Glycemic Index on Nutrition Facts panel

[Kev]

The US FDA should add the glycemic index to the nutrition
facts panel.

Australia has a program for putting glycemic index info on
packaging. Its at <http://www.gisymbol.com.au

[Wuzzy]

Kev <KevinRog@aol.com> wrote in message
news:<B9A4060F.189C7%KevinRog@aol.com>...
> The US FDA should add the glycemic index to the nutrition
> facts panel.
>
> Australia has a program for putting glycemic index info on
> packaging. Its at <http://www.gisymbol.com.au>

Thanks, they have updated it since I was last there: they have
added the insulin index of foods. I have converted it to a
spreadsheet.

the correlation coefficient of GI with Insulin Index is
0.37 R^2=0.61

meaning that carbs and GI combined explain about 60% of
variance. This leaves alot unexplained, likely due to the
type of sugar (fructose) and maybe whether protein and fat
are added.

I have also sorted them in order, curiously toronto-sugar is
in the middle of the insulin index table.. oat porridge is
towards the extreme insulin index..

[Kev]

When I go to the grocery store, here in the US, I want to look
at the label and know what the glycemic index is for every
packaged food I buy. This would be very helpful. Its as
important to know as the carbs and protein for some of us,
especially diabetics, but also for anyone interested in
optimizing their health.

Lets all get together and push the FDA to add the glycemic
index!! Whos with me??!!!!

=)

in article d996c21a.0209111803.5e4d4c58@posting.google.com,
wuzzy at mypcos@hotmail.com wrote on 9/11/02 8:03 PM:

> Kev <KevinRog@aol.com> wrote in message
> news:<B9A4060F.189C7%KevinRog@aol.com>...
>> The US FDA should add the glycemic index to the nutrition
>> facts panel.
>>
>> Australia has a program for putting glycemic index info on
>> packaging. Its at <http://www.gisymbol.com.au>
>
> Thanks, they have updated it since I was last there: they
> have added the insulin index of foods. I have converted it
> to a spreadsheet.
>
> the correlation coefficient of GI with Insulin Index is 0.37
> R^2=0.61
>
> meaning that carbs and GI combined explain about 60% of
> variance. This leaves alot unexplained, likely due to the
> type of sugar (fructose) and maybe whether protein and fat
> are added.
>
> I have also sorted them in order, curiously toronto-sugar is
> in the middle of the insulin index table.. oat porridge is
> towards the extreme insulin index..

[Wuzzy]

ok, if I had bothered to read my printout I'd catch my error
and variance explained is only 13%. Out of interest I tried to
explain the missing variance, by sorting by the *difference*
between insulin and glycemic indices.

out of total 233 foods, there is an exhorbirant-_high_ insulin
compared to glucose with: -sirloin chops with potatoes
-beefburger with rice, cheese and butter -chicken with rice
-cows milk, several varieties -ice cream, several -pinto beans
with tortilla -glucose with beefburger -yoghurt -lentils and
rice -hummus

There is a artificially _low_ insulin release in the
following: -spaghetti -sultanas -some rice varieties.

when I excluded the above foods the variance explained moved
from 13% to 26%, showing how carb synergy with protein (milk
is also unique) might explain alot of the variance.

The reason for the persistently low explained variance is
possibly an artifact of using different methods of testing the
foods. Maybe due to higher intraindividual differences in GIP
hormones (incretins) and insulin responses (eg., due to serum
fatty acids, time of day, insulin sensitivity).

However, reference (1) came to similar conclusions. Notably
the same foods (red meat, dairy and beans) had
disproportionate insulin. Also their R^2 when including
protein, carb and GI in the model was 32.8%, which is close
the 26% here.

I will check that these are independent studies with
independent researchers/equipment. They may be independent
since the "names" are different - eg., "sirloin" was studied
as "beef"? in (1) etc. Nonetheless both studies are from
australia, so I have an incling that they are by same person -
particularly given above similarities in the results. Will
check tommorrow..

REFERENCES:
1.An Insulin index of foods. AJCN 1997:66:1264-76. Susanne HA
Holt. JC Brand Miller, University of Sydney,
Australia.(only 38 foods)
2.The website quoted by other poster.

[Dr. Dickie]

wuzzy wrote:

> ok, if I had bothered to read my printout I'd catch my error
> and variance explained is only 13%. Out of interest I tried
> to explain the missing variance, by sorting by the
> *difference* between insulin and glycemic indices.
>
> out of total 233 foods, there is an exhorbirant-_high_
> insulin compared to glucose with: -sirloin chops with
> potatoes -beefburger with rice, cheese and butter -chicken
> with rice -cows milk, several varieties -ice cream, several
> -pinto beans with tortilla -glucose with beefburger -yoghurt
> -lentils and rice -hummus
>
> There is a artificially _low_ insulin release in the
> following: -spaghetti -sultanas -some rice varieties.
>
> when I excluded the above foods the variance explained moved
> from 13% to 26%, showing how carb synergy with protein (milk
> is also unique) might explain alot of the variance.
>
> The reason for the persistently low explained variance is
> possibly an artifact of using different methods of testing
> the foods. Maybe due to higher intraindividual differences
> in GIP hormones (incretins) and insulin responses (eg., due
> to serum fatty acids, time of day, insulin sensitivity).
>
> However, reference (1) came to similar conclusions. Notably
> the same foods (red meat, dairy and beans) had
> disproportionate insulin. Also their R^2 when including
> protein, carb and GI in the model was 32.8%, which is close
> the 26% here.
>
> I will check that these are independent studies with
> independent researchers/equipment. They may be independent
> since the "names" are different - eg., "sirloin" was studied
> as "beef"? in (1) etc. Nonetheless both studies are from
> australia, so I have an incling that they are by same person
> - particularly given above similarities in the results. Will
> check tommorrow..
>
> REFERENCES:
> 1.An Insulin index of foods. AJCN 1997:66:1264-76. Susanne
> HA Holt. JC Brand Miller, University of Sydney,
> Australia.(only 38 foods)
> 2.The website quoted by other poster.

Thanks wuzzy. Good job! An insulin index to go with a
glycemic index would be a treasure-trove to us diabetics. Too
bad there are only 38 foods listed, I may have to do some
checking myself.

--

Dr. Dickie Skepticult member in good standing #394-00596-438
Poking kooks with a pointy stick
------------------------------------------------------
"The important thing is not to stop questioning. Curiosity has
its own reason for existing."
E. Einstein

[Wuzzy]

wuzzy wrote;
> Thanks, they have updated it since I was last there: they
> have added the insulin index of foods. I have converted it
> to a spreadsheet.

ie., search for insulin >0 on:
http://www.calvin.biochem.usyd.edu.au/GIDB/search.htm

It is from the same group as the reference I posted so the
similarities between the 288 foods in 2002 and 38 foods from
1997 are the same, I comment on the above. Still the fact that
various types of beef with sugar or beef with rice or potatos
shows that there is synergism happening..

also the only way to do this test is isocalorically, ie 1000kJ
portions which is a technique innovated by this group..

[Kev]

You know, you can't tell the potential effect a food has on
insulin by looking at a nutrition label. Without knowing the
glycemic index (or insulin index), all you can do is look at
the ratio of protein to carbs and guess. If you happen to know
the glycemic index for a bunch of foods, you can adjust for
those that you know. But, its all a bunch of guess work.

It would be a lot easier if the GI was listed on the Nutrition
Facts Panel. Don't you agree?

in article d996c21a.0209120501.49884723@posting.google.com,
wuzzy at mypcos@hotmail.com wrote on 9/12/02 7:01 AM:

> wuzzy wrote;
>> Thanks, they have updated it since I was last there: they
>> have added the insulin index of foods. I have converted it
>> to a spreadsheet.
>
> ie., search for insulin >0 on:
> http://www.calvin.biochem.usyd.edu.au/GIDB/search.htm
>
> It is from the same group as the reference I posted so the
> similarities between the 288 foods in 2002 and 38 foods from
> 1997 are the same, I comment on the above. Still the fact
> that various types of beef with sugar or beef with rice or
> potatos shows that there is synergism happening..
>
> also the only way to do this test is isocalorically,
> ie 1000kJ portions which is a technique innovated by
> this group..

[Ian Davies]

yeah, wish they'd do it in the UK too "Kev" <KevinRog@aol.com>
wrote in message news:B9A5656A.18BD3%KevinRog@aol.com...
>
> When I go to the grocery store, here in the US, I want to
> look at the
label
> and know what the glycemic index is for every packaged food
> I buy. This would be very helpful. Its as important to know
> as the carbs and protein
for
> some of us, especially diabetics, but also for anyone
> interested in optimizing their health.
>
> Lets all get together and push the FDA to add the glycemic
> index!! Whos with me??!!!!
>
> =)
>
>
>
>
> in article d996c21a.0209111803.5e4d4c58@posting.google.com,
> wuzzy at mypcos@hotmail.com wrote on 9/11/02 8:03 PM:
>
> > Kev <KevinRog@aol.com> wrote in message
> > news:<B9A4060F.189C7%KevinRog@aol.com>...
> >> The US FDA should add the glycemic index to the nutrition
> >> facts panel.
> >>
> >> Australia has a program for putting glycemic index info
> >> on packaging.
Its
> >> at <http://www.gisymbol.com.au>
> >
> > Thanks, they have updated it since I was last there: they
> > have added the insulin index of foods. I have converted it
> > to a spreadsheet.
> >
> > the correlation coefficient of GI with Insulin Index is
> > 0.37 R^2=0.61
> >
> > meaning that carbs and GI combined explain about 60% of
> > variance. This leaves alot unexplained, likely due to the
> > type of sugar (fructose) and maybe whether protein and fat
> > are added.
> >
> > I have also sorted them in order, curiously toronto-sugar
> > is in the middle of the insulin index table.. oat porridge
> > is towards the extreme insulin index..