July 20, 2009

The FDA Notices Meter Inaccuracy

Here's some evidence that "change" is more than a campaign slogan. After eight years when the FDA behaved like a wholly owned subsidiary of the drug industry, there are small signs that the incoming leadership might use its power to help patients, rather than those who prey on them.

I'm referring of course to the report published in the New York Times this weekend that the FDA is considering taking steps to do something to remedy the abysmal accuracy of the blood sugar meters sold to the public.

You can read the report here:

Standards Might Rise on Monitors for Diabetics

What other high tech product can you think of where the markup on the manufacturing cost is in the 10,000% range while well conducted research has proven that the accuracy can be off by as much as 32%?

The drug companies who sell us these obscenely expensive and highly inaccurate devices cover themselves by stating they meet an international standard. What they don't tell you is who set that standard and who ensures it doesn't change. Three guesses and the first two don't count.

Now the FDA is suggesting that it will publish its own standard if the international standard setting body doesn't act to tighten the accuracy standard from its current abysmally wide setting. The government is a huge customer for blood sugar meters, since so many of them are sold to Medicare patients, so if the FDA were to set a new, tighter, standard it would impel meter makers to improve the woeful performance of their products.

I've been using a meter for just shy of 11 years. The Aviva meter I tried and threw out last winter was no more accurate than the ExacTech brand meter I started out with in 1998. In fact, given the wide variation from reading to reading, the Aviva may have been worse. The 2009 Aviva varied from the 1998 meter in three particulars. It took a lot less blood to get a reading, the reading appeared faster, and the strips cost $1.07 apiece retail where the ExacTech strips had cost about $.40 each.

I have long questioned why the huge "charitable" organizations that collect money in the name of people with diabetes have kept silent on this issue. But you have only to look at the ads from meter companies in the publications of organizations like the ADA and JDRF to get your answer. These "sponsors" are cash cows for these organizations, so they are not about to challenge them, even though as the NYTimes article points out, meter manufacturers' greed and resistance to improving their products' quality is not only beggaring people with diabetes, it is killing them.

The NYTimes article points out something I hadn't known: A major difference between the recent study that found that keeping blood sugar tightly controlled in the ICU preserved lives and the one that found that keeping tight control in the ICU caused more deaths was this: the first study used a highly accurate blood sugar meter available only to hospitals. The other used the crappy meters and strips you and I are paying all that money for each month.

When a meter is off by 32% it is impossible to use it to achieve tight control because a reading of 80 mg/dl, a safe, normal number might actually represent a blood sugar level of 54 mg/dl which is very far from safe--especially if it is heading down.

I learned this the hard way a few years ago. When I started insulin I went out and got a brand new meter--an Aviva since I understood that using insulin would require that my readings be as accurate as possible. Since my doctor thought, incorrectly, that I was a "classic Type 2" he started me at a much higher insulin dose than would have been right for me. I developed the classic symptoms of hypo--pounding pulse, waking from sleep at 4AM in a cold sweat, but my meter readings were in the normal range. It was only when I took my meter to the lab and tested within a minute of the blood draw that I learned that my shiny new meter was giving a result that was over 40 mg/dl higher than the lab reading. That was a post-meal test where I was tested with a high blood sugar since my doctor was questioning that I needed insulin and I had to demonstrate why. But given the 40 mg/dl accuracy gap I saw with a reading over 200 mg/dl, who knows what the 90s my bright shiny new Aviva meter was spitting out--to the tune of over $1 a test--really were?

When I contacted Roche, the Aviva's manufacturer, to complain about the poor performance of my Aviva, the customer rep assured me that the reading was "within the accepted standard" which, since I knew what the standard was, I knew wasn't true.

The rep immediately transferred me to someone who tried to get met to say, on a recorded line, that I had not been harmed by the inaccurate meter reading--at the same time that the company was protesting that the meter reading was within an acceptable range of accuracy. When I refused I was transferred to yet another company operative who tried to force me to make the statement exonerating the company.

The company's priorities were crystal clear. That I ended up with three doctor visits trying to find the cause for my nonstop high pulse was seen only as a threat to company profits.

I was using only 1/2 the dose of insulin my doctor had prescribed because I doubted I was a Type 2. If I had been less knowledgeable about diabetes and had started out with the full dose the doctor prescribed, I might well have ended up unconscious, or worse, with a blood sugar meter reading that would have given little clue as to why.

I was lucky, but as the NYTimes article makes clear, many others aren't. And even more people with diabetes decide it is foolish to pay $1.07 for test strips whose results may be so inaccurate as to be worthless. Since it is extremely difficult to maintain normal health if you don't keep track of what your blood sugar level is, this is another way that poor meter quality harms people with diabetes.

It this concerns you, drop a line to the FDA explaining that you are a consumer who uses a blood sugar meter and that you are sick of paying an exorbitant price for a product so inaccurate it may cause you serious health problems. Demand that the FDA stand up to the meter companies and tighten the standards for blood sugar meters.

The Ombudsman for the Center for Devices and Radiological Heath might be one place to start: ombudsman@cdrh.fda.gov.

The press officer for FDA regulation of devices is:
mary.long@fda.hhs.gov (Peper Long, according to the FDA Contact information page).

Finally, to save myself answering a LOT of email here are my current recommendations for blood sugar meters. Though they have poor accuracy, some meters are better than others, and using those meters properly can make a huge difference in your health. The email I receive from readers month after month suggests that using a meter can drop your A1c from as high as 14% to the 5% range. I often hear from people who have done just that using the technique described HERE. Even with an accuracy gap of 25%, a reading of 100 mg/dl is a lot better for your long-term health than one of 300 mg/dl.

I use an Ultra Mini, which is the most consistent meter I could find. It isn't perfect by any means, but it agrees with itself and with another, older Ultra I own pretty well. I have discarded my Ultra II which was not as consistent.

I recommend the Wal-mart Relion meter to people who are paying for their own strips as the Relion strips are much more affordable. The Relion strips are okay as long as you use them within a short period of opening the vial. Over time they seem to lose their accuracy.

If you are just starting blood sugar testing, don't panic if you see an unusual reading. Wash your hands, dry them, and test again to make sure that you didn't just get a bad strip. There are bad strips out there. Lots of them. Let's hope the FDA moves swiftly on this issue.



Jim Purdy said...

Let me distinguish between inaccuracy and inconsistency.

I do use a cheap Reli-On meter from Wal-Mart. Now suppose that my meter is inaccurate by 30 points, so that it may say that my blood glucose is 152, when it is really 122 or 182.

As long as I use the same meter all the time, and if (a big if) I can watch for signs of highs or lows, should the meter at least be consistent with itself?

If I can determine what numbers ON THAT METER are safe for me, can I assume that I can at least use the readings on that meter to know that I am within the right range?

Jenny said...

You are correct that consistency is valuable--which is why I spend the money for Ultra strips.

But many of these meters are not consistent. Test the same drop of blood twice and you see two readings quite far apart.

Alternatively, the accuracy at a low reading may be far different than the accuracy at a high reading, so you can't just apply a known factor.

If there were such a factor the companies could adjust their meters to give accurate readings. It is because they aren't that they insist on defining "accurate" in such a way that a watch that read 6:00 AM at noon one day and 2:00 PM at noon the next day could be sold as "accurate."

Jim Purdy said...

You're depressing me. I never woulda guessed that I couldn't trust big businesses.

; )

trinkwasser said...

Somehow I can guess what the outcome of this will be: meter manufacturers will withdraw from the market all strips for current meters and thus force everyone to buy more expensive meters with even more expensive strips.

Prove me wrong!

Venkat said...


You are absolutely correct. I have been using One touch Utlra Deluxe meter for the past 2-3 years.

When I switched over to Low Carb diet a year ago, I was testing 12-13 times a day and recording the BG values in an excel sheet. Based on the ingested carbs I knew it should be around 90. It showed me 112. I tested second time and it showed me 89. Third time, it displayed 95. When I contacted customer support, she parroted the same, "our meters meet the standards stipulated".....

So it does not matter which one you use - we have to take it with a pinch of salt. Only way to know is by going for complete labs once in 3 months atleast to see if the numbers do correlate with one another.



Krista said...


Even calibrating/testing one's glucose meter against lab tests every 3 months is not reliable as home glucose meters measure capillary blood whereas blood drawn for lab tests is venous blood. I understand that difference alone will result in significant differences in the readings/results.

Venkat said...


Thanks for educating me on this one.



country mouse said...

Maybe, just maybe we should contact an Indian manufacturer to stamp out test strips for us at a substantially reduced cost. If were lucky, we will even find the factory where the original test strips are manufactured.

ravnhawk said...

Thanks, Jenny, a timely post because I"m in the market for a new meter.

Unfortunately, just before I'd made that decision I bought 100 strips for my current meter, Freestyle lite--which I'd had replaced by the company because my readings seemed off. Now I've started testing the new meter for consistency and find it's hit or miss. What a merry-go-round.

I'm leaning toward a Wavsense Presto. The strips are less--I have to pay for my own--and the comments I read say the accuracy is better. Does anyone have experience with it?

Jim Purdy said...

Jenny, I've been very slow in checking my ReliOn Ultima meter and blood glucose test strips for consistency.

Here are 3 blood glucose measurements I got this morning in a span of only 6 minutes, using the same meter and same batch of strips:

851am -- left forefinger -- 233

855am -- right forefinger -- 190

857am -- left forefinger -- 193

With wild fluctuations (40 POINTS!) like that making the readings useless, why should I even bother to test my blood glucose?

Maybe this would be overkill, but could you start a new topic about meter inconsistency, and address my question about whether it's even worth testing if the numbers are useless?

Jim Purdy said...

A little while ago, I commented about my ReliOn meter being wildly inaccurate -- 40 points in 6 minutes. After that, I called ReliOn customer service, and the rep told me that that much variation was "acceptable."

Hogwash! That much variation can result in life-threatening differences in decisions about insulin, pills, diet and exercise.

How can companies get away with this stuff? Any class action lawyers out there?

Jenny said...


Even though there is a wide range in your readings they are ALL much too high to maintain health. So the reading is useful in letting you know you have to do some more work.