Great Cholesterol Lie?

[Red Baron Farms]

The problem is actually quite subtle. I haven't gone through all RBF's links, but I suspect that most of them will be observational studies. The problem with these is that it can be damned near impossible to reliably extract causation from correlation in such studies.

Here's an interesting article about the health effects of eating meat, the conventional wisdom of why it's not good for you, and why the conventional wisdom cannot be relied upon when it's only backed up by observational studies:..... Well, you probably can't with an observational study.

Gary Taubes has an interesting point. But actually I disagree with him on this issue. He has one hypothesis as to why the Harvard study found correlation between beef and various health issues heart disease, cancer, chronic inflammatory diseases etc...(the observational study thing) But there is one major flaw in his argument. Beef today and beef even 30-50 years ago are radically different products with regards to the things I mentioned above from the feeding studies. (Due to CAFOs) Taubes makes the unequivocally false assumption that meat is meat is meat and that there are no significant nutritional or qualitative differences between CAFO and grass fed. Any arguments he makes after that false assumption are then flawed.

The studies I listed include mostly feeding trials and clinical trials. There is a causative link, not just a correlation with a multitude of confounding factors.

I don't think we are anywhere near complete understanding, but it is getting pretty clear that CAFO supplied animal products from a nutritional standpoint are at least partly to blame for the correlation between meat and certain chronic health problems increasing significantly in the last several decades. (corn and soy oils being another big one that also affect lipid balance) Ironically what we need are MORE studies in the general population because they are lacking. It's one thing to test these lipid balance effects clinically and in livestock feeding studies. Quite another to see what the long term effects in the general population might be to switch back.

 

[casebro]

My point was rather that you could not make any conclusion at all, casebro:
Pharmaceutical company acknowledgement: 2
No pharmaceutical company acknowledgement: a unstated number
(plus some you could not access)

....

Here is what I said, " And then tried to look up the full text of the articles in the journals for six trials. The first two gave acknowledgement to the drug companies, Merck/Sharp/Dohm, and Pfizer. "

I DID SO state a number. There it is for for the third time.

I felt half a dozen was reasonable effort in a casual platform like this.

But stop being such a prickety ****. Your stats agree with my principle and you know it. .

 

[Reality Check]

Here is what I said, ... .

You stated a number but you did not have enough data to state a %: 2/unknown number does not compute, casebro .
That 100% of the trials you could find information on were sponsored is trivial and as we have found out slightly wrong.

 

[Pixel42]

Aren't pharmaceutical companies legally obliged to conduct clinical trials on the medicines they develop before putting them on the market? Why would someone else do it for them, when it's so incredibly expensive?

Of course I know that universities also conduct clinical trials, especially in the early days of developing a promising new treatment, but surely it's inevitable that the vast majority will be done by the manufacturers of those treatments. It's not evidence of something nefarious going on.

 

[bit_pattern]

It's not evidence of something nefarious going on.

Certainly not on its own but there is a disturbing tendency for pharma companies to conduct pretty shoddy research in the interests of profit - take Vioxx for example, Merck knew it was deadly but systematically covered the fact up while completely fabricating its efficacy.

http://en.wikipedia.org/wiki/Rofecoxib

 

[casebro]

Aren't pharmaceutical companies legally obliged to conduct clinical trials on the medicines they develop before putting them on the market? Why would someone else do it for them, when it's so incredibly expensive?

Of course I know that universities also conduct clinical trials, especially in the early days of developing a promising new treatment, but surely it's inevitable that the vast majority will be done by the manufacturers of those treatments. It's not evidence of something nefarious going on.

But even those universities get much of the funding from the drug companies.

Just look at the meta study that Reality Check and I have been discussing. Twelve out of the thirteen used in the study were funded by the drug companies. Now I have to wonder how the meta study was funded? Though I think Metas are cheap, and might be done independently by some "publish or perish" professor.

Hey Reality Check, did you see the chart of the many trials included in the meta study? It had various columns of qualities, like "selection bias of subject", proper double blinding, publication selection, etc. Most columns had a (+) in them. I'm not sure how to interpret that, did you see a key?

 

[Pixel42]

Who do you think should pay for the hugely expensive clinical trials of new medical treatments if not their manufacturers, who are legally obliged to conduct them?

 

[casebro]

Who do you think should pay for the hugely expensive clinical trials of new medical treatments if not their manufacturers, who are legally obliged to conduct them?

So do we have a new debating tactic, Appeal to Cost?

But the SOP in the scientific world is independent replication, something we have NOT been shown with Statins, at least not in this thread.

I've asked for that up-thread a couple times. Nobody has come up with anything but Pharma sponsored studies.

 

[Reality Check]

Aren't pharmaceutical companies legally obliged to conduct clinical trials on the medicines they develop before putting them on the market?

That is certainly true in the USA (the FDA enforces this) and probably many other countries.
However this is a step in the known bias between sponsored and not sponsored clinical trials. The companies test their product. If the result is negative then they do not release the drug for outside clinical trials. If the result is positive then they sponsor outside clinical trials for a product that has already shown to have positive results. The obvious effect is that overall sponsored trials get more positive results than non-sponsored trials. Where the bias comes in is that sponsored trials tend to positive a lot more than non-sponsored trials on the same product.

The cause of this bias is hard to pin down. Could it be that less money = smaller trials and so a higher probability of negative results? Is it confirmation bias (sponsored trials are expecting positive results, non-sponsored trials are expecting negative results)? Is it good old publication bias (positive trials are more likely than negative to be published)?

 

[Reality Check]

But the SOP in the scientific world is independent replication, something we have NOT been shown with Statins, at least not in this thread.

casebro, you seem to have a different definition of "independent replication" than is used in science. In science, independent replication is replication of experiments or clinical trails by any other team than the original team. The sources of funding for the other teams does not matter. If they do the clinical trail correctly then the results are valid no matter what they are. This is one case where scientists implicitly trust other scientists .
The correctness of the clinical trial is a matter of peer review at publishing, peer review post-publication and meta-studies.

So we have independent replication with Statins as shown in the meta-study containing many independent studies: Statins for the primary prevention of cardiovascular disease.
This paper includes at least one non-sponsored trial of Statins:

Caution also needs to be taken regarding the fact that all but one of the trials had some form of pharmaceutical industry sponsorship. It is now established that published pharmaceutical industry-sponsored trials are more likely than non-industry-sponsored trials to report results and conclusions that favour drug over placebo due to biased reporting and/or interpretation of trial results (Als-Nielsen 2003).

"Figure 1. Methodological quality summary: review authors’ judgements about each methodological quality item for each included study." has no key but the symbols look obvious: green with "+" for feature present, yellow + "?" for unclear, red + "-" for feature not present.

 

[casebro]

casebro, ...


"Figure 1. Methodological quality summary: review authors’ judgements about each methodological quality item for each included study." has no key but the symbols look obvious: green with "+" for feature present, yellow + "?" for unclear, red + "-" for feature not present.

I've been locked out of the full versions. I guess my free sample window has expired. IIRC, those 'features' were things like "subject selection bias", so a green (+) there means the trial DID HAVE selection bias.

This is what you said it meant, isn't it? And it is why I asked.

Talk about a "Great Cholesterol Lie"

 

[Reality Check]

I've been locked out of the full versions.

Do a Google - there is a PDF available elsewhere.

Your recall is wrong. See the other version of Figure 1: "Figure 2. Methodological quality graph: review authors’ judgements about each methodological quality item presented as percentages across all included studies."
Green = Yes (low risk of bias)
Yellow = Unclear
Red = No (high risk of bias)
Also the paper contains a little thing called text .

We judged 13 of the trials to be free from selective bias. (Figure 1; Figure 2).

and there are 13 green pluses in Figure 1 for the "Free of selective reporting" column, i.e. the feature (*freedom of selective reporting") is present in the trial. Thus Green = good, Red = bad, Yellow = not known.

Talk about a "Great Cholesterol Lie", casebro - a conclusion based on vague memories of the contents of a paper!

 

[casebro]

Talk about a "Great Cholesterol Lie", casebro - a conclusion based on vague memories of the contents of a paper!

If that was my conclusion I would not have asked you about it would I?

 

[Reality Check]

If that was my conclusion I would not have asked you about it would I?

You wrote "IIRC, those 'features' were things like "subject selection bias", so a green (+) there means the trial DID HAVE selection bias."
That is a conclusion based on memory. The features actually were things like "a lack of subject selection bias".

And finishing with "Talk about a "Great Cholesterol Lie"" based on a IIRC is not good.

 

[bigred]

You wrote "IIRC, those 'features' were things like "subject selection bias", so a green (+) there means the trial DID HAVE selection bias."
That is a conclusion based on memory. The features actually were things like "a lack of subject selection bias".

And finishing with "Talk about a "Great Cholesterol Lie"" based on a IIRC is not good.

You forgot to say casebro

 

[stevea]

It never was consumed cholesterol, we knew that decades ago.

If "we" means you and me I might agree, but seemingly authoritative medical sources are STILL promote low dietary cholesterol.
http://www.mayoclinic.org/diseases-...terol/expert-answers/cholesterol/faq-20058468
http://www.nhlbi.nih.gov/files/docs/public/heart/chol_tlc.pdf
http://www.johnshopkinshealthalerts.com/alerts_index/heart_health/380-1.html

When you eat saturated fats, your body produces cholesterol to transport it in the blood stream.

True enough, but also in response to eating carbs in excess of immediate needs (which are converted to fats). However much of the LVL/vLDLD appears to be related to spills from the liver as non-alcoholic fatty-liver-disease NAFLD occurs during metabolic syndrome.

And apparently partially saturated fats (turns oils to solid fats) called trans fats turned out worse than naturally saturated fats. You're still better off with oils (except soy and corn oil that unless it's organic can have herbicide residues that are associated with breast cancer).

There is no such thing as "partially saturated fats". This is nonsense terminology belies ignorance of the basic chemistry. There are saturated fats, and fats w/ various degrees of desaturation (carbon-carbon double bonds on the constituent fatty acid).

Trans-fats are indeed very bad wrt CVD & LDL, however they are unsaturated fats with some point(s) of trans- (rather than the typical cis-) configuration the point of desaturation. They result in quantity from 'hydrogenation' process; a chemical process to reduce the extent of desaturation, and therefore increase the freezing point of unsaturated fats.

As for inflammation, C reactive protein is a known risk for heart disease. Can't remember what makes it better or worse but I do recall some infections, especially gingivitis, are suspected as playing a role in the risk of heart disease.

CRP is a marker for systemic inflammation, not a risk factor per se. A "marker for risk" is the current terminology. High CRP can be due to infection, auto-immune disease like some forms of arthritis, allergic reactions. One theory is that high omega-6 : omega-3 fat ratios in modern diets predisposes one to such inflammation. Omega-3/omega6 refers to the exact point point the last point of saturation on a FA chain. Grasses, phytoplankton&algae and the things that east these tend to have higher omega-3 levels. We could devolve the discussion to EFA/DHA (specific FAs), but this is a bit far afield.

Statins are worth taking most of the time. You should take them if you've already had an MI. But prior to known cardiac disease there are still a few unknowns. It's a case by case basis. I went off them when I developed an autoimmune disease and my cardiac panel didn't change much so I didn't go back on them.

According to the most current Cochrane reviews - statins have decided value for those at risk of CVD, and small marginal value to those w/ no specific risk and inconsequential negative side-effects. However it's worth noting that statins have anti-inflammatory & blood thinning features, so they *might* be no better than aspirin for a population w/o cholesterol risk.

 

[ThunderChunky]

I don't think we are anywhere near complete understanding, but it is getting pretty clear that CAFO supplied animal products from a nutritional standpoint are at least partly to blame for the correlation between meat and certain chronic health problems increasing significantly in the last several decades.

Doesn't chicken have a worse lipid profile than beef? This cant explain the whole red meat issue.

 

[Red Baron Farms]

Doesn't chicken have a worse lipid profile than beef? This cant explain the whole red meat issue.

Depends how it is raised. Chicken fat can be very healthy indeed. It can also be very bad, as you say.

 

[ThunderChunky]

Depends how it is raised. Chicken fat can be very healthy indeed. It can also be very bad, as you say.

Use optimal conditions for the chicken and the worst case scenario for the cow. Chicken will still contain more omega 6 than the beef. This appears to immediately debunk your hypothesis about why red meat is associated with disease. If good chicken has a worse profile than bad beef, certainly bad chicken should be what is associated with disease and not bad beef.

 

[Red Baron Farms]

Use optimal conditions for the chicken and the worst case scenario for the cow. Chicken will still contain more omega 6 than the beef. This appears to immediately debunk your hypothesis about why red meat is associated with disease. If good chicken has a worse profile than bad beef, certainly bad chicken should be what is associated with disease and not bad beef.

Actually no. That particular statement is wrong for a couple reasons. And to understand why you need to look at causation. The reason for the problem is an enzyme that is used for both metabolising n-3s and n-6s It preferentially will be used for n-3s but even if preferential will still be overwhelmed at about 4:1. But it isn't just the meat, it's the ratio in the whole meal and diet. You actually can grow a chicken under that 4:1 ratio pretty easy, but fry it in corn oil for southern fried chicken and you just blew the whole thing. Fail to eat your fruits and vegetables and you just blew the whole thing as well. Omega 3's are not the only part of the lipid profile that matters either. Chicken fat can be very low in saturated fat too, even lower than beef fat. It also can be quite lean overall having less fat in total. There is a whole chain of events that make a heart attack and a whole range of dietary factors that can either raise or lower that risk and chicken can be quite healthy or not depending on a whole set of factors either in raising it, or in preparation.

But as a general rule yes. Beef is probably better for you than chicken, assuming it is raised correctly. We evolved eating red meat and it is healthy for us to eat. However, fish is probably better than both, if it wasn't for all the water pollution.