Cohort studyindependent cross-checkabstract checked★ key study I-PPI-108 in-depth review
Long-term PPI use linked with a small rise in stomach cancer
Proton pump inhibitors and risk of gastric cancer: population-based cohort study.
Overview
Abrahami D et al.. Gut 2022 · PMID 34226290 · DOI 10.1136/gutjnl-2021-325097
Compared with H2 blockers, PPI users had a 45% higher relative risk of stomach cancer, but only about 1 extra case per 1,200-2,100 users over 5-10 years.
Does the study support the specific claim? Partly · Methodological quality: moderate
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Compare stomach cancer risk of PPI and H2RA users.
- How was the study done?
Large UK cohort.
- How many people, animals or samples were studied?
Over 1.1 million new users.
- What was actually found?
HR 1.45; dose-response.
- How large was the observed effect?
NNH 2121 (5 y), 1191 (10 y).
- What do the results mean in everyday life?
If real, the extra risk is small; use PPIs only when indicated.
- What are the study's limitations?
Observational; H. pylori not fully known.
- What does it say about Elawady's specific claim?
Partly supports a cancer concern; contradicts 'deadly poison'.
- Do the results apply to Germany and Europe?
UK data, applicable to Europe.
- What can explicitly not be concluded from this study?
Cannot prove causation.
Scientific in-depth review
1. Full reference
- Authors
- Abrahami D et al.
- Journal
- Gut
- Year
- 2022
- Identifier
- PMID 34226290 DOI 10.1136/gutjnl-2021-325097
- Verification
- verified – PubMed efetch 2026-10-26 + Crossref 2026-10-26
- Retracted
- no
- Corrections
- none
- Extent checked
- abstract only
- Conflicts of interest and funding
CIHR funding; authors report advisory/consulting roles unrelated to the study
2. Research question
Are new users of PPIs at higher risk of gastric cancer than new users of H2RAs?
3. Study design and population
- Study design
- Cohort study – Population-based new-user active-comparator cohort (UK CPRD)
- Population
New users of PPIs or H2RAs in UK primary care
- Country
UK
- Period
1990-2018
- Sample size
973,281 PPI and 193,306 H2RA new users
- Inclusion and exclusion criteria
New users; standard exclusions per protocol (not detailed in abstract)
4. Methods and endpoints
- Exposure / intervention
PPI
- Comparator
H2RA
- Endpoints
Gastric cancer
- Statistical methods
Cox models with standardised mortality ratio weights from calendar-time-specific propensity scores; NNH from Kaplan-Meier
5. Actual results
- Results
Median follow-up 5.0 years: HR 1.45 (1.06-1.98); NNH 2121 at 5 years and 1191 at 10 years; HR increased with cumulative duration, dose and time since initiation; consistent sensitivity analyses.
- Absolute risks
NNH 2121 (5 y) and 1191 (10 y): about 47 and 84 extra cases per 100,000 users
- Authors' conclusion
PPI use is associated with increased gastric cancer risk compared with H2RAs, although the absolute risk remains low.
6. Strengths
- Active comparator new-user design
- Dose-response
- Very large
7. Weaknesses
- Observational
- H. pylori status incompletely recorded
- Indication differences between PPI and H2RA users
8. Risk of bias
- Orienting tool
- ROBINS-E
- Type of assessment
- simplified assessment (abstract only)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
Residual confounding by indication and H. pylori possible.
9. Assessment of the statistical interpretation
HR CI log-symmetric; absolute excess small (stats/C-PPI-09_cancer.py).
10. Relevance to the specific claim
Best-quality support for a cancer signal: a modest relative increase with very low absolute risk - far from 'deadly poison' or a doubling of cancers.
11. Independent contrary findings
Sawaid 2026 (I-PPI-107); COMPASS no excess cancer over 3 years (I-PPI-01).
12. Transferability to Germany and Europe
Directly applicable to Europe.
13. Overall methodological judgement
Real, small, uncertain signal for gastric cancer only.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | some concerns |
| Precision of results | moderate |
| Directness | direct |
| Reproducibility | partly replicated |
| Consistency with the overall evidence | mixed |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Causality and role of H. pylori.