Diet Studies - Reader Beware

An interesting read if you are so inclined. It’s been discussed here before.


/nyti.ms/2QiSnmP?smid=nytcore-ios-share

“While very few scientists are accused of misconduct or misreporting data, critics have long contended that nutrition research is plagued by a credibility problem. They argue that an alarming number of food studies are misleading, unscientific or manipulated to draw dubious conclusions.

What about vitamins? ;-)

Not subscribing to the nutrition, diet, thing but still have to wonder why some countries have waaaaaay less PC than we do?????

browntrout said...
Not subscribing to the nutrition, diet, thing but still have to wonder why some countries have waaaaaay less PC than we do?????

Well, it can certainly make one reconsider the importance of diet, Some of those countries are more vegetarian than we are, some are not. But I am willing to bet a $ that all have lower average blood insulin levels than we do.

I think Bohemond covered well the near impossibility of accounting for the variables in studying the diets of a large group of people. Another variable I don’t think anyone else mentioned is the truthfulness of the participants. Are they really eating what they are reporting they are eating?

Bohemond said...

BillyBob@388 said...
When it comes to diet(and maybe also vitamins), sometimes you have to be your own study. For example, we could determine if a diet high in carbs, low in fat and especially saturated fat, and with a nice helping of trans-fat(the margarine they pushed on us for so many years, and cooking with vegetable oils) actually helped our waist line, HDL, triglycerides, blood sugar etc, and if we actually got healthier. Or if, horrors, we actually got worse. And then if we go against most advice, supposedly backed up by the science, and do the opposite of what our doctors and dieticians told us to do. Did we get worse or better?

If we actually got better, then that means there are a bunch of so called studies out there that are pure crap.

The point of the NYT article is exactly that many of these studies are products of p-hacking: "the process of running exhaustive analyses on data sets to tease out subtle signals that might otherwise be unremarkable. Critics say it is tantamount to casting a wide net and then creating a hypothesis to support whatever cherry-picked findings seem interesting"

However this doesn't mean that any study that conflicts with your or my personal anecdotal experience is pure crap. It only means that studies that cherry pick data to support a desired conclusion are crap. Sad to say, but your or my personal experience is no more than a one-off conclusion drawn upon whatever aspects of our personal lives we choose to cherry pick - what we eat, where we live, what chemicals we may or may not have inhaled or ingested, our genetic heritage, how much or what kind of exercise we have, how much or little stress we have or think we have, etc. The potential variables and combinations of variables are virtually endless for any one individual. And one individual's experience can't isolate or control for enough variables to support any kind of reasonable conclusion.
Jim

I agree. Just some of the studies are crap. That is why, as the article says, one study tells us one thing, the next week another study tells us the opposite. But probably, one of those studies is right. Which one?

And I totally agree with " your or my personal experience is no more than a one-off conclusion drawn upon whatever aspects of our personal lives we choose to cherry pick - what we eat, where we live, what chemicals we may or may not have inhaled or ingested, our genetic heritage, how much or what kind of exercise we have, how much or little stress we have or think we have, etc. The potential variables and combinations of variables are virtually endless for any one individual.". Certainly. As well as only being a study of N=1. And in my case, that probably means more than 1 variable in play at any given time, and another variable is the way those variables interact with each other.

But none of that is going to keep me from going with the result of my personal little study. For example, if I follow my doctors orders for a year and eat low fat, lower calorie, exercise more, etc, and at the end of 1 year I have lost no weight and maybe gained a little, some of my lab #s(like maybe LDL/TCL) have improved a tad but others are definitely worse (HDL/TGL/blood sugar), OK, there is clinical trial 1, as far as I am concerned. If then I start low carb eating and/or intermittent fasting, and in a few months I have lost weight, my gut has shrunk, my LDL/TCL is just a little worse or the same but my HDL is much improved and my TGLs have been cut in half and my blood sugar and BP have gone down, and I have not really been hungry(minimal suffeng), which one do you think I am going with? If you had the same experience, which one would you go with? I am definitely going with the latter, and I don't really care if my docs have a study telling me I should do the opposite. I will more likely become very suspicious of their study.

All though, for the most part, they don't really have any studies when it comes to diet. At least not controlled trials showing big benefits for low fat vs low carb. They mostly are just going on popular opinion and conventional wisdom. But when they do have a study supporting the opposite of what has been working for me - as well as quite a few friends who have tried it - I am not going to pay much attention to it. And if I look at it I will often find a problem with it, like not truly low carb, they are just comparing not low carb to somewhat lessor carb, or something else not right. Then they will declare no difference between low carb vs whatever. IOW, I am probably going to figure out there is something wrong with their study, where I am asking "why the heck did they do that?". Maybe a bias? But if it has been working for me, I am unlikely to accept a study saying it does not work or is bad for me. Would you? Would you discount your personal experience in a study of N=1?

Of course, we are mainly talking diet here. But with drugs, it is going to be harder to go against your docs recs based on their studies. But it does happen, like the time they put my wife on one of those anti bone loss drugs, and she felt like heck from the 1st pill. And rightfully reasoning that anything that made you feel that bad just could not be good for you overall said "forget that". So they did, and she went at it from a vitamin approach, and has not worsened any in the following years. But generally, it is probably going to be harder to tell if vitamins are helping or harming, compared to the dramatic results that can be seen from dietary changes. But there was that one time for me with my shoulder and liposomal B12. As well as one member here who tried it with at least initial success. (I don't know if it lasted a long time or not, like it did me) Those results were fairly dramatic and obvious. In my case, no more talk of shoulder surgery for well over 10 years now. But I doubt there are any modern studies on that one.

Post Edited (BillyBob@388) : 9/30/2018 3:39:42 PM (GMT-6)

This is nothing new. A few years ago I posted about a book, Doctoring Data, by Malcolm Kendrick, that discusses flawed medical research in depth.

Here are a some of the apropos chapter titles to whet your interest:

"Relative mountains are made out of relative molehills"
"Facts can, and often are, manufactured out of thin air"
"Things that are not true are often held to be true"

We want to be certain. Consider the perfervid certainty with which we tout our favorite diets on this forum and hold our views with nearly religious zeal.

It has become increasingly difficult to enjoy eating as one of life's simple pleasures. Deciding what to eat and what not eat to avoid deleterious effects on our health can end up being a full-time job.

Pratoman said...
Tim, this (your comments) is why I stopped analyzing every little morsel I put in my mouth. Now, I eat a lot of cheese. And it makes me happy. It’s that simple

Me too!

All studies are wrong...for obvious reasons

GoBucks said...
I am in a clinical trial of 1. Since being diagnosed and after surgery I decided to enjoy life more, especially ice cream. Since that decision my PSA has been zero. So I suggest you all get a Handel's Ice Cream franchise. (coincidentally I started Lupron, Zytiga & Prednisone at the same time as my ice cream trial). Clearly ice cream is the difference maker.

It has always worked for me. In fact, I had a dose tonight.

jim nyc said...
There was an article in The Economist a few years back about the large amounts of questionable articles in scientific journals due to the pressure to publish. I don't remember the numbers but it was quite surprising. I think it is clear from the nutrition studies that they are trying to infer causation from correlation and that's always a stretch. I think cultural studies shine some light on diet but there are too many variables to pin it down.

Personally, I've gravitated to a Mediterranean type diet to get my LDL down and that's been effective. I tend to avoid red meet and cheese-laden food; I've increased fish. I also find that rich food makes me sluggish after I eat. I'm not strict about it but a gradual change has been helpful for me.

+1

browntrout said...
Just my take, for what it is worth. India, for example has way less cancer"S" than US including PC. Is it their diet? I don't know but keep thinking, what else could it be? Here is an interesting take. And please remember, just another take that "might" point the finger at diet.

https://nutritionfacts.org/2015/05/05/why-are-cancer-rates-so-low-in-india/

Some conclusions would indicate eating way less meat, a lot more spices, fruits, and vegetable. So, pretty much a plant based diet. The paragraph showing the cancer rate is something to really think about.

I seem to have read somewhere else that folks who come from countries with lower cancer rates and later move to the US tend to increase their cancer rates.

Although when I linked to this fellow in another recent thread, I was informed about what a biased zealot he was(probably true), none the less, he does link to some very interesting info. This article included, thanks for the link!

Even with the vastly lower PC rates, curcumin is only one of several possible causes, of course. For example, I would bet the average citizen of India- whether by way of less pasta and pie and pancakes, or simply by way of significantly lower average calories available to them- inadvertently have a low insulin diet.

If it is the curcumin even in part- and I bet it is- then what about what they tell us about Curcumin being horribly absorbed when eaten? If it is the Curcumin, they seem to be absorbing enough to be a huge help.

One thought is: they also eat plenty of black pepper. Which greatly increases the absorption rate of curcumin. I try to always take my curcumin with a meal with black pepper. Actually, usually I buy supplements that already contain the black pepper. The expensive brand name- Bio something- is not really needed as it is just black pepper.
/nutritionfacts.org/video/boosting-the-bioavailability-of-curcumin/

Post Edited (BillyBob@388) : 10/2/2018 12:53:43 PM (GMT-6)

Let's add some clarity: It's foolish to attempt draw meaningful conclusions about prostate cancer in the US versus prostate cancer in some place like India. Not going to be meaningful apples-to-apples. Here's a few points, and feel free to pooh-pooh any of them individually, but taken in aggregate the differences are far to vast.

• PC is, and remains, an "old man's disease." Yes, there is a long, thin tail on the distribution which results in a very small percentage of 50 YOs who die from PC, but generally it is an old man's disease. Average life expectancy in India is 10-years less than the US...this factor alone is the most significant contributor to the lower PC death rate in India.
  
• Hundreds of millions of Indians are in poverty; well over a quarter of the country. It's a joke to think there is any meaningful comparison of average diets.
  
• Attribution; in a country where some of the leading causes of death are diarrhoeal dieases and tuberculosis. There ARE some good hospitals in India where they accurately track cause of death, but far more people die every day without much understanding of why and with numerous confounding health issues.

You are kidding yourself, BillyBob, when you make comparisons about PC death rates and diets between countries such as these.

Post Edited (Normal59) : 10/2/2018 2:28:06 PM (GMT-6)