Eswaran SL, Chey WD, Han-Markey T, Ball S, Jackson K. A randomized controlled trial comparing the low FODMAP diet vs. modified NICE guidelines in US adults with IBS-D. Am J Gastroenterol. 2016;111(12):1824-1832.
To compare low fermentable oligo-, di-, and monosaccharides and polyols (FODMAP) dietary advice to conventional irritable bowel syndrome (IBS) dietary advice for patients with IBS with diarrhea (IBS-D).
In this randomized controlled trial, researchers used a modified version of the National Institute for Health and Care Excellence (NICE) guidelines for IBS dietary guidance as an active control. The participants and dietitians were not able to be blinded, but remaining study personnel were. Participants were not told which diet was the true intervention diet. There was a 2-week run-in phase and a 4-week intervention phase.
Ninety-two participants (71% female) with IBS-D were randomized; 84 participants completed the trial and were included in the primary analysis.
The low FODMAP diet was the intervention. The comparison diet, acting as an active control, was the IBS diet recommended by NICE, modified by removing legumes and onions (mNICE).
Study parameters assessed
Food intake, IBS symptomatology including abdominal pain and bloating, stool frequency, and stool consistency were assessed.
Primary outcome measures
The primary outcome was the proportion of participants who experienced “adequate relief.” This was defined as those patients who reported “adequate relief” of their IBS symptoms ≥50% of the time during the second half of the 1-month trial.
Fifty-two percent of the participants counseled with a low FODMAP diet reported “adequate relief” for their IBS-D symptoms, compared to 41% of those given NICE IBS dietary advice. This did not reach statistical significance (P=0.31). Statistically significant benefits were seen in some secondary outcome measures, such as abdominal pain, bloating, and stool form and frequency.
This was a well-conducted study overall. Some points in the study’s favor include adequate randomization, prospective registration, clear a priori primary outcomes, a reasonable active control, and few dropouts (even if unbalanced).
My biggest concern clinically comes with the difficulty patients have in following a low FODMAP diet.
As with any trial, there are some concerns as well. The participants and the dietitians of course could not be blinded. However, the study authors took care to obfuscate which diet was the intervention diet, so placebo effects likely occurred in both groups, possibly canceling each other out. The same cannot be said for the dietitians, so performance bias is still possible.1 Meta-epidemiologic research has shown us that we need to be most concerned about such biases in subjectively reported outcomes such as the “adequate response” primary outcome of this study.2 Therefore, this is a concern; however, there is no way to avoid this in dietary intervention trials for IBS, because they cannot be blinded and IBS outcomes are necessarily subjective. The sponsors of the study were mostly academic centers and did not seem to have a direct mechanism for biasing the result. However, one of the senior authors received funding from Nestlé, which currently is heavily marketing low FODMAP food products.
This is actually the second randomized controlled trial comparing low FODMAP diets to conventional IBS dietary advice for IBS. The first, conducted in Sweden and published in 2015, also did not find a statistically significant improvement in primary outcome.3 That study also used NICE guidelines for IBS dietary advice as the control diet. This American study by contrast modified the NICE guidance for its control group by removing language around avoiding high FODMAP foods, such as onions and beans. In this way, only the intervention group would be avoiding high FODMAP foods.
The results for the primary outcome in this trial were negative, which suggest that avoiding high FODMAP foods may no effect above that seen with standard dietary advice for IBS. It is also possible that there is an effect above standard IBS dietary advice, but that it is a smaller effect than anticipated and this study was not powered to detect it (type II error, also known as a false negative). Larger trials or meta-analyses of multiple smaller trials would be needed to clarify this.
While the primary outcome did not reach statistical significance, there does seem to be an improvement in some of the individual symptoms such as bloating, abdominal pain, stool form, and consistency. Apparently, however, these improvements were not enough to convince patients that they were getting adequate symptomatic relief from their IBS.
I believe this study was conducted as well as it could have been. I would have been content with an effect size in the 10% range had it been statistically significant. My biggest concern clinically comes with the difficulty patients have in following a low FODMAP diet. If indeed the authors found an effect in the 30% range, as they had originally anticipated, getting a patient to undertake such a challenging diet might seem reasonable.