So I heard back from both Alan Aragon and the author of the article (props to both for being responsive.) She was also kind enough to send me a copy of the study in question. It was from one month earlier than I looked… I was lazy, I guess.
Here’s a link to the abstract of the study, Greater Healthful Food Variety as Measured by the US Healthy Food Diversity Index Is Associated with Lower Odds of Metabolic Syndrome and its Components in US Adults, published March 1, 2014. it was first published electronically this past December.
I’m not going to go into a ton of detail unless asked, but after reading the study, I stand by my earlier speculative position. This wasn’t a study that carefully compared low variety vs. high variety of comparably healthy foods. Rather, they used a broad index - the relatively new US Healthy Food Diversity Index - and found that high scores on this index are associated with less metabolic syndrome for Non-Hispanic Whites. It didn’t tie as well with less metabolic syndrome for Non-Hispanic Blacks, but it does associate well with less obesity (which is one aspect of metabolic syndrome). And it didn’t tie at all with lower metabolic syndrome for Hispanics.
It’s a useful approach for doing large-group analysis, but the HFD index has some weaknesses. “Healthy Food Diversity” sound like’s measuring diversity, but ti’s really a combination of two things: score is a function of the healthiness of food and of the diversity of food (and the proportions). More variety raises your score, but choosing healthier foods also raises your score. If you need a “one number describes it all” index, it works, but it’s losing the detail… for example, for two people with the same score, one may have gotten there with a broader variety of less healthy foods, while the other got there with a smaller variety of healthier foods. Is one better?
(I’m side-stepping the question of whether the assumed Healthiness Factors are correct - every food needs a Healthiness Factor to work into the equation.)
Also, this study didn’t take into account salt intake or calorie intake, because neither is captured by the HFD index. If two people consume the exact same foods in the exact same proportions, they both have the same score - even if the second person consume three times as much of everything, and serves it all with 7000 mg of sodium.
I think this particular study was not trying to answer the question, “is a variety of foods better for you.” It was not built for that. The answer to that question is already assumed to be, “yes,” based on prior studies.
I think this study was built more to answer questions like, “does a higher HFD index suggest a healthier diet?” In essence, testing the new HFD. And the study suggests answers to questions like, “does allowing people’s natural inclination toward variety tend to lead them to an overall healthier diet?” This is an interesting question, partly because a lot of people have been advocating more limited dietary options (don’t eat this, don’t eat that), hoping to guide people to a healthier overall diet.
Most importantly, this study is not a good basis on which to downplay the value of a carefully balanced diet which is designed to meet all nutritional needs. That’s simply not what it was looking at.