Diets
A1 beta-casein worsens gastrointestinal symptoms
In plain terms: Does regular (A1) milk upset your stomach more than A2 milk?
Leans support Diets 💰 Industry COI noted🔬 Includes disconfirming
Part of: 🧪 a1 beta-casein
RefutedContestedStrong support
consensus score 0.24
Yes — but much of the supporting evidence comes from A2-milk marketers, so weigh that conflict of interest.
📅 Last reviewed: 2026-07-15 ⓘ
Evidence ladder
How far up the ladder this claim has climbed. A high consensus on a low rung means "consistent so far," not "proven in people."
Top evidence so far: Human trials (RCT / n-of-1)
MechanismIn-vitroAnimalObservationalRCTMeta-analysis
How the studies fall
8 support 2 contradict 4 tested null 5 mixed · 19 sources, 8 independent groups
What the evidence shows
Whether A1 (vs A2) beta-casein worsens GI symptoms is CONTESTED and confounded: the positive trials are funded by the a2 Milk Company, while the best independent/objective-endpoint trials are null or find symptoms track LACTOSE, not casein type. A real distinct 'A1 intolerance' in lactose-tolerant people is not established.
The evidence (26)
| Source | Grade | Stance | Quality | Finding |
|---|---|---|---|---|
| Yu W, Wang W, Sheng X 2025 · Journal of pediatric gastroenterology and nutrition | RCT | supports | low | RCT in 180 toddlers (A1-protein-free vs conventional formula, 90 days): ARI/diarrhea relative risk did NOT differ significantly between groups, but ARI duration was shorter (3 vs 5 days, p=0.012) and bloating/gassiness/regurgitation were le |
| Li J, Yang T, Sheng X 2025 · Food science & nutrition | RCT | supports | moderate | RCT in 280 Chinese infants (n=140/arm): A1-beta-casein-free formula (A1PF) had significantly lower IGSQ total/stooling/spitting-vomiting/crying/fussiness scores and fewer crying episodes vs conventional (A1+A2) formula at weeks 2 and 4 (p<0 |
| Borş A, Borş SI, Floriștean VC 2024 · Veterinary sciences | observational | mixed | low | Narrative review stating A2 milk 'may have potential health benefits' but explicitly notes 'few clinical studies have been conducted in humans, and the results have been unsatisfactory,' urging caution interpreting the literature. |
| Cieślińska A, Fiedorowicz E, Rozmus D, Sienkiewicz-Szłapka E, Jarmołowska B, Kamiński S 2022 · International journal of molecular sciences | observational | supports | low | Review update: 'newest clinical studies on humans showed a negative effect of variant A1 on serum glutathione level, digestive well-being, cognitive performance score in children, and mood score in women,' while calling for more well-design |
| Ramakrishnan M, Zhou X, Dydak U, Savaiano DA 2023 · Nutrients | RCT | tested-null | low | MRI crossover: A1+A2 milk emptied faster from stomach and raised abdominal pain vs A2-only in lactose maldigesters. |
| Liu B, Qiao W, Zhang M, Liu Y, Zhao J, Chen L 2022 · Frontiers in Nutrition | animal | supports | low | A2-only milk improved villus/crypt morphology and gut immune/SCFA profile vs A1A2 milk in mice. |
| Greenway FL, Hsia DS, Rebello CJ 2026 · Nutr J | RCT | contradicts | low | Double-blind randomised crossover pilot (2 wk per milk, 750 mL/d), 50 completers with self-reported milk intolerance; a2 Milk Company-sponsored. GI symptom severity A1PF vs conventional milk: 'odds ratios ranged from 0.971 to 1.4, with all 95% confidence intervals crossing 1.0; p -values ranged from 0.288 to 0.869' (abdominal pain OR 0.971, 0.68 to 1.38); SmartPill transit, stool frequency/consistency also no significant difference. Null on the directional claim -> contradicts. Quality low: 'no formal power calculations', pilot, retrospectively registered (NCT06763185). The within-A1PF-arm U-gal correlation (r = -0.296 with pain) is exploratory and not a between-milk contrast. |
| Jeong H, Park YS, Yoon SS 2024 · Food science and biotechnology | observational | mixed | low | Narrative review: cites animal studies showing A1 beta-casein increases digestive transit time and myeloperoxidase activity, and states BCM-7 'can lead to inflammation and discomfort in sensitive individuals'; also repeats unproven links to |
| Jianqin S, Leiming X, Lu X, Yelland GW, Ni J, Clarke AJ 2016 · Nutr J | RCT | supports | moderate | A2-only milk vs A1+A2 reduced PD3 GI discomfort, inflammation markers and worsened cognition in self-reported lactose-intolerant subjects. ⚠️ correction-on-file (Crossref) - kept, corrigendum not retraction |
| Ali AH, Hachem M, Najjar Z, Liu SQ, Bamigbade GB, Ayyash M 2026 · NPJ science of food | observational | supports | low | Narrative review of βCM-7 (the A1-derived opioid peptide): notes potential advantages on immune/GI health but also 'concerns persist regarding its adverse effects' — no quantitative synthesis given in abstract. |
| Sheng XY, Mi W, Yuan QB, Liu BY, Carnielli V, Ning YB, Einerhand AWC 2024 · Frontiers in nutrition | RCT | mixed | low | RCT (n=120 infants + 60 breastfed comparator, 4 months): A2 beta-casein formula (with added CPP/palmitate) showed softer stool, more frequent stooling, less abdominal distention/burp/flatus, less constipation, and fewer adverse events vs co |
| Meng Y, Zhou Y, Li H, Chen Y, Dominik G, Dong J, Tang Y, Saavedra JM, Liu J 2023 · Nutrients | RCT | mixed | low | A2-only growing-up milk improved digestive comfort/GI tolerance vs conventional milk in toddlers (China RCT). |
| Kamiński S, Cieślińska A 2025 · International journal of molecular sciences | observational | tested-null | low | Analytical study quantifying A1 and BCM-7 in raw/processed milk (ELISA); A1 variant more abundant than A2 in samples, but BCM-7 levels 'did not vary significantly with CSN2 genotype or processing method' — no GI symptom data collected. |
| Robinson LA, Cavanah AM, Lennon S, Mattingly ML, Anglin DA, Boersma MD, Roberts MD, Frugé AD 2025 · Nutrients | RCT | tested-null | moderate | 5-week double-blind crossover RCT (n=48) comparing lactose-free conventional milk (A1-containing), A2 milk, and lactose-free A2 milk; overall symptoms minimal, but A2 milk had LOWER bloating/flatulence than lactose-free A2 on some days, and |
| Choi Y, Kim N, Song CH, Kim S, Lee DH 2024 · Journal of cancer prevention | RCT | mixed | moderate | Double-blind crossover RCT (n=40, GI-discomfort-prone Koreans): A2 milk INCREASED bloating (p=0.041) and loose stools (p=0.026) vs A1/A2 milk on GSRS, but caused LESS abdominal pain (p=0.05), fecal urgency (p<0.001), and borborygmus (p=0.00 |
| Song CH, Kim N, Choi Y, Kim S, Kim KS, Park MH, Lee SH, Lee DH 2025 · PLOS ONE, 2025;20(5):e0323016 | RCT | supports | low | Secondary microbiome sub-analysis of the SAME RCT cohort as s38957588 (n=35 of 50 enrolled provided stool samples): 'This study analyzed microbiome data as a secondary outcome of a previous study.' A2-only milk shifted gut microbiota beta diversity vs baseline (PCoA, p=0.04) and vs A1/A2 milk post-consumption (p=0.031), with LEfSe enrichment of Bifidobacterium longum/Blautia wexlerae; alpha diversity unchanged. The GI-SYMPTOM outcome this claim asserts ('A2 milk caused less abdominal pain (P=0.050), fecal urgency (P<0.001) and borborygmus (P=0.007)') is not this paper's own measurement — it is recapped from the parent report [27], which already votes on this claim as PMID 38957588 (stance: mixed). |
| Ramakrishnan M, Eaton TK, Sermet OM, Savaiano DA 2020 · Nutrients | RCT | mixed | low | Single-meal crossover: A2-only milk gave lower abdominal pain scores than conventional A1+A2 milk in lactose-intolerant subjects. |
| Barnett MP, McNabb WC, Roy NC, Woodford KB, Clarke AJ 2014 · Int J Food Sci Nutr | animal | supports | moderate | A1 diet raised GI transit time, colonic myeloperoxidase and jejunal DPP-4 vs A2 in rats; opioid-dependent and -independent effects. |
| Fernández-Rico S, Mondragón ADC, López-Santamarina A, Cardelle-Cobas A, Regal P, Lamas A, Ibarra IS, Cepeda A, Miranda JM 2022 · Foods (Basel, Switzerland) | observational | supports | low | Review noting A2 milk releases much smaller amounts of BCM-7 'linked to harmful effects on human health' but frames this as an industry/technology update without independent quantitative synthesis of GI symptom trial data. |
| Brooke-Taylor S, Dwyer K, Woodford K, Kost N 2017 · Adv Nutr | observational | supports | moderate | Systematic review of 39 studies: A1 delays intestinal transit and triggers inflammation (rodents) and looser stool/discomfort (humans) vs A2. |
| Ul Haq MR, Kapila R, Sharma R, Saliganti V, Kapila S 2014 · Eur J Nutr | animal | supports | moderate | A1-containing milk feeding induced Th2-mediated gut inflammatory response in mice vs A2A2. |
| Gao X, Ju Z, Wang X, Wei X, Gao Y, Yang C, Shi Y, Huang N, Liu W, Jiang Q, Wang J, Zhang Y, Xiao Y, Huang J 2025 · Food science of animal resources | animal | tested-null | low | Mouse study (n=32, 4 groups gavaged with A2, A1/A2, A1 milk, or saline x5wk): A2 group showed increased goblet cell number and occludin expression (barrier markers) in colon vs other groups; gut microbiota/metabolite differences also seen b |
| He M, Sun J, Jiang ZQ, Yang YX 2017 · Nutr J | RCT | supports | low | n=600 RCT: A1/A2 milk -> higher GI symptom scores vs A2-only (a2 Milk Company funded) |
| Escobar-Charry MA, Conde-Martínez N, Peterson DG, Tello E, Quintanilla-Carvajal MX 2026 · Journal of Dairy Science | in-vitro | mixed | moderate | In vitro INFOGEST digestion model of Colombian commercial milks: BCM-7 release from A1/A2 casein varies by geographic origin/phase (gastric release 5-50 ng/mL, intestinal release 30-1050 ng/mL) — a compositional/kinetic study, no symptom or |
| Sheng X, Li Z, Ni J, Yelland G 2019 · Journal of pediatric gastroenterology and nutrition | RCT | supports | moderate | A2-only milk reduced GI symptoms of milk intolerance vs conventional milk in Chinese preschoolers, with cognitive gains. |
| Mannila E, Hokkanen L, Ahonen E, Turpeinen AM, Kalliomäki M, Kortesniemi M, Linderborg KM 2025 · (crossover) | RCT | contradicts | moderate | Genotype-stratified RCT: symptoms tracked LACTOSE; A1/A2 difference vanished once lactose controlled |
Disagree, or know a study we missed?
We grade by evidence, not opinions. The way to weigh in is to point us to a study we haven't cited (check the evidence table above first), or to flag a problem with one we have. Every submission is reviewed; if it holds up, the grade updates and shows in Science Changes Its Mind.
Opens a short form. You'll sign in with Google so submissions are tied to a real account — we don't display your identity, and we only accept a link we can verify (PubMed, DOI, ClinicalTrials.gov).
Educational only, not medical advice. Grades and scores reflect published evidence weighted by study design and quality; see the methodology.