optimal foods for different goals

A number of attempts have been made to rank foods based on their nutrient density or some other measure.

Useful parameters that can be used to optimal foods for different goals include:

  • nutrient density / calorie,
  • nutrient density / cost,
  • nutrient density / weight,
  • fibre / calorie,
  • fibre / weight,
  • calorie / weight,
  • cost / calorie, and
  • percentage insulinogenic calories.

This article details a new system that combines these parameters to identify optimal foods for different goals such as:

  • weight loss,
  • diabetes and nutritional ketosis,
  • therapeutic ketosis, and
  • athletes and the metabolically healthy.

My hope is that all this number crunching will help take the some of the guess work and ambiguity out of nutrition.

If we agree that we should focus on nutrient dense foods that don’t overload our pancreas’s ability produce adequate insulin, then we can move closer to agreeing which foods are optimal for an individual’s individual needs.

If you want to skip the detail, the end result of is a number of simple lists of optimal foods for different goals that you can access via the links below. If you want more detail then read on.

goal blog cheat sheet detailed list
therapeutic ketosis visit download download
diabetes and nutritional ketosis visit download download
fat loss visit download download
athletes and metabolically healthy visit download download

Firstly let’s take a look at a number of approaches that have previously been used to rank and prioritise foods.

low carbohydrate diets

As popularized by Dr Robert Atkins, limiting carbohydrates is a simple way to prioritise foods to reduce insulin demand.

By restricting carbohydrates intake, a range of foods are excluded, particularly those that are highly processed and contain added sugars.

While a low carb approach will reduce the insulin load of our food, no specific consideration is given to nutrient density or food quality.

image001

Aggregate Nutrient Density Index (ANDI)

In contrast to Akins’ approach, Joel Fuhrman’s Aggregate Nutrient Density Index (ANDI)[1] ranks foods based on micronutrients per calorie.[2]

I think there is an element of genius to Fuhrman’s nutrient density ranking system.  However when you look in the detail you find it is based on a select range of vitamins and minerals without any consideration of beneficial amino acids or fatty acids.

Fuhrman’s nutritarian approach has come under criticism for excluding a number of essential nutrients and placing extra emphasis on more fringe measures such as “oxygen radical absorbance capacity”.

image002

To determine the ANDI scores, an equal-calorie serving of each food was evaluated. The following nutrients were included in the evaluation: fiber, calcium, iron, magnesium, phosphorus, potassium, zinc, copper, manganese, selenium, vitamin A, beta carotene, alpha carotene, lycopene, lutein and zeaxanthin, vitamin E, vitamin C, thiamin, riboflavin, niacin, pantothenic acid, vitamin B6, folate, vitamin B12, choline, vitamin K, phytosterols, glucosinolates, angiogenesis inhibitors, organosulfides, aromatase inhibitors, resistant starch, resveratrol plus ORAC score.

While claiming to be “evidence driven”, without the inclusion of amino acids or fatty acids Fuhrman’s “nutritarian” approach ends up being heavily biased towards plant based foods.[3]   

Another issue with Furhman’s ANDI is that it can be skewed by a single nutrient present in very high quantities. For example, kale ranks at the top of Furhman’s list primarily due to its massive amount of Vitamin K.  Unfortunately, a mega dose of Vitamin K, which is a fat soluble vitamin, may have limited use by itself.  Rather than finding foods that are high in one nutrient it would be ideal to identify foods that were high in a broad range of nutrients.

Ranking foods in terms of nutrient density per calorie also tends to prioritise leafy veggies, which is great if you are trying to lose weight but not ideal if you’re an athlete trying to fuel up for an intense workout on kale and watercress.

While I think most people would benefit from consuming more green leafy vegetables, in the long term I think they will also benefit from foods with adequate protein protein and beneficial fatty acids.

In the short term someone who is obese has plenty of excess fatty acids and amino acids to spare so they will likely feel great as they are losing weight, however as their weight loss slows and they stop feasting off their own protein and fat the benefits of the a very low fat, very low protein approach may diminish.

NuVal

Professor Dr David Katz and an auspicious group of friends have developed the NuVal[4] food ranking system which uses the following sixteen positive ‘numerator nutrients’ to compare and rank common foods:

  • Fibre
  • Folate
  • Vitamin A
  • Vitamin C
  • Vitamin D
  • Vitamin E
  • Vitamin B12
  • Vitamin B6
  • Potassium
  • Calcium
  • Zinc
  • omega-3 fatty acids
  • total bioflavonoids
  • total carotenoids
  • Magnesium
  • Iron

image003

The sum of the ‘numerator nutrients’ is divided by the sum of the ‘denominator nutrients’ listed below to calculate a score of between one and one hundred:

  • saturated fat
  • trans fat
  • sodium
  • sugar
  • cholesterol

image004

The NuVal system also considers the following ‘additional entries’:

  • protein quality
  • fat quality
  • glycemic load
  • energy density

It’s interesting to note the foods to which it gives a score of 100 including:

  • non-fat skim milk,
  • sweet potato,
  • tomatoes,
  • beans,
  • bananas,
  • blueberries,
  • mango, and
  • wheat bran.

While the stated goal of the NuVal system is to combat diabetes, the food insulin index[5] shows that many of these foods will be problematic for a diabetic trying to maintain normal blood glucose levels.

Some of the more puzzling scores thrown up by the system include:

  • shrimp – 40
  • lobster – 60
  • coconut – 24
  • chicken – 57
  • beef – 46

Other concerns with the NuVal system include:

  • Because it biases heavily against saturated fat, some diabetic friendly foods like beef and coconut are further down the list.
  • The number of foods analysed is fairly limited.
  • Only sixteen vitamins and minerals are included in the analysis.
  • Dietary cholesterol is penalised by the NuVal system although dietary cholesterol does not necessarily lead to cholesterol in the blood or heart disease.
  • The NuVal algorithm has been calibrated to fit the views of the panel of experts, hence it is likely that it will simply reinforce previously held views.
  • Considering added sugar and the glycemic index are a good start, however I think using the food insulin index would be more useful as it is a better measure of the actual amount of glucose being metabolised.

Dave Asprey’s Bulletproof Diet Roadmap

Dave Asprey has developed the Bulletproof Diet Infographic[6] which is a simple ranking of foods to avoid, and preference based on both nutritional density and toxins.

image006

While I think Asprey’s ranking system is excellent, the downside is that it features only a select range of foods and does not explain why each of the foods has been given a particular ranking, although there is a good discussion of the toxins and various other considerations in his Bulletproof Diet Book.[7]

Asprey’s list also doesn’t differentiate between what would be most appropriate for someone with diabetes versus an athlete, or someone aiming for therapeutic ketosis or wanting to lose weight.

Soylent

Another noteworthy foray into the realm of optimising nutrition is Rob Reinhardt’s Soylent.[8]

Reinhardt set out to produce a manufactured food that ticked off all of the micronutrient Recommended Daily Intake (RDI) values, while reducing the cost and the hassle of food preparation.

While Reinhardt notes that his creation would be healthier than the ramen noodles that he was living on before creating Soylent[9], there are a number of downsides to this food replacement which is basically a protein shake on steroids.

Using manufactured foods leaves you exposed to not getting all of the non-essential micronutrients or even the beneficial nutrients that haven’t made it to the current list.  Eating real whole foods seems to be a safer option to ensure you are getting all the nutrients you need.

Mat Lalonde’s nutrient density

After reviewing the various options available and finding them lacking, Dr Mathieu Lalonde developed an excellent ranking of foods based on nutrient density per weight of food using the USDA food database.[11]

Lalonde also included a broader range of nutrients than Fuhrman or Katz by also considering beneficial amino acids and fatty acids.

This analysis identified organ meats as one of the more nutritious foods, followed by herbs and spices, nuts and seeds.

image008

In this video of his AHS2012 presentation Lalonde noted that people wanting to lose weight may wish to prioritise in terms of nutrient density per calorie, however he chose to analyse nutrient density in terms of weight as that might be more relevant for athletes (Lalonde is a CrossFit athlete as well as a biochemist). [12]

After watching this video and hearing about his quantitative approach to nutrient density I was left excited, yet a little unsatisfied, wondering what the ranking might look like in terms of nutrient density / calories.

fibre per calorie

One of the more interesting concepts in the area of nutrition recently is that what you eat could affect your gut bacteria.

Typical daily fibre intake is around 17g for those of us in western civilisation compared to the Recommended Daily Intake (RDI) of 25 to 30g per day.[17]

It is said that African hunter gatherer children obtain more than 150g of fibre per day from eating unprocessed foods in their natural state[18], and before the invention of fire and cooking our ancestors were eating more than 100g of fibre per day.[19]

Fibre is not digestible by the human gut and hence it does not provide energy or cause a rise in blood sugar or insulin.  Fibre in our food neutralises the insulinogenic effect of carbohydrate.[20]

If we rank for fibre per calorie we end up with a few spices such as cinnamon, curry powder, or cocoa at the top of the list along with veggies such as turnip, artichoke, sauerkraut, and cauliflower.

  1. cinnamon
  2. turnip greens
  3. artichoke
  4. curry powder
  5. sauerkraut
  6. cauliflower
  7. raspberries
  8. lettuce
  9. blackberries
  10. lemon peel

Again, this list is interesting, but not something you can live by.  Somehow we need to combine all these approaches to arrive at a more useful list that balances all of these considerations.

what are the “essential nutrients”?

So after reviewing these ranking systems I thought it would be interesting to design my own that would build on these previous approaches as well as considering the insulin response to food to make it more useful for people with diabetes.

The obvious starting point is to agree on the nutrients that should be included.  Listed below are the commonly accepted list of essential amino acids, vitamins and minerals.[21]

vitamins

  1. Choline
  2. Thiamine
  3. Riboflavin
  4. Niacin
  5. Pantothenic acid
  6. Vitamin A
  7. Vitamin B12
  8. Vitamin B6
  9. Vitamin C
  10. Vitamin D
  11. Vitamin E
  12. Vitamin K

minerals

  1. Calcium
  2. Copper
  3. Iron
  4. Magnesium
  5. Manganese
  6. Phosphorus
  7. Potassium
  8. Selenium
  9. Sodium
  10. Zinc

amino acids

  1. Cysteine
  2. Isoleucine
  3. Leucine
  4. Lysine
  5. Phenylalanine
  6. Threonine
  7. Tryptophan
  8. Tyrosine
  9. Valine
  10. Methionine
  11. Histidine

fatty acids

The list of essential and conditionally essential fatty acids is shorter than the other lists and is largely made up of omega 3 fats that the human body cannot manufacture in sufficient quantities. We need to go out of our way to incorporate these into our diet.

  1. Docosahexaenoic acid (DHA) (22:6 n-3)
  2. Eicosapentaenoic acid (EPA) (20:5 n-3)
  3. Docosapentaenoic acid (DPA) (22:5 n-3)
  4. Alpha-linolenic acid (18:3 n-3)

Given that a large part of my focus is to create a system that prioritises diabetic-friendly foods, I thought it would be good to give some more detailed consideration to other ‘good fats’, given that fat typically comprises more than half of the calories for someone following a reduced carbohydrate approach.  Listed below are the additional fatty acids that research shows to be beneficial.

  1. Arachidonic acid (20:4)
  2. Oleic acid (18:1)
  3. Lauric acid (12:0)
  4. Capric acid (10:0)
  5. Pentadecanoic acid (15:0)
  6. Margaric acid (17:0)

You can read more on the reason for inclusion of these additional good fats the Good Fats, Bad Fats article.

nutrient density score

Building on Joel Fuhrman and Matt Lalonde’s nutrient density approach, the nutrient score score is a relative score calculated by comparing the amount of a particular nutrient in each food against all of the foods.

For example, if a particular food has an average amount of Vitamin C compared to the 8,000 other foods in the database it will get a score of zero because it is zero standard deviations from the mean.  If it has a large amount of a certain nutrient then it will receive a high score.

If the amount that a particular nutrient is two standard deviations from the mean then it will get a score of two for that nutrient.  If however it is five standard deviations from the mean it gets a maximum score of three in order to avoid prioritising foods that have massive amounts of one single nutrient versus foods that have solid amounts of a range of essential nutrients.

image011

One example of where this limitation comes into play is kale, which has a massive amount of Vitamin K versus spinach which has a high amount of Vitamin K but also has a range of other nutrients.  Because of the upper limit on the score for a single nutrient the system gives a higher priority to spinach, which has a more well-rounded nutrient profile rather than simply being an overachiever in one or two nutrients.

image013

The nutrient score for a food is the sum of the individual nutrient scores across the forty three nutrients.  The higher the score the more nutritious that food is in comparison to the other foods in the database.

Weighting one nutrient as more important than another could be useful for an individual with a particular goal or health condition (e.g. DHA for someone battling brain cancer).  However I have chosen to keep ‘clean’ to avoid arguments about bias with equal weighting given to each nutrient.[22]  This system will simply highlight foods that have a wide range and a high quantities of nutrients.

weighted multi criteria analyses

Ranking foods by an individual parameter is interesting, however it doesn’t produce a balanced list of foods that you can live by.  Where things start to get interesting is when we combine the different parameters using a multi criteria analysis to suit different goals.

As an engineer I often use a multi criteria analysis (MCA) to analyse a lot of data.  A numerical MCA is a useful way to make sense of a large amount of data and shortlist from a wide range of options.

 

The available parameters from the USDA foods spreadsheet are:

  • nutrient density / calorie,
  • nutrient density / weight,
  • fibre / calorie,
  • fibre / weight,
  • calorie / weight, and
  • percentage insulinogenic calories.

The table below shows the weightings given to each criteria refined to create a shortlist of foods to suit different goals.

goal

ND / cal

ND / weight fibre / cal fibre / weight calories / weight

insulinogenic (%)

fat loss

40%

5% 5% 5% 25%

20%

athlete

5%

30% 10% 5% 5%

45%

diabetes & nutritional ketosis

5%

20% 10% 5% 10%

50%

therapeutic ketosis

5%

20% 5% 5% 0%

65%

  • Someone aiming for therapeutic ketosis will want to minimise their insulin load while maximising nutrition in the context of a very high fat diet.
  • Someone with diabetes or trying to achieve nutritional ketosis will also want to minimise their insulin load, however they should also look to maximise nutrient density and obtain adequate fibre.
  • Someone who has control of their blood glucose levels but is still trying to achieve fat loss will likely benefit from a diet with a reduced calorie density while still maximising fibre and nutrition.
  • An athlete’s primary priority will be to maximise nutrients without as much concern for calorie density or insulin load.

reality check

I have refined these weightings used in the MCA by reviewing the top 500 foods (of the 8000 foods in the USDA foods database) for each scenario.

goal

fibre (g) 

weight (g)  % protein % net carbs % insulinogenic

% fat

fat loss

45

1614 29% 13% 33%

31%

athlete

25

436 26% 12% 31%

56%

diabetes & nutritional ketosis

25

413 30% 4% 21%

58%

therapeutic ketosis

13

357 14% 3% 14%

80%

average all foods

26

899 26% 38% 52%

31%

It’s interesting to see that the net carbohydrates ends up being relatively low for all scenarios when we maximise nutrient density.  It appears that starchy carbs (e.g. grains and sugars) have a relatively low nutrient density compared to other available foods.

image015

The big differentiator across the approaches is calorie density.  If someone has stabilised their blood glucose and insulin levels then the next step in the journey may be to decrease calorie density to naturally manage food intake.  The fat loss approach is slightly more insulinogenic however practically it will be difficult to fit in all the food.

the results

While this process is somewhat convoluted the end result is a fairly simple list of foods that are ideal for different goals.  I have included a shortlist of the highest ranking foods on the blog here along with ‘cheat sheets’ that you can print and stick to your fridge or compile your food lists from.

It’s been great to see many people benefit from focusing these shortlists.  If you’re inquisitive and like to ‘peek under the hood’ I have also included links to a more detailed list that shows the basis of the rankings for each food.

goal blog cheat sheet detailed list
therapeutic ketosis visit download download
diabetes and nutritional ketosis visit download download
fat loss visit download download
athletes and metabolically healthy visit download download

references

[1] http://www.wholefoodsmarket.com/healthy-eating/andi-guide

[2] http://www.wholefoodsmarket.com/healthy-eating/andi-guide

[3] http://www.westonaprice.org/book-reviews/eat-to-live-by-joel-fuhrman/

[4] https://www.nuval.com/

[5] https://optimisingnutrition.wordpress.com/the-insulin-index/

[6] http://www.bulletproofexec.com/wp-content/uploads/2014/01/Bulletproof-Diet-Infographic-Vector.pdf

[7] http://www.amazon.com/The-Bulletproof-Diet-Reclaim-Upgrade/dp/162336518X

[8] https://www.soylent.com/

[9] http://www.newyorker.com/magazine/2014/05/12/the-end-of-food

[10] http://robrhinehart.com/?p=424

[11] http://ketopia.com/nutrient-density-sticking-to-the-essentials-mathieu-lalonde-ahs12/

[12] https://www.youtube.com/watch?v=HwbY12qZcF4

[13] http://solvingnutrition.com/engineering-the-cheapest-and-healthiest-diet-on-a-budget/

[14] http://blog.paleohacks.com/ultimate-guide-paleo-diet-budget/

[15] https://www.youtube.com/watch?v=VvfTV57iPUY

[16] http://perfecthealthdiet.com/

[17] https://www.nrv.gov.au/nutrients/dietary-fibre

[18] http://www.abc.net.au/catalyst/stories/4067184.htm

[19] http://www.gregdavis.ca/share/paleo-articles/academic/The%20Ancestral%20Human%20Diet%20by%20S.%20Boyd%20Eaton.pdf

[20] https://optimisingnutrition.wordpress.com/2015/03/30/what-about-fibre-net-carbs-or-total-carbs/

[21] http://ketopia.com/nutrient-density-sticking-to-the-essentials-mathieu-lalonde-ahs12/

[22] http://www.westonaprice.org/book-reviews/eat-to-live-by-joel-fuhrman/

24 thoughts on “optimal foods for different goals”

  1. Of COURSE Katz’s and Fuhrman’s systems are biased toward vegetables–both these gentlemen are VEGETARIANS. Both Fuhrmans, actually–yes, there’s more than one.

  2. I feel that you’re overcomplicating things. Why not just tell people to do the following:-?
    Base their diets on animal (~15% by weight) produce and vegetable (~85% by weight) produce and tweak it to suit themselves, allowing the occasional treat.
    That would minimise their consumption of over-refined, moreish, over-marketed food products.
    What say you?

    1. Hey Nigel. What you’re saying sounds like a pretty good rule of thumb. I was just one of those annoying kids that always wanted to know why and work out how it works. I figure if we can quantify nutrient density and insulin response we can then compare different foods and different foods and determine which ones are more appropriate for different goals. In the end all this demonstrates why a lower carb nutrient dense whole foods paleo style approach might be the optimal approach compared to the other stuff that we’re marketed on the basis of not omitting whole food groups (like sugar and processed grains). Cheers. Marty

      1. I used to be like that, too! I think that I’ve mellowed as I’ve gotten older (I’ll be 61 next March, so I’m doing reasonably well) 😀

        Cheers, Nige

  3. Hey there. Excited to find this. Thanks for the data and time you took to collate it all. I’m attempting a 50 day challenge combining your athletic and fat loss protocol, essentially testing the viability of insulin timing in a semi-vigorous setting, so finding this resource is great.

    At any rate, thanks again and take care.

Leave a Reply