# Re: [IP] A1C question related to BS average

```Avner Kornfeld <email @ redacted> wrote:

> HbA1C is a good measure of average BG. But many experience the
> "rollercoaster" effect which may lead to reasonable HbA1C (6-7) while be
on
> a 50 - 250 on daily basis. What I am looking for is the figure for mean BG
> (HbA1C) and its variance which will give us a clue on the distribution of
> our BG over the last three months.

HbA1c is not a mean in any statistical sense.  It's the measurement of a
biological process.  It is *the* measurement of BG over an extended period.
I would have to do some literature review to find the statistical variance
distribution of HbA1c (if it's even been done).  HbA1c is not normally
distributed, so it would have to use some pretty sophisticated methods.

> HbA1C act as an integral over the last 12 weeks of our BS. What I would
like
> to see is an histogram of our BS. When we do aggressive BS monitoring - 12
> tests / day we approximate this histogram. Glucowatch, or continuous BS is
> the ultimate thing. Much better than one over whole figure that does not
> comprehend a possible rollercoaster with fast transitions from low / high
> BGs.

Generating a histogram of recorded blood glucose measurements is pretty
simple.  Most software associated with meters will do that.  There are also
published approximations of how HbA1c translates to "average" blood glucose
(i.e. HbA1c = 6.0 translates approximately to BG = 140 mg/dl).  But, as has
been pointed out, this association is not linear, so it's not a matter of
plugging in a formula.

Part of my point is that the swings are *not* as important as the HbA1c in
predicting collateral health problems from diabetes.  The process of
glycation is not an immediate one, so that spikes that are quickly corrected
do not have the same import as longer term highs (assuming that the spike is
not so high as to bring on DKA).

Another part of my point is that home test meters do not have the accuracy
and precision to do what I think you are asking for.  To have that degree of
inference, they would need to be fully qualified lab tests (and be true
serum glucose, for that matter) in order to bring the variance into a
meaningful range.

Jim Handsfield
email @ redacted

The opinions expressed are mine and do not necessarily represent those of my
wife who runs our house and makes more important decisions than I do.
----------------------------------------------------------
for HELP or to subscribe/unsubscribe, contact: HELP@insulin-pumpers.org
send a DONATION http://www.Insulin-Pumpers.org/donate.shtml

```