[Previous Months][Date Index][Thread Index][Join - Register][Login]
  [Message Prev][Message Next][Thread Prev][Thread Next]

Re: [IP] Freestyle accuracy

Laura F. wrote:

> To me, testing isn't something that should take a
> huge amount of skill and dexterity to perform.  It should be something
> you can do even if you're having a bad hypo.  If someone who's
> experienced with bg testing on other meters needs a lot of "practice" to
> get it right with the Freestyle, that's a problem with the meter!  (and
> who wants to waste a ton of test strips "practicing"?)
I agree wholeheartedly!  That is *exactly* why I like the Freestyle
so much.  Ok, the tiny sample size is nice too. :-)

Several weeks ago I participated in a diabetes research study,
and received IV insulin and had BG measured *four* different ways
for 8 hours.  One method was a continuous BG monitor (yep, 2 IV's),
one was a standard piece of hospital lab equipment (hemocrit?), one was an
Accucheck (because that's what that hospital's nurses have to use),
and one was my Freestyle (alternate site) meter.  Bringing the Freestyle
was my idea; this was a golden opportunity to learn more about its accuracy.

The continuous BG monitor in this study is a device used in hospital
rooms.  Every 5 minutes it draws a sample through the IV, recalibrates,
measures, then returns the blood sample, and finally rinses the
sensor with saline.  It barely qualifies as "portable", mounted on an
IV-stand.  Sensors aren't small, and need about 20 minutes to calibrate
before first use!  The equipment really is suited to in-hospital use.

The Accucheck needed a *gigantic* drop of blood.  And if the
sample was merely huge, its BG reading was substantially lower
than the other methods.  When compared to the standard lab
device, its readings were the most variable.  This could easily be
attributed to variations in (droplet) sample size IMHO (i.e. my technique).
I also wasted a number of Accucheck test strips trying to become accustomed
to its "proper technique".

My Freestyle, on the other hand, had very consistent readings compared
against the standard lab device.  Only occasionally do I waste a strip due
to poor technique.

> As for accuracy with the Freestyle -- whenever I've done back-to-back
> tests, the results have usually been within a couple points of each
> other.  I think the largest variation I've gotten was 12 points...
A 12-point variation is probably well within the repeatability specification
for the meter itself - technique notwithstanding.  That is not the same
as accuracy.

It is well documented that measurements of BG from samples taken at
different locations in the body do not exactly match.  One would hope for
a high degree of repeatability from each method, producing a constant
difference in values obtained from alternate methods.  The comparisons
I made above refer to a variable amount of discrepancy.

The researcher is to share data with me (probably after publishing in the
medical journals first), and I'll share them with you.  It is admittedly not
a large sample set, but it was obtained under lab conditions, using a strict
protocol, conducted by a non-commercial third-party.

for HELP or to subscribe/unsubscribe, contact: HELP@insulin-pumpers.org
send a DONATION http://www.Insulin-Pumpers.org/donate.shtml