Meta-analysisindependent cross-checkabstract only checked★ key study I-INTB-14 in-depth review
How accurate are H. pylori tests?
Non-invasive diagnostic tests for Helicobacter pylori infection.
Overview
Best LM et al.. Cochrane Database Syst Rev 2018 · PMID 29543326 · DOI 10.1002/14651858.CD012080.pub2
Breath tests are highly accurate; blood and stool tests somewhat less so; about 46 false positives per 1000 tested in a model scenario.
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?
Compare H. pylori tests
- How was the study done?
Cochrane diagnostic review
- How many people, animals or samples were studied?
101 studies, 11,003 people
- What was actually found?
Breath test most accurate
- How large was the observed effect?
sensitivity 0.94 at specificity 0.90
- What do the results mean in everyday life?
A positive breath test is usually correct; blood tests can mislead more often.
- What are the study's limitations?
Indirect comparisons; abstract only
- What does it say about Elawady's specific claim?
Partly true that tests can be falsely positive; false that most positives are wrong.
- Do the results apply to Germany and Europe?
High
- What can explicitly not be concluded from this study?
Cannot give exact false-positive rates for a particular lab in Egypt.
Scientific in-depth review
1. Full reference
- Authors
- Best LM et al.
- Journal
- Cochrane Database Syst Rev
- Year
- 2018
- Identifier
- PMID 29543326 DOI 10.1002/14651858.CD012080.pub2
- Verification
- verified – PubMed efetch 2026-10-25
- Retracted
- no
- Corrections
- none
- Extent checked
- abstract only
- Conflicts of interest and funding
NIHR Cochrane Programme Grant (record)
2. Research question
What is the diagnostic accuracy of urea breath test, serology and stool antigen test for H. pylori?
3. Study design and population
- Study design
- Meta-analysis – Cochrane diagnostic test accuracy review and meta-analysis (HSROC)
- Population
People suspected of H. pylori infection
- Country
Multinational
- Period
search to March 2016
- Sample size
101 studies, 11,003 participants
- Inclusion and exclusion criteria
Index test vs histopathology reference standard
4. Methods and endpoints
- Exposure / intervention
Urea breath test (13C, 14C), serology, stool antigen test
- Comparator
Histology (reference)
- Endpoints
Sensitivity, specificity, diagnostic odds ratio
- Statistical methods
HSROC, bivariate models
5. Actual results
- Results
DORs: UBT-13C 153 (95% CI 73.7-316), UBT-14C 105 (74.0-150), serology 47.4 (25.5-88.1), stool antigen 45.1 (24.2-84.1). At fixed specificity 0.90, sensitivity 0.94, 0.92, 0.84, 0.83. In a hypothetical 1000 people (prevalence 53.7%, specificity 0.90): 46 false positives; false negatives 30, 42, 86 and 89 respectively. 53 studies excluded people recently on PPIs or antibiotics.
- Absolute risks
46 false positives per 1000 tested at specificity 0.90
- Authors' conclusion
Urea breath tests had high diagnostic accuracy; serology and stool antigen tests were less accurate (based on indirect comparison).
6. Strengths
- Cochrane DTA methods
- Large evidence base
7. Weaknesses
- Indirect comparisons
- Variable thresholds
- Unclear risk of bias in all studies
- Abstract only read
8. Risk of bias
- Orienting tool
- none applicable
- Type of assessment
- simplified assessment (abstract only)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
QUADAS-2 used by authors; unclear risk of bias in every study.
9. Assessment of the statistical interpretation
Specificity was fixed at 0.90 for comparison; real false-positive rates vary by test and threshold.
10. Relevance to the specific claim
Contradicts '700 of 1000 positive results are false'; supports that no non-invasive test is perfect and that serology/stool tests are less accurate than breath tests.
11. Independent contrary findings
Serology cannot distinguish past from current infection (general knowledge, not quantified in abstract).
12. Transferability to Germany and Europe
High.
13. Overall methodological judgement
Best summary of H. pylori test accuracy.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | high |
| Risk of bias | some concerns |
| Precision of results | moderate |
| Directness | direct |
| Reproducibility | replicated |
| Consistency with the overall evidence | consistent |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Performance in high-prevalence settings such as Egypt.
15. Linked fact checks
- The common test for H. pylori ('the stomach germ') is the test with the highest rate of false-positive results. – Studies contradicting or limiting the claim