Sunday, February 24, 2008

What’s Wrong with Health Studies?

Like many vegetarians, I enjoy reading news reports about new health studies that come out. Most of these studies, of course, directly or indirectly support the vegetarian cause by showing that some aspect of eating plants is good for us, and/or some aspect of eating animal products isn't so good for us. I've come to count on these scientific findings as continuing confirmation of my good judgment to be a vegetarian.

But what about those other studies—the ones that you occasionally see that find eating meat or dairy products may be healthy? What do we do about those?

I used to think that any studies that found meat-eating to be healthy were aberrations that should be relegated to mere footnote status in the great tide of favorable literature. They must be written by people with loose scientific morals and a vested interest somewhere, I thought. Lately though, I've read some pretty troubling findings from some pretty credible sources. I've had to delve a little deeper and rethink my position.

What I've learned is that the whole system of scientific inquiry into the relationship between diet and health is hugely unfavorable to vegetarianism. More importantly, though, I've learned not to take any study results as gospel. Here's why:

  • People really do have loose scientific morals and vested interests. The meat, dairy, and egg industries have the money and organization to fund a lot of research. Cantaloupe and lettuce growers don't. No matter how honest scientists try to be, their results are going to favor the people paying the bill. If you need an example, just look at our court system. "Experts" can be hired to support any position.
  • Cause and effect relationships are difficult to establish. Whether you're doing research in the lab or conducting an epidemiological study, it's incredibly difficult to isolate the effect of just one dietary factor and study its health consequences. There are simply too many environmental and genetic variables, and too many interactions between variables to get in the way. Results from even the best-designed studies can vary widely.
  • Animals may be lousy models for people. When you take animals in the lab and pretend they are people, you've introduced a potential for serious error right off the bat. Just last week I happened to be reading the literature on a new drug. Testing found it caused cancer in mice, but not in hamsters. Okay… so what can we do with that useful information? It should be obvious that any results based on animal testing are suspect, and should be subject to additional scrutiny.
  • Test subjects lie. Maybe they don't intent don't intend to, but it's human nature. In any study that requires people to report on their own diets there's going to be an impetus to report things just a little healthier than they really were. (I mean, who's really going to own up to eating a whole box of Ding Dongs for dinner?) The errors are magnified when, as in many studies, people have to report on their dietary patterns in the past. Of course, this exaggeration is always going to make the results of "healthy" food choices seem a little less healthy.
  • There aren't enough "vegetarians." It's especially difficult to do studies of vegetarians, because there often aren't enough of us to be a good sample. More importantly, we all know that people lie about their vegetarianism. (A recent Time Magazine poll found that 37% of the people calling themselves "vegetarian" had eaten red meat in the last 24 hours!) Finally, there are huge variations in vegetarian diets, and studies may not compensate for that. They may lump the raw foods folks in with the people who live on Ding Dongs. Again, the result of this is going to make "vegetarians" seem a lot more like everyone else, and a "vegetarian" diet seem a lot less healthy.
  • There aren't any vegans. Vegans are so few in number and so misunderstood, that it is going to be a rare study indeed that will properly analyze and/or give meaningful results that apply to this group. Because of the problems cited above, results for "vegetarians" may be totally inapplicable.
  • Since practically everyone eats meat, dietary studies often suffer from the "lesser of the evils" syndrome. For example, we've all seen lots of studies that cite the benefits of eating fishes. But if the "control" group is made up of folks who eat other kinds of flesh foods, it's hard to draw absolute conclusions. Sure, fish-eaters may be healthier than hamburger-eaters, but if fish-eaters were compared to vegetarians, the results might look entirely different.

    Another recent example was a study that found that, among overweight and obese adults, those who ate massive amounts of dairy products were less likely to develop insulin resistance syndrome. The conclusion, widely reported in the media, was that 5 servings a day of dairy products protects us. But how did the people who ate less dairy make up those calories? Almost certainly by eating more meat. (I doubt that any vegetarians were in their sample group.) Could the conclusion of the study just as easily have been "avoid meat at all costs—even if you have to eat lots of dairy?"

  • Since practically everyone eats meat, results of dietary studies may not apply to vegetarians. Results from dietary studies may only apply to people with diets similar to the test group, and that rarely includes vegetarians. For example, I recently read a report raving about the benefits of fish oil (omega-3 fatty acids). The primary benefit noted was fish oil's ability to thin blood and thus prevent heart attacks and strokes. But thinner blood is only of benefit to people already sick with cardiovascular disease (i.e., most people on the Standard American Diet). For many vegetarians and vegans, and others with healthy cardiovascular systems, this blood thinning could be a health detriment rather than a benefit.
  • There is scientific evidence out there to support everything. A librarian friend of mine once told me "doctors like to write." It's true. You can find a study out there to support almost anything you can imagine. If it isn't there yet, it's on the way. No individual study, nor the credence that someone gives to it, should be taken as definitive.
  • People give more authority to later studies than earlier ones. While the conclusions expressed in the scientific literature may vary wildly, people tend to believe the latest news they hear. They mistakenly believe that new scientific studies override the results of those that came before. This is unfortunate indeed when poorly designed or aberrant studies conflict with what has been established in numerous previous studies.
  • People like good news. How much press has "red wine is good for you" received? How many millions of books has Dr. Atkins sold by telling people it's healthy to eat lobsters and butter? People want to believe, and they tend to cite, studies that favor their own positions. We vegetarians are guilty of this all the time. On the other hand, since the vast majority of people in our society are addicted to animal products, any study at all hinting that meat and dairy products might somehow be good for us will receive considerably more notoriety than a study finding the opposite.
  • The press exaggerates. (And that's an understatement!) Combine this with the fact that people like good news, and the potential for the public to be misled in favor of meat and dairy products is enormous. For example, a report on the fish oil study mentioned above, written by a registered dietician, starts off proclaiming the "good news," and ends with the pronouncement that "Including rich fish as a regular eating habit is one of the most positive steps anyone can take to protect against the development of heart disease. Best of all, it is good nutrition at its tastiest." Objective medical reporting? Hardly.
  • Doctors still believe in what they grew up hearing and doing. Old ways die hard for most of us, and that's true for doctors and medical researchers as well. Anything that goes against what their mothers told them as children is going to be looked upon with additional skepticism. That doesn't bode well for vegetarianism. Just consider how Dr. Spock was almost drummed out of the medical community when he had the audacity to suggest that cow's milk isn't good for growing children. Consider how many doctors are happy to heed medical evidence and recommend eating more vegetables and less meat (Mom would have liked this). Consider how few are willing to heed that same medical evidence and recommend vegetarianism (Mom definitely wouldn't approve).


    Given all of the problems inherent in medical studies of diet and disease, it's amazing that we can learn anything at all. Yet, the fact that some results (saturated fat is bad; fiber is good, etc.) keep coming up over and over again, means that we can probably trust them and rely upon them. The best advice may be to use our common sense (if it sounds too simple or too good, ignore it) and take a big picture/long term approach. (I'm starting to sound like my stockbroker!)

    Considering how the odds are stacked against vegetarianism, it's equally amazing that we get any good news at all from the medical community. And yet, there they are, week after week—new scientific studies lending support to the vegetarian way of eating. No, they may not be the only things we read, but that's good enough for me. They are more than enough to make me feel confident that my decision to be a vegetarian was a good one.

1 comment:

Anonymous said...

I bet your stockbroker also says to diversify!