In 1983, a Cornell University nutritional biochemist named T. Colin Campbell embarked on what would become one of the most ambitious dietary studies in history, a sweeping collaboration between Cornell, Oxford University, and the Chinese Academy of Preventive Medicine that examined the eating habits and disease rates across 65 rural Chinese counties. The New York Times called it the Grand Prix of epidemiology. Campbell called it the most comprehensive study of nutrition ever conducted. He was not wrong about the scope. What he did with the findings is where things get interesting.
The core observation was genuinely compelling. The more animal protein consumed, the greater the toll of disease. Rural Chinese populations eating predominantly plant-based diets had dramatically lower rates of heart disease, cancer, and diabetes than their Western counterparts gorging on meat and dairy. Campbell published his findings in 2005 in a book called The China Study, which became a massive bestseller and a cornerstone text of the plant-based movement. People who ate the most animal-based foods got the most chronic disease, and people who ate the most plant-based foods were the healthiest. The conclusion Campbell drew was that a whole foods plant-based diet could prevent and even reverse a wide range of chronic illness. So far so good.
Then came the rats.
Before the China study itself, Campbell had been conducting laboratory research in which he fed rats a protein derived from cow's milk called casein, first poisoning them with aflatoxin, a mold-based carcinogen, and then varying how much casein they received. The poisoned rats given a 20 percent casein diet all developed cancer or cancer-like lesions, while the rats fed a 5 percent casein diet did not. Campbell found he could essentially switch cancer on and off by adjusting the casein level. He found this so remarkable that he concluded animal protein, all animal protein, promoted cancer growth. He then used this rat data to frame and reinforce the epidemiological findings from China, and the book proceeded from there.
This is where the scientific community raised its hand.
What powdered, isolated casein does to rats tells us little about what traditionally consumed forms of milk will do to humans, and tells us nothing that we can generalize to all animal nutrients. Casein is one protein derived from one source in one highly processed form, fed to rats who had already been poisoned with a carcinogen. The leap from that to "all animal protein causes cancer in humans" requires, as one critic memorably put it, several somersaults of logic and possibly some cartwheels of delusion. Whey, the other major protein found in milk, appears to have qualities that actually protect against cancer, but Campbell did not include whey or any other animal protein in his experiments.
The methodological problems with the broader China data were equally pointed. Campbell presents an almost exclusively unilateral analysis of the data, overlooking many potentially confounding variables such as alcohol consumption, sugar intake, wheat intake, activity levels, and geographical location. The rural Chinese populations eating plant-based diets were also eating far fewer calories overall, doing considerably more physical labor, consuming virtually no processed food, and living in ways that differed from Western populations in dozens of variables that had nothing to do with animal protein. Attributing their health outcomes to the absence of meat while ignoring everything else is a little like crediting a monk's longevity to the color of his robe.
Campbell does not look into carbohydrate consumption or sugar consumption, and the main mistake he makes is lumping casein and meat together as though they are the same thing. They are not even close to the same thing.
None of this means the book's broader instincts were wrong. They were not. The push toward whole foods and away from processed junk is correct. The observation that Western diets laden with animal products and refined carbohydrates produce disease is well supported by evidence far beyond this study. The problem is not the destination. It is the reasoning used to get there, and the certainty with which Campbell arrived, which was considerably stronger than his data warranted.
What the China Study actually proved, at its most honest, is that populations eating simple, unprocessed, plant-heavy diets tend to be healthier than populations eating the opposite. That is a valuable and important finding. It did not prove that a single gram of animal protein is a biological threat, that casein in rats equals meat in humans, or that veganism is the only viable path to health. Those were conclusions Campbell wanted to reach, and the data was selectively organized to support them.
The book remains worth reading. The core message, that food is medicine and that the Western diet is slowly killing people, is as relevant today as it was in 2005. Just read it the way you should read any passionate argument from a true believer: with genuine interest, and one eyebrow raised.