Nutrient Density Cheat Sheet
Not sure how many of you guys would be interested in this, but here it is. I've been working on this Nutrient Density Cheat Sheet for the last few months. Basically, it organizes foods by relative nutrient density in the most objective way possible. In addition to stratifying foods by nutrient density, foods are also organized into a series of different scores. These scores range from 0 to 100. They're meant to capture a number of different dietary goals and philosophies. Some I agree with, some I don't. But I wanted to make sure that everyone could get something out of this.
I've created scores for everything from FODMAP-friendly foods to foods most appropriate for people with certain MTHFR genetic polymorphisms. There is even a column that organizes foods by their average market price per 100g. That way you can target the most nutrition for the least money. I even included check-boxes for those who want to use the spreadsheet as a grocery list. I even have a personal score that stratifies foods by my own biases, haha. Personally, I've never seen anything like this other than Mat Lalonde's Nutrient Density Spreadsheet. However, mine covers a wider range of nutrition data and stratifies foods in multiple different ways. It is also more condensed and easier to read.
I've put hundreds of hours into this, and it's totally free for anyone who wants to use it. There's also embedded documentation with instructions and a detailed overview of my methods. I encourage everyone to share it. I also encourage people to make suggestions. The more people suggest, the more useful the Cheat Sheet becomes. As long as there is sufficient data to be integrated, I will do my best to get it done. Anyway, enjoy!
PS. For anyone interested in the actual nutritional content of specific foods, the little plus sign above column GX will unhide all of the raw nutrition data, normalized nutrition data, and calculated data.
Comments
-
Wow! This is incredible. Great job @BRBWaffles
Out of curiosity, where did you source your nutrient information from?
Susan Macfarlane, MScA, RD
Registered Dietitian Nutritionist
cronometer.com
As always, any and all postings here are covered by our T&Cs:
https://forums.cronometer.com/discussion/27/governing-terms-and-disclaimer -
Thank you!
There is documentation linked on the spreadsheet that explains where I sourced my data. But, the actual essential nutrient data is tricky.
The USDA database is insufficient, as it lacked a number of essential nutrients, essential fatty acids, and essential amino acids. I wanted to use the NCC database because it includes all of it. But you need a license to get that data.
So, I exploited a loophole in Cronometer's data export feature. I included 100g servings of every food on the spreadsheet into a single day, and used Cronometer's "export serving data" to generate a spreadsheet that included all of the essential nutrient data. That spreadsheet is the foundation of all of the calculations I used.
-
Hi, Where is the actual spreadsheet? I would love to have a copy. I don't see an attachment or a link? But Maybe I missed it. I just joined and I am trying to work on a health issue I have (excessive artery clogging), I have elevated Lipoprotein (a) and I also have a Compound Heterozgous MTHFR Gene mutation. I am on special B Vitamins to help with that piece. And following the Linus Pauling Method for my heart right now. So this would be a useful document. Thank you so much for sharing your hard work.
-
It's the very first link at the very top of the page. But here is the link again.
https://docs.google.com/spreadsheets/d/1VnG4oMoHyjMZLvSVgoKYzNFnZUKXlAuvipIghhUk-QQ/edit?usp=sharing
-
Is there a filter option to filter out vegan food only ?
-
Unfortunately not. However, non-animal foods are colour-coded green so they're easy to spot.
-
-
Sorry, but your spreadsheet is completely wrong.
Satiety index is defined by scientists [1] according to how satiated you feel PER CALORIE consumed. So, if you are trying to lose weight, and not feel hungry, you should eat foods that have a high satiety.
In your little spreadsheet, it's completely backwards, with veggies having the lowest "satiety". I guess you didn't know what satiety index meant, or its utility in choosing foods. Useless.
[1] Holt SH et al, A satiety index of common foods. (The board doesn't let me post links, so you'll have to google the article)
-
The satiety index is the Fullness Factor calculation pulled from Nutrition Data. It's not my formula, I just used it because the results correlate nicely with reported satiety scores in feeding experiments.
-
I would like to know what sort of things people use the spreadsheet for.
Nothing
-
Tell that to the 8500 people who upvoted it on Reddit and the 2 platinum, 2 gold, and three silver I got for it. Also there are at least ten people from around the world viewing the spreadsheet as we speak.
Hate on me all you want because we disagree, but I created a resource that thousands of people have used, and there are two acclaimed Reddit threads showering it with praise and reverence.
-
Also, I did remove calories from the satiety equation and then added them back so that I could separate satiety by weight and satiety by calorie. Satiety by calorie is called the Guyenet score, and you're correct in that it does favour plant foods.
-
Just got home and checked. The Guyenet scores calculates satiety as you suggested-- satiety index divided by calories. However, when you organize the list, high-calorie plant foods are at the bottom of the list, fatty animal foods are intermediate, lean animal foods are even better, and non-starchy vegetables are at the top.
If you organize the list by satiety index divided by weight, high-calorie plant foods are still at the bottom, fruit is now intermediate, followed by animals foods, and high fiber plant foods are at the very top. So, satiety by weight isn't even "backwards" as you claim. It's just a different stratification. Both approaches puts plant foods at the top and bottom with animal foods somewhere in the middle.
Listen. If you're going to criticize someone's hard work, at least take the time to figure out if those criticisms are actually valid. So far everything you've said has been utterly baseless, and just cheap attempts to get under my skin. Grow up.
-
-
THANK YOU so much for creating this spreadsheet!
I think the "LINEAR NUTRIENT DENSITY CALCULATIONS" for Vitamin B12 and Iron are off. This appears to be the % RDA for these nutrients for a 100 g serving of the food. There are a LOT of foods at exactly 62.5% in the Vitamin B12 category, and none which are higher. This makes me think that 62.5% might have accidentally been made a maximum value at one time for some reason. For example, "Eggs, Dried" shows as 62.5% for B12 on the spreadsheet, but 123% on Cronometer.
Iron is also way off by a factor of about 9. Some of this might be that my daily requirement is different than the amount used as the default, but not to this extent. Dried Eggs on the spreadsheet provides 10% of iron, while in Cronometer, it shows 90%. It seems to be this way for all foods.
I'm getting slightly different numbers for B6, Copper and Zinc, but I'm not sure if that might be caused by me having different requirements than the default human.
-
Thanks for the support! I modified the nutrient yield of foods based on certain criteria. For example, heme iron has a maximum absorption cap of around 2.2mg per serving. That's why iron looks funny. B12 has a maximum absorption cap of 1.5mcg per serving, based on average intrinsic factor activity. I tried my best to factor in bioavailability, nutrient conversion capacity, and nutrient absorption capacity. I hope that helps.
-
My rationale is detailed in my blog post. The link is available on the sheet itself. Essentially I wanted the scores to represent nutrient yield, not content. Nobody should actually care what the absolute nutrient content of a food is, in my opinion. What we should care about is the sort of nutrient yield we can expect from our foods. Beans are a good example. They're loaded with iron, but you only absorb about 1% of that iron from beans.
-
In general, I strongly disagree with the idea that a "Nutrient Density" rating should be calculated based on the weight of the food vs the amount of food that a person is likely to eat. If this is the case, then all you would need to make a food more "nutrient dense" is to dehydrate it, and if you add water to the food, then you destroy its "nutrient density".
In addition, 100 grams of one food might be vastly different from 100 grams of a different food. I buy dried basil in 7 gram jars and that jar might last my family of 3 for 2 years. But 2 cups of cantaloupe (320 grams) might be a side dish at breakfast.
I personally prefer to look at nutrient density per calorie, or per 1600 calories, since I tend to eat about 1600 calories per day. I also like to look at nutrients per serving. I'd have to consume about 69 jars of dried basil to get 1600 calories. However, a 1/4 tsp serving of dried basil would provide so few nutrients, that I wouldn't even bother to log it into Cronometer.
-
I disagree. Without standardizing the unit of comparison, like 100g of weight, there is no way to compare foods fairly. Any judgement about serving size should be an ad hoc consideration.
The "Calories" column gives you what you want, nutrient density per calorie. It is also confounded by the hydration state of the food, as clam juice and coffee are ranked highest in that column. The hydration state of the food confounds every score. It's unavoidable.
-
A lot of that confounding goes away when you look at calories (or fat / protein / carbohydrate if you prefer) rather than weight. 2 dehydrated eggs (153 calories) weighing 26 grams should contain about the same amount of nutrients as 2 non-dehydrated eggs (155 calories) weighing 100 grams, or two eggs added to a liter of wate r(155 calories) weighing 1100 grams.
My impression is that the RDA for nutrients is already adjusted for the bio-availability of those nutrients in common foods. So, the human body might need 5 units of nutrient x, and a certain pile of food might contain 10 units of nutrient x, and the nutrition authorities recommend that you eat enough foods to get 20 units of nutrient x just to be on the safe side.
That's why you always see headlines like "98% of Americans aren't getting enough potassium". Yet, you see those same people just walking around. The emergency rooms aren't overcrowded with people checking in for potassium deficiency. When their serum levels are checked, most people have potassium levels in the healthy range. The nursing homes have people living to 105 years old, all of which were spent in a potassium deficiency.
Anyway, I very much appreciate all of the work and thought you've put into this spreadsheet, and I found it to be enormously helpful. I was just trying to point out some instances where the numbers on the sheet didn't match up with the numbers I get in Cronometer, but if those numbers were deliberately adjusted for some reason, I'm fine with that.
There are some foods which I would like to see added. Do you take requests? If so, I would definitely like to see frozen peas and canned peas added. I see snow peas and split peas, but these have very different nutritional profiles, particularly if you're going by weight.
Thank you again for taking the time to make this spreadsheet, and for making it available to the general public!
-
-
I can certainly see the value of having a separate column that adjusts for average serving size. I do agree that is useful information. I aso agree that it would be useful to have a separate column that adjusts by 100 kcal servings instead of weight. But, I feel like there would still be a large amount of confounding. For example, dried parsley would still be favoured to an unreasonable extent, and ad hoc value judgements would still be necessary. There isn't really a way to get around the confounding introduced by the hydration status of the food. Oatmeal is a good example on the sheet. Raw oats are near the top, whereas cooked oats are near the bottom. The only difference is the water added during cooking. Plus, it is simply standard to measure nutritional content of food by weight. The USDA and NCC both use weight in their databases. So, I work with what I've got.
The only DRI that actually accounts for bioavailability is calcium. This is because calcium bioavailability is very uniform between foods. Usually 30% for animal foods and 50% for plant foods. with some exceptions. So, the DRI is meant to capture total calcium intake required to ensure that calcium needs are met, not necessary how much actual calcium you need to take in. As a result, no adjustments are made the calcium in the spreadsheet.
The DRIs are set to capture the needs of 97% of people based on experimental studies. Potassium doesn't have an RDA. It has an AI, which is the replacement for the RDA when research is inadequate to establish an RDA.
I will absolutely take requests. I will add those foods as soon as I get a chance. I'm always open to ways to improve the spreadsheet! Thanks again for the support. It was not my intention to come across as adversarial. I just don't personally agree that serving sizes or fixed calories are necessarily the best way to judge the differences between foods, and they're all highly vulnerable to confounding and interperative problems.
-
I added a column for nutrition per serving size.
-
Wow, this is awesome @BRBWaffles. I had a lot of this info spread out via different web sites but you've consolidated everything I wanted to know into a single spreadsheet. Great job!
1) What does the color coding mean exactly? Pink, blue or green?
2) Is nutritional yeast a whole food? When I've looked at it in stores I see it's fortified via ingredients list containing chemical equivalents for vitamins and minerals.
-
Thank you! I tried to consolidate as much info as possible.
Red foods are animal foods, green foods are non-animal foods, blue foods are refined foods, and orange foods are processed foods. My rationale for everything is explained in my blog post here.
You can get unfortified nutritional yeast. Most are fortified. The entry in the spreadsheet is unfortified.
-
Blog = jackpot!! Fantastic info and work.
-
Thank you! I'm glad you found it useful!
-
-
Thank you for the time you have put into this information, I’m hoping to go through it and find some low carb high protein vegan foods.... that I don’t need to eat masses upon masses amounts of 😁
-
No problem! I may have some swift answers to your question. Any solid processed soy products would likely suit your needs. Tofu, tempeh, natto, and soy protein isolate are all relatively low-carb, vegan, and tremendously high protein.
-
Yeah thanks for that. I’m aware of that however those products aren’t cheap (especially if I’m trying to ensure where they are sourced from and organic!) and before buying pulses and grains were an easy cheap option to bulk out my meals and get plenty of fibre, protein and other nutrients. I may just end up with a lot less variety in my meals but if I can stick with it long enough to get the hang of it and get significant health benefits then I will do. Hopefully your cheat sheet will make logging and planning much easier.
-
The cheat sheet also includes a grocery list that estimates the cost your shopping trip (for my local prices it can guess the cost of the grocery trip to within three dollars). You can use that to manage the expense of your shopping. There is also a meal planner that allows you to see exactly what a given selection of foods contributes to the diet. Entries in the meal planner are adjusted per standard serving size. In order to use these features you need to make your own copy of the cheat sheet in Google Sheets. Got to File, make a copy, and that's it.
-
Vital wheat gluten can also be a fairly cheap and versatile source of protein that is low in carbs.
Best of luck!
Susan Macfarlane, MScA, RD
Registered Dietitian Nutritionist
cronometer.com
As always, any and all postings here are covered by our T&Cs:
https://forums.cronometer.com/discussion/27/governing-terms-and-disclaimer -
@BRBWaffles haha haha to “got to file, make a copy and that’s it” I appreciate that it is probably very simple to the majority of people but I am doing this all on a smart phone and even if I wasn’t, I have no clue how to make full use of a spreadsheet. I am not IT literate in the most basic of things that I really should be... but I can take time to write out the parts that make sense to me for a shopping list and search around for good prices. I think it’s going to take a good few weeks for me to get the hang of this but you have definitely helped.
@Susan_RD_101 thank you for the suggestion, unfortunately I have a wheat intolerance that seems to have gotten worse with age. Every now and then (usually hormone driven) I’d cave in and have bread or something with wheat in it and then soon regret it with a host of unpleasant side effects.
