Nutrient density chart

In our scores, chicken liver (100) and beef liver (90), followed by cooked spinach, oysters and beet greens. Liver leads because a small serving covers a day's B12 and vitamin A, and much of the day's choline and folate.

Cooked spinach (87), beet greens (81), broccoli (70), Swiss chard (66) and asparagus (64). Leafy greens do well per calorie because 100 calories of them is a big bowl. Spinach and chard are high in oxalate, which binds much of their calcium and iron (which we've taken into account).

How much of the nutrients you need a food gives you for the calories it brings. We went a step further and counted what the body absorbs from each food, across 22 vitamins, minerals and amino acids.

Heme iron is found only in meat, fish and poultry and is absorbed far better: roughly 15 to 35%, against 2 to 20% for non-heme iron from plants, eggs and dairy (Hurrell and Egli 2010). Phytates in grains and legumes and oxalates in spinach lower non-heme absorption; vitamin C raises it. In our scores heme iron from red meat counts in full and non-heme iron from spinach a quarter.

On our predicted satiety score, braised meats and legumes: braised pork ribs (70) and beef chuck (65) for their protein, then split peas (64) and lentils, white beans and mung beans (63) for protein and fiber together.

Per normal serving: cooked artichoke (1,370 mg), Swiss chard (840 mg), blackberries (825 mg), red cabbage (676 mg) and coffee (641 mg). Beans, lentils and chestnuts read even higher, but they were measured dry and scaled to the cooked food, so their values are an upper bound. Adzuki beans top the chart from a single sample.

We kept polyphenols out of the nutrient density score. There is no daily target, and absorption varies a lot between compounds and people. The values are total polyphenols by the Folin assay, from Phenol-Explorer (Neveu 2010; Rothwell 2013), which also reacts a little with vitamin C and sugars.

We took 100 calories of each food, or a big real portion if that was less (250 g for most vegetables, 100 g for raw leafy greens, 10 g of nori). For that amount we worked out how much of each of 22 nutrients the body absorbs, as a share of an adult woman's daily need, capped at 100% so no single nutrient could carry a food. We averaged the 22 shares with equal weight and set the top food to 100.

We built absorption in: heme iron from red meat counts in full; non-heme iron from grains and legumes about a third, from spinach a quarter, with sprouted and fermented foods in between. We counted plant omega-3 at 5% and plant B12 as zero. Where USDA had no value, we filled the gap from a related USDA entry or left that nutrient out of the food's average.

We calculated it from what a normal serving does in the body. We took protein (30 g is a meal's worth), soluble fiber (3 g), insoluble fiber (7 g) and the water held in the food (250 g) each as a share of a meal's worth, added fat (20 g) at a quarter weight, and averaged them. Fiber counts up to one meal's worth, so beans rank on their protein and fiber together.

The best-known one, Holt's satiety index, comes from 38 foods, eaten by a dozen people at equal 240-calorie portions, who rated how full they felt for two hours. That is a small sample and a subjective self-report on a seven-point scale. Real servings differ a lot from 240 calories, and two hours ends before the post-meal dip that drives the next bout of hunger.

We used Holt's foods as a sense check. Where both had a value we compared the two orders and adjusted our formula until they broadly agreed (rank agreement 0.72, where 1 is the same order).

Sulfur-rich: brassicas, the onion family and eggs. High histamine: rated 2 to 3 on the SIGHI list. High oxalate: about 10 mg or more per serving, from the Harvard oxalate table. High FODMAP: at the serving shown, following Monash University. Dairy and gluten as marked.

Live bacteria, polyphenols, K2 and the minerals in broth have no score. We used US average values (USDA SR Legacy); soil, breed and season change them. Servings are typical portions.

USDA FoodData Central, SR Legacy (2018). Holt SHA et al. Eur J Clin Nutr 1995;49:675–690. Rolls BJ et al. Am J Clin Nutr 1999;70:448–455. Marlett JA. J Am Diet Assoc 1992;92:175–186. Atkinson FS et al. Diabetes Care 2008;31:2281–2283; Am J Clin Nutr 2021;114:1625–1632. WHO/FAO/UNU 2007. Meléndez-Hevia E et al. J Biosci 2009;34:853–872. Hurrell R, Egli I. Am J Clin Nutr 2010;91:1461S–1467S.