Trusting Data, Part 2: Evidence on Both Sides

Last Friday, I posted a qualified criticism of empiricism. Primarily, I intended to denounce the political, personal, and social biases inherent to much of today's academic research. These biases include pharmaceutical companies funding a majority of medical research and then giving kick backs to general physicians (statins), liberal bias in research on intelligence, and political bias in economics for ideologues like Paul Krugman, 2008 Nobel Prize winner in Economics. I promoted a "commonsense" approach to making life decisions mostly based in evolutionary wisdom and anecdotal experience.

A few commenters took issue with my argument and blasted my presumed lack of respect for statistical rigor. I agreed somewhat and thus specified my approach should be applied almost exclusively to those fields not inundated with potential bias, like physics, astronomy, and engineering. I didn't illustrate my point further, namely that a layperson could have a very hard time meandering through all the specious data available. While those of us in the reactionary sphere, the iconoclasts who've justifiably rejected much of mainstream "wisdom", understand the basics, a person who hasn't yet taken the red pill could easily find research buttressing the mainstream. In almost every post on a controversial issue, I try to embed links to a few research papers supporting my arguments, but even I must admit that someone on the other side could do just the same.

So let's say someone has just discovered the paleo diet. He simply doesn't understand how the opposing paradigm, the one that lauds vegetarian, high-carb, low-fat as optimal, could promote such wrong information, how this lie could disseminate so thoroughly throughout society. A curious sort, he goes online to consider the data both ways and he starts with what he knows - the vegetarian diet is the best. Here's a sampling of what he'll find:

The Mayo Clinic promotes vegetarian diet as healthy alternative, including phytate-heavy beans, six servings of grains per day, incomplete-protein nuts, and spaghetti.

Website containing a large amount of research promoting health of vegetarian diet:

A significant body of population-based research documents the health benefits of a vegetarian diet. For example, a paper published in 1999 summarized the results of a study associating diet with chronic disease in a group of nearly 35,000 Seventh day Adventists living in California. The members of the group who followed a vegetarian diet (defined as eating no red meat, poultry, or fish)had lower incidences of many diseases, including obesity, hypertension, diabetes, arthritis, colon cancer, prostate cancer, and ischemic heart disease than the nonvegetarians (Fraser, 1999). Also in 1999, Key, et al., analyzed the combined results from five studies involving a total of more than 76,000 people that compared the incidence of disease among vegetarians (defined as eating no red meat, poultry or fish) to that of nonvegetarians with similar lifestyles. Mortality from ischemic heart disease was 24% lower in vegetarians than nonvegetarians (Key, et al).

Study showing a Low-Fat Vegan Diet Improves Glycemic Control and Cardiovascular Risk Factors in a Randomized Clinical Trial in Individuals With Type 2 Diabetes

Famous China Study by scholars at Cornell and Oxford claiming eating animal products leads to significant increase in modern disease (see destruction of this study here)

American Dietetic Association on vegetarian diet:

The results of an evidence-based review showed that a vegetarian diet is associated with a lower risk of death from ischemic heart disease. Vegetarians also appear to have lower low-density lipoprotein cholesterol levels, lower blood pressure, and lower rates of hypertension and type 2 diabetes than nonvegetarians. Features of a vegetarian diet that may reduce risk of chronic disease include lower intakes of saturated fat and cholesterol and higher intakes of fruits, vegetables, whole grains, nuts, soy products, fiber, phytochemicals.

Of course, one could also find plenty of research supporting the paleo diet and debunking nutritional myths like the connection between total cholesterol and heart disease as well as almost startingly quotes like this from the former president of the American College of Cardiology:

The low-fat, high-carbohydrate diet may well have played an unintended role in the current epidemics of obesity, lipid abnormalities, type 2 diabetes, and metabolic syndromes.

Now, I can almost guarantee each vegetarian study has an obvious flaw, usually that heavy meat-eaters tend to also eat lots of bad foods (think of the people at McDonald's). Essentially, none of these studies adequately compare the two major nutritional paradigms: low-fat/high-carb vs. high-fat/high-protein. And even I'll admit a vegetarian diet low in processed foods is a step-up from the typical American diet that includes meat but also a ton of horrible extras.

Of course, one could look at the data itself, but one can't expect a layperson to wade through the endless amount of data out there. And even if he had the time to do so, how many are capable of understanding the relevant arguments, dissecting faulty scholarship, or having the fortitude to reject the mainstream's maxims based on his own skepticism? In the end, it's exceedingly difficult because so much data exists on both sides. And that's why I consider commonsense, informed by paleo/evolutionary wisdom (generally defined) and anecdotal observation, so powerful.