This article is more than 5 years old. Information may no longer be current.
Researchers question quality of data supporting low FODMAP diet in IBS
Click Here to Manage Email Alerts
Randomized controlled trials of the low FODMAP diet for irritable bowel syndrome carry a high risk for bias and have insufficient follow-up, and the improvements in symptoms observed in these trials could be driven by a placebo response, according to new research published in Alimentary Pharmacology and Therapeutics.
“This relatively new diet has gained much attention, and information about the diet is available on the internet in excess,” Laura Rindom Krogsgaard, MD, of the department of gastroenterology at Zealand University Hospital in Denmark, and colleagues wrote. “Experts recommend the diet as a first-line treatment for IBS and refer to the diet as being supported by a high level of evidence,” but the quality of available evidence has not yet been evaluated, they added.
Therefore, they systematically reviewed relevant medical literature published up to October 2016, and included nine randomized controlled trials involving a total of 542 patients in their analysis, assessing the quality of trials by estimating the risk for bias and evaluating the trial’s general methodology.
Five trials reported the low FODMAP diet was superior to a control intervention for improving IBS symptoms, and four reported a comparable effect between the diet and control. The investigators identified a high risk for bias in several components of all the trials, particularly due to lack of blinding and choice of control groups.
High risk for bias
First, they noted that the trials’ intervention periods were short, ranging from 2 days to 6 weeks, which “are generally considered insufficient for long term treatments.” Only one trial reported 6-month follow-up after the dietary intervention.
Second, the trials evaluated small numbers of patients who were “primarily recruited from tertiary care, which may compromise generalizability,” as they “constitute a selected population that might be especially motivated for dietary intervention,” the investigators wrote.
Importantly, they noted a high risk for bias in trial design based on the choice of control groups in all trials, which “can lead to overestimation of the treatment response.”
In three trials, meals were provided to the study participants and typical American/Australian diets high in FODMAPs served as control interventions. According to the investigators, using a diet high in FODMAPs as a control is not a clinically relevant comparison, “as the control arm does not serve as a placebo group nor reflects standard of care.”
Further, a dietician provided low FODMAP diet instruction in six trials with varying control interventions that had “limited established efficacy,” and no trials included the FODMAPs reintroduction phase in the intervention period, an important part of the diet, they noted.
“The control groups in the trials, which estimated the effects of consultation by a dietician, unfortunately, do not allow the discrimination of a proper effect of the diet from effects induced by other factors,” the researchers wrote.
Additionally, while they acknowledged that blinding patients to the low FODMAP diet can be difficult due to the abundance of available information on it, they emphasized that the “lack of blinding can severely affect the interpretation of results, because the effect of the intervention is measured as subjective symptom outcomes highly susceptible to performance bias.”
Trial investigators did attempt to blind participants in six trials, but only one “evaluated the completeness of blinding, and reported that 83% of the participants were able to identify the diet and were thus unblinded,” Krogsgaard and colleagues wrote. Further, only one trial was double blinded.
Finally, the researchers noted that “several of the trials present data, which indicates that the postulated effect of the low FODMAP diet is primarily placebo driven.” These factors include considerable variation in time to symptomatic effect on the diet, a generalized symptomatic effect across stool subtypes and different symptoms, and a pattern of symptom improvement in one trial that included gluten provocation that suggested a placebo driven response “raised by the expectation of symptom relief.”
They concluded that “future trials should establish the symptomatic effects of a dietary intervention in subgroups of IBS patients with a specific phenotype, eg, IBS-C and use as the control arm an active treatment with proven efficacy eg, linaclotide. Ideally, a double blind, double-dummy design should be used to account for differences in the administration of the intervention (drug vs. dietary advice).” – by Adam Leitenberger
Disclosures: The researchers report no relevant financial disclosures.
Click Here to Manage Email Alerts