Originally published January 1993. Published on KurzweilAI.net August 6, 2001.

[Just in time to help you with your New Year’s resolutions, this month’s Futurecast will change tack to illustrate that the future involves not only the workings of the mind but those of the body as well. Though columnist Raymond Kurzweil is best known for his unparalleled contributions in the sphere of artificial intelligence, he has conducted extensive research over the past ten years in the area of health, and his book The 10% Solution for a Healthy Life: How To Eliminate Virtually All Risk of Heart Disease and Cancer (LJ 12/92) is being published this month. In writing the book Kurzweil pored over virtually every related study available in the nation’s libraries and worked with a distinguished medical review team. The book has been endorsed by leading physicians at the Harvard Medical School and other prominent medical institutions. -Ed.]

When collecting data for my book, my laboratory was not the expected setting of a buzzing hospital or healthcare center. Instead, I found everything I needed at the Harvard Medical School Library and other research libraries around the country. The experience gave me a new perspective on the value of libraries in conducting research and the depth of information that is readily available on vital topics that influence our lives.

New research tools such as online information services and research compilations on CD-ROMs have become enabling tools that allow the synthesizing of diverse research materials that would have been impractical only a few years ago. And you don’t have to write a book to appreciate the benefits of computerized knowledge sources. I know eighth graders who routinely use similar library services to conduct research on topics ranging from immigration to zoology.

Distorted Facts

Many people consider themselves well informed on the topic of health and nutrition, and it certainly is a subject that affects the quality of our everyday lives. Yet virtually all of the information that people receive on this subject is obtained courtesy of the media: newspapers, brief television segments, and, of course, ubiquitous advertisements that are hardly unbiased in their perspectives. Even if we ignore the obviously commercial attempts to influence the public debate, the information available in articles and news broadcasts can be equally distorted. I have often found it extraordinary how the common wisdom contained in today’s headlines can vary from the truth that is readily discovered at the library.

Let’s consider some examples. A recent headline in a major daily stated that “Nurses Study Shows No Link Between Fat Consumption and Breast Cancer.” The article was consistent with the headline: nurses eating “normat” diets and “low-fat” diets apparently did not vary in their rate of breast cancer. Well, so much for that controversy; the article made it clear that this case was closed.

But if you visited your local research library, you would have discovered (as I did) a very different story. After searching for “nurse’s study,” “fat consumption,” and “cancer, breast,” the appropriate abstract appeared on the library’s computer screen. The abstract then made it clear that the study compared two groups: one eating the normal American diet of approximately 37 percent calories from fat and the other restricting their fat intake to no more than 30 percent of calories. Consistent with the newspaper articles, the abstract stated that the two groups did not vary to a significant degree in breast cancer mortality.

Dropping the keywords “nurse’s study” uncovered other articles on the same topics (“fat consumption” and “breast cancer”). These other studies were consistent with the nurse’s study in that reducing fat consumption to 30 percent of calories does not significantly alter the rate of breast cancer, but they also stated that reducing fat consumption further, to ten percent of calories, does make a substantial difference. In fact, it makes all the difference in the world. With fat consumption at ten percent of calories, the rate of breast cancer (both the incidence of disease and the death rate) was reduced by a remarkable 90 percent. Thus lowering fat consumption sufficiently virtually eliminates the risk of this disease.

I then wanted to see if the nurse’s study researchers had examined levels below 30 percent. The abstract on the MedLine screen did not comment on this, so I needed to get the full text of the article. For this I had to rely on 19th-century technology and obtain an actual paper document (the journal itself). The full journal article made it clear that the researchers had not examined levels of fat consumption below 30 percent of calories and acknowledged that lower levels just might make a difference.

People who read the articles on this new study walked away with the impression that fat is not a factor in breast cancer. Yet the opposite is the case – reducing fat consumption to ten percent of calories (as opposed to 30 percent) eliminates most of the risk of this disease. Sometimes people feel too well informed by the media. A trip to the library could save your life.

The fat of the land

WeII, back to the Harvard Medical School Library and MedLine. Examining other cancers, such as colon cancer and prostate cancer, revealed the same phenomenon. There’s not much benefit to reducing fat to 30 percent of calories, but if you reduce fat consumption to ten percent of calories, virtually all risk of contracting these diseases, let alone dying from them, is eliminated.

For heart disease, we see the same relationship, although there is some reduction in risk at 30 percent of calories. At the 30 percent calories from fat level, which is the recommendation of the American Heart Association, heart disease is reduced by 20-30 percent. That’s noticeable, but if you look up the rate of heart disease in the rural counties of China, where fat consumption is ten percent of calories, you find a rate of heart disease that is one hundred times lower than in the United States.

With that intriguing result, I became curious about demographic data. I found one online service that provides demographic data on health from the World Health Organization, but again I was directed to the paper stacks to obtain the food balance sheets from the Food Agricultural Organization for nutridonal information. That data are even more remarkable.

Rather than just a simple comparison of several groups, I discovered that if one plots a graph with one point for every society in the world (for which such data are available) on a graph with one axis for fat consumption and one axis for rate of disease, the graph makes an almost perfectly straight line. In other words, the more overall fat consumption in a society’s diet, the higher the rate of disease.

You can obtain almost the same chart for breast cancer, colon cancer, prostate cancer, heart disease, and other conditions. There’s the rural regions of Cbina and other Asian societies at the lower left-hand corner of the graph with ten percent calories from fat and almost no incidence of these “degenerative” diseases. And there’s the good old United States at the upper right-hand corner of the graph with high fat consumption and an epidemic of disease. I plotted the graphs using the same computers available at the library, and they were so dramatic, I included them in my book.

Trends not to follow

One gains an enormous appreciation from this process of the efficiency of doing research when information is online. You can follow up ideas and leads in minutes; the same task can take hours when you search through large stacks of paper documents. A few strokes of the keyboard (too bad these library terminals are not voice activated!), and I can change the year, or the disease, or the country.

Fortunately, the disease data were available online, so I examined the changes over the last ten years. This revealed a disturbing trend. Heart disease was indeed coming down. It is down about 30 percent compared to ten years ago. That is a result of some reduction in fat consumption and a growing consciousness of the role of lipid (blood) cholesterol (but it does remain our biggest killer). The rate of breast cancer and other cancers, however, is soaring. What is going on here?

In examining what other factors have changed in the past ten years, I thought that the increasing rate of cancer might be due to earlier diagnosis, but presumably that should reflect itself in reduced death rates, and death rates have been going up as well. I therefore wanted to see if there were any undesirable nutrients for which consumption was increasing.

Unfortunately, the nutritional consumption data are not online, so the data had to be examined the old-fashioned way. To get the same information from the paper documents on nutritional trends that I had obtained online for the disease trends took at least 50 times as long. But the data were needed, so I waded through and discovered that the consumption of polyunsaturated fat has been dramatically increasing and that this consumption trend looks exactly like the cancer disease trend.

Is the cure the cause?

Could it be that the trend toward so-called “heart healthy” oils was causing a dramatic increase in the rate of cancer? Going back to the general scientific and medical studies, it became clear that the consumption of polyunsaturated fat is as much a risk factor for cancer as saturated fat is for heart disease. Carcinogens and other environmental factors can cause the initiation of cancer (which has occurred for the majority of the female population by the young adult years), but the crucial promotion phase in which cancer becomes clinical and the critical progression phase in which cancer metastasizes (and thereby becomes deadly) are fueled by consumption of polyunsaturated fat. It is the primary fertilizer or growth promoter advancing the development of a tumor during these stages.

So why does the American Heart Association and the American Cancer Society recommend a limit of 30 percent calories from fat when the benefit from reducing fat consumption to this level is marginal at best? In leading some of the discussions on what public recommendations should reflect, it seems to be justified by something called the “theory of maximum compliance.”

In other words, it is felt that there will be more overall benefit from getting a large group to comply with a supposedly easy recommendation that will result in a small benefit than to get a somewhat smaller group to obtain a much larger benefit, even if that larger benefit is virtual elimination of risk. Of course, no one communicates to the public (except for an occasional book author like myself) that these public recommendations are far from optimal but rather reflect a considerable compromise.

Have your cake and eat it, too

Another argument that is put forth is that going down to ten percent calories from fat is “too difficult.” To test that out, I found that a research library was the wrong place to be. Going down the street from my house to my local library, however, turned out to be a suitable setting to research this issue.

In the food and recipes section, I found a wealth of recipes and eating ideas that comply with the ten percent guideline. Having tried out hundreds of these recipes for many years now and having developed many of my own (which I share with my readers). I can attest to the fact that you can eat a remarkably diverse diet of everything from meat to dessert and readily stay at around ten percent calories from fat.

People will find encouraging another fact that is available back at the research library: if you eat food that is very low in fat, you can actually eat an enormous quantity of food and stlll lose weight. I personally lost over 40 pounds on a low-fat diet while eating well over 2000 calories per day – that is a large amount of food.

So who says you can’t have fun and be healthy at the same time? Thanks to my local library (and a little high-tech research assistance), I discovered how it can be done.

Reprinted with permission from Library Journal, January, 1993. Copyright © 1993, Reed Elsevier, USA

Other Futurecast columns