I’ve measured my fasting blood sugar (= blood sugar before breakfast) for about four years. I began out of curiosity but became alarmed when the values approached “pre-diabetic” (> 100 mg/dl, diabetic is 126 mg/dl or so). Eventually I learned that walking an hour/day put them in the 80′s consistently. Perhaps 84 is optimal, who knows.
I have used Abbott test meters and strips. They need so little blood that testing is painless. Recently (January 2012?) Abbott introduced new “butterfly” test strips that “wick” the blood. The meters stayed exactly the same. The new test strips are certainly better. I started to use them. I started using them after a gap (a month?). All of a sudden my scores were about 5 mg/dl better — for example, 84 instead of 89. I assumed this was due to lifestyle changes on my part. I was walking more, I was more muscular, whatever. These were plausible explanations. Surely Abbott had not corrected a huge mistake (given the size of the business, the importance of diabetes, and the need for accurate test strips, to be consistently off by 5 mg/dl would be a huge mistake).
Now I wonder. I recently found some old-style test strips, barely expired (2012/04). I have compared them to the new-style strips (expiration 2013/06). Here are my results:
Morning 1. New: 81, 84. Old: 99, 91, 100.
Morning 2. New: 80. Old: 96, 94, 95.
Morning 3. New: 84. Old: 104, 97, 105.
Morning 4. New: 86. Old: 101, 100, 100.
These results involve three different meters. The old strips come from three separate vials. It is clear that the old strips produce readings much higher (about 15 mg/dl higher) than the new strips.
The old strips are expired but I doubt they got 15 mg/dl worse in 1-2 months. I expect they are accurate when they leave the factory and slowly get worse. Now I have some idea of how much worse (and in what direction). Apparently there is a big increase in bias with little increase in variability. I’ve gone from batch to batch before and never noticed a difference. Only when comparing the new strips with the old has a difference been clear. The earlier comparison, with a 5 mg/dl difference, compared unexpired old strips with the new strips.
I conclude that with the old strips, deterioration with age is worse than I expected and I should pay more attention to test strip age.
Cool. Now you are a connoisseur of test strips.
It’s one of the Miracles of Modern Medicine that no test reading ever comes with a plus-or-minus interval attached. To ask them also to record potential bias is therefore a hopeless cause.
Seth: I don’t mind the absence of a plus-or-minus interval. I can figure that out myself by doing repeat tests. I cannot figure out bias and 15 points is really bad.
I am using a Reli On meter that uses strips by Abbott that take a larger amount of blood, and they too suffer from this effect – and it starts happening before they expire. These strips are individually wrapped. I originally noticed this problem with my original meter (that came from Target) that used strips in a bottle, so I switched to a meter that used individually wrapped strips. It still happens, just more slowly. By the end of the pre-expiration period, they are 10-15 points higher.
I bought a digital fever thermometer a few years ago. The first time I had to use it, I measured my temperature about five or six times in a row. I was appalled by the variability in the readings — they differed by as much as a degree. I’d prefer to use one of those old-fashioned mercury thermometers, but I think that they’re not available anymore. Maybe I could get a used one from eBay.
I believe consistency is a big problem with glucose test strips, second only to price (alleged markups 1000%+ over production costs).
You might find this post about Diabetes Technology (and how little has changed) quite interesting: https://www.hanselman.com/blog/TheSadStateOfDiabetesTechnologyIn2012.aspx
Tim ferriss compared sugar strips of various makers, foinding large differences (i faintl think 10-20)
I think it is reported in four hour body