Sometimes you may hear these two terms used interchangeably.
What is the difference between precise and accurate?
Here is a quick example that makes it easy to understand (and doesn’t use a bunch of math terms like Mean Absolute Deviation (MAD), variance, or standard deviation that makes normal people’s eyes bleed. ).
Suppose your blood sugar is currently 120.
If you test it multiple times and you see results such as 132 and 131, those results are more precise but less accurate.
If you test it multiple times and you see results such as 125 and 116, those results are more accurate but less precise.
In the first example, the numbers were closer to each other, but farther from your actual BG value of 120. More precise, but less accurate.
In the second example, the numbers were closer to the actual BG value, but farther from each other. More accurate, but less precise.
The image below shows the difference between accurate and precise:
Obviously, a meter that is BOTH accurate and precise is the best. But which is better between accurate and precise? The answer may surprise you. That will be the topic of another post.
I remember a discussion on these terms in a freshmen engineering class.
A clock that is always running 5 minutes late is more accurate then a clock with a dead battery.
But the dead battery clock is more precise twice a day during the time where it is stuck.
With BG testing, there are also human error or bad product factors that can also contribute. (Hmmm, as I recall the class on human factors, and the story we heard about the guys who got hurt when they lifted a lawnmower and tried to trim their bushes. There were no safety guards or warning labels to tell them not to do it, and they tried to sue the mfg.
this is extremely helpful to me, with one exception:
how do i actually know what my precise BG is before testing?
for example, i test my BG on meter “A” and it reads 120.
i test my BG on meter “B” and it reads 132. so i re-test on both meters.
meter “A” says 127, while meter “B” reads 134.
it would appear that i should trust Meter “B” b/c it seems consistent (only a 2 point difference). but then i come to find that my actual BG is 124. in your example, which i fully understand, it would be meter “A” that is more accurate and meter “B” which is more precise.
so, how on earth, w/out a veinapuncture, would i know which meter reading to trust? is there a solution to this that i am missing? or is this something that you could do at your doctors office when they draw blood from your vein for your A1c?
I am putting together a wiki on this topic. Here is the specific item for that particular subject:
It is helpful validate your meter with a lab blood test. This is easy to do. When you are getting your lab tests done during your endocrinologist visit, right before they draw the blood you should check your BG with your meter twice. Write the numbers down! When you get your endocrinologist’s lab blood test results back, they will usually have the blood glucose value as part of the blood test panel they performed. Compare the lab BG value with the results you got from your meter right at the same time as the blood test (which is why you wrote those numbers down!). Those numbers should be close, particularly if you were in normal range.
Wow, they must be using an in-office A1c without a blood draw. Every three months we get the full blood draw with all of the results. Different strokes…
Usually serum glucose levels are only “truly” measured if they do serious labs like a comprehensive panel… if I was being charged for this without indication more than about once a year I’d be pretty pissed about it…
We aren’t super happy about it, but it is their practice not ours. We are in an HMO, so no choice really. Fortunately the cost is only $5 extra per visit. But we do get a full panel every visit.
I understand for that fingerstick BG test, but I mean outside of that. Like they are not doing any kind of blood test panel. No standard complete blood count, no creatinine, no WBC and RBC count, no hematocrit, vitamin D. All those things.
I don’t know if they never do those tests for kids, or just sometimes, or what. But if you are getting any blood taken from a vein, it seems that glucose would be one of the things they would test - especially for a diabetic!
Only if they order a “comprehensive metabolic panel” is your glucose level actually determined-- which is probably not selected every time… I’m not really sure how often it is. Glucose levels aren’t tested in CBC (shelly just corrected me). Vitamin d levels are a seperate and individual lab test…
I don’t always get blood drawn, but I figure if I have to go through the hassle, they might as well make it worth my while. So those are always part of the results I get.
).
