Insulin and mortality

[Guest guest]

> For me, this is so counterintuitive that

> I would have to know everything about

> this study before I would be willing to

> consider it.

You are right to be sceptical because it has big implications but

there is no way that you are going to get to know EVERYTHING about

it!

> Shown by whom?

The study was conducted by a Dr. Gregg Fonarow:

http://www.healthcare.ucla.edu/institution/physician?

personnel_id=8623

who is a professor of cardiology at UCLA and director of the

Ahmanson-UCLA Cardiomyopathy Center.

> Who sponsored this study?

It was sonsored by a private foundation, the Ahmanson Foundation of

Beverly Hills, CA.

> How many test subjects were there?

554

> How were the test subjects chosen?

They were all patients hospitalized for advanced heart failure.

> For all I know it was sponsored by Glucatrol...

Dr. Fonarow is research consultant and speaker for Glaxo Kline,

Bristol-Myers Squib, Pfizer and Merck, all manufacturers of oral

diabetes medication, but they did not have any part in sponsoring

the study as far as I can make out.

I presume that it would be impossible to persuade an insulin

manufacturer to sponsor a study of that kind. It would not be to

their advantage (and, if they are anything like the cigarette

manufacturers, they probably knew the results already!)

There is a PDF version of the full report available but I would need

to take out an annual subscription to the American Heart Journal for

US$258 to see it!

Regards

T.

[Guest guest]

Thanks for the information, .

>Dr. Fonarow is research consultant and speaker for Glaxo Kline,

Bristol-Myers Squib, Pfizer and Merck, all manufacturers of oral diabetes

medication, but they did not have any part in sponsoring the study as far as I

can make out.<

Perhaps not formally. At this point, I may be a bit cynical, but I wonder how

popular he would have been with his bread and butter clients if he discovered

that orals caused problems and insulin didn't.

Hugs, Dianne

__________________________________________

Yahoo! DSL – Something to write home about.

Just $16.99/mo. or less.

dsl.yahoo.com

[Guest guest]

> Do you know what kind of insulin was used?

> Regular or synthetic? I think studies like this

> where they take people with preexisting

> conditions and come to the conclusions that

> it was the insulin that made them sick is silly

Nobody has suggested that " it was the insulin that

made them sick " , I don't know where you get that

idea from. The study established a statistical

connection between treatment of DM2 with insulin

and the chances of survival for more than one year

after a heart attack, nothing more than that.

There is a clear statistical connection between

taking too long to get a heart attack patient to

hospital and survival for more than one year after

but that does not mean that they are suggesting

that the delay causes the heart attack!

If you are DM2, use insulin but never have a

heart attack then you are not concerned. If you

are DM2, use insulin and are considered to be

at risk for a heart attack then you might want to

reconsider returning to oral medication. That

is the conclusion of the study, not that " the

insulin made them sick " !

Regards

T

[Guest guest]

> My personal take is that achieving and maintaining

> glycemic control is paramount and supersedes the

> choice of insulin versus oral meds. Please note

> that this opinion flies in the face of the results

> of a second correlational study I cited, performed

> by one of the authors of the study under current

> discussion, which found, very roughly interpreted,

> that bad control was associated with *better* outcomes

> for heart patients.

Wouldn't it be interesting if the research leads to the conclusion

that maintaining good control leads to minimum complications with

maximum mortality and bad control leads to maximum complications and

minimum mortality?

That would be the ultimate diabetic lifestyle choice, a short life

and a merry one (if you can see a life spent maintaining good

control as being " merry " ) OR a long life and being constantly

doctored for one thing or the other until the end of your days!

That could lead to a compromise therapy target of " down the middle " ,

a strategy not of frantically trying to obtain the best possible

control regardless of the consequences but of aiming for " just

enough control " . I do something like that already in that I could

easily get HbA1c of 4.5% (I once had it for a while) but cut back on

my medications to maintain a fairly constant 5.5% to obtain an

acceptable hypo risk.

I admit that I have never thought about it that way before but my

physician keeps telling me not to forget " quality of life " ; he

reminds me practically every visit when I lay out my charts and

results that there are far, far worse things than diabetes waiting

out there and not to lose sight of the real goal which he describes

as living a full life in spite of diabetes and avoiding

obsessive behavior patterns.

> The rule I'm following, a simple application

> of a probability theorem known as Bayes Law,

> could be stated as follows: surprising conclusions

> demand strong evidence. If enough good studies

> contradict my existing mental models, I'll

> eventually choose to discard them in favor of what

> I can hope are better models.

The flaw to both the Bayesian and the frequentist schools of

statistics is that neither of them hold true if there is no symmetry

among the alternatives. According to Bayes you would confidently

expect a coin tossed a large number of times to come down as often

tails as heads. Try that with an ordinary US cent and you find that

it comes down more often tails than heads which is a surprising

conclusion to most people. I get a feeling that many of the options

in diabetes management are just as asymmetrical as a red cent.

Bayes defined his law nearly 250 years ago and he died before

his work could be published so his own decisions were prima facie

not exactly optimized!

Regards

T.