends though rather than worry too much about single readings and hopefully with technology moving so fast, the margins of error will reduce even further in the future.

dannyw said:

Although I completely agree with this the fact that certain meters read consistently high or low does make a difference if on insulin and are a driver. We are all aware of the DVLA guidelines and how strict they are ( as they should be ) so a meter that read lower than it should can make a difference.
I too look for trends though rather than worry too much about single readings and hopefully with technology moving so fast, the margins of error will reduce even further in the future.

Click to expand...

This is why there's the five to drive guidance, and equally, if your meter shows you at 4.8, the DVLA guidelines say you can still drive. Even if the meter is reading 20% high, and your number is 4.5, with a real number that is 3.6, your meter says 4.5 (so no issue on that point) and you aren't technically or clinically in a hypo state at 3.6!

I too use trend analysis to base my major decisions on, I also use averages to detect and compensate for strip pack changes, However, as a driver on max dose of Gliclazide, I do need to be predictive when I get my readings so I can be confident of not having a hypo in the next hour or so. The other day, I had pre-hypo warnings, My main meter read 4.7, my backup meter read 6.5, While I was in the supermarket my hypo developed and I had to take evasive action. Checking my spreadsheet back at base, I found that my main meter had just changed strips, and was reading 1.6 mmol/L higher than it had been on the previous pack. Neither of my meters had given me a warning,and both apppeared to not meet lthe 0.8 mmol/L ISO requirement. Do I ditch them and hope that the replacement meters are better? Nope, they will suffer the same problem wrt stripchange,

I try to be careful, but can still be fooled. I have had misreads where the two meters differ by up to 10 mmol/L on the same blood drop. I have had repeat readings that reflect the same misread values, but then give more believable results on a third re-try. Where the error is so obviously a misread then it sticks out like a sore thumb (excuse the pun) but how often does it occur day to day, and which meter is misreading?

I think the problem is worse if the meter is reading higher than it should, This can mask a possible hypo approaching, or prompt an increased dose of insulin. My friend (T1D on pump) reacts immediately to his meter, and instantly ups his bolus if he gets a high reading. He is limited on strip allocation, so cannot afford to do much retesting. He averages a significant hypo at least once a week

Oldvatr said:

He is limited on strip allocation, so cannot afford to do much retesting.

Click to expand...

If he is in the UK, then this shouldn't be the case, especially on a pump. I've tagged @noblehead who can provide a letter for him to take to his GP to resolve that situation.

tim2000s said:

If he is in the UK, then this shouldn't be the case, especially on a pump. I've tagged @noblehead who can provide a letter for him to take to his GP to resolve that situation.

Click to expand...

This is the letter that was sent out from the Dept of Health to all gp/diabetes clinics in 2013:

https://www.gov.uk/government/uploa.../Safe-care-of-people-with-type-1-diabetes.pdf

Hi @tim2000s I am not sure where the problem lies, but I was very aware that he ran out of strips just before Xmas holiday, and it took him 2 full days to get enough to last to the New Year. He was desperate and immensly frustrated with his GP practice, but did manage to reach his Consultant. I must say that some of the delay was in physically getting a scrip in front of a pharmacist rather than in getting agreement. However, our County has just been named and shamed in the news yesterday as being the worst in the UK for diabetes care with the highest incidence of amputations and retinopathy. Personally, I have myself been receiving fairly good support (it was not my GP practice above) I will print off the letter for my buddy since he is not able to access the net.