Real Food (2006)
Nina Planck
343 pages, $23.95
Everyone has heard of the scientific method, but few people seem to understand what it is. As a former science teacher, I feel a sense of responsibility about this, so here’s the one-minute version. The scientific method works like this:
1. Make an observation.
2. Form a hypothesis that fits the observation.
3. Design and perform an experiment to test the hypothesis.
4. Based on the results of the experiment, either toss out the hypothesis or say that the experiment supports the hypothesis. If many experiments performed by many different investigators support the hypothesis, you reach a scientific consensus–as close to “fact” as you can get with science.
But that’s no fun, is it? Doing experiments is expensive and messy. Here’s a more popular version of the scientific method.
1. Make an observation.
2. Form a hypothesis that fits the observation.
3. There is no step three.
This version is awesome. It saves a bundle of cash and you get to spend your time writing books and appearing on talk shows instead of pipetting. It doesn’t work at all, but hey.
It’s the streamlined scientific method on display throughout Nina Planck’s _Real Food_, an appealing but thoroughly frustrating polemic. It’s not a cookbook: it’s a renegade (sort of) scientific analysis of what we should and shouldn’t eat. There are chapters on meats, fats, dairy, vegetables, and so on.
_Real Food_ is appealing because Planck likes to eat the same kinds of things I do. She’s in favor of lard, cheese, eggs, organic produce, and other tasty things. Basically she likes all the things you get at the farmers market, which is perhaps not surprising, since she heads a farmers market organization.
Planck is admirably skeptical about some nutritional research. Much of this mainstream research, the kind that suggested that low-fat diets protect against heart disease, is based on observational studies. You recruit a group of people, examine their behavior, and correlate their behavior with the medical outcome you’re studying. At the same time, you attempt to control for confounding factors like sex, race, income, and so on.
This approach is a great start. It’s step one of the scientific method. If you notice an interesting correlation (maybe people who eat lots of yogurt get the flu less often), that takes you to step two. If you then form conclusions based on your hypothesis, you are in a state of scientific sin.
When mainstream researchers make this mistake, which they do all the time, Planck pounces on them. She then promulgates fringe research that, while based on interesting observations, is on even less sure footing than the mainstream research.
For example, on page 93, Planck claims that breast, testicular, and prostate cancers are have risen substantially since 1950 because of synthetic hormones in cattle. Well, that’s certainly an intriguing hypothesis. Let’s take a look at the research she cites. Oh, wait, there’s no citation.
If hormones are the enemy, omega-3 fatty acids are the white knight. They cure depression, schizophrenia, obesity, heart disease, and diabetes. Here Planck does give a citation–to a book by a psychiatrist called _The Omega-3 Connection_.
Look, I’m not saying that cow hormones are good for you and omega-3s aren’t. At least in the case of omega-3s there are actual intervention studies which have found that, for men, increased consumption of fatty fish lowered death rates–although it’s impossible to say from that whether the improvement was due to omega-3s or something else.
This is a book about food, and Planck seems to think this makes her free to ignore any health effects not related to food. She spends a lot of time talking about the remarkable health of hunter-gatherer societies compared to industrial societies. Let’s accept that assertion (although Planck doesn’t offer a whole lot of evidence). Wouldn’t you expect traditional and industrial societies to differ in ways other than food? What if all of the effects she’s talking about are related to stress rather than diet? I certainly don’t know if this is the case. Neither does Planck.
Let’s go back to the issue of dietary fat for a minute.
A causal relationship between consumption of dietary fat and cardiovascular disease has been suspected since the 1950s. There have been hundreds of observational studies on the topic, but the first large intervention study was completed in 2006. The study cost $415 million, which is why studies like this don’t happen often. I know you heard about this in this news, but the finding was entirely negative: that is, there was no evidence that reducing dietary fat intake reduced the risk of cardiovascular disease.
The study was criticized on several grounds. Maybe the researchers didn’t look at saturated fat intake. (They did.) The results only apply to older women. (True, but based on what we know about lipid metabolism, they’ll probably generalize to older men.) Finally:
> But the scientists declined to call the $415 million venture a failure, pointing to signs of less breast cancer in women who cut out the most fat, and in less heart disease in women who ate low amounts of the worst kinds of fats.
That’s called cherry-picking, rummaging around in your data for *something* statistically significant. Bad idea–it means you’re looking for an answer that the study wasn’t designed to provide.
What does this have to do with _Real Food_? Well, the link between fat and heart disease is one of the best-established tenets of our nutritional orthodoxy. There is a ton of evidence suggesting a relationship between saturated fat, cholesterol, CHD, and stroke. It was smart to make dietary recommendations based on this very suggestive data.
It cost $415 million to determine that these recommendations were probably wrong. Maybe not. Maybe the subjects were too old to make a difference. Maybe the hypothesis was correct but the results weren’t statistically significant. You can’t make an all-encompassing conclusion on the basis of one study.
But this is the *best evidence we have*. It’s much better than the evidence in favor of fish oil or blueberries. It’s much better than the evidence against white sugar and trans fats (two more of Planck’s demons).
I wanted to like this _Real Food_ as much as I like lard, eggs, raw-milk cheese, and fresh marionberries. There is scant evidence that any of these things are bad for you, and if you enjoy them, you should eat them. If only Planck had stuck to that message.