The Centers for Disease Control seems to have taken up the nutrition density scoring gauntlet to rate high-value fruits and vegetables for their “powerhouse” value. A research paper in this month’s Preventing Chronic Disease journal derives a nutrient density formula that’s not a million miles away from the ANDI scoring scheme Whole Foods was touting a couple of summers ago. The author presents a table of 41 plain, raw and unadorned fruits and vegetables that made the cut by delivering more than 10 percent of your recommended daily value of a combo of 17 major nutrients for 100 grams of raw weight and/or (this part wasn’t quite as clear) 100 kcal worth of food.
The fact that the CDC is now publishing this kind of study lends nutrient density scoring more legitimacy than perhaps it really deserves.
On the plus side:
- The author, Jennifer Noia of William Paterson University in New Jersey, is an actual trained nutrition researcher with a Ph.D. in the field.
- She’s not making a pitch or selling special dietary supplements. Her stated goal is to help the CDC develop practical guidelines for public health reduction of cancers and cardiovascular disease by rating vegetables and fruits for their general nutrition-worthiness.
- She doesn’t bias her formula in favor or disfavor of her favorite name-dropping superfoods or taboos, as Joel Fuhrman and the admirers who started the ANDI scoring empire did.
- Avocado doesn’t score big; it’s not even included (too caloric for what it delivers).
- Noia does not include trendy components with questionable or untested nutritional value, things like selenium, antioxidants (unspecified groups of) and phytochemicals (unspecified groups of) among the 17 well-tested nutrients she counts in for the composite formula.
So far, so good.
But the specific method she derives is still kind of muddled, and the logic behind the nutrient density comparisons is too.
First, are we going for 100 grams or 100 kcal (also known outside the lab as “100 calories”) as the standard amount of each food for comparison? The article switches back and forth without clarifying and the formula does not normalize to one or the other as a uniform standard.
Second, how does that amount, whichever one is in use, compare with a likely normal serving of the specific fruit or vegetable? Arugula’s way up near the top for nutrient density–but if you ate 100 grams of it, or worse yet 100 calories’ worth, at a sitting, you’d be trying to eat an entire plateful or maybe several platefuls of it. Very bitter. Most people include a small handful, maybe a quarter cup per serving, in a mixed salad for interest, or (as I do) on a sandwich. With mustard or vinaigrette and some other veggies.
Same for watercress. And both are expensive per serving compared with romaine, bok choy, Chinese cabbage, turnip greens…parsley??? Who’s going to eat 100 grams OR 100 calories’ worth of parsley? Scallions? Maybe if you grill them, but then again, 100 grams? 100 calories?
The list is also a little arbitrary and incomplete in terms of what’s included. Green beans make no appearance on the list, for good or ill. Maybe it’s because they don’t deliver a ton of vitamins per se, even though they have some fiber and potassium and are low in calories. Mostly, though, they don’t happen to fit into one of the four broad categories (cruciferous, leafy greens, citrus, and yellow/orange) included in the selected list. That’s not a nutrition criterion, it’s a plant classification criterion, even though it is based on some generalizations that those four categories are the most nutrient-dense of the common vegetables and fruits. But at least the author acknowledges that limitation in her study and isn’t saying green beans have no worth in one’s diet.
Of what is included in the list, the rankings by nutrient density score are a bit counterintuitive. Broccoli and cauliflower are, perhaps disappointingly, rated a lot lower on the scale than watercress–in the 20-25 range, not the 100 (maximum score). So are carrots and tomatoes.
One of the reasons for this is, as Dr. Noia writes, “As some foods are excellent sources of a particular nutrient but contain few other nutrients, percent DVs were capped at 100 so that any one nutrient would not contribute unduly to the total score.”
So the scoring formula is purposely handicapped toward well-rounded performers. Is that realistic or meaningful? Some of the foods that scored lower within each of the four broad categories may provide large amounts of one critical nutrient–vitamin C or A, or fiber, or potassium, or iron–but perhaps not loads of B vitamins or calcium.
Well–that’s the way it is. Very few single vegetables–and almost no fruits–deliver so many different nutrients at high density in an edible portion. It’s why we eat a variety of vegetables and fruits and don’t just gravitate toward one impossible or hard-to-eat-exclusively jack-of-all-trades food.
And that’s the major flaw in this approach to defining nutrient-worthiness through a catch-all formula. The author of this study, the ANDI Score folks, Dr. Fuhrman and countless others really are looking for a magic bean. They want “AND”, not “OR”, a vegetable or fruit that delivers everything by itself. Even if they think they’re making it simpler for the average consumer to get better nutrition advice, they’re working from a false premise.
Still, at least Dr. Noia isn’t overreaching as much as the commercial popularizers of “superfoods” schemes. She admits the list is limited, and that the formula she’s derived from previously validated major studies is still preliminary. The correlation between her nutrient density score and established nutrients with some cardiovascular disease and/or cancer-prevention effect is predictably high–well, there’s a lot of overlap to begin with, so what it really tells you isn’t a great deal.
But there is some value in looking at the list of what scored at least a 10 out of 100. What can you really learn from this list?
First, “Bitter is Better.” Sort of, anyway. You wouldn’t want to make a whole dish of arugula or watercress, but you might want to throw a good handful or so into your salad or sandwich or pasta.
Notably, though, “More Expensive and Trendy isn’t Necessarily More Nutritious.” Kale isn’t as high up on the list as ordinary unglamorous spinach or turnip greens, or even darker leafy lettuces–isn’t that interesting?
Third, “Green is Good.” Darker greens are closer to the top of the list, and even between broccoli and cauliflower there’s a slight pan-nutrient decrease, though it’s not meaningful enough to start shunning cauliflower. Which I happen to like nearly as much as broccoli, and sometimes in combination with it. And although brussels sprouts are marginally more nutritious than cauliflower, they’re also more of a pain to peel and trim, and there’s a lot more waste.
The real Green Effect here has to do with calories. You notice that none of the citrus or berry fruits are up in the top 10 in the list. Apples and bananas don’t make the list at all, and melon is right out. So are summer stone fruits. Even apricots, which have a fair amount of vitamin A. That’s because the greens on the list are very low-calorie, with almost no carbohydrate or sugar. And because the nutrient density scoring formula accounts for nutrients per 100 calories (at least sometimes, when there are countable calories involved), fruits are naturally going to score lower.
From the perspective of diabetes prevention and weight control, this is a reasonable way of looking at how we get critical vitamins and minerals. The common phrase “fruits and vegetables” leads people to assume that fruits should be thought of first when you shop, and vegetables are kind of an afterthought. But it’s obvious from this kind of scoring that fruits should be considered dessert rather than the major source of vitamins and minerals.
However–given that citrus fruits, berries and stone fruits deliver large doses of vitamins C and A respectively, plus potassium and vitamin E and fiber on occasion, they shouldn’t be shunned for not “doing it all” and doing it carb-free. We need some carbohydrates, and if you eat an orange, a half a grapefruit, or even a nectarine at breakfast instead of a piece of coffeecake, so much the better. Just notice that you’re also eating some carbohydrate in the form of fruit and don’t eat three at a time. Or dump sugar and butter on it and still think it’s righteous because it’s nominally fruit.
I think we can all handle that. Even without calling anything a superfood.