--- topic: "Plant Milks" title: "Are Plant Milks a Healthy Swap?" order: 1 description: "How oat, almond, soy and rice milks compare with dairy on protein, calcium, iodine and added sugar, and who should be careful, especially young children." question: "Is oat, almond, or soy milk a healthy swap for dairy?" answer: "Fortified soy milk is likely a fair swap for adults, but most other plant milks are low in protein and iodine, so check the label." actions: - "Pick unsweetened, calcium-fortified versions, plus vitamin D and B12 if it is your main milk: in a US audit, under 5% of 641 products met proposed nutrient standards [4][5]." - "For protein, pick soy milk, which comes closest to dairy: a serving of almond or rice milk gives only about 2% or 8% of the protein in cow's milk [1][2]." - "If plant milk replaces dairy, cover iodine another way (iodized salt, seafood, or an iodine-fortified brand), because unfortified drinks contain almost none [3]." cautions: - "Do not use plant milk instead of breast milk or formula in a baby's first year, and ask a pediatrician before making it a toddler's main drink [2][6]." - "Avoid rice milk for infants and young children because of its inorganic arsenic content [7][8]." - "If you are pregnant or breastfeeding, unfortified plant milks can leave you short of iodine, which a baby's brain needs [3]." confidence: 3 confidenceNote: "Composition surveys agree with each other and a few small trials test calcium absorption, but no long trials compare health outcomes of swapping dairy for plant milk." evidenceSummary: rctCount: 1 observationalCount: 1 metaAnalysisCount: 0 totalParticipants: "5,034 children (largest cohort)" replicated: true newestStudyYear: 2021 references: - id: "1" pmid: "30937581" title: "Nutritional and health attributes of milk and milk imitations" authors: "Scholz-Ahrens KE, Ahrens F, Barth CA" year: 2020 journal: "Eur J Nutr" - id: "2" pmid: "32732790" title: "North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition Position Paper: Plant-based Milks" authors: "Merritt RJ, Fleet SE, Fifi A, et al." year: 2020 journal: "J Pediatr Gastroenterol Nutr" - id: "3" pmid: "28946925" title: "Iodine concentration of milk-alternative drinks available in the UK in comparison with cows' milk" authors: "Bath SC, Hill S, Infante HG, et al." year: 2017 journal: "Br J Nutr" - id: "4" pmid: "34746213" title: "Proposed Nutrient Standards for Plant-Based Beverages Intended as Milk Alternatives" authors: "Drewnowski A, Henry CJ, Dwyer JT" year: 2021 journal: "Front Nutr" - id: "5" pmid: "33806688" title: "International Analysis of the Nutritional Content and a Review of Health Benefits of Non-Dairy Plant-Based Beverages" authors: "Craig WJ, Fresán U" year: 2021 journal: "Nutrients" - id: "6" pmid: "28592600" title: "Association between noncow milk beverage consumption and childhood height" authors: "Morency ME, Birken CS, Lebovic G, et al." year: 2017 journal: "Am J Clin Nutr" - id: "7" pmid: "25536328" title: "Arsenic in rice: a cause for concern" authors: "Hojsak I, Braegger C, Bronsky J, et al." year: 2015 journal: "J Pediatr Gastroenterol Nutr" - id: "8" pmid: "18385862" title: "Inorganic arsenic levels in rice milk exceed EU and US drinking water standards" authors: "Meharg AA, Deacon C, Campbell RC, et al." year: 2008 journal: "J Environ Monit" - id: "9" pmid: "34444815" title: "An Overview on Nutritional Aspects of Plant-Based Beverages Used as Substitutes for Cow's Milk" authors: "Fructuoso I, Romão B, Han H, et al." year: 2021 journal: "Nutrients" - id: "10" pmid: "16177199" title: "Calcium bioavailability of calcium carbonate fortified soymilk is equivalent to cow's milk in young women" authors: "Zhao Y, Martin BR, Weaver CM" year: 2005 journal: "J Nutr" - id: "11" pmid: "10799379" title: "Bioavailability of the calcium in fortified soy imitation milk, with some observations on method" authors: "Heaney RP, Dowell MS, Rafferty K, et al." year: 2000 journal: "Am J Clin Nutr" - id: "12" pmid: "10749030" title: "Consumption of oat milk for 5 weeks lowers serum cholesterol and LDL cholesterol in free-living men with moderate hypercholesterolemia" authors: "Onning G, Wallmark A, Persson M, et al." year: 1999 journal: "Ann Nutr Metab" --- A plant milk can be a fine swap, but not a like-for-like one. Fortified soy milk comes closest to dairy on protein [1]. Almond and rice drinks supply only a few percent of the protein [2], and most unfortified plant milks contain almost no iodine [3]. Choose unsweetened, fortified versions, and do not rely on them as a main drink for babies and toddlers [2]. ## How it works Plant milks are plants blended with water, so what is in the carton depends on the ingredient, the processing and what the maker adds. Across 122 beverages, protein ranged from almost nothing to over 12 g per 100 mL and calcium from none to over 1,200 mg per 100 mL, and salt was the most common added ingredient [9]. Across 148 products from three continents, protein ran from 0 to 10 g a serving. On average they were low in sodium, saturated fat and calories, over half were fortified with calcium at or above dairy levels, and vitamin D and B12 fortification was low [5]. **Protein.** Soy approaches the protein content of cow's milk, but most plant drinks do not match its protein quality or its vitamins and minerals [1]. **Calcium.** In young women, calcium from calcium carbonate-fortified soy milk was absorbed as well as calcium from cow's milk, while tricalcium phosphate-fortified soy milk was absorbed less well [10]. An earlier study in men found only 75% as much absorbed from soy milk [11]. Shake the carton, since the added calcium is a suspension of particles [11]. **Iodine and B12.** In the UK, milk is the main source of iodine. Of 47 plant drinks there, only three were fortified with iodine, and the unfortified ones had about 2% of the iodine in cow's milk [3]. B12 and vitamin D fortification is also often low [5]. **Sugar and additives.** Dairy-free does not mean sugar-free. Nutrient standards proposed for these products cap added sugar and salt, and fewer than 5% of 641 US products met them [4]. **Oat and rice.** In a 5-week trial, an oat drink rich in beta-glucan fiber lowered LDL cholesterol by about 6% compared with rice milk, but that was a specially made drink, so check the label of what you buy [12]. Our [oats article](/oats/heart-health-blood-sugar-and-gut) covers beta-glucan. Rice milk can contain inorganic arsenic, and a 2008 survey found 80% of tested samples above the drinking-water limit [8]. **Children.** In 5,034 Canadian preschoolers, each daily cup of non-cow milk went with children being 0.4 cm shorter [6]. A pediatric society warns of poor growth, rickets and other deficiencies when plant drinks replace milk without a plan [2]. If your goal is dairy-free, see how plant drinks compare in our [dairy article](/dairy/is-dairy-healthy). For iodine, see [iodine](/iodine/thyroid-essential). ## The evidence in detail **Nutrient surveys.** Fructuoso and colleagues compiled 122 beverages from 22 plant sources (soy the most common, 28%): energy ran from 6 to 183 kcal per 100 mL, protein from 0.06 to 12.4 g and calcium from 0 to 1,253 mg, with salt the most frequent additive [9]. Craig and Fresán analyzed 148 labels from the US, Australia and Western Europe, including 49 nut and seed drinks, 38 grain, 36 legume, 10 coconut and 15 blends. Averages looked good for sodium, saturated fat and calories, but protein ranged from 0 to 10 g a serving and B12 and vitamin D fortification was low [5]. Drewnowski and colleagues proposed standards (at most 85 to 100 kcal per 100 g, at least 2.2 g of quality protein, limited added sugar, and fortification with calcium and vitamins A, D, B2 and B12) and applied them to 641 products in a US branded-food database. Fewer than 5% met them [4]. These are label audits, not trials, and products change. **Minerals.** Heaney's team gave 16 healthy men 300 mg calcium loads from cow's milk or soy milk with labeled calcium. Absorption from soy milk was 75% of that from cow's milk [11]. Zhao and colleagues gave 20 young women 250 mg calcium in a crossover trial: fractional absorption was 0.211 from calcium carbonate-fortified soy milk and 0.217 from cow's milk (no difference), but 0.181 from tricalcium phosphate-fortified soy milk, which was lower [10]. For iodine, Bath's team analyzed 47 UK drinks in 2015. The median of 44 unfortified drinks was 7.3 micrograms per kilogram, 1.7% of conventional winter cow's milk (438), while one brand's iodine-fortified soya, oat and rice drinks had about 270 to 290 [3]. **Oat.** Onning and colleagues randomized 66 men with moderately high cholesterol (52 analyzed) to 0.75 L a day of oat milk, which had 0.5 g of beta-glucan per 100 g, or rice milk for 5 weeks each in a double-blind crossover. Oat milk lowered total and LDL cholesterol by 6% each, with no change in HDL, while triglycerides rose after rice milk [12]. That is one small, old trial of a specific product. In our searches we found no trial of the blood-sugar response to commercial oat drinks. **Rice.** Meharg and colleagues found all 19 rice milks in a supermarket survey exceeded the EU limit for total arsenic in drinking water (10 micrograms per liter), up to threefold, and 80% of the 15 tested for inorganic arsenic exceeded 10 micrograms per liter [8]. The ESPGHAN Committee on Nutrition recommends avoiding rice drinks for infants and young children [7]. **Children.** In a cross-sectional study of 5,034 children aged 2 to 6, each daily cup of non-cow milk was associated with being 0.4 cm shorter (95% CI 0.2 to 0.8). A 3-year-old drinking 3 cups of plant milk was 1.5 cm shorter than one drinking 3 cups of cow's milk. Cow milk intake explained only part of this, and causation is unproven [6]. The NASPGHAN position paper lists failure to gain weight, short stature, kwashiorkor, rickets and scurvy among harms of misusing plant drinks, and says commercial formula may be the better dairy-free option after the first year when a child's diet is otherwise short of protein, calcium and vitamin D [2]. **Conflicts.** Averages in Craig's survey look favorable while Drewnowski's audit finds few products meet a nutrient standard; both are right because they ask different questions. Heaney's 75% and Zhao's equivalence differ by fortificant and method. Scholz-Ahrens and colleagues conclude soy approaches dairy on macronutrients, yet plant drinks are generally not equivalent in protein quality or micronutrients [1]. **Bottom line (confidence 3 of 5).** Plant milks can fit a varied adult diet when fortified and unsweetened. Soy is closest to dairy. Evidence on long-term health effects is thin, and young children are the clearest risk.