All Red Meat is Bad for You

[jt512]

How would you account for the grossly wrong results of the massive study linking simvastatin and AD that I linked above? There were also earlier smaller studies indicating a similar benefit. A 50% reduction in AD incidence due to the use of simvastatin using a population of 4.5 million! And it was later proved to be wrong!
See, the key to scientific evidence is to understand and demonstrate a causality. Anything short of that is not conclusive as so many past medical missteps have demonstrated. Drink the kool aid if you like; I prefer to wait for real scientific evidence.

What's that? The imprecautionary principle?

 

[Perpetual Student]

Yeah, I'd want to know that also. Entirely possible for the hypothesis to be correct, and a study to be flawed.

I don't know if these links have already been posted, but they raise some good points.


http://garytaubes.com/2012/03/science-pseudoscience-nutritional-epidemiology-and-meat/

http://www.zoeharcombe.com/2012/03/red-meat-mortality-the-usual-bad-science/

Yes, thanks. I'm looking forward to hearing from those who have defended the study in question -- after they take the time to read those commentaries.

 

[jt512]

Yes, thanks. I'm looking forward to hearing from those who have defended the study in question -- after they take the time to read those commentaries.

Zoe Harcombe, the author in the second link, has no idea what she's talking about. I don't even know where to start. Her analysis is simply nonsensical. And any shred of credibility that Taubes may have had just went out the window by saying that Harcombe did a "wonderful job of dissecting the article." How embarrassing for both of them.

 

[casebro]

I just read the article from the OP.

Interestingly, it says the fish lowers the rate by 7%, less than the other dietary interventions mentioned.

And "(red meat) also be linked to an overall risk of death from any cause," Accidents? Infections? Lung conditions? Ulcers? This kind of absoluteness gives me doubts. Like the way the Woos tell us that (____) cures EVERYTHING.

Maybe there s some kind of selection bias here? Like "Unhealthy people crave meat, and eat more of it. And since unhealthy people have a higher mortality rate..."

I also wonder whether some of the groups they controlled for have a negative correlation? What if smokers who eat more meat live longer than smokers who eat less meat?

Until they come out with the defining mechanism, I'll keep my doubts. And I suspect it will be something indirect- like digesting red meat depletes some reserve of an enzyme, and with out it the white cells don't do something to control some unrelated activity. And we'll all be eating more rutabagas to keep those reserves filled. Or something.

 

[Pi_314]

For the most part, eat whatever you want and live between 70 and 80. Live any longer and be a burden to society.

 

[crimresearch]

Zoe Harcombe, the author in the second link, has no idea what she's talking about. I don't even know where to start. Her analysis is simply nonsensical. And any shred of credibility that Taubes may have had just went out the window by saying that Harcombe did a "wonderful job of dissecting the article." How embarrassing for both of them.

Wow... that certainly proves everything beyond any doubt whatsoever.

I mean, who needs facts or examples when we've got your Godlike declarations?

 

[jt512]

And "(red meat) also be linked to an overall risk of death from any cause," Accidents? Infections? Lung conditions? Ulcers? This kind of absoluteness gives me doubts. Like the way the Woos tell us that (____) cures EVERYTHING.

You're misinterpreting "any cause." It just means "total mortality." That is, the more red meat there was in the diet, the higher was the risk of death per unit time. It doesn't mean that red meat was associated with death from every cause, such as homicide, vehicular accidents, and whatnot.

Jay

 

[Kevin_Lowe]

Zoe Harcombe, the author in the second link, has no idea what she's talking about. I don't even know where to start. Her analysis is simply nonsensical.

She clearly knows something about statistics, but at the same time some of the other things she says are indeed simply nonsensical and I'm baffled as to how anyone who knows anything about statistics typed those words.

Is she a meat industry PR person, or did she learn statistics by thumbing randomly through an old first year textbook, or what? Does she just not understand the idea of controlling for a confounding variable?

And any shred of credibility that Taubes may have had just went out the window by saying that Harcombe did a "wonderful job of dissecting the article." How embarrassing for both of them.

Indeed.

 

[Perpetual Student]

I just read the article from the OP.

Interestingly, it says the fish lowers the rate by 7%, less than the other dietary interventions mentioned.

And "(red meat) also be linked to an overall risk of death from any cause," Accidents? Infections? Lung conditions? Ulcers? This kind of absoluteness gives me doubts. Like the way the Woos tell us that (____) cures EVERYTHING.

Maybe there s some kind of selection bias here? Like "Unhealthy people crave meat, and eat more of it. And since unhealthy people have a higher mortality rate..."

I also wonder whether some of the groups they controlled for have a negative correlation? What if smokers who eat more meat live longer than smokers who eat less meat?

Until they come out with the defining mechanism, I'll keep my doubts. And I suspect it will be something indirect- like digesting red meat depletes some reserve of an enzyme, and with out it the white cells don't do something to control some unrelated activity. And we'll all be eating more rutabagas to keep those reserves filled. Or something.

I agree. Ultimately, looking for causality is how real science is done. Population studies can yield clues to give us a direction for further study. However, basing conclusions only on population studies has misled researchers and the public too many times for us to simply slip into this kind of flawed thinking once again. We have many confounding factors (both those identified in the study and those that remain unknown), we have the exigencies of data from questionnaires, we have statistical variability with quite tenuous statistical results, and we might have the potential of researcher bias.
Some months ago on another thread, I asked medical professionals on this forum to explain how a study of 4.5 million people conducted over some considerable number of years about simvastatin and Alzheimer's can show a 50% reduction in that disease and end up being dead wrong. The only answer I got was some vague mantra about the problematic nature of population studies. If a study showing a 50% reduction of the incidence of some disease can be so wrong, how could we give any credibility to a study with the thin results of this one concerning eating red meat where there is no demonstrated causality?

 

[epepke]

How about you just go to Pubmed, type in "red meat" and sit back and go "oh holy tits!" You cannot be expected to be taken seriously when there are MANY nutritional reports that show overconsumption of red meat to lead to higher mortality.

The problem here is that, in this study, high mortality was associated with high consumption of red meat and lower serum cholesterol levels. So you're going to have to make a choice between which conventional wisdom you want to go with.

 

[Kevin_Lowe]

Some months ago on another thread, I asked medical professionals on this forum to explain how a study of 4.5 million people conducted over some considerable number of years about simvastatin and Alzheimer's can show a 50% reduction in that disease and end up being dead wrong. The only answer I got was some vague mantra about the problematic nature of population studies.

It and similar incidents being so damned common (see Contradicted and Initially Stronger Effects in Highly Cited Clinical Research by John P. A. Ioannidis, MD) is a puzzle I'm mulling over at the moment.

Possibly gold standard studies are too often asking the wrong questions. Possibly the desk drawer effect is in full swing and many studies which fail to achieve statistical significance aren't getting published. Possibly outright cheating is substantially more common than we'd like to think. Possibly other things. Almost certainly a combination of several things is responsible.

Whatever it is though, at least one in six or seven top-tier studies with statistical significance get totally overturned in a relatively short time frame in medical research. I believe some specific biomedical fields (I think it might have been medical genetics) are even worse.

 

[Nasty Neil]

"False DICHOTOMY.

You DO not have to GIVE up either to pursue the OTHER!"

MAYBE so, in your MICKEY MOUSE OPINON!

 

[Nasty Neil]

"For the most part, eat whatever you want and live between 70 and 80. Live any longer and be a burden to society."

REALLY?

I would like to be around when you are my age (77) and see IF you are still saying the same thing.

AND, IF you had a CLUE, you would know there are more than a few people over 80 who still contribute a LOT.

 

[tsig]

It is... Including moderation itself.

Absolutely, you need the occasional binge.

 

[gabeygoat]

"For the most part, eat whatever you want and live between 70 and 80. Live any longer and be a burden to society."


AND, IF you had a CLUE, you would know there are more than a few people over 80 who still contribute a LOT.

Word!

 

[crimresearch]

It and similar incidents being so damned common (see Contradicted and Initially Stronger Effects in Highly Cited Clinical Research by John P. A. Ioannidis, MD) is a puzzle I'm mulling over at the moment.

Possibly gold standard studies are too often asking the wrong questions. Possibly the desk drawer effect is in full swing and many studies which fail to achieve statistical significance aren't getting published. Possibly outright cheating is substantially more common than we'd like to think. Possibly other things. Almost certainly a combination of several things is responsible.

Whatever it is though, at least one in six or seven top-tier studies with statistical significance get totally overturned in a relatively short time frame in medical research. I believe some specific biomedical fields (I think it might have been medical genetics) are even worse.

Ahhh.. so when Taubes says that sort of medical study has a poor track record for holding up under replication, he has zero credibility...

But when you copycat the same idea, you're an unimpeachable expert who doesn't have to provide facts and figures because no one but you could possibly understand them?

Same old same old.

 

[ThunderChunky]

It should be considered red meat. But it should also be considered a bunch of other things including a bunch of food additives and processed who knows what.

The separated processed from unprocessed meat.

Interesting study. It will be equally interesting to see if they eventually investigate the differences between processed and unprocessed, or pastured vs CAFO, meat, as they made a note in the intro that they made no distinction between the two.

Not sure exactly what you mean here, but they did separate processed from unprocessed. It's mentioned in the LA times report and in the abstract of the paper.

 

[jt512]

Wow... that certainly proves everything beyond any doubt whatsoever.

I mean, who needs facts or examples when we've got your Godlike declarations?

Since you have obviously not given any critical thought yourself to the Harcombe piece, it seems unlikely that you are sincerely interested in a detailed critique of it. However, if I'm wrong and you are, then insults and sarcasm are not likely to persuade me to invest the time and effort to produce one.

Jay

 

[Lowpro]

Since you have obviously not given any critical thought yourself to the Harcombe piece, it seems unlikely that you are sincerely interested in a detailed critique of it. However, if I'm wrong and you are, then insults and sarcasm are not likely to persuade me to invest the time and effort to produce one.

Jay

I'm interested in hearing it though, and am not in the least sarcastic.

 

[Beerina]

It should be considered red meat. But it should also be considered a bunch of other things including a bunch of food additives and processed who knows what.

How do you statistically separate it out? For example if fast food was extra bad for you and people who ate "red meat" are really just eating a lot of fast food how can you simple blame "red meat", instead of blaming processed crap?

I think that's one of the things they tried to sort out with this study -- processed (i.e. nitrate-laden) vs. unprocessed meat, and they both were about equally bad for you.


Of course, they really need to separate this out for obesity -- is a 250-lb. vegetarian really doing a lot better than a 250-pound McDonald's addict? Or a home-made hamburger addict, for that matter?




One study claimed nitrates dissolved into chemicals that specifically kill off the insulin-generating cells in the pancreas, which is what helps lead to type II diabetes. However, nobody seems to have ever done an autopsy on a type II diabetic to see if their pancreas is sub-optimal.