 There is a Japanese dish called fugu. It's prepared from the porcupine fish, genus diodon. This fish contains a poison that is lethal if ingested. It's especially concentrated in the liver. You're relying on a highly trained chef to prepare a meal that is delicious, yet non-lethal. Now here's the question. How good does fugu have to taste to risk your life to eat it? Annually, this dish is responsible for up to 176 deaths worldwide. Can you imagine their weeping families, their children or parents mourning their passing? There are many things I would risk my life for. My family, my friends, but I can't imagine risking my life for a flavorful fish. You know what else is good? Tilapia, salmon, cod, herring are good. Sashimi-grade tuna or snapper or perch or halibut all have lovely distinctive flavors and don't contain deadly neurotoxin. Now, what about raw milk? I'm sure it tastes quite good. I'm sure whole milk cheese tastes better than manufactured cheese to some. That's great. Now, what are the chances of getting sick or dying from raw milk? Very, very small. And unlike fugu, it doesn't kill very many people. Raw milk, for those of you unfamiliar with the term, is milk that has not been processed beyond extraction. Milk you buy in most grocery stores has been passed through four processes. Pasteurization, which briefly heats the milk to a high temperature, killing microorganisms. Homogenization, which makes the milk fat globules a uniform size that can stay suspended in the fluid milk so it doesn't separate on the shelf. Standardization, which adjusts the milk fat ratio to a standard 3.25% for whole milk or less for other kinds. Fortification, which adds vitamin D and sometimes vitamin A. The most important of these processes is pasteurization, named for the famous scientists who first applied it to wine to prevent spoilage. It increases shelf life, kills bacteria and viruses, and makes the milk safer to drink. It was first introduced in the late 1800s. But started gaining acceptance in the U.S. around 1910, especially in the cities, where milk had to be transported from high-density dairies to all the stores and households by room temperature, horse cart, or car. Even before we used pasteurization to render our milk safe, the number of people killed by milk-born illness was never very high. Barely a few tens of thousands a year. Far less than 1% of the population, and it was only about 25% of total foodborne illness. Today, milk accounts for less than 1% of foodborne illness. Most illness goes unreported. A very long day of vomiting or diarrhea is the worst most of us will have to experience. A child, the elderly, those who are sick, AIDS, cancer, or transplant patients, on the other hand, may risk serious injury or death. Before I give you exact numbers, I want you to answer a very simple question. How good does raw milk have to be to risk your life to drink it? How far over the edge of the cliff are you willing to lean to get the best view? Here are the actual figures. In the years 1922 to 1944, during the period that first saw the popularity of pasteurizing milk for safety, there were a total of 37,965 cases of reported milk-born disease, with an average of 1,726 cases per year. That's a deceptive statistic, though, as milk was considered a beverage for babies, not adults. And these are only the reported cases, most likely severe, that managed to be confirmed as milk-born. From 1998 to 2005, the CDC identified 45 outbreaks of foodborne illness that implicated unpasteurized milk, or cheese made from unpasteurized milk. These outbreaks accounted for 1,007 illnesses, 104 hospitalizations, and two deaths. They estimate this is only a small percentage, as most illness is unreported. Most of the diseases are caused by fecal coliforms, like E. coli and Salmonella, although campylobacter is really the big one. And soft, unripe cheeses, like farmer cheese or quesifresca, are major sources of listeria monosatogenes. Where do the bacteria come from? They come from cow feces, cow skin, and of course human sources. Cows are not, by nature, clean animals, and they harbor bacteria that we find harmful. Pasteurization renders milk largely free of bacteria, or greatly decreases bacterial load. It can still be contaminated after pasteurization, and that's why even pasteurized milk can still be a source of harmful pathogens. You can certainly go the Fugu route and have a very skilled dairy technician prepare your possibly contaminated milk. All the risk can be minimized by various ways, like using free-range grass-fed cows grown on very clean, very small farms, but only pasteurization eliminates those known risk factors. A California survey in 1997 indicated that young Hispanic males with no high school diploma were more than twice as likely as the general population to drink raw milk. However, the raw milk movement is rapidly gaining support among affluent white alt-med hipsters, who tout raw milk as edible medicine, preventing everything from cancer to asthma. I think part of the appeal is the counterculture edge of being rebellious against medical establishment, and the medical and agricultural experts are against raw milk consumption. The CDC, NIH, FDA, USDA, the AMA, and the National Association of State Departments of Agriculture have all issued statements against raw milk consumption. Even the future farmers of America, the group I suppose might have the most vested interest in this industry, have an education program on risks. Yet it's still legal to purchase in 28 states, completely illegal in 15, and in the remainder there are legal exemptions, or cow-sharing programs, or farmer direct sales. It's still against U.S. federal law to transport it for sale across state lines. However, some large raw milk operations ship their product labeled as pet food to subvert this protection. Canada and Australia also have laws outlawing direct sales. I want to use this issue of raw milk to demonstrate how to use rationality and critical thinking to determine the smartest policy on a controversial issue. I want you to imagine a scale. On one side will have costs, and the other side will be benefits. We'd normally be assigning quantitative values to each side, so a sheet of paper can be helpful for this. For raw milk, the proponents will tell us that the benefits are nutrition, flavor, probiotics, and I'll add one extra item that we should always include, liberty, the freedom to eat what you want. On the cost side, we have foodborne illness. Under that, we're going to have a number of sub items, direct cost of medical care, lost productivity and lost lives, and risk of epidemic outbreaks. There are some adjustments that need to be made, though. There is very little difference between raw and pasteurized milk in terms of nutrition, and in a peer-reviewed publication, I was able to locate. The fat levels in raw milk are often higher than 3.25% of homogenized whole fat milk. Pasteurized milk is going to have vitamin D added. The protein content is the same as well as sugar. There are going to be additional vitamins found in the contaminating bacteria of raw milk. However, if there are other unknown differences, there are going to be lower risk ways to obtain that nutrition. So we'll cross out nutrition. I didn't even put it up, but we can also rule out reduced cancer risk or heart disease risk. Both have been studied, and unpasteurized milk consumption has no strong correlation with cancer and heart disease in repeated studies. Asthma treatment is a bit more complicated. Breast-fed infants are less likely to develop asthma, but there's still no strong case that raw cow's milk has an equivalent effect on human babies. We know that exposure to certain farmyard bacteria can have an anti-allergy effect in a small percentage of children, but when raw milk is isolated from the total farm experience, the effect disappears. There is a controversial idea called the hygiene hypothesis that proposes an inverse link between the things we are exposed to as children and our later in life allergies. People who grow up in a sterile environment are more likely to develop these conditions. Probiotics are good bacteria that colonize our gut and protect us from bad bacteria. Just one problem. There's not a lot of evidence supporting a probiotic effect of raw milk. In fact, just the opposite. You're more likely to end up with a harmful toxin-producing pathogen that has no effect on the cow. Let's be really clear on this issue. Milk is sterile at the source. It is contamination with cow fecal bacteria that result in bacteria in the milk. High levels of this fecal bacteria are always a bad thing, and it's high levels that are needed for any probiotic effect. If you do want probiotics, active culture yogurt is a great way to go and carries a much lower risk. So what are we left with in benefits? Flavor and liberty. In the costs, we just have foodborne illness. Now in case you have doubts that raw milk is really more risky than pasteurized, here's a statistic that really stands out. Raw milk constitutes 1% of milk consumption, but 56% of milkborne disease outbreaks since 1973. That's 44 times as much as we would expect if they were of equal risk. Fortunately, the number of people affected is much lower because the raw milk can only reach the small number of people who have arranged for it. So let's try to quantitate the social cost. Let me quote from the CDC website. From 1993 to 2006, 69 outbreaks of human infections resulting from consumption of raw milk were reported to CDC. These outbreaks included a total of 1,505 reported illnesses, 185 hospitalizations, and two deaths. End of quote. Both deaths included were children, as were a majority of the hospitalizations. That's more than five outbreaks per year. 115 reported sick, 14 hospitalized, all from less than 1% of the U.S. population that consumes raw milk. We can attach some dollar values to these losses. Treating hemolytic urinary syndrome in a hospital setting costs an average of $15,000. Deaths cost on average about $6.1 million, and loss productivity is a broad estimate at $2,500 per case. So in 13 years, raw milk cost $18.7 million in direct costs, or $1.4 million per year in health damage. Again, this is for 1% of the U.S. population. It's also led to unquantifiable concerns about the return of milkborne epidemics. Many of these can also become waterborne pathogens, or directly transmitted from sick people among the community. So here's how I see the issue. On one side we have liberty, and a delicious flavor for about 30 million people to enjoy. On the other, we have $1.4 million U.S. dollars worth of suffering and death, as well as a possible risk of future epidemics. As with Fugu, that better be some really delicious milk you're drinking, because you're taking a measurable risk and drinking it, and we're all bearing the cost of that risk. Thanks for watching.