Equation for A1c vs BG

I used to remember the simple formula for converting an average BG value to an implied A1c but I forgot! Can someone please refresh my memory? Thx.

From this American Diabetes Association website:

The relationship between A1C and eAG is described by the formula 28.7 X A1C – 46.7 = eAG.

This website displays a conversion calculator where you can input your A1c and it will calculate your estimated average blood glucose. It allows you to choose your blood sugar denominated in mg/dL or mmol/L. Conversely, you may input your average blood sugar and the calculator will give you the equivalent A1c.

I found it by googling “A1c average blood sugar.”


Thx. I gotta remember to Google!

Thx again Terry. I did find the article on the latest thinking on how to convert A1c or eAG to BG and vice versa. I didn’t know the latest thinking is to do it differently than I was accustomed to, so thanks for that too.

The only wrinkle it poses for me now is that I’m a numbers guy who has kept track of many years’ values of A1c from every quarterly appointment with my endo, and it has always been in the old A1c and mg/dl units. So if I start recording the new way of reporting it, I’ll need to remember I’ve switched from measuring apples to measuring rutabagas! That’s ok as long as I do remember that every time I look at my own tables and graphs!

Are you saying that you used a different formula in the past? The one listed on the ADA website is the only one I knew. By the way, my A1c and actual average measured by my CGM do not correlate well. My average blood glucose is equivalent to an A1c 0.5% lower than this formula calculates.

I’m now not very interested in an A1c number and put much more value in my CGM time in range, glucose variability and average glucose.


Not exactly, Terry. I’m saying that every time I visit my endo she does the usual blood drop and puts the sample into a gizmo on the shelf in the examingroom. It whizzes and whirs for about five minutes and eventually spits out a number I’ve always been told is my A1c. For the past several years that figure has ranged between 7.4 and 6.6 with a peak of 7.6 and 6.6 at the lowest since mid-2017.

My point about apples and rutabagas was only to say that IF she tells me later today that, using my current 90-day avg BG of 163 (as defined for me at the moment by Dexcom”s clarity app on my iPhone) and the formula you just gave me, she’ll tell me my eAG is 9.4, whereas the formula I’m used to (the one I can’t remember at the moment) would tell me my A1c is 7.5. If my numbers show a big bump up I’d just have to remember it’s because I’m measuring a different parameter a different way. Neither is wrong; they’re just apples and rutabagas. :laughing:

Terry, I should also note that I don’t worry so much any,ore either about A1c. I’m much more interested in my TIR, like you.

Terry, one more from me: I should also point out that today’s appointment with my endo is by phone or zoom, so the only A1c OR eAG I’d get from her would be an implied value using either formula. Hence my interest today in what the current formula is. Nothing more than that.

Well, dang. Me again. I’ll stand by everything I’ve said above except my math! In the msg above where I said 9.4 I was completely wrong mathematically! Sorry.

I continue to want an A1c test because it is the only way I have to compare my current status to previous years. Time in range is a great measurement but I don’t have statistics on that going back many years. I have a great TIR according to my endo and that is a pat-on-the-head for working hard. But I have a streak of A1c’s going back almost 20 years and I like to see that number. My A1c tends to be about .5 lower than the GMI from Dexcom.

I have A1c results going back to 1985 and bills for similar tests from the late 1970’s. But the tests and ranges were different. It seems to be that 1988 was the year that the ranges became consistent with today’s tests.


Laddie, I agree with you. My historical data (not decades worth but nearly ten years anyway) is also all in one format measured one way, and the whole reason for my OP was just to observe that if the way it’s always measured is now different, those of us who do keep track ourselves just have to remember that the data after some time is different than the data before that time.

1 Like

I’m fluent in typo. :upside_down_face:

1 Like

Spell check!!! It’s both a blessing and a purse.

If this is based on average bg meter values, then the calculated A1C can be way off based on when and how frequent bg is tested. If all bg tests are before meal, none after meal, the estimated A1C would likely be much lower than direct A1C.

Most accurate A1C estimate is based on cgm with readings every 5 minutes.

If you want more frequent A1C numbers, there are OTC kits that you can buy and test yourself.

Mine as well, A1c is about .5% lower than Dexcom GMI. I am much more interested in time in range. I get really annoyed during the wacky 24 hours of a new sensor tending high and the odd false lows. I wish the Dexcom app would allow for a finger stick event and Clarity would allow for notes tied to the graphs. Also exercise could be more well defined than light, medium and heavy. Perhaps data from activity trackers could be sent over.

That.5 difference is common. It’s because probably your control is tighter at night. Most of our red blood cells are produced at night. So it will skew the results.
Red cells are glycated when they move from marrow into blood stream.
My dexcom tells me 6.0 and my a1 c is generally 5.5

That’s why I rely more heavily on TIR

1 Like