False positives and false negatives
By BeatSIBO editorial team · Updated August 7, 2026
Accuracy is usually discussed as one number. It is really two different problems with different causes, and conflating them is how people end up confident about the wrong thing.
The two directions are not symmetric
The short answer
A positive can be produced by fermentation that started in the colon rather than the small intestine, which is a measurement reaching the wrong place. A negative can be produced by a test that physically cannot detect the gas in question. These are different failures with different fixes, and no single accuracy figure describes both.
How a positive goes wrong
The lactulose test uses a sugar humans do not absorb, so it travels the full length of the gut and can reach the colon, where fermentation is expected and normal. If that happens inside the collection window, the rise it produces can look like the rise the test is meant to detect.
Scalese G et al., Journal of Clinical Medicine 2025 Peer-reviewed studyThat study also found clinical symptoms did not significantly differ between test-positive and test-negative groups, which removes the obvious fallback. You cannot use how you feel to decide whether a positive was real, because symptoms did not separate the groups in the first place.
This is also why the collection window is part of the method rather than an administrative detail. A longer run is more likely to capture something that began once the sugar reached the colon.
How a negative goes wrong
The short answer
Most breath tests sold measure two gases. Hydrogen sulfide is a third, and a two gas test cannot detect it at any concentration. A negative from that test is not a finding about hydrogen sulfide, it is silence about it, and the report does not usually make that distinction visible.
The distinction matters because the two read identically on paper. A result that says negative and a result that says not measured look the same to someone who was not told which gases the test covers.
Rezaie A et al., American Journal of Gastroenterology 2017 Clinical guidelineThe consensus that sets the thresholds covers hydrogen and methane. It contains no hydrogen sulfide criterion, so even a test that measures the third gas returns a number with no published cutoff behind it. That is a real limitation rather than a technicality, and it applies to the best case rather than the worst.
What is not settled here
We are not going to publish a list of things that skew a result, with effect sizes, because we could not find one we would stand behind. The literature is thinner than the confident lists circulating online suggest, and several of those lists appear to be copied from each other rather than from a source. What is established is the structure of the problem: the substrate decides where the test looks, and the gases covered decide what it can see.
What to do with a result you doubt
Do not retest hoping for a cleaner answer without first knowing what changed. Establish which substrate was used, which gases were covered, and how long the collection ran. If a second test differs from the first on any of those, it is answering a different question rather than checking the first one.
Common questions
- Does a negative result rule out overgrowth?
- It rules out what the test can measure, at the thresholds published for it. Most tests sold detect two gases and cannot register hydrogen sulfide at all, so a negative from one of those leaves that gas entirely unexamined. That is not a false negative in the usual sense. It is a question the test never asked.
- How accurate is the test?
- Accuracy is the wrong shape of question here, because it implies one number covering both error directions against a settled reference. There is no such reference. In one study of patients who took both, under one in ten overgrowth diagnoses were made by both tests, which tells you more about the state of the measurement than any single accuracy figure would.
- Should I trust a positive result?
- Treat it as one input rather than a verdict, and take it to a clinician with its substrate, its collection window and its date. A positive is genuinely informative. It is just not, on its own, the settled fact that the number on the page looks like.
