Almond milk: almost everything in it was added

The strongest thing anyone can say about almond milk is that it has barely been tested. When a network meta-analysis pooled the randomised trials comparing plant drinks with cow milk, it found 14 trials in 543 people covering soy, rice and oat drinks, and it could not include almond at all, because trials of it do not exist. What has been tested is what the manufacturer added. A pilot in 18 people with advanced kidney disease raised blood vitamin D by about 3.4 nmol/l, on four ounces a day for three weeks. The drink they were given carried vitamin D3 in soy protein nanoparticles, which is a laboratory formulation rather than the fortification in a carton on a shelf. The bone claim Europe allows on the carton belongs to the added calcium and says nothing about almonds. Even that calcium is the difficult part. In a digestion model covering twenty-five plant foods, plant drinks landed in the lowest band, under 10 percent of their calcium released, which the authors put down to the tricalcium phosphate used to fortify them. How much was added is the whole question, and across the USDA Foundation samples it moves further than any finding here. Calcium runs from 7.5 to 289 mg per 100 g over 14 samples, vitamin B12 from 0 to 3.588 µg over 10, and vitamin E from 0 to 8.226 mg over 14. A zero at the bottom of those ranges is an unfortified drink, not a measurement error.

The numbers, per 100 g

How much is actually in it

• Plant drinks are built to taste like milk rather than to match it, and the nutrition follows the taste. — plant-based milk products, not people (n = 20 studies) Evidence D source
  The most data were available for soy and almond milks and the most common measure was greenhouse gas emissions; the review states that plant milks try to reproduce the sensory properties of dairy and often do not match its nutrient profile, and that there is not yet enough data for conclusive statements about the environmental impact of all types

What your body absorbs

• A pilot in 18 people with advanced kidney disease raised blood vitamin D by about 3.4 nmol/l in three weeks, on four ounces a day of almond milk carrying vitamin D3 in soy protein nanoparticles rather than the fortification a carton gets. — adults with advanced-stage chronic kidney disease (n = 18) Evidence B source
  Serum 25(OH)D3 rose in the fortified group from 33.0 to 36.4 nmol/L (p<0.05) while falling in the control group from 49.2 to 44.5 nmol/L (p<0.05), with a significant time by treatment interaction (p<0.01); serum calcium and parathyroid hormone did not differ between groups
• Plant drinks came last of twenty-five plant foods for how much of their calcium a digestion model could release, and the reason is the calcium salt they are fortified with. — food products in a simulated digestion, not people (n = 25 products) Evidence D source
  Gross calcium ranged from 7.48 to 959 mg per 100 g fresh weight, with about half the products matching or exceeding skimmed milk, while bioaccessibility ranged from about 0.1 to 50%; plant-based beverages fell in the lowest band under 10%, alongside spinach, tofu, dried figs and tahini, attributed to oxalate and phytate in some products and to the low solubility of the tricalcium phosphate used to fortify the beverages

What cooking and storage change

• The smoothness of a plant drink comes from forcing it through a valve at hundreds of atmospheres, and that step changes the proteins as well as the texture. — plant-based food matrices, not people Evidence D source
  The treatment inactivates microbes, improves texture and consistency, reduces lipid oxidation and alters protein secondary and tertiary structure, which raises emulsifying properties and protein hydrophobicity, while forming micro-sized particles

What it does to your health

• Swapping milk for an almond beverage made no difference to muscle building in older women, and walking more did. — older women, mean ages 63 to 69 years (n = 22 (7 whole milk, 7 skim milk, 8 almond beverage)) Evidence B source
  Daily muscle protein synthesis rates were not differentially affected by whole milk, skim milk or the almond beverage; collapsed across groups, synthesis rose from phase 1 to phase 2 by 0.133%/day (95% CI 0.035 to 0.231) and again from phase 2 to phase 3 by 0.156%/day (0.063 to 0.250)

Safety, limits and interactions

• A glass of almond milk carries about a fiftieth of the protein of a glass of cow milk, and paediatric societies treat that as a hazard for small children. — infants and young children fed plant-based drinks instead of cow milk Evidence E source
  An 8 oz serving of almond milk provides about 2% of the protein of the same serving of cow milk, and rice milk about 8%; documented adverse outcomes from misuse of such beverages include failure to gain weight, decreased stature, kwashiorkor, electrolyte disorders, kidney stones, iron deficiency anaemia, rickets and scurvy

What people believe that the data does not support

• The bone claim on a carton of almond milk belongs to the added calcium, not to the almonds. — — Evidence E source
  The wording calcium is needed for the maintenance of normal bones is Authorised; the register contains no entry under the name almond milk at all, and of the claims mentioning calcium 14 are authorised against 42 refused

What is still unknown

• The network meta-analysis that compared plant drinks against cow milk could not include almond at all, because trials of it do not exist. — adults in randomized trials (n = 14 randomized trials, 543 participants, covering soy, rice and oat drinks and cow milk) Evidence A source
  Few differences appeared between plant drinks or against cow milk; replacing 500 mL/day of cow milk with soy drink may lower LDL cholesterol by 0.47 mmol/L (low certainty) and systolic blood pressure by 8.23 mmHg (low certainty), while oat drink slightly lowered total cholesterol at very uncertain evidence

The bottom line

Drink it because you like it, and read the label for what was added. If you are drinking it in place of cow milk for protein, an 8 oz glass gives about 2 percent of the protein of the same glass of milk. That gap is why paediatric societies treat these drinks as a hazard for small children. The outcomes they document from using them instead of milk include failure to gain weight, shorter stature, kwashiorkor, electrolyte disorders, kidney stones, iron deficiency anaemia, rickets and scurvy. For a young child this is a question for a paediatrician rather than a shopping decision. In older women, whole milk, skim milk and an almond beverage all did the same thing to muscle building, which was nothing much, and what moved muscle protein synthesis in that trial was more activity. A review of the whole category puts it fairly: these drinks are built to reproduce the taste and feel of dairy, and they often do not match its nutrient profile. The smoothness itself is engineering, from forcing the liquid through a valve at high pressure, which also changes the shape of the proteins. Nobody has shown almond milk does anything to your health. Nobody has shown it does nothing, either, and the difference between those two sentences is the whole state of the evidence.

Sources

1. amk-r1-05 · scoping review
   PBMs attempt to replicate the organoleptic properties of dairy but often do not exactly match the nutrient profile of dairy.
   Adv Nutr, 2022 · checked 2026-09-22
2. amk-r1-02 · double-blind randomized controlled pilot trial
   The consumption of almond milk fortified with vitamin D3 dispersed in soy protein nanoparticles increased 25(OH)D3 in adults with advanced stages of CKD over 21-days.
   J Nephrol, 2025 · checked 2026-09-22
3. amk-r1-08 · in vitro digestion study
   The lowest bioaccessibility (<10 %) was found in spinach, plant-based beverages, tofu, dried figs and tahini and was attributed to: 1. the high content of oxalate and phytate in some of the products, and 2. the low solubility of tricalcium phosphate which was used for fortification in the plant-based beverages.
   Food Res Int, 2024 · checked 2026-09-22
4. amk-r1-06 · systematic review of processing technology
   This study focuses on recent advances in using HPH/UHPH on plant-based milk sources such as soybeans, almonds, hazelnuts, and peanuts.
   Compr Rev Food Sci Food Saf, 2024 · checked 2026-09-22
5. amk-r1-03 · randomized controlled trial
   Daily MPS rates were not differentially affected by the addition of WM, SM, or AB to a standardized diet.
   J Nutr, 2022 · checked 2026-09-22
6. amk-r1-04 · professional society position paper
   For almond or rice milk, an 8 oz serving provides only about 2% or 8%, respectively, of the protein equivalent found in a serving of CM.
   J Pediatr Gastroenterol Nutr, 2020 · checked 2026-09-22
7. amk-r1-07 · regulatory register entry
   POL-HC-6346 Calcium Calcium is needed for the maintenance of normal bones Authorised
   EU Register on nutrition and health claims, знімок 2026-09-21 · checked 2026-09-22
8. amk-r1-01 · systematic review and network meta-analysis of randomized trials
   Fourteen RCTs involving 543 participants were included, comparing soy-, rice-, oat drinks, and cow milk.
   Adv Nutr, 2026 · checked 2026-09-22