You are currently browsing comments. If you would like to return to the full story, you can read the full entry here: “Where's the beef?”.
Related posts:
- Liver: nature's most potent superfood Most people might think of a green spirulina drink or a "nutrition bar" when they...
Tags: archives, beef, Cancer, internal, meat, medicine, red, study
-
well, I did pretty darn good…I saw a good couple of dozen of those articles, didn’t bother reading any of them and speculated everything you document here in your piece!!
sometimes it’s just to easy to see through the crap!!
but why are there so many people who simply don’t??
-
Most consumers don’t study research methodology in school. Understandably, the average Joe or Jane doesn’t really know how to read a study and critique it.
For that, they depend on the media. Oops! I’m not sure where they find the “science reporters” at major news outlets, but clearly they either 1) are not reading the actual studies, and just parroting whatever they get off the AP news wire; 2) don’t know how to properly analyze and critique a study; or, 3) share the agenda of the study authors and aren’t interested in getting to the truth.
I’m not sure which of those possibilities is worse.
-
well…I’ve never been formally taught how to review a study, nor have a studied research methodology and I figured out all the stuff you were talking about…WITHOUT reading the study. I figured, well, this is not organic grass fed beef…and this is average Americans who eat a ton of processed meat and white flour…blah blah blah…I DIDN”T NEED A SPECIFIC EDUCATION on how to look at a study!!
I don’t know…I guess a good general education and a naturally somewhat analytical brain is all you need and most people don’t have that…but journalists certainly should.
My mind is constantly boggled by what passes as news.
-
Chris,
Do you know of a paper that I can read that teaches me how to critique a study and determine if it is a “good” study?
I do understand that there are whole classes that teach this type of thing.
Thanks
-
David,
You could check out the Statistics for Clinicians article series at your local medical library. I’m not aware of any other free resources, but you might see what you can find with Google. I’d recommend taking a class at your local junior college if you’re really interested, though.
Best,
Chris -
Say Hey-
How many times have we been advised to just plot the data ourselves? Read the data tables for yourself; here is one that made my jaw drop…
Women
Red meat(g/k-kcal) 9.1 21.2 31.2 42.8 65.9
Cancer (Deaths) 2134 1976 1784 1687 1348This is their data on red meat and cancer in women. If you want to expose how the study’s correlation between processed meat and men’s cancer is not a casual link, their own data on women and cancer is all you need to consider.
By their own logic, red meat reduces cancer in women; even processed meat reduced cancer in women. White meat was about neutral. So a key to this study is the difference between men and women. Why not come out and tell women red meat is good for them?
Examining the data format reveals that the red meat data is both red meat and processed meat. They did not plot red meat alone. Why include processed meat? Because that’s what generates the correlation. There is a nice Irish study (British Journal of Nutrition (2005), 93, 933–942) on red meat consumption and disease that pursues this relationship between men and processed meat.
Want to guess the health aspects of a lifestyle based on bologna and hot dogs?
-Rich
-
Hi Rich,
Those are all very good points, and they highlight the ridiculousness of inferring causality from observational studies. Observational studies are useful for indicating correlations that might be worth investigating in future clinical trials, but you can never prove causality with an observational study. The study authors surely learned this in their undergraduate level research methodology course, but apparently their ideology and agenda got in the way of good science. Another issue is that meat consumption was based on the participants self-reporting in food diaries, which are notoriously inaccurate. Most people can hardly remember what they ate yesterday, let alone several months ago.
The study is a joke. It’s embarrassing that it even got published. Even more remarkable is that not a single science reporter in the mainstream media (that I’m aware of) analyzed the study before splashing headlines of its nonsensical conclusion across the internet and newspapers. They should be ashamed of themselves. Business as usual, I guess. This kind of thing is exactly what prompted me to write this blog.
-
David,
I’ve recently come across a couple of resources you might find helpful.
The first is an article on basic statistical literacy in relation to health care.
The second is an independent health news review website that evaluates health news headlines and the studies they reference. For example, here is their review of the red meat / mortality study that is the subject of this blog post.
-
Also, here’s a list of what people should know about health news stories, from the HealthNewsReview site.
-
Thank you for this, Chris.
Really, point #7 is my big beef with most studies that call one food or another bad. There is just too much focus on calories and fat, and it seems like only the holistic community is taking into consideration the quality of the food and whether or not preservatives were present.
I am reminded of the somewhat infamous egg study from I believe the 50’s that told everyone that eggs would raise their cholesterol. This created a wonderful market for egg white products, coincidentally. In Ann Louise Gittleman’s book Your Body Knows Best, she explains that the original study was done on powdered egg yolks and was funded by the Cereal Institute.
I really wish researchers would start putting out some more realistic studies.
-
Detailed description, good.
-
Sorry to be contrary, but T. Colin Campbell’s “The China Study” should put this issue to rest. Please consider the information presented there. The methodology is impressive.
-
The methodology of The China Study is anything but impressive. The book is riddled with inconsistencies, inaccurate generalizations and fallacious arguments. See Chris Masterjohn’s review of Campbell’s book for details. Make sure to read Campbell’s response, and Masterjohn’s subsequent rebuttal.
-
Chris, Thank you for the links. I found them very interesting and informative.
-
Thank you for constantly reaffirming the things I believe in!
-
You’re welcome! Thanks for your support.
-
I truly appreciate the level of your posts here in the Healthy Skeptic. It is refreshing and inspirational to see your well thought out discussions. I enjoy all your posts and look forward to more.
-
Sabra,
I’m glad you’re finding the blog to be helpful. Welcome!
Chris
-
You say, “Observational studies can show an association between two variables (i.e red meat consumption and death), but they can never show causation (i.e. that eating red meat caused the deaths). A simple example of the difference between correlation and causation is that elevated white blood cell count is correlated with infections. But that doesn’t mean elevated white blood cell counts cause infections!”
It’s an important point. However, though correlation does not imply causation, would you agree it implies a relationship between the two? In your example of white blood cells and infection, a relationship does exist; infections cause an increase in white blood cells (right?). When considering red meat and death, if there is a statistically significant correlation between the two, one might infer a relationship… I would hope an impartial scientist would then ask the question: What is the nature of this relationship? -
Sure, correlation implies a relationship but it’s dangerous and irresponsible to assume that the relationship is causal. That’s my point.
A study done in the 50s showed that television ownership is highly correlated with heart disease. Does that mean buying a TV causes heart disease? Or that heart disease causes one to buy TVs? Hardly. But it does suggest that there may be some other unknown factor that explains why people with televisions tend to have more heart disease. In this case it was very likely that people with TVs exercised less then those who didn’t have them.
If people who are eating more red meat in general are also eating more refined oils, sugar, and other processed foods that actually do contribute to heart disease, then it wouldn’t be a surprise at all to see a correlation between red meat consumption and heart disease. But it would be a tremendous mistake to assume that red meat was the cause – without actual proof that this is the case.
20 comments
Comments feed for this article