Small company-funded trial: A2 milk caused fewer stomach complaints

Overview

Jianqin S et al.. Nutr J 2016 · PMID 27039383 · DOI 10.1186/s12937-016-0147-z

People who felt milk did not agree with them had more symptoms with ordinary milk than with A2 milk.

Does the study support the specific claim? Partly · Methodological quality: low

Original publication

Plain-language summary

Scientific in-depth review

1. Full reference

Authors

Jianqin S, Leiming X, Lu X, Yelland GW, Ni J, Clarke AJ

Journal

Nutr J

Year

2016

Identifier

PMID 27039383 DOI 10.1186/s12937-016-0147-z · PMC4818854

Verification

verified – PubMed E-utilities efetch, 2026-10-25

Retracted

no

Corrections

PubMed lists: ErratumIn (Nutr J. 2016 Apr 29;15(1):45. doi: 10.1186/s12937-016-0164-y.)

Extent checked

full text

Conflicts of interest and funding

Funded by The a2 Milk Company Limited (including medical-writing support); test products supplied by A2 Infant Nutrition Limited; trial conducted by CRO SPRIM China. Co-author AJC is an employee of The a2 Milk Company and conceived and designed the study and selected the variables (not involved in conduct or analysis). Other authors declare no competing interests. Erratum published (Nutr J 2016;15:45).

2. Research question

Effects of A1+A2 milk vs A2-only milk on GI symptoms, inflammation and cognition in self-reported intolerant subjects.

3. Study design and population

Study design

Randomised controlled trial – Single-site, double-blind, randomised (computer-generated, gender-stratified, sealed envelopes), 2x2 crossover trial; 2-week dairy washout, 14-day intervention periods, 2-week washout between periods; exploratory, no sample-size calculation.

Population

Han Chinese adults aged 25-68 (mean 46.6 +/- 14.0 y; 21 men, 24 women) with irregular milk consumption and self-reported mild-to-moderate digestive discomfort after milk; 23/45 lactose intolerant by urinary galactose test. Recruited via community-hospital noticeboards; IBS, constipation and unmedicated IBD excluded.

Country

China (Xin Hua Hospital, Shanghai)

Period

October 2014 - December 2014

Sample size

45 randomised (sequence 1 n=22, sequence 2 n=23); 104 screened; smart-pill transit data n=40 (only 80 pills purchased), smart-pill inflammation n=40

Inclusion and exclusion criteria

Age 25-68; irregular milk consumption (FFQ); self-reported intolerance to commercial milk with mild-moderate discomfort; normal ECG and blood pressure; no medication, supplements or other dairy during study. Excluded IBS, constipation, unmedicated IBD (other exclusions in supplement).

4. Methods and endpoints

Exposure / intervention

250 mL milk twice daily after meals for 14 days, conventional milk with A1:A2 beta-casein about 40:60 (about 400 and 600 mg per 100 mL).

Comparator

Otherwise identical milk containing only A2 beta-casein (same protein, 4.8% lactose in both), supplied by A2 Infant Nutrition Ltd.

Endpoints

PD3 symptoms (bloating, abdominal pain, flatulence, heavy stomach, borborygmi; 0-3 Likert), stool frequency and Bristol scale; smart-pill transit times and capsule-image inflammation grading; SCIT response time and error rate; serum hs-CRP, Hb, IL-4, IgG, IgE, IgG1; faecal SCFA; adverse events. BCM-7, GSH and MPO listed in the study design figure.

Statistical methods

Mixed-effects ANOVA (intervention and period fixed, subject random nested in sequence) with baseline as covariate; carry-over tested via intervention x period; GEE for symptom scores versus baseline within sequence; no adjustment for multiple comparisons.

5. Actual results

Results

Conventional (A1+A2) vs A2-only milk, between-product estimates: IL-4 +2.53 ng/L (P<0.0001), IgG (sqrt-scale estimate 0.14, P=0.0007), IgE +5.97 IU/mL (P=0.0253), IgG1 (log-scale, P=0.0037); hs-CRP not significantly different (P=0.0608); faecal total SCFA -0.13% (P=0.0009). Colonic transit +6.6 h and whole-gut transit +6.3 h longer (both P<0.0001), small-bowel transit no difference (-0.20 h, P=0.59). Symptoms: bloating, flatulence and borborygmi significantly worse than baseline during A1+A2 weeks in both sequences (e.g. bloating log-OR 1.75, 95% CI 0.51-3.00); no worsening with A2-only milk; stool frequency increased by about 3 stools/week with A1+A2. Small-bowel inflammation on capsule images improved in 36.4% after switching A1+A2 -> A2 vs 11.1% in the reverse sequence (P=0.042); stomach P=0.427. SCIT response time (P=0.0013) and error rate (P=0.0004) higher with A1+A2. Diarrhoea: 8 events in 5 subjects attributed to A1+A2 milk, 3 events in 3 subjects to A2 milk. BCM-7 concentrations are claimed in the abstract but are not reported in the main-text tables.

Absolute risks

not reported in full text (diarrhoea: 5/45 subjects with events attributed to A1+A2 milk vs 3/45 to A2 milk)

Authors' conclusion

Milk containing A1 beta-casein worsened GI symptoms, prolonged transit, increased serum inflammation markers, lowered faecal SCFA and slowed cognitive processing; A2-only milk did not; some symptoms attributed to lactose intolerance may stem from A1 beta-casein rather than lactose.

6. Strengths

• Randomised, double-blind crossover with identical-looking repackaged products and blinded capsule assessor
• Registered (NCT02406469); same lactose content in both milks isolates the beta-casein difference
• Objective transit measurement by smart pill
• Lactose-intolerance status tested

7. Weaknesses

• Industry-funded; company employee conceived the study and selected outcomes; company-supplied products and company-funded medical writing
• Small exploratory trial (n=45), no sample-size calculation, 2-week periods
• Many outcomes without multiplicity adjustment; symptom analyses mainly within-sequence vs baseline rather than direct head-to-head
• Selective reporting: BCM-7 (stated in abstract), GSH and MPO listed in design but not reported in main results; hs-CRP non-significant
• Smart pill not used at baseline; 5 subjects without smart pill
• 'Inflammation' biomarkers are mainly Th2/immunoglobulin markers with small changes; clinical relevance unclear
• Self-selected Han Chinese population with self-reported intolerance; low transferability
• Cognitive (SCIT) findings small and of uncertain clinical meaning

8. Risk of bias

Orienting tool

RoB 2 (crossover)

Type of assessment

structured assessment based on the full text by a single reviewer (no formal double review)

Full formal assessment

no

Overall risk of bias

high

Rationale

Simplified single-reviewer RoB 2 from full text; formal assessment needs two reviewers and the registered protocol. Industry sponsorship is not itself a RoB 2 domain but adds to concerns about outcome selection and interpretation.

Full-text review (25 October 2026): Added period (Oct-Dec 2014), randomisation details, dose (250 mL twice daily), A1:A2 40:60, effect sizes (colonic transit +6.6 h, whole-gut +6.3 h, IL-4 +2.53 ng/L, IgE +5.97, hs-CRP P=0.0608 n.s., SCFA -0.13%), capsule inflammation (36.4% vs 11.1%, P=0.042), adverse events, no multiplicity adjustment, no sample-size calculation. Found that BCM-7 (claimed in abstract), GSH and MPO are not reported in the main-text results. RoB 2 overall changed some_concerns -> high (D5); methodological_quality moderate -> low.

9. Assessment of the statistical interpretation

Several statistically significant but mostly small differences (e.g. IL-4 +2.5 ng/L, IgE +6 IU/mL) from an exploratory crossover with many unadjusted tests; the most robust finding is longer colonic transit (+6.6 h) and more self-reported bloating/flatulence with conventional milk vs A2 milk. hs-CRP was not significantly affected. Results compare two kinds of milk; they do not show that milk per se is harmful, because A2-only milk produced no worsening versus the dairy-free baseline.

10. Relevance to the specific claim

Core evidence for C-DRY-07; supporting C-DRY-02.

11. Independent contrary findings

I-DRY-04 (most RCTs on markers null; low certainty beyond digestive symptoms).

12. Transferability to Germany and Europe

Low (specific population).

13. Overall methodological judgement

Industry-designed small exploratory trial at high risk of bias from selective reporting; suggests conventional milk may cause more GI symptoms than A2-only milk in self-reported intolerant Chinese adults. It provides no evidence that milk in general inflames the gut, and the BCM-7 claim cannot be verified from the main text.

14. Open questions

Independent, adequately powered trials with prespecified primary outcome; full reporting of BCM-7; effects in unselected and European/Egyptian populations; relevance for cheese and fermented products.

15. Linked fact checks

• Dairy products harm health and cause inflammation. – Studies with mixed or unclear results
• A1 beta-casein in conventional milk releases beta-casomorphin-7, which triggers gut inflammation, slower transit and worse symptoms (shown in a 2016 randomised trial). – Studies supporting the claim

Use by the Tayebat archive

Note: according to the archive, the studies were compiled by the archive operators. Whether Dr. Elawady himself cited this study is not established.

What the archive says about the study

A1 beta-casein milk produced more gut inflammation, slower transit, beta-casomorphin-7, and worse cognitive performance than A2-only milk; narrative says the trial 'proved' it.

A. What was actually studied?

A1+A2 milk vs A2-only milk in intolerant Chinese adults.

B. What did the authors find?

Worse symptoms, longer colonic transit, higher IL-4/IgE/IgG/IgG1 and lower faecal SCFA with conventional A1+A2 milk vs A2-only milk; hs-CRP not significantly different; A2-only milk did not worsen anything versus dairy-free baseline. BCM-7 values not shown in main text.

C. What conclusions do the authors themselves draw?

Some intolerance symptoms may stem from A1-triggered inflammation; avoidable with A2 milk.

D. What does the archive derive from it?

Used to argue that conventional milk inflames the gut.

E. Is this derivation scientifically justified?

Partly overinterpreted: small, short, industry-designed and funded trial with selective outcome reporting; 'proved' is not justified. It supports a possible A1-vs-A2 difference in self-reported intolerant people, not dairy avoidance, and A2-only milk showed no adverse effects. Casein (incl. A1 beta-casein) is also concentrated in the cheeses the system allows.

Adequacy of the interpretation (by the archive)

partly correct but over-interpreted