- advertisement -

Meter Accuracy - Rant!

Discussion in 'Parents of Children with Type 1' started by JackyH, May 10, 2013.

  1. JackyH

    JackyH Approved members

    Joined:
    Jan 7, 2010
    Messages:
    386
    From the time we started pumping/CGMing we have had an issue with our A1C being no where near what the Meter/CGM would suggest it should be. Up to now, we have always used the Contour Link meters that came with our pump - it's just so convenient because, you know, they link with the pump and I cannot trust my almost teenager to enter the BG manually! Anyway, I read a thread a few months ago that pointed to the Freestyle Lite / Freestyle Freedom Lite meters being almost 100% accurate in meter/lab comparisons and I thought ... as soon as my current supply of strips is out, I'm getting me one of those meters. I picked it up this week and we did some testing.

    Two tests / same blood sample for each test one minute apart:-

    Contour Link: 5.9/106
    Freestyle Lite: 6.7/120
    Percentage Difference: 12.7%

    one minute later

    Contour Link: 5.3/95
    Freestyle Lite: 6.7/120 (What!! I've never had the Contour give me the same result on two different blood samples)
    Percentage Difference: 21.5%

    two minutes later on me (non-D)

    Contour Link: 3.9 / 70
    Freestyle Lite: 4.9/ 88
    Percentage Difference: 22.7%

    We have done more simultaneous tests since then and every time, the Contour is almost a point (18 points) lower than the Freestyle - it has never been even close! Obviously these aren't lab tests but I've been suspicious for a while that the contour meters have been reading lower. However, we have our meters compared with the lab blood test every 4 months and, apart from once a year or so ago, they always pass. I'm grumpy - basically we've been using crappy meters for three years. I didn't realise the 20% either way variant was for every test!!! So basically for three years we have been overtreating "lows" that weren't lows and not treating highs as aggressively as we should have been. Funnily enough back at the beginning when I was always trying to get his average down, if I had to verify a high I wouldn't do it on his main meter in case I messed with the average :rolleyes: so I would test with his Precision Xtra and every time it would be up to 2 or 3 (50) points higher! I always assumed the Precision Xtra was rubbish .... duh!

    Has anyone else noticed this? I never really took the 20% variance seriously until now that I have realised what a huge impact it has had on his A1C.

    Anyway, anyone else out there with A1Cs that never match up - look to your meter. That's my rant done. If it helps anyone else question their meter accuracy then I hope it was worth the post.
     
  2. badshoe

    badshoe Approved members

    Joined:
    Jun 28, 2006
    Messages:
    2,153
    The standard for meter accuracy is 95% in the range of +/- 20%. Newer standards that are coming out from ISO this year are said to be +/- 15%.

    100% would still be with in the ranges.

    The new contour strip that works with the new meter that connects with a Minimed is I believe an ISO 2013 meter or +/- 15%.

    Also when two meters differ which one is right? Temperature maters, strip storage matters all kinds of stuff impacts BG readings.

    Diabetes Forcast has a nice piece on the subject:
    http://forecast.diabetes.org/meters-jan2013

    Hope that helps.

    I share your frustrations. Meters are only one part of the juggling act. Carb counting on food labels have a range of accuracy, activity is hard to estimate, insulin curves difficult to predict and ratios hard to estimate. Our lives as Parents of CWD are a crazy juggling game of imprecision. I just try to keep the balls in the air and consider that success.
     
  3. Sarah Maddie's Mom

    Sarah Maddie's Mom Approved members

    Joined:
    Sep 23, 2007
    Messages:
    12,521
    ^ all that Bennet noted.

    We build our house on imperfect foundations; meters, test strips, carb counts, the dozens of factors effecting blood glucose that we cannot know, much less calculate. It's frustrating, but it is what it is.:cwds:
     
  4. swellman

    swellman Approved members

    Joined:
    Jul 30, 2008
    Messages:
    3,544
    I think the Verio IQ conforms to the new ISO as well.

    I have a dozen meters in our drawer and I have no confidence that if I tested a "true" 150 I would get half of them within 20%.

    I "feel" more confident with the Verio IQ based on unscientific comparisons.

    By the way, that's what control solutions are for.
     
  5. cdninct

    cdninct Approved members

    Joined:
    Jul 29, 2011
    Messages:
    888
    I'm not sure if the Canadian Medtronic-linked Contour is the same as the new American one, so there might be a problem with that particular brand that those of us in the US are not really aware of?

    A few months ago, I tested every brand of meter in the house on the same blood samples at around 100 and then again above 200. There was definitely variation, especially at higher numbers, but it was "reasonable":

    OT Ultra2 - 100
    Contour NextLink - 102
    Verio - 103
    Aviva - 107
    OT UltraMini - 111
    Freestyle Lite - 114

    Verio - 219
    NextLink - 234
    Aviva - 242
    Freestyle Lite - 243
    Ultra2 - 249
    UltraMini - 264


    In general (though not at all in in my not-so-scientific trials!), I've found that the Freestyle Lite we love reads a bit lower than the Aviva that we are not so keen on, which could possibly contribute to higher a1cs. It doesn't mean that we have stopped using it--because it is reliable and consistent (especially at lower numbers). It just means that we treat highs a bit more aggressively than the meter reading would suggest.

    As others have said, we are forced to put a lot of faith in the precision of technology that is not particularly precise. I think as long as we keep that thought in mind, we will not be led too far astray!
     
  6. Cookie Monster

    Cookie Monster Approved members

    Joined:
    Jan 5, 2009
    Messages:
    131
    Hi Jacky,

    Yours is a very timely...er ...rant!

    As mentioned by others the new ISO standard will be coming into effect soon. The technical committee working on the new standard completed the final voting early last month and the standard is expected to be published, I think, in July. The new standard will be 95% of readings to be within 15% of laboratory reference for readings above 100mg/dl (5.6mmol/l) or within 15mg/dl (0.83mmol/l) of laboratory reference for readings below 100mg/dl (5.6mmol/l).

    Because the new standard is coming out there have been several studies looking at meter performance against the new standard, such as this one from Germany last year.

    I won't regurgitate everything in the study as it will be copyright but of those meters mentioned above in the thread the Freestyle Lite performed extremely well, as did the Aviva (and Aviva Nano).

    The Verio was good in this study (although the Verio Pro was less good!) and the One Touch Ultra, although not included in this study performed well in a different study.

    I couldn't see the Contour Link in the study but the Contour USB was there but performed less well (although by no means the worst) with less than 95% of readings meeting the standard. I'm not sure if the Contour USB compares to the Contour Link but that may explain some of what you see.

    This was a German study looking at meters common to the European market. There may be differences in the meters between markets.

    This study was funded by Roche but the results for the Accu-Chek meters is similar to results in other non-Roche funded studies so I am comfortable that they are true results.

    It will be interesting to see how funding bodies and insurance companies react to the new standard as it is clear that many of the meters currently on the market will fall well short.
     
  7. JackyH

    JackyH Approved members

    Joined:
    Jan 7, 2010
    Messages:
    386
    Thanks everyone. I'm just cranky because we base the majority of our dosing treating decisions off the numbers on the meter and to find out it's so off is a tad irritating. The average on the contour for the last three days was 7.1 (128) - the average on the Freestyle 8.8 (158)! I am very curious to see how this meter change affects our A1C - I have 6 weeks to tweak things.

    It will definitely be interesting to see how the meter companies relay these new ISO standards to the end user. There is no way my contour link would pass. I haven't heard anything about a new linked meter being available yet in Canada for the Veo but I will give them a call.
     
  8. ecs1516

    ecs1516 Approved members

    Joined:
    Dec 11, 2007
    Messages:
    7,028
    When we go to endo appointments they always have us check our meter against the office meter to see what the difference is.
     
  9. JackyH

    JackyH Approved members

    Joined:
    Jan 7, 2010
    Messages:
    386
    Update

    I just wanted to post an update on this. Since we started using the Freestyle Lite meter 6 weeks ago his A1C has come down 1.1% :D even though the meter and sensor average is much higher than it used to be. The other unexpected bonus ... he has been consistently feeling and catching lows in the 70s - maybe because now he is actually low!!! 30s are a thing of the past - we haven't had one in 6 weeks. The endo (very tongue in cheek) thanked us for helping to bring down the clinic's A1C average and said she could use him as a poster boy for teenagers. We are still all over the place numbers wise, he's in the thick of puberty with basals changing like the weather (he has grown almost 4 inches since his last appointment in Feb) but at least now I feel we have a tool we can rely on and his A1C is right where it should be based on the numbers - yay! Funnily enough I had both meters checked against the lab draw and both passed - however the Contour read 5.4 (97) and the Freestyle read 6.8 (122) ...
     
  10. Momontherun

    Momontherun Approved members

    Joined:
    Nov 25, 2012
    Messages:
    103
    Thanks for the update. I love how the contour is linked to the pump but I see it runs a little lower then the accu-check nano. So I am going to check the accuracy.
     
  11. JackyH

    JackyH Approved members

    Joined:
    Jan 7, 2010
    Messages:
    386
    That is the only downside - I definitely miss the linked feature:cwds:
     
  12. wilf

    wilf Approved members

    Joined:
    Aug 27, 2007
    Messages:
    9,652
    We like the Aviva here. It has always stood up well to lab comparisons, is consistent, and our average numbers compare to what we'd expect from the A1Cs.. :cwds:
     
  13. Mary Jayne

    Mary Jayne Approved members

    Joined:
    Sep 2, 2007
    Messages:
    48
  14. Helenmomofsporty13yearold

    Helenmomofsporty13yearold Approved members

    Joined:
    Oct 5, 2008
    Messages:
    702
    Michelle'sMom posted this earlier today

    http://www.integrateddiabetes.com/Articles/ADA 2013 meeting notes.pdf

    which contained this:

    "In a recent study of meter accuracy conducted at Rainier Clinical Research Center, researchers evaluated the MARD (Mean Absolute Relative Difference) of a number of current meters, comparing meter values to simultaneous lab values. A low MARD indicates better meter accuracy. Here are the results:

    ? Contour Next: 3%
    ? Accu-Check Nano: 4%
    ? Verio Pro: 5%
    ? Freestyle Lite: 10%
    ? One Touch Ultra: 11%
    ? TrueTrack: 12%

    Of note: the One Touch Ultra system was particularly inaccurate in the hypoglycemic (<70 mg/dl) range with a MARD of 21%. The Freestyle Lite system tended to underestimate BGs consistently in the higher range? the higher the glucose, the greater the discrepancy. "
     
  15. badshoe

    badshoe Approved members

    Joined:
    Jun 28, 2006
    Messages:
    2,153
  16. JackyH

    JackyH Approved members

    Joined:
    Jan 7, 2010
    Messages:
    386
    So does anyone know which is the most consistently accurate meter available today in Canada? Is it the Aviva? We used the Accuchek Mobile for a while but we hated it - too chunky and lots of failures - every second test seemed to give an error so I kind of lost faith in Accucheck after that.
     

Share This Page

- advertisement -

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice