Thursday, October 14, 2010

Why science alone can't tell you what's best to eat

We live in an age of incredible scientific discovery about all kinds of things, including about the connection between diet and health.

Literally every day, new discoveries are announced that improve our understanding of human biology, and about the complex interaction between DNA, diet, behavior and environment to determine how long and how well we can live.

There is, however, an unfortunate disconnect between what scientists do and what the public thinks they do. As a result, media stories about new findings can confuse as much as they enlighten.

Here's what you need to keep in mind when considering scientific discoveries about diet and health:

1. Scientists are never finished understanding anything.  

The pace of new discoveries appears to accelerate with the application of new research technologies. Yet no final understanding of anything has ever been achieved. There is always more to learn.

For example, all the major dietary vitamins were discovered by scientists between 1913, with the discovery of Vitamin A, through 1941, with the discovery of Vitamin B9. These discoveries spawned an industry of vitamin pills, which were viewed by many consumers as an alternative to eating plenty of fruits and vegetables.

The assumption by the lay public was that vitamins and minerals were all fresh produce had to offer. Another assumption was that, say, the vitamin you get in a pill is as good as the vitamin you get in food.

In the last couple of decades, researchers have discovered the importance of phytochemicals, including polyphenol antioxidants. Even newer research has identified not only that vitamin pills can't substitute for food, they can even damage health.

Those substituting vitamin pills for produce all those years were robbing themselves of important nutrients because they had irrationally assumed that science was finished understanding how fruits and vegetables maintain health.

This is a lesson for the future. All the discoveries that will be made about food in the next hundred years already apply to the foods you're eating today.

Scientists can tell you what they know, but they can't tell you what they don't know.

2. The baseline is usually "normal," not "healthy." 

The average human subject in dietary research is unhealthy. Scientists find representative samples from the general public. The average person is overweight, severely deficient in Vitamin D, chronically under-hydrated, out of shape, stressed out and subsists on a highly inflammatory diet that's too high in fat, sugar, salt and man-made chemicals and deficient in most key nutrients.

Even in rigorous double-blind studies, both the experimental and control groups are likely populated by unhealthy people. So when some effect is attributed to a food or drink, it's not necessarily applicable to those eating plenty of fresh, whole organic foods and avoiding processed foods and addictive substances. 

For example, numerous research studies have linked the drinking of coffee to lowered risk of one disease or another. One study found that drinking four or more cups of coffee per day may lower the risk of gout. A casual health-conscious consumer of this news might conclude that drinking a lot of coffee is a good idea. And for someone with a horrible diet, metabolic syndrome and a refusal to embrace a healthy lifestyle, four daily trips to Starbucks might delay gout.

But for a non-coffee drinker who eats a great diet, the introduction of coffee degrades, rather than improves, health.

The scientific assumption of bad health has another curious effect. It leads researchers and journalists to express improved health from eating a healthy food as if that food was a kind of drug that improved a natural unhealthy condition. For example, a recent study found yet another link between the amount of vegetables people eat and their likelihood of getting cancer. Here's the lead sentence from the institution's press release:
"Investigators from the Slone Epidemiology Center at Boston University School of Medicine have reported that African American women who consume more vegetables are less likely to develop estrogen receptor-negative breast cancer than women with low vegetable intake."
See how that finding has been expressed? To paraphrase, they're saying that eating more vegetables lowers the risk of a specific breast cancer.

A more accurate but less scientific expression of that finding might be that not eating enough vegetables increases the risk of a specific breast cancer. It's more accurate because our DNA simply expects us to eat vegetables. (In fact, another study has determined that cancer may be an entirely modern and made-made disease, caused by environment, industrial diets and other recent phenomena.)

However, scientists must take the world as they find it. In the actual world, normal must be the baseline, not historically normal, or healthy normal, or biologically normal, but normal as an expression of average or typical. 

3. Science must isolate factors, but in the real world nothing is isolated.

Much diet-related health research involves identifying specific ingredients in food that affect specific aspects of health.

Let's say a food company develops a new preservative, and wants to test whether it is safe for human consumption. Researchers might do this by eliminating all other unhealthy substances from the diet of laboratory animal subjects so that any ill effects can be faithfully attributed to the preservative. But let's say the researchers find the preservative safe, and so does the government, so the preservative is approved and added to the company's products.

But when people buy and eat the product, they are doing so in conjunction with an incomprehensibly large number of other factors not tested in the lab. Scientists found the preservative safe in isolation, but what about when combined in the bloodstream with the chemicals from toxic pesticides on non-organic fruit, household cleaners, personal-care products, car exhaust inhaled during jogging and excessive alcohol?

In other words, scientists must test foods and food ingredients in isolation. But we don't eat them in isolation. It's likely that some diseases, such as some cancers, may often be caused entirely by the long-term combination of a large number of substances proved safe in isolation.

Also: Natural foods vary in their biochemical and nutritional makeup. But scientists need precision in testing. So often they'll test not with a whole food, but with an extract, or a concentration, or a powder, or even a single isolated substance from food.

When they announce their finding, they report the truth, which is that a specific food extract had a specific measurable effect. But that doesn't mean the extract is better to eat than a whole food. It means only that it's better for testing.

And finally, scientists try to nail down a specific effect. For example, the Boston University study mentioned above tested for the effect of vegetable consumption on one very specific form of breast cancer. A reasonable person can conclude that if eating more vegetables lowers the risk of one form of breast cancer, it probably also has a wide-ranging benefit to overall health. But scientists can't say that unless they tested for it.

4. Science is biased in favor of new products or policies.

Broadly speaking, scientific research about health tends to result ultimately in the development of a new product, such as a pill, or in a change in public policy. There are two reasons for this.

First, science doesn't happen unless it's funded. While some funding is made by grants or taxpayer dollars, a lot of funding comes from companies that make the investment with the intention of monetizing it through product development. In other words, projects that are seeking a new pill that can be sold for profit are much more likely to be funded.

Second, scientists are aware that human behavior is very difficult to change. The greatest number of people will benefit from research not by giving advice that most will ignore, but by making the consumption of a substance either very easy or mandatory.

By advocating a pill, rather than a wholesale change in diet or lifestyle, researchers can do the most good for the greatest number. In cases of major public health, lawmakers often mandate the introduction of substances, such as the addition of fluoride to water, or vitamin D to milk -- and they do so often on the recommendations of researchers.

That doesn't mean a new pill or new law is best for you, personally.

Scientific research is the best tool we have for understanding our world, including the part of our world we eat every day. But by itself, science can't tell you what's best to eat because of the limitations listed above.

The best answers come from combining scientific findings with reason and common sense -- which is what we do in the Spartan Diet. That, plus a little help from our friends in ancient Sparta.