The China Study

Sponsored by the Pine Street Foundation, this transcript gives more important information regarding the link between nutrition and disease.


The China Study demonstrates the link between nutrition and heart disease, diabetes, and cancer. Referred to as the “Grand Prix of epidemiology” by The New York Times, this study examines more than 350 variables of health and nutrition with surveys from 6,500 adults in 65 counties, representing 2,500 counties across rural China and Taiwan. While revealing that proper nutrition can have a dramatic effect on reducing and reversing heart disease, diabetes, cancer, and obesity, this study calls into question the practices of many of the current dietary programs, such as the Atkins diet, that enjoy widespread popularity in the West. The impact of the politics of nutrition and the efforts by food industry lobbyists on the creation and dissemination of public information on nutrition is also discussed.


In the late 1970s and early 1980s, there were two principal observations suggesting a relationship between diet and cancer. First, rich Western diets (high in fat and meat, low in dietary fiber) were strongly associated (correlated) with incidence of colon and breast cancer. Second, migrants moving to areas of different cancer risks acquired the risk of the country to which they moved, regardless of their ethnic or genetic backgrounds.


In 1981, the Chinese Academy of Medical Science published an Atlas of Cancer Mortality on the 1973-75 mortality rates for about a dozen different cancers for 2,400 counties in China. These maps showed that cancer was highly localized in specific geographic regions. Residents of these regions tended to live in the same regions all their lives and to consume the same diets unique to each region each and every year. Their diets (low in fat and high in dietary fiber and plant material) also were in stark contrast to the rich diets of the Western countries.



Colin Campbell, PhD, is the project director of the China-Oxford-Cornell Diet and Health Project (the China Study), a 20-year study of nutrition and health. He is a Jacob Gould Schurman Professor Emeritus of nutritional biochemistry at Cornell University. In more than 40 years of research, he has received more than 70 grant-years of peer-reviewed research funding and authored more than 300 research papers. He lives in Ithaca, New York.


Lawrence H. Kushi, ScD, is Associate Director for Etiology and Prevention Research, Division of Research, Kaiser Permanente. Internationally recognized for his expertise in nutritional epidemiology, Dr. Kushi’s research interests have focused on the role of food and nutrition in the development and prevention of coronary artery disease and breast and other cancers. Dr. Kushi is also is the second son of Michio and Aveline Kushi, the foremost proponents of macrobiotics and leaders in the development and acceptance of unconventional and lifestyle approaches to cancer. Dr. Kushi is collaborating with the Pine Street Foundation on the Pine Street Survival Study, a 10-year follow-up study of people with breast, lung, and colon cancers treated with an integrative vitamin/herbal protocol in combination with standard chemotherapy.


Dr. Lawrence H. Kushi: You are a strong advocate of essentially vegetarian or even vegan dietary choices. One of the things that I find interesting is that you actually grew up on a dairy farm in Virginia.

Dr. T. Colin Campbell: As a youth, I was a meat and potatoes kind of guy. I was raised on a farm and milked cows until I went away to school. I was the first member in my family on either side to even go to college, let alone graduate school, so I really went into the whole scientific profession rather naïvely. My initial graduate work at Cornell University was on how to grow cows more efficiently so we could eat more of them and drink more of their milk. My doctoral dissertation was specifically focused on making sure we all got enough protein, especially so-called “high quality” animal protein. I would have preferred to have continued on that path, but as I got involved in the Philippines working with malnourished children and then in the laboratory doing a lot of basic research, I started to get some results that began to question my whole upbringing, especially what I believed about protein.

LK: Can you give an example of your work in the Philippines that triggered this?

CC: My job in the 1970s was to coordinate a nationwide program to feed malnourished children in the Philippines. In those days, and still to some extent today, the notion was that malnourishment largely resulted either from not enough calories, not enough protein, or some combination of the two. The protein issue was one of my principal interests when I first got involved.

One day, I was playing golf with a medical advisor to President Marcos who told me that he and some other doctors had recently been operating on children four years old and younger for primary liver cancer, which I thought was very interesting. I then started investigating where these children were likely to come from and they seemed to be from families who were the best fed and who were getting the most protein. All of a sudden, I began to question what I was doing there; I was trying to get more protein to the kids and, in some cases, they were ending up worse off. It was exactly the opposite of what I thought I was going there to do.

LK: How did you then take this observation into your work?

CC: I first told one of my senior colleagues back at MIT about this and he thought it was crazy. Then I saw a paper in the Indian journal Pathology that showed that when rats were exposed to a carcinogen and then fed two different levels of animal-based protein, the animals fed the regular levels of protein essentially all got tumors whereas the animals fed the lower level – the so-called “inadequate level” – did not. That was essentially what I thought I was seeing in the children, too. The interesting thing about that study was that the level of protein being fed that caused tumors to develop was not exorbitant; it was well within the range of what humans might ordinarily consume (the protein requirements of rats and humans, as a percentage of total calories, are fairly identical).

LK: The type of protein used in that study was casein, the principal protein found in milk.

CC: In the beginning, I didn’t pay a lot of attention to the kind of protein we were using in our research, but eventually had to come to terms with the fact that the type of protein I was using all along – as everyone else was doing in those days – was casein. So I then tried soy protein and wheat protein and they didn’t have the same tumor growth effect. I also found that the casein effect only existed when the level of protein in the diet was above 10% of total calories. In other words, once the animal satisfied its need for protein and then started consuming excess of that as casein, tumors started to grow. In the case of plant proteins, however, exceeding 10% and even up to 20% of total calories as protein, tumors did not grow. So, suddenly there emerged this dichotomy between the two kinds of protein, animal and plant.

LK: Would you say casein is a carcinogen?

CC: Chemicals that cause cancer are called carcinogens and here we had a situation where casein fed at levels that could be anticipated in human diets was, in a sense, having a very strong effect in promoting tumor growth; 20% protein (as casein) diets generated tumor growth whereas 5% didn’t. Although we didn’t do the test at the time, there was also all sorts of evidence to suggest that this effect was probably a property of animal protein in general, not just of casein.

LK: How did the China Study first begin?

CC: At the time, the Chinese government had just released a study showing that cancer tended to occur more in certain Chinese counties and much less in others, resulting in great differences in mortality rates. Intrigued by this, we decided to go to China and measure various factors with respect to diet and lifestyle experiences and then compare those measurements with the disease rates that the Chinese government had already obtained for the years of 1973 to 1975 to see what relationships between the two, if any, existed. We selected 130 villages in rural China, which tended to be a very stable population (people lived in the same village all their lives) and tended to produce and use locally grown foods. From an epidemiological point of view, it was very nice.

LK: What were some of the major findings from the China Study?

CC: Essentially that animal-based foods were a problem. In the China Study, we found that as soon as animal-based proteins started to appear in the diet in certain Chinese counties, blood cholesterol levels, which were very low by Western standards (about 80mg per deciliter) started going up and that various cancers and heart disease started to appear. The association between the increase in animal-based foods and the increase in those diseases was highly significant, so much so that it made me get to a point where I started to question the way I was raised and made me switch to consuming a purely plant-based diet.

LK: What is your opinion of the current promotion of low-carbohydrate diets?

CC: When the promotion of these diets first started, I reacted in a negative way because the proponents took that term and essentially confused the public – on purpose – to make their point. The whole genesis for the notion of low-carbohydrate (and therefore higher in protein and higher in fat) being better was basically a frontal attack in a very simplistic, albeit very effective, way on the recommendations that I have become accustomed to because plant-based diets are inherently “high carb” diets. In the China Study, it was the diets highest in carbohydrates that were associated with the least cancer and the least heart disease.

LK: What about studies that indicate that low-carbohydrate diets do appear to have some short-term health benefits, such as lowering cholesterol levels and assisting in weight loss?

CC: Low-carbohydrate diets have been shown to cause some weight loss, especially among those who are substantially overweight. To some extent, cholesterol levels have also been shown to decrease. But not all studies have produced those effects, however, and the weight loss and lower cholesterol benefits are minimal as compared to what is possible with a plant-based diet. Furthermore, people who go on low-carbohydrate diets usually consume fewer calories, at least initially, so we’re talking about a calorie effect there that can’t be sustained.

LK: What about the “Mediterranean Diet”?

CC: There was research that showed that people who lived in various Mediterranean countries – Crete, southern Greece, southern France, and so forth – tended to have lower rates of breast cancer, colon cancer, and heart disease as compared to people living in the United States and England, despite the fact they were consuming diets that were fairly high in fat, mostly from olive oil. It turns out that in those Mediterranean countries where the fat intake is quite high, the proportion of their total food as plant food was very similar to what we saw in rural china. However, when you compare the rural Chinese to people in Mediterranean countries, it turns out that the heart disease rates in the Mediterranean are quite a bit higher than in rural China. So, really the question you should be asking is, “Why are the disease rates so high in those Mediterranean countries as compared to rural China?” Is it because of the consumption of olive oil?

LK: What about soy protein?

CC: Soy is a legume and a good source of protein, as far as plants are concerned, so it should be part of a healthy diet. But soy has now been processed into so many products and this processing can lead to things that one might raise some questions about.

Another issue is that we have people who are quite possibly consuming too much soy. One element that has been discussed is the presence of phytoestrogens in soy; these estrogens might be just as mischievous as people’s own estrogens or estrogens they might consume from animal foods. At certain consumption levels, these phytoestrogens are probably beneficial because they are antiestrogens and tend to diminish the effect of otherwise high levels of mammalian estrogens. But I think soy protein, if it were fed at high enough levels, might end up doing some of the same things as animal proteins…it’s simply a matter of quantity.

Regarding the soy debate itself, I first saw it emerge in the 1970s when I was living in the Philippines. The Philippines wanted to ship soy to the West and, almost immediately, there erupted all these news releases about the hazards of soy that ultimately proved to be coming from the dairy industry. I know that the dairy industry has not been happy over the years with the erosion of their territory because of soy products, so when I see so-called “reports,” I don’t know how to react to them because I just have this inherent skepticism as to where it might really be coming from.

LK: Are there differences between protein from fish and protein from beef?

CC: In one study, Dr. Kenneth Carroll compared a whole variety of animal-based proteins, including fish protein, to a whole variety of different plant proteins in their ability to generate cholesterol levels. What he found when he examined them in a systematic way in experimental animals was that animal proteins tended to increased cholesterol and plant proteins tended to decrease cholesterol. Fish protein was the one animal protein that was the least effective in increasing cholesterol levels, but it was still far higher than all the plant proteins.

LK: Because you advocate a plant-based diet, one of the questions you are often asked is, “How are you supposed to get enough protein?” It seems there has been a large fixation on protein, both in the nutrition science community as well as in the general public.

CC: The Recommended Daily Allowance for protein, ever since 1943 when we were making such recommendations, is set at 10% of total calories. This amount allows for some variation among individuals and is supposed to be enough. This is also about the level of protein that a good plant-based diet provides. However, there’s such a fixation on protein in this country that, as a population, we’re consuming somewhere between 11% and 23% and the average now is around 17% or 18%. If we listen to the food pyramid recommendations, we’re being encouraged to go even higher.

LK: Is there an over-emphasis on cholesterol measurements?

CC: Yes. Generally speaking, the higher the cholesterol levels, the greater the risk for heart disease and stroke. But when reducing that to the individual level, we know there are lots of exceptions; some people with relatively low cholesterol levels have heart disease while some people with relatively high cholesterol levels don’t have heart disease.

Cholesterol measurement is a crude instrument. It was refined, to some extent, when we started measuring HDL and LDL – “good cholesterol” and “bad cholesterol” – and now has become more refined by measuring oxidized LDL, but that hasn’t really been taken into consideration clinically yet. But I think we should recognize the limitations, especially for individuals, of measuring cholesterol as an indicator of heart disease and recognize that there’s a host of other factors that, when put together – perhaps as an index – is eventually going to turn out to be a better estimate of disease risk than just measuring cholesterol alone.

LK: You have said that “the distinctions between government, industry, science, and medicine have become blurred and the distinctions between making a profit and promoting health have become blurred.”

CC: That’s a view that I think a lot of people share, too. Unfortunately, I’ve seen things that I find to be deeply, deeply troubling and I think it’s getting worse. I think our academic science is being severely corrupted by commercial interests. Right now, for example, the most recent food and nutrition board report from the National Academy of Sciences is funded, in part, by the food and drug companies.

The food pyramid committee has been similarly corrupted. For example, when someone requested information as to what conflicts of interest the panel members on the previous board may have had, the USDA refused to make that information publicly available, even though it’s required. So after about eight months, a judge forced the USDA to release that information, which showed that six of the eleven members, including the chair, had an association with the dairy industry. And now in the new food pyramid that was just released, we’re getting recommendations to increase milk consumption. I find it deeply troubling that we can’t be honest about the science without having to worry about who’s paying the bills.

Comments are closed.