Exercise Training and Insulin

how many people actually want to decipher your overly
complex, wordy posts, professor?

dave

Jim Dumas said:

Tonya said:

In short, you're trying to solve a problem with too many
variables which change due to factors you cann never
control with the accuracy you're striving for. The math
and equatations are interesting but ultimately probably a
waste of time, given the inherent uncontrollable
variables.

Hi Tonya,

I agree that there are many unknown variables in SC
dosing. But with short-acting insulins like Humalog and
Novolog, the injection site issues or delays are almost
nonexistant. Next, in my case, I've estimated the insulin
action profiles for Humalog, Novolog and R for my
metabolism, i.e., with antibody binding delays via the
Glucose Transform. So I have eliminated much uncertainty,
(in my case, antibody binding delays govern insulin action
[this is the case with most long-term T1s, BTW, since
antibody concentrations continue to rise for T1s on pumps
or MDI]), by jumping directly to the glucose uptake stage
in the cascade of insulin transport to target cells. What
is interesting about my antibody binding to R, for
example, is the peak insulin action of 5.7 hours post dose
is extremely predictable, thanks to the antibodies in the
bloodstream. This tends to eliminate other less important
influences like injection site location or temperature of
the injection site, for example.

In any case, I beg to differ. My argument is: _Any_ math
model is better than tossing darts for dosing post-
exercise, as we T1s seem to do now.

So let's refine the math to improve this "back-of-the-
envelope" dosing approach many T1s are currently using.
Moreover, by discussing this on MHD, we exchange ideas and
move toward the best model for this effect.

Finally, this is a good time to discuss exercise effects
on insulin sensitivity with summer activities on the
horizon. (There are certainly gender specific hormonal
influences that I will not address. But the framework of
the male model will still apply to the female as a
starting point.)

And thanks for input, of course.

"Jim Dumas" <[email protected]!mindspring.com> wrote in message
"]news:[email hidden]...

Quoted message said:

In any case, I beg to differ. My argument is: _Any_ math
model is better than tossing darts for dosing post-
exercise, as we T1s seem to do now.

Quoted message said:

Jim Dumas T1 4/86, background retinopathy, rarely
hypoglycemic: <1/mo. lispro+R+U+NPH daily, moderate
exercise, typically <6% HbA1c

I do a lot of exercise. My working day is often
unpredictably strenuous. I run and cycle Insulin dosing has
to take this into account.

My mental calculations are all 'back of the envelope' and
pretty good. I find that 'darts dosing' is very effective
and sums up the whole 'dafne' approach. We learn to
adjust insulin levels through trial and error improving
with practice.

Citizen_DM

"Jim Dumas" <[email protected]!mindspring.com> wrote in message
"]news:[email hidden]...

Quoted message said:

Beav wrote:

Quoted message said:

No thank you, Beav. My wife an I are quite happy. But
ask BAD.

You got married then Jim? I'm sure you mentioned it before,
but I forget things. Anyway, congrats.

And I don't think I'll bother asking BAD for a quickie, you
were ALWAYS my first choice :-))

Beav

oldal4865 said:

I'm a retired engineer. I use my wife's cast-off
computers; I don't even use a digital watch.

That ain't gonna work, Dude. The PeaBrain bicycle computer
fits into my Levi's jean watch pocket. So your cast-off
computers are too big to mount on the handlebars.

Quoted message said:

I don't "handle" the variable workout issue. Instead, I
strive for constant workout at constant time-of-day
with constant food in my belly and constant H + U in my
fat layer.

I want total flexibility to be spontaneous. I also want an
empty stomach during my exercise to keep side-stitches (side
pains that force you to stop your workout) to a minimum.

And nothing is really constant no matter how hard you try.
--
Jim Dumas T1 4/86, background retinopathy, rarely
hypoglycemic: <1/mo. lispro+R+U+NPH daily, moderate
exercise, typically <6% HbA1c

you are an idiot! Of COURSE I exercise.

dave

steinar_m said:

Simple, he does not excercise.

steinar_m said:

I myself find this really hard. Every morning, I bike 24
km to the train, the train uses 35 mins into town, then I
walk for about 25 mins to work. Same rutine on the way
home. Often I have a low, which in turn brings a high when
I am on my way to work.

Share your finding and I will be happy!

Hi Steinar,

I currently ride 14 miles (22.5 km) every other day. I'm
just starting to collect the necessary heart rate versus
insulin dosage data to undestand the effect. I plan to post
the results/model in the September time frame.

This model must work for no exercise as well as exercise
effects. This will provide a broad range of use to cover or
predict excessive workout stress for weekend warriors, as an
example. (What if I ride 100 miles [161 km] on a Saturday at
an average heart rate of 135 BPM? How strong of an insulin
sensitivity effect will I experience before I try this?)

I also have a dosing computer model (HP-48 calculator) that
uses the date-time of the end of the exercise period to
estimate the decline in insulin sensitivity over a two day
post exercise period. I wrote this about 6-7 years ago and
am now finally testing it with the heart rate data of the
bicycle computer. (Life gets in the way.)

In any case, I'll post again when I find the correct model,
--
Jim Dumas T1 4/86, background retinopathy, rarely
hypoglycemic: <1/mo. lispro+R+U+NPH daily, moderate
exercise, typically <6% HbA1c

Jim Dumas <[email protected]!mindspring.com> wrote in
:"]news:[email hidden]:

Quoted message said:

I've been thinking about using two exponential decays
summed to simulate glycogen store rebuild (one for muscle
and the other for liver) after a training session, as the
basic insulin sensitivity modifier for exercise effects.
One could have 5 tau (time constants) set for 12 hours and
the other for 36 hours, as an example. Then adjust the
ratio for the mass of glycogen in each type of tissue.
Lastly, normalize the function using AUC to duplicate the
"observed time averaged insulin sensitivity" we see in our
dosing adjustments. But I haven't put pencil to paper on
this yet.

AUC? Many physical processes are essentially exponential,
but as I said, for me, on an experiential level things seem
to act as threshold, step function, and beating wave events
rather than simple linear or even exponential sums. I am
sensitive to the multiple time scale lags because my
exercise varies on multiple time scales as well. As in all
of diabetes consistency increases control and decreases
control effort.

Quoted message said:

So what is your method to compensate for exercise effects?

I use a combination of pre exercise basal reduction and bg
adjustment.

Quoted message said:

Glycogen stores increase with training. But I don't think
this will have much impact on the double exp model
mentioned above.

In any case, I still think the total distance * average
BPM is a minimum measurement to capture the required
stress level data. How do you quantify the stress level
for your current exercise method? You must have some
numerical approach if you're adjusting your dose for
exercise.

Seems like a reasonable metric to me, although I would be
inclined to use estimated calories expended, hard to do on a
bike ride with elevation changes. As fitness increases I
would expect ave BPM to fall for constant work.

I don't use a numerical approach, or at least not a rigorous
one. I cut back on insulin more or less heavily based on
recent experience and vary the balance between insulin
changes and bg increases based on how long in advance I know
I will make the gym on a given day. My metric for stress
level is perceived effort which is a very subjective metric.
Based on perceived equal effort for a constant time, the
various aerobic exercise machines will calculate caloric
expenditures that vary by a factor of 2. Early on, I tried
things along the lines you are pursuing, but simpler. I
never came up with anything that worked for me. I hope you
have better luck. Keep us informed of your progress. I will
confess a bias. I've seen many free parameter models that
can be tuned to match an existing time series very well, but
can't predict their way out of a paper bag. And since it is
not something I am good at, I gave up easily.

-------
Charly Coughran [email hidden]

Beav said:

"Jim Dumas" <[email protected]!mindspring.com> wrote in
message -"]news:[email hidden]-
thlink.net...

Quoted message said:

mike gray said:

I think yer trying to measure insulin/glucose
utilization to a level of accuracy that is impossible
with the human body and the Florida weather.

Hi Mike,

You use insulin. So how do you handle this variable
workout issue now?

Hi Jim, I'm not Mike, but.. :-)

I work out the variable workout issue REALLY
simplistically. I work 'til I feel just a TINY bit "off",
then I test. I'm ALWAYS going to be low, so I eat
something which I normally wouldn't DREAM of eating. A
cream and jello filled doughnut for example. WONDERFUL
they are and if I work too hard at trying to STOP that
low, I'm not going to have that wonderful doughnut am I?

Does it make me fat? Not on your life, because I've
burned, and will continue to burn all that lovely
glucose:-)

So... I enjoy the work that brought about the need for the
doughnut and I enjoy the doughnut even more and not ONCE
did I need to do even simple maths LOL. All in all, a MUCH
more fun way to deal with BG's.

That's about it.

My exercise regimens vary considerably from day to day
(problem one). Things like mowing the lawn I can plan for
from experience (as Coughran suggested). I still do 80 - 100
dives a year, and that is highly unpredictable (the two
biggest variables being water temp and current).

I have collected a list of beginning dosages for various
activities, and I usually (always, when diving) carb up
(mmmmmmm, donuts) and exercise with elevated glucose levels.

I don't worry about high bg levels for these periods. I
don't believe the damage, if any, is relevant.

Beav said:

So... I enjoy the work that brought about the need for the
doughnut and I enjoy the doughnut even more and not ONCE
did I need to do even simple maths LOL. All in all, a MUCH
more fun way to deal with BG's.

Dear Beav,

Doughnuts are wonderful accelerators of CHD-CVD.

DMs have one artery of the heart that gets plugged
without fail.
--
Jim Dumas T1 4/86, background retinopathy, rarely
hypoglycemic: <1/mo. lispro+R+U+NPH daily, moderate
exercise, typically <6% HbA1c

Hey Beav,

How's the Beavette doing and the cholipops?

Jim could make his life a whole lot simpler if he eased back
on injecting INDUSTRIAL strength doses of insulin.
Carboloading might go handtohand with this mania.

There's another IDDT 09October2004 in Brum.

Still looking forward to eating some Mr. Brain's [censored].

Let's hear it for England against those Frogs on Sunday
in Portugal!

Cheers,

BGN

Beav said:

"Bay Area Dave" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

tee hee! you wrote "sort of" what I was thinking about
all the complex scientific jargon he throws around. I
wonder if he thinks that jazz impresses anyone? All it
tells me is he is obsessed.

Jim's comfortable with his maths, but I doubt many others
are as comfortable. I don't think he's trying to impress
anyone, but I've been wrong before now :-)

I think that obsessive behaviour IS something he should
consider though.

Beav

I hardly need to listen to your ranting. I'll be out
exercising shortly, you dumass.

dave

Jim Dumas said:

Bay Area Dave said:

All it tells me is he is obsessed.

It tells me you don't have a training schedule for
exercising. This variability in insulin sensitivity, post
variable exercise period, is a major problem that you
haven't a clue about. So take a seat.

Some people just can't be quiet (i.e., don't know when to
listen).

Bay Area Dave said:

how many people actually want to decipher your overly
complex, wordy posts, professor?

Dear Dude,

Please plonk me again. Thanks in advance,
--
Jim Dumas T1 4/86, background retinopathy, rarely
hypoglycemic: <1/mo. lispro+R+U+NPH daily, moderate
exercise, typically <6% HbA1c

Citizen DM said:

My mental calculations are all 'back of the envelope' and
pretty good. I find that 'darts dosing' is very effective
and sums up the whole 'dafne' approach. We learn to adjust
insulin levels through trial and error improving with
practice.

Hi CDM,

If you can reliably adjust your dose, then you must have
some math model to do so. Please tell us how you arrive at
your dose values.

Thanks,
--
Jim Dumas T1 4/86, background retinopathy, rarely
hypoglycemic: <1/mo. lispro+R+U+NPH daily, moderate
exercise, typically <6% HbA1c

Bay Area Dave said:

you are an idiot! Of COURSE I exercise.

Dear Dude,

Please increase your Prozac dose. Best wishes,
--
Jim Dumas T1 4/86, background retinopathy, rarely
hypoglycemic: <1/mo. lispro+R+U+NPH daily, moderate
exercise, typically <6% HbA1c

"Jim Dumas" <[email protected]!mindspring.com> wrote in message
"]news:[email hidden]...

Quoted message said:

oldal4865 said:

I'm a retired engineer. I use my wife's cast-off
computers; I don't even use a digital watch.

That ain't gonna work, Dude. The PeaBrain bicycle computer
fits into my Levi's jean watch pocket. So your cast-off
computers are too big to mount on the handlebars.

Quoted message said:

I don't "handle" the variable workout issue. Instead, I
strive for constant workout at constant time-of-day with
constant food in my belly and constant H + U in my fat
layer.

I want total flexibility to be spontaneous. I also want an
empty stomach during my exercise to keep side-stitches
(side pains that force you to

stop

Quoted message said:

your workout) to a minimum.

And nothing is really constant no matter how hard you try.

Doesn't that sort of make a pigs ear out of your formula
then Jim? btw, doughnuts are constant🙂

Beav

Charly Coughran said:

AUC? Many physical processes are essentially exponential,
but as I said, for me, on an experiential level things
seem to act as threshold, step function, and beating wave
events rather than simple linear or even exponential sums.
I am sensitive to the multiple time scale lags because my
exercise varies on multiple time scales as well. As in all
of diabetes consistency increases control and decreases
control effort.

Hi Charly,

AUC is Area-Under-the-Curve for the health-care
professionals lurking. This is the integral of the double
exp insulin sensitivity function that we would see in our
insulin dose adjustments.

My HP-48 open-loop feedback dosing model provides for
dynamic dosing, thereby permitting tight BG control with
variable diet, exercise and insulin. So I no longer require
consistency in my DM management. But admittedly, it takes
years to master this juggling act and I do not recommend it
to the DM public (but maybe for T1 athletes).

Quoted message said:

I've seen many free parameter models that can be tuned to
match an existing time series very well, but can't predict
their way out of a paper bag.

This just means there was no physiological basis in that
time series model. But the double exp sensitivity model
mimicks glycogen rebuilding in the two primary tissues:
muscle and liver. It should therefore be closer to the real
phenomenon we're attempting to predict. Also note the
initial step function in the double exp model right when the
exercise session ends. This likewise mimicks the physiology
as well as your observations. (One of my problems now is to
decide how to handle multiple exercise sessions within a 48
hour window. Do I only use the last double exp or do I sum
the effects of the closely spaced exercise sessions? So I'm
planning a series of experiments to sort this out. It's also
important to note that the model only needs n=1 to set up
the math framework. [Any physiologist will attest to this.]
So I don't need to run a clinical trial to say that specific
model is the best one to use.)

Thanks for your ideas,
--
Jim Dumas T1 4/86, background retinopathy, rarely
hypoglycemic: <1/mo. lispro+R+U+NPH daily, moderate
exercise, typically <6% HbA1c

Bay Area Dave said:

I hardly need to listen to your ranting. I'll be out
exercising shortly, you dumass.

Good.

It's nice to know I'm having an influence on you. Best
wishes,
--
Jim Dumas T1 4/86, background retinopathy, rarely
hypoglycemic: <1/mo. lispro+R+U+NPH daily, moderate
exercise, typically <6% HbA1c

excellent idea. done!

dave

Jim Dumas said:

Bay Area Dave said:

how many people actually want to decipher your overly
complex, wordy posts, professor?

Dear Dude,

Please plonk me again. Thanks in advance,

"Bay Area Dave" <[email hidden]> skrev i melding
"]news:[email hidden]...

Quoted message said:

you are an idiot! Of COURSE I exercise.

dave

steinar_m said:

Simple, he does not excercise.

The fact that you can't take a little joke, does not suggest
that I am the idiot here.

You have a pump, I don't. I can't get a pump. It makes me
extreamly happy that people like Jim Dumas is around, they
actually help people out by explaining and figuring out how
to manage this [censored] better. You don't.

r, Steinar

"Jim Dumas" <[email protected]!mindspring.com> wrote in message
"]news:[email hidden]...

Quoted message said:

Beav said:

So... I enjoy the work that brought about the need for
the doughnut and

I

Quoted message said:

Quoted message said:

enjoy the doughnut even more and not ONCE did I need to
do even simple maths LOL. All in all, a MUCH more fun
way to deal with BG's.

Dear Beav,

Doughnuts are wonderful accelerators of CHD-CVD.

I'm sure they are Jim, but I don't want to tread carefully
all my life right up to the end, sometimes it's beneficial
to charge at life headon and accept that no matter WHICH
route you choose, the destination is the same, but one way
is a lot more fun than the other. And of course, I think the
number of doughnuts eaten has to make a difference. There's
probably even a formula for it :-))

One doughnut per week = 10 minutes off the end of your
days maybe?

Quoted message said:

DMs have one artery of the heart that gets plugged
without fail.

So why bother with all your work? If it happens to us all
without fail, then you're as likely to have a clogging as
anyone else aren't you?

Anyway, I'll do some research, find out which artery it is,
then get it checked. I'll do it every year and if it starts
clogging, I'll get it by-passed, then recuperate with a
nice cuppa tea and a ..... doughnut:-))). Preventative
medicine eh? :-))

Beav