How credible is The China Study?

January 2, 2008 3:00 PM

How credible is The China Study by Dr. T. Colin Campbell and Thomas M. Campbell II? We would like opinions from individuals who have read the book, although opinions based on other factors are welcome so long as you mention it.

My boyfriend and I, I think, want to believe what the book says, and that is interfering with our ability to read it critically. Are any of those experiments/studies bogus? Was their design flawed? Was Campbell jumping to conclusions?

We just want to hear from the clever cookies on mefi how much weight we ought to give to the message of The China Study.

posted by mjao to science & nature (8 comments total)

I don't know about the book, but here is the NY Times article about the original study that they cite prominently on their site. The study appears to be fairly credible.
posted by uaudio at 3:13 PM on January 2, 2008

I've been planning on reading for quite awhile. Thanks for reminding me.

I remember I was reading the Amazon reviews/discussion and those were absolutely fascinating.

But it also reminded me why I like Micheal Pollan and I'm more likely to read his new book before I pick this up. His article Unhappy Meals and his new book address something he calls "nutritionism." According to Pollan, reducing diet to stats is ridiculous at best and harmful at worst.

The China Study is a perfect example of reducing thing to stats without the context of human evolution or culture. Give me an ethnography over this kind of thing any day. Plenty of cultures have low rates of "diseases of civilization" without eating a vegan diet. It's pretty clear that people can eat animal products without getting diabetes/heart disease from studies of other populations like the Inuit. Also see this NYT article.

The cancer problem is the interesting one.

The biggest problem with the study is that none of the people were eating the diets in contention. Some were eating more animal products, other were eating less, but none of them were vegan or paleo or any of the other diets being debated on the Amazon page.

I will read it eventually though and since I'm trained in stats, it should be interesting. In the end though, it's just stats and the deeper picture will emerge if/when they find how certain foods cause cancer.
posted by melissam at 3:25 PM on January 2, 2008

Well, I read it and I eat a high protein diet to this day.

Certainly there exists much criticism of his study and his conclusions for example, here.
posted by Comrade_robot at 3:30 PM on January 2, 2008

Here is more criticism. He cites pretty well, but it's up to you to follow the citations.
posted by Kwantsar at 4:12 PM on January 2, 2008

My understanding of the China Study is that the data were pooled at the county level. This type of study design is called an "ecological study" in the nutrition epidemiology literature.

Such studies can suffer from problems like the ecological fallacy. For example, countries that have the highest Wii possession per capita have the highest rates of obesity. This does not mean at the INDIVIDUAL LEVEL, however, that owning a Wii causes obesity. In fact, the opposite might be true at the level of the individual.

Many of these early studies were essential in developing nutrition/disease relationships that were later tested by different study designs.

In general, to see if certain diets or nutrients are significant predictors of disease risk for individual people, look for studies in which individual people were followed prospectively, and their disease outcomes then measured.
posted by tiburon at 5:20 PM on January 2, 2008

Sichuan Province has the highest population in China and the diet features meat very prominently. I doubt a study of 6500 people accounted for regional dietary differences.
posted by markovich at 7:19 PM on January 2, 2008

The China Study cites a lot of research, not just the one big study in China, all supporting the same conclusion: a plant-based, whole foods diet provides tremendous health benefits over traditional American/European diets.

Another book you really should read is The Food Revolution by John Robbins (he wrote the forward to The China Study). I considered myself pretty knowledgeable about the state of American food production, but this book shocked and utterly horrified me. It has a pretty clear agenda, but it's very well-researched and makes extensive use of citations. I wish I'd read it years ago.
posted by brain at 10:36 PM on January 2, 2008

According to Pollan, reducing diet to stats is ridiculous at best and harmful at worst.
I agree that there is some harm in reducing diet to statistics, but I think Pollan's reducing ethics to convenience is much more harmful.

To me, the China study is one more data point that a diet that is low in meat can be very healthy. If you want to read more about health arguments for strongly reducing meat and dairy, you may want to read something by Dr Fuhrman. Eat to Live is not a very well written book, but it shocked many people I know into starting to eat a vegetable based diet. Fuhrman is credible and cites lots of research. He also does not advocate a vegan diet per se, just a diet low in animal products.
posted by davar at 2:10 AM on January 3, 2008