Yes, I understand the context, and the Keys study suggests that Bannister et al. (2014) is wrong.
My point is that you seem to be saying that the Keys study shows that metformin doesn't increase life expectancy. I am basing this understanding on two things you wrote:
"found no reduction in all-cause mortality by metformin treatment"
"Mortality rates are worse among metformin treated subjects."
But the Keys study does not show that Metformin doesn't increase life expectancy (or reduce all-cause mortality). It shows that Metformin isn't enough to counteract the impact (on lifespan) of diabetes. But it seems to say nothing about whether people (whether diabetic or not) would live longer with metformin or without.
I realize the Keys study isn't the only piece of research in this area. But the topic of my original comment (and this one) are simply that particular study, and what you say it shows.
If I have misunderstood anything I am happy to be corrected/educated.
I think you summarize this well. Banninster’s claim was that even the T2 diabetics lived longer on metformin than even healthy controls. That is clearly debunked. I do not know of any heslth controls being put on metformin at say 70 years of age to find out if is lowers all-cause mortality. Mouse data is a strong No. The cell paper is a study of about 12 monkeys and a lot of bioinformatic hocus-pocus of the type I do. But I prefer much larger numbers and much stronger effects.
My point is that you seem to be saying that the Keys study shows that metformin doesn't increase life expectancy. I am basing this understanding on two things you wrote:
"found no reduction in all-cause mortality by metformin treatment"
"Mortality rates are worse among metformin treated subjects."
But the Keys study does not show that Metformin doesn't increase life expectancy (or reduce all-cause mortality). It shows that Metformin isn't enough to counteract the impact (on lifespan) of diabetes. But it seems to say nothing about whether people (whether diabetic or not) would live longer with metformin or without.
I realize the Keys study isn't the only piece of research in this area. But the topic of my original comment (and this one) are simply that particular study, and what you say it shows.
If I have misunderstood anything I am happy to be corrected/educated.