
REFIT2 changed gut microbes without a significant IBS symptom advantage over placebo. Learn what the results mean for microbiome claims and food choices.

On 2026-11-02, Medical Republic reported that the Norwegian REFIT2 phase 3 trial found no significant benefit from donor fecal microbiota transplantation for irritable bowel syndrome (IBS) compared with placebo. The treatment changed recipients’ gut microbes without producing a corresponding advantage in overall symptoms.
Before REFIT2, earlier studies of fecal microbiota transplantation (FMT) had reported possible benefits for IBS. A clinical summary of the trial discussion described that earlier evidence as limited by small sample sizes and less stringent definitions of improvement. That history explains why a larger trial with a clear symptom target mattered.
The earlier evidence did not amount to a settled medical consensus that donor stool reliably treats IBS. There was a promising idea that needed stronger testing: transferring microbes from a healthy donor might produce a useful change in a recipient’s gut. The medical question was whether that change would make symptoms meaningfully better.
Microbiota means the community of microbes being measured. Altering that community is a biological effect. Relief from discomfort or irregularity is a patient outcome. These are related questions, but one does not automatically answer the other.
That distinction helps define what a successful study needs to show. A report of changed microbes addresses whether the intervention affected its biological target. A comparison of symptom improvement addresses whether it helped participants more than the alternative tested. REFIT2 brought both questions into the same trial.
According to Medical Republic, REFIT2 was a Norwegian phase 3 study conducted at five hospitals. It enrolled 450 adults with moderate-to-severe IBS. Two participants were excluded after their allocated treatments were switched, leaving 448 in the analysis. Participants had a median age of 36, and 65% were women.
Participants were assigned in a 2:1 ratio to donor FMT or placebo. Both groups received a single rectal enema. The donor group received stool from a healthy donor, while the placebo group received their own stool. This gave researchers a comparison between donor material and the trial’s own-stool procedure.
The study was double-blind and placebo-controlled. Its primary outcome was set in advance rather than chosen after the results were known. That matters because a study may measure several things, but its main success test should remain the one it was designed to answer.
The prespecified target was an improvement of at least 75 points on the IBS Severity Scoring System (IBS-SSS) at 90 days. In Medical Republic’s account, 40% of the donor-FMT group and 38% of the placebo group reached that threshold. The reported absolute difference was 1.9 percentage points and was not statistically significant.
The important comparison is between those groups. Reading the donor-group result by itself would make the treatment appear helpful to many participants. Reading it alongside the placebo result shows why the trial did not demonstrate an added overall symptom benefit from donor FMT.
Follow-up did not reveal a later advantage. Medical Republic reported improvement at one year in 44% of the FMT group and 45% of the placebo group. There was also no significant between-group difference in quality-of-life improvement at 90 days or one year.
These findings address more than a short-term change in a laboratory sample. They compare symptom improvement at the main assessment point and again at longer follow-up. The quality-of-life result adds another patient-centered measure to the picture.
The trial also looked at bowel patterns within IBS subtypes. It found no significant improvement in stool consistency among participants with diarrhea-predominant IBS. Among those with constipation-predominant IBS, it found no significant improvement in complete spontaneous bowel movements.
At the same time, donor FMT produced measurable changes in recipients’ gut microbiota. That biological result did not translate into a significant overall symptom advantage. The finding challenges the assumption that a changed microbial community is enough to establish therapeutic success.
For readers of gut-health research, this is the central change in understanding. A microbiome measurement and a symptom score are separate pieces of evidence. A claim about the first needs evidence about the second before it becomes a claim of symptom relief.
The same distinction is useful when keeping microbial measurements separate from symptom results. It helps readers identify what an article actually measured rather than treating every biological change as an improvement in daily life.
REFIT2 tested a particular approach, not every possible way of changing the microbiome. An accompanying commentary summarized in coverage of the trial highlighted the single-enema route and the use of three donors as limits on how broadly the result can be applied. Repeated treatments or other delivery routes remain different research questions. Their being untested here is not evidence that they work.
The participants also had moderate-to-severe IBS. This result should stay attached to that condition and population rather than being extended to every source of bloating or irregularity. It is not a finding about every use of microbiota-based therapies. Nor does it establish that all nutrition-related or microbiome-related approaches are ineffective.
Another boundary concerns individual improvement. Both groups included people who reached the symptom threshold, but the comparison does not establish why any particular participant improved. A nonsignificant difference is not a statement that nobody felt better. It means this trial did not demonstrate an overall symptom advantage for donor FMT over its comparator.
Medical Republic reported 24 adverse events in the FMT group through 365 days, including six serious events. The placebo group had eight events and none was serious. The groups were not allocated in equal numbers, so these raw counts should not be read as equal-size comparisons.
Four FMT recipients and three placebo recipients had events considered possibly or probably related to treatment. None of those events was serious, and no deaths occurred during follow-up. Events occurring after a procedure and events judged related to it are different categories. The report does not justify labeling every serious event as caused by FMT.
A fair assessment therefore needs both the symptom comparison and careful safety reporting. A laboratory response is not a substitute for either.
REFIT2 was not a dietary trial. It offers no reason to add or remove a particular food based on its FMT result. For everyday decisions, the useful lesson is to ask whether a claim shows a meaningful symptom benefit in people with the relevant condition, not just a change in gut microbes.
Personal decisions about ongoing pain or bowel changes belong with a qualified healthcare professional. Background reading on why similar digestive symptoms need careful evaluation can help frame that conversation. The realistic implication is a higher standard for evidence, not a more restrictive plate or a promise that microbiome alteration will solve IBS.
An altered gut microbiome can sound like a successful IBS treatment even when symptom results do not support that conclusion. DigestGenius uses free educational articles and in-depth resources to make digestive research clearer, helping adults distinguish an interesting biological finding from evidence relevant to comfort and bowel function.
DigestGenius publishes research-led guidance on digestion, the gut microbiome, fiber, probiotics, gut-brain signaling, inflammation and everyday digestive wellbeing.



Explore clear, research-led guidance on digestion, the microbiome, food, fiber, probiotics and the gut-brain connection.
read the blog