I’ve run out of reasons to test my BG 28 times a day, so I’ve been forced to try to get my fix from home A1c tests… Now unfortunately, I do not have a lab draw to compare against. So I know that’s a real bummer… because you can’t really get too much from the tests then. However, I figured I could see how reliable the value is. Kind of. So when I ran the first test and saw 6.0, I decided to see what a second one would reveal. I will say that although I once thought of these tests as completely reliable and accurate, my last two tests that I compared against lab draws showed as much as a .5 difference, with the lab draw coming in higher. I was pretty disappointed. So once I find an endo and get a lab a1c drawn, I’ll repeat this to find… the standard deviation… Or something like that, but for now I’ll just show what 4 back to back home tests have me…
Test 1: 6.0
Test 2: 5.9
Test 3: 5.6
Test 4: 5.5
So if I can be up to .5 points higher or lower from the top and bottom scores, I could be anywhere between a 5.0-6.5.
Your results don’t surprise me at all. Once you work in an analytical lab and realize all the work that goes into calibration, internal standards, automating sample size using robotics, etc etc. The more you realize that home tests of any shape are going to be approximations that carry a considerable amount of error. Good to know about the A1c tests. btw, which ones are you using?
I’d not agree with this assessment based on the numbers you saw.
Here’s how I’d see it instead: you took 4 tests. The mean was 5.75. The max was 6.0. The min was 5.5.
This leaves you at a .25% +/- spread from mean. I actually find this to be extremely tight and I’d consider it an extremely successful series of tests myself. I’d consider my home-tested a1c to be 5.75 if I was you and I’d be satisfied with its repeatability.
+/- 2.5%= 99.5% accurate. I only speak for myself but I consider 99.5% acceptable test result.
Had to. My particular multi-test pack included only one tester. I thought about that first but there was no way around it. Not only did I have to do 4 separate pricks, but there’s obviously a 5 minute wait for each test— plus an additional 15-20 minutes for the tester to reset. Maybe I could’ve stuck another disk thing in there to force it to retest, but I didn’t try— not at $10 a pop. So anyway, 4 separate finger pricks over a span of probably an hour and a half. Ish.
Obviously I can’t say exactly, but I was very careful to fill to the tip on each. So as much as is humanly possible, they were filled equally. But I understand the point, and the actual answer is no.
Real answer or long one? I shook each to a count of 8. In my mind, I like to think of that as shaking perfection. So no. I did not.
This one I can’t even play with. I get nervous every single time I do one of these tests that I’m not pressing it down long enough or that I’ve held it too long. Every time. So I’m going to go ahead and go with a probably not on that one.
I know you know all these things. It must be sad not believing in any of the magic? All jokes aside, I used home a1c tests for almost 2 years on a fairly regular basis, and the times a home test coincided with a lab test, they were always within .1 of each other. Maybe there was a .2 once… but it felt like the most accurate test ever. This last couple of times I’ve compared home with lab, there’s been a much bigger difference. But to see something like this… a difference of .5 over the course of 4 back to back tests… well, it kind of makes it feel less magical. I’ve got like 6 more tests in that box though and a lab draw coming up soon enough. I think I’m going to go all in on lab draw day and see what happens. Because I’m a little hard headed like that.
I’m actually not sure I do either. I knew the +/- .5 was kind of BS, but I also just wasn’t willing to do any more tests. Is 4 tests really a large enough sample to do this with? I had a box of 10… what are the chances one read a 6.3?? I’ll also tell you, for whatever it’s worth—which really is probably nothing, that my last lab a1c was a 6.2, and the reason I was testing is because I’m very concerned I’ve continued to rise. I was expecting a 6.4… So I also don’t really agree with my analysis of the numbers, but I think I still feel pretty safe in saying it’s not the most reliable test?? Trust me, I don’t want to believe that. I have had a very good trusting relationship with the home a1c in the past… It’s just a real financial decision now. I don’t want to spend $20 a test if there’s no use in getting the number, and I don’t want to spend $140 or whatever on 10… if there’s no use in getting the number.
You are absolutely correct. On paper, you are absolutely right. So maybe I’m wrong, but if this is typical— and especially considered extremely tight, I don’t think I’d buy another. I’d be bummed, but I wouldn’t. My last draw was a 6.2. I’m struggling a little with keeping the right mentality, and I’d like to get a number to wrap my mind around. I know I’m no 5.5, and I really highly doubt I’m a 6… or even, and I hope I’m wrong, a 6.2. I don’t usually mind waiting until the next lab draw, but I happen to be out an endo… so was trying to keep myself in line until I’ve got something arranged. But you make great points, and after I blow the other 5 tests on lab draw day, I’ll probably save myself the money and skip another order.
Relax. You tested 4 times and got a mathematically proven 5.75. Don’t drive yourself crazy.
All of those numbers are essentially equally excellent within the margin of error. You can hope for perfect results all you want but that’s not how health, chemistry, or life works… but the good news is that your results are repeatably excellent! And that’s cause for celebration instead of uncertainty.
Honestly, I know you felt through a small sample size (<100) that the test would be as accurate as a lab value, but really, it is pretty amazing that those tests exist at all, and are as accurate as they are.
Since CGMS and trends, TIR, std dev, etc, I think A1C is only important to dr.
My lowest lab A1C was 4.6, but that was during chemo so my RBCs had high turnover rate.
Lab wise, it was extremely accurate, but meant nothing relative to my bg control.
I was trying to get fancy and quote you and @MM2 here (with the “why do you define yourself by a number?”), but that didn’t work. Thank you both for the reality check and for the supportive words. I do realize it sounds like I’m obsessing over this one number, but there is a ton of context left out of that. It’s not this specific a1c as much as it is the trend… the slope. And I’m not just slipping in the a1c department but in the overnight BG department, the pre-run scores department, the whether I should have another brownie even though I really don’t even want it department… the chips department, the late night coffee department, whether I even give a rat’s ass department. I’m also slipping in the symptom department and a touch in the mood management and even depression department. I have a lot of departments I’m not running particularly well currently. So.
I just know myself too well, and I’m trying to keep my head right. BUT… a big part of that for me is not sabotaging my efforts once I feel like I’ve already made a big mistake. So even though I just told you a whole lot of things you didn’t ask to hear, I do want to say thank you again for the reality check… because a 6.4, even if it’s that, is not exactly a big mistake. So I can put my sabotage plans away for now.
And probably the chips… but I also don’t want to get crazy.
Balance… its all about balance and reminding yourself about the goals in the big picture! And sometimes you just do the best you can do at the current moment.
I am amazed at all the great accomplishments you have posted, and the journey you’ve been through.
This is great… but I like how you say it much better. I never did tell you, but I repeated your words all throughout my race… and then didn’t die or give up. So I like your words a lot, and I like even more that they are always the same.
Question: What about home kits where you send the blood in? For example, when I joined the Loop study, the JAEB people sent me a home collection kit where I pricked myself and sent the sample in. Is this as accurate as going to a lab and having them prick me and processing it?
If you send you blood in to a lab, it would be a lab value. I would assume that is as good as any other lab assuming they are doing the work to keep their equipment up to snuff.
