by Marion Nestle
Jan 20 2010

The perils of interpreting food composition

Thanks to Sodium Girl (see comments under Feedback) for giving me a chance to talk about one of my favorite nutrition topics: how do we know what’s in food.  Her question:

I am on a very low sodium diet…I am beginning to have less and less faith in the nutritional labels – who is regulating them and what process/research do they use to define the amounts? And it is not just produced goods. I find it hard to know what information to trust when it comes to whole foods as well. I know that the USDA reports nutritional values which are the standard. Even with their documents though, a raw egg is 70 mg of sodium while a boiled egg is upwards of 120 mg of sodium – are they taking salted water into account?

This comment sent me straight to the USDA’s nutrient composition data base.  Despite the daunting home page, it’s a lot of fun to use once you get the hang of it.  Start with where it says Search the data base online.  Type in Egg (not Eggs) at Keyword.  Click on Dairy and Egg where it says Select Food Group, pick what you are looking for (I checked “egg, whole, raw;” on the next round, I checked “egg, whole, cooked, hard boiled”), click on Submit, and then decide how you want the data presented.  I chose one large egg.  Bingo.

USDA reports large eggs as 50 grams.  A 50-gram raw egg has 70 mg sodium and a cooked one has 62.  My interpretation: no significant difference.

Here’s the deal on food composition tables: you have to consider these numbers as ballpark figures, not as something engraved in stone.

  • Foods grown and raised in different places under different circumstances have different nutrient compositions, so the food you are eating is unlikely to be identical to the ones tested by USDA.
  • Nutrient amounts depend on weight; if your egg is a little bit bigger or smaller, the nutrient numbers change accordingly.

So you need to interpret food composition numbers leaving a lot of wiggle room.  That’s why I think reporting calories the nearest calorie is silly.  A 50-gram hard-cooked egg is 78 calories?  Plus or minus 10 maybe.

The USDA figures are the most authoritative available.  The office in charge of the nutrient composition data base is an unsung treasure of American government.  The scientists who work there are first rate, but they struggle daily with two problems: (1) not enough money to do their own testing, and (2) food companies know quite well what is in their products but they won’t give the USDA any information about nutrient composition beyond what is on the food label; they consider that information “proprietary” and don’t have to.

When it comes to sodium, which we eat in gram amounts per day, the difference between 70 and 62 mg is trivial.  I use the USDA figures as ballpark estimates and don’t pay any attention to small differences.

Sodium Girl: unprocessed foods like eggs are all relatively low in sodium so you don’t have to worry about it if that’s what you routinely eat.   Things start getting salty when you eat foods like cheese, pickles, and soy sauce, or anything commercially processed or prepared for you by others.  That’s why I’m for getting food companies and restaurants to cut down on salt so it will be easier for you to follow your doctor’s orders.

  • good topic !!!

  • Pingback: uberVU - social comments()

  • Mom of Two

    I’ve been asking this question of chef friends for months without a satisfying answer — maybe now is my chance!

    Does the caloric value of a food change when it’s cooked?

    I know that, in many cases, the weight/volume changes — 6 oz. or 1 cup of uncooked rice is nothing at all like 6 oz./1 cup of cooked rice. I know, too, that the sugar/starch content can change. [And I understand the caveat above about wiggle room — I don’t expect the numbers to be terribly accurate for raw or cooked food.]

    But if I were to weigh all my food before cooking it and use the ‘raw’ values in the USDA database, would my data be consistent once I cook it? [It’s much easier to weigh raw food than cooked food.]

    Any insight on this would be very welcome. Thanks.

  • Organic Grocer

    Mom of Two: In his latest book, “Catching Fire: How Cooking Made us Human,” Harvard Primatologist Richard Wrangham argues that cooking foods changes the available nutrient content and actually raises the available calories.

    Listen to an interview with him on NPR:

    Marion: Do you have thoughts on the validity of this claim?

  • Cathy Richards

    Always read nutrition labels “with a grain of salt” and a jaundiced eye. In my life, and my work with hospitals, schools, and vending machine operators, I frequently see nutrtion facts labels that I know are wrong.

    Eg. nuts with salt in the ingredient list, but the sodium is declared as 2 mg per 1/4 cup serving? A salted snack product whose nutrition label claims 23 mg of sodium, but whose on-line nutrition information says 1000 mg for the same portion size? A blueberry cheesecake that claims 0 g of sugar, and lists ingredients for an entirely different product?

    I always alert our food inspection agency when I see these discrepancies. They can’t monitor everything on their own.

    Marion has good advice. It’s hard to go wrong with unprocessed food that you prepare yourself.

  • Daniel

    For being in its 22nd release, the USDA ARS database is still quite limited. A policy initiative mandating that companies either lab verified nutrition information for all SKU’s or fund USDA lab work would reduce the burden on this currently underfunded lab and make the database more useful for the target demographic (which I would assume is all U.S. citizens, at least those that eat).

  • My journey is going over two years as a vegan and vegetarian 20+years. The last few years is when I decided to gather knowledge about what I’m really eating, how it gets onto my plate and what it does to the planet. The horrid life of 97% of the animals we keep captive made my decision to go vegan very very easy. For me, responsibility is the action that follows knowing. Your book and blog is one of the many books I’ve read over the past few years and suggest others read it as well. Thank you for keeping a blog with articles and updates.

  • annie

    i will continue to insist that there is a vast difference between what is called “salt” i.e. table salt, iodized salt, etc. and what salt was before we killed it with over-processing. in my opinion, the dirtier the salt, the healthier it is for me, including all those necessary trace minerals my body requires. i do not trust any “salt” product whose ingredients include any form of sugar…. the easier it pours, the more likely i am to avoid it.

  • “(2) food companies know quite well what is in their products but they won’t give the USDA any information about nutrient composition beyond what is on the food label; they consider that information “proprietary” and don’t have to.”

    That this is legal says so much.

  • Cathy Richards

    Lab verified nutrition labels would be more accurate, but making it mandatory would be beyond the financial capacity of small local food producers and so would give the multinationals even more of an edge over innovative local and perhaps healthier products = something none of us want to see.

    It’s indeed a difficult issue.

  • Pingback: January Links at 10,000 Monkeys and a Camera()

  • Marion – thank you so much. I just caught your answer to my question and I have become even more resolute to stick to whole foods that I make at home. I do like that USDA site, though, as a foundation of making my dietary decisions safely and in an educated manner (or at least as educated as I can).

    All thebest