Cohort studyindependent cross-checkfull text checked★ key study I-CHK-07 in-depth review
Meat and 25 common conditions (UK)
Meat consumption and risk of 25 common conditions: outcome-wide analyses in 475,000 men and women in the UK Biobank study.
Overview
Papier K et al.. BMC Med 2021 · PMID 33648505 · DOI 10.1186/s12916-021-01922-9
More poultry was linked to slightly more reflux, gastritis and diabetes, partly explained by body weight, and to less anaemia.
Does the study support the specific claim? Partly · Methodological quality: high
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Meat intake and 25 non-cancer conditions
- How was the study done?
Prospective cohort (UK Biobank), outcome-wide
- How many people, animals or samples were studied?
474,985; Middle-aged UK adults
- What was actually found?
Poultry per 30 g/day: GERD HR 1.17 (1.09-1.26), gastritis/duodenitis 1.12 (1.05-1.18), diverticular disease 1.10, gallbladder disease 1.11, diabetes 1.14 (1.07-1.21); lower iron-deficiency anaemia 0.83 (0.76-0.90). Associations substantially attenuated by BMI.
- How large was the observed effect?
Cox, multiple-testing correction
- What do the results mean in everyday life?
Small associations, not proof of harm.
- What are the study's limitations?
Observational; associations small and largely explained by body weight; total energy and cooking method not measured; meat types not mutually adjusted; hospital-recorded outcomes only.
- What does it say about Elawady's specific claim?
Mixed.
- Do the results apply to Germany and Europe?
High (UK).
- What can explicitly not be concluded from this study?
Cannot show causation or that chicken is intrinsically indigestible; no absolute risks reported.
Scientific in-depth review
1. Full reference
- Authors
- Papier K et al.
- Journal
- BMC Med
- Year
- 2021
- Identifier
- PMID 33648505 DOI 10.1186/s12916-021-01922-9
- Verification
- verified – PubMed efetch 2026-10-25
- Retracted
- no
- Corrections
- none
- Extent checked
- full text
- Conflicts of interest and funding
Wellcome Trust (Our Planet Our Health - LEAP), Cancer Research UK, UK Medical Research Council; authors declare no competing interests.
2. Research question
Meat intake and 25 non-cancer conditions
3. Study design and population
- Study design
- Cohort study – Prospective cohort (UK Biobank), outcome-wide
- Population
UK Biobank participants aged ~40-69 at recruitment (2006-2010) from 22 assessment centres, excluding prevalent cancer
- Country
UK
- Period
2006-2017 (mean follow-up 8 years)
- Sample size
474,985 (of 503,317 recruited); poultry analyses n=473,011; ~69,000 with >=3 repeat 24-h recalls (Oxford WebQ) used to calibrate intakes
- Inclusion and exclusion criteria
Excluded study withdrawals, prevalent cancer (except non-melanoma skin cancer), sex discordance; per outcome, participants with prior relevant diagnosis/procedure excluded; missing/unknown meat intake excluded per exposure.
4. Methods and endpoints
- Exposure / intervention
Baseline touchscreen frequency of unprocessed poultry (and red/processed meat), with category means assigned from repeated 24-h recalls; HR per 30 g/day poultry
- Comparator
Lower poultry intake (trend per 30 g/day)
- Endpoints
Incident hospital admission or death for 25 leading non-cancer causes of hospital admission (plus any diabetes diagnosis), via HES/Scottish/Welsh linkage
- Statistical methods
Cox models stratified by sex, age, region; Model 1 sociodemographic/lifestyle, Model 2 + fruit/vegetables, cereal fibre, fish, Model 3 + BMI; Bonferroni threshold P<0.002; sensitivity analyses excluding first 4 years and in never smokers; meat types not mutually adjusted; total energy not available.
5. Actual results
- Results
Poultry per 30 g/day (Model 3, Bonferroni-robust): GERD HR 1.17 (1.09-1.26); gastritis and duodenitis 1.12 (1.05-1.18); diverticular disease 1.10 (1.04-1.17); gallbladder disease 1.11 (1.04-1.19); diabetes 1.14 (1.07-1.21); iron-deficiency anaemia 0.83 (0.76-0.90). Most positive associations were substantially attenuated by BMI adjustment; similar results excluding the first 4 years and in never smokers. For comparison, unprocessed red + processed meat per 70 g/day: IHD 1.15 (1.07-1.23), pneumonia 1.31 (1.18-1.44), diverticular disease 1.19 (1.11-1.28), colon polyps 1.10 (1.06-1.15), diabetes 1.30 (1.20-1.42).
- Absolute risks
not reported in full text (only relative hazard ratios; case numbers in supplementary tables not available)
- Authors' conclusion
Higher BMI accounted for substantial proportion; residual confounding or mediation possible.
6. Strengths
- Very large prospective cohort with linked hospital and death records
- Outcome-wide design limits selective outcome reporting
- Repeated 24-h recalls to correct for regression dilution
- Bonferroni correction and sensitivity analyses for reverse causation and smoking
7. Weaknesses
- Residual confounding, especially by adiposity and energy intake (total energy not measured)
- Meat types not mutually adjusted
- Single baseline frequency questionnaire; cooking method/type of poultry (e.g. fried, processed) not distinguished
- Hospital-based outcomes miss milder disease; GERD/gastritis admissions may reflect health-care use
- Low response rate/healthy-volunteer selection of UK Biobank
- Associations are small (HR 1.10-1.17 per 30 g/day)
8. Risk of bias
- Orienting tool
- ROBINS-E
- 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
- some concerns
- Rationale
Unchanged (some_concerns). Single-reviewer structured judgement; a formal ROBINS-E needs two independent reviewers and the protocol.
| Domain | Judgement | Rationale |
|---|---|---|
| Confounding | some concerns | Extensive adjustment including BMI, but no energy intake, no mutual meat adjustment; strong attenuation by BMI indicates residual confounding risk. |
| Exposure measurement | some concerns | Brief touchscreen frequency questions calibrated with 24-h recalls in a subsample; no cooking method. |
| Selection of participants | some concerns | UK Biobank volunteers (healthy-volunteer bias); exclusions of prevalent disease per outcome. |
| Post-exposure interventions | not applicable | Not applicable. |
| Missing data | low | Covariate missingness <10% handled as categories; linkage-based follow-up. |
| Outcome measurement | some concerns | Hospital admission/death records; non-hospitalised cases missed. |
| Selection of reported result | low | Outcome-wide pre-defined set of 25 conditions with Bonferroni correction. |
Full-text review (25 October 2026): Added exclusions, exposure calibration with WebQ, model sequence and Bonferroni threshold, sensitivity analyses (similar excluding first 4 years and in never smokers), no energy/mutual adjustment, funding; confirmed HRs; absolute risks not reported. Bias unchanged.
9. Assessment of the statistical interpretation
Precise but small relative associations that the authors attribute substantially to adiposity (residual confounding or mediation). Causation is not established; the authors speculate on handling/cooking (e.g. H. pylori on raw poultry) for gastritis.
10. Relevance to the specific claim
11. Independent contrary findings
Attenuation by BMI.
12. Transferability to Germany and Europe
High (UK).
13. Overall methodological judgement
High-quality observational study finding small positive associations of poultry intake with some digestive conditions and diabetes, largely explained by BMI, and lower iron-deficiency anaemia. It does not show that chicken is intrinsically indigestible or harmful.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | high |
| Risk of bias | some concerns |
| Precision of results | high |
| Directness | direct |
| Reproducibility | not replicated |
| Consistency with the overall evidence | unknown |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Causality; role of cooking method and processed poultry; replication in other cohorts.
15. Linked fact checks
- Chicken is forbidden because it is harmful in itself, not because of how it is raised (it is 'the pig of birds'); its fibres and waste products are hard to digest. – Studies with mixed or unclear results
- Modern chicken breast is indigestible and damages the digestive system and gut microbiome. – Studies with mixed or unclear results