irritable bowel: Mediterranean diet is best

A recent study found that a Mediterranean diet was superior to the traditional diet in the treatment of patients with irritable bowel syndrome, IBS  (see irritable bowel Mediterannean  Diet helps AnnIntMed2025 in dropbox or doi:10.7326/ANNALS-25-01519)

 

Details:

-- 139 persons from across the United Kingdom aged 18 to 65 with IBS per the Rome IV diagnostic criteria were enrolled after having received a score of at least 75 on the IBS Symptom Severity Scale (IBS-SSS)

    -- participants were enrolled into 2 groups

        -- traditional IBS diet (TDA), a sensible eating habit to avoid excess fatty foods, spicy foods, processed foods, caffeine, fizzy drinks, and alcohol (all of which can be triggers to IBS)

        -- Mediterranean diet, with principal components being rich in fruit, vegetables, pulses (dried edible seeds of legume plants such as beans, lentils, and peas), whole grains, nuts, fish, and olive oil.

    -- patients were randomized into groups reflecting severity of IBS (mild, moderate, or severe), and its subtypes (mix pattern, diarrhea-predominant, constipation-predominant)

    -- this trial was conducted from October 2023 to December 2024 using an online virtual platform

    -- all patients were told about their assigned diets and received dietary education from 1 or 2 dieticians, but no information about the diet of the other group. these online dietary education sessions lasted 30 minutes, with a slide presentation and questions/answers. participants were subsequently emailed supporting information

    -- patients were given baseline demographic questionnaires, were allocated to their diet for 6 weeks, and every 2 weeks they reported their dietary adherence reflecting how often they followed the dietary advice, rated as sometimes, frequently, or always,

    -- they also filled out the IBS-SSS questionnaire, wherein the person rated their symptoms over the last 10 days for abdominal pain severity, pain frequency, bloating, bowel habit dissatisfaction, and life interferences related to their bowel symptoms. This score has a maximum of 500 points

        -- IBS-SSS interpretation: if no symptoms (score <75), moderate symptoms (score 75-174), and severe symptoms (score 300-500)

    -- also a variety of other self-administered questionnaires:

        -- Patient Health Questionnaire-4: this assesses anxiety and depression over the preceding 2 weeks with a total score ranging from 0 to 12, and a higher score representing worse mood

        -- Patient Health Questionnaire-12, recording the 12 bothersome non-gastrointestinal symptoms often associated with IBS over the past month, with a score of 0 to 24,  higher scores representing more somatic symptom severity. These non-gastrointestinal symptoms include: backpain, headaches, chest pain, dizziness, fainting spells, feeling one's heart pound or race, shortness of breath, pain or problems during sexual intercourse, pain in arms/legs/joints (eg knees, hips, etc), feeling tired or having low energy, menstrual cramps, trouble sleeping with menstruation: https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2036.2010.04402.x

        -- Short Form-8 Health Survey: an 8-item instrument assessing physical and mental quality of life over the preceding month with a mean combined score of 0-100, higher scores represent better quality of life

        -- Mediterranean Diet Adherence Screener (MEDAS): a 14 item instrument with total scores ranging from 0 to 14 that assesses the intake of food groups relevant to a Mediterranean diet, where a 2 point increase in scores is associated with better health outcomes

        -- Comprehensive Nutritional Assessment Questionnaire: an instrument to assess macronutrient and micronutrient intake, with a record over the past 6 weeks. This questionnaire was optional given that it comprised more than 200 questions and takes almost 30 minutes

-- the intention here was to determine if the MD was not inferior to the TDA by more than 5%, making MD a viable first-line option; this was also to avoid reaching placebo response rates of 35% as documented in other studies

-- mean age 40, 80% women, 92% white, 77% employed

-- mean BMI 27.6

-- IBS severity per IBS-SSS score: mild in 12 (9%), moderate in 52 (37%), severe in 75 (54%); mean IBS-SSS was 309.2

-- IBS subtype: mixed patter in 67 (48%); diarrhea-predominant in 43 (31%); constipation-predominant in 29 (21%)

-- questionnaire responses at baseline (see descriptions above for more detail):

    -- mean PHQ-12 (nongastrointestinal somatic score): 7.7

    -- mean PHQ-4 (anxiety and depression): 3.8

    -- mean SF-8 (quality of life): 53.1

    -- mean MEDAS (dietary adherence assessment): 4.8

-- primary endpoint: proportion of people achieving a clinical response to their diet, defined as at least a 50-point reduction in their IBS Symptom Severity Scale (IBS-SSS)

-- secondary outcomes: changes in IBS-SSS scores, psychological health, somatic symptom reporting, quality of life, diet satisfaction, and the Mediterranean Diet Adherence Screener (MEDAS)

 

Results:

-- 168 participants: 82 assigned to traditional dietary advice (TDA) and 86 to Mediterranean DIet (MD)

-- primary endpoint of response to assigned diet:

    -- MD: 62% (50%-73%)

    -- TDA: 42% (31%-55%)

        -- difference in response: 20 percentage points (4-36 percentage points), p=0.017

        -- this difference confirmed noninferiority of the MD, and further demonstrated actual superiority of MD

--reduction of IBS-SSS:

    -- MD: -101.2

    -- TDA: -64.5

        -- difference of -36.7 (-70.5 to -2.8), p=0.034

--of note:

    -- the main improvement in IBS-SSS was found within 2 weeks for both of the diets (p<0.001)

    -- the largest difference between the treatment groups was seen at week 6 (p=0.034)

 

-- changes in mood: no difference between diets

-- changes in somatic symptoms: no difference between diets

-- changes in quality of life: no difference between diets

-- changes in diet satisfaction: no difference between diets

-- MEDAS score: significantly increased after MD (5.1 to 8.9) vs TDA (4.6 to 5.8), p<0.001

    -- ie, the MEDAS showed that the MD diet was followed, with the 2-point or more increase suggesting it was a clinically relevant level of change

--IBS subgroup analysis:

    -- severe IBS group from baseline to 6 weeks who experienced a 50-point reduction in IBS-SSS:

        -- MD: 63% (24 of 38)

        -- TDA: 43% (16 of 37)

            -- difference 20 percentage points, p=0.08

    -- moderate IBS group from baseline to 6 weeks who experienced a 50-point reduction in IBS-SSS:

        -- MD: 70% (16 of 23)

        -- TDA: 38% (11 of 29)

            -- difference 32 percentage points, p=0.02

    -- mild IBS group from baseline to 6 weeks, experiencing a 50-point reduction in IBS-SSS:

        -- MD: 29% (2 of 7)

        -- TDA: 60% (3 of 5)

            -- nonsignificant benefit of TDA, but very few individuals in this mild IBS group, limiting statistical relevance

    -- individuals with IBS-mixed pattern from baseline to 6 weeks:

        -- MD: 65% (22 of 34)

        -- TDA: 39% (13 of 33)

            -- difference of 26 percentage points, p=0.04

    -- individuals with IBS-diarrhea predominant pattern from baseline to 6 weeks:

        -- MD: 59% (10 of 17)

        -- TDA: 50% (13 of 26)

            -- insignificant difference, with p=0.57

    -- individuals with IBS-constipation predominant pattern at baseline to 6 weeks:

        -- MD: 59% (10 of 17)

        -- TDA: 33% (4 of 12)

            -- insignificant difference, with p=0.18, but few individuals in this IBS group

-- per-protocol analysis confirmed higher response rate with MD than TDA  in achieving a 50-point and 100-point reduction in IBS-SSS, as well as a greater mean reduction at week 6

-- sensitivity analyses:

    -- not much difference when statistically using inverse probability weighting analysis to minimize the significant baseline imbalances in the groups (age, BMI, somatic burden and quality of life; also incorporating sex, baseline IBS severity score, and baseline MEDAS)

    -- for the primary outcome and considering missing outcomes as nonresponders, there was no difference in results; similarly if used multiple imputations to account for the missing outcomes

Commentary:

-- IBS is remarkably common, affecting 5% to 10% of the population, with chronic symptoms of abdominal pain and altered bowel habits (diarrhea and/or constipation). It typically affects young to middle-aged adults and can lead to reduced quality of life and work impairment. It is also the most frequent GI condition found in primary and secondary care settings

-- >80% of people report foods as a trigger or aggravater of symptoms, leading to treatments targetting consumption of these foods

    -- historically, the traditional dietary advice (TDA) has been the first-line therapy, which encourages a balanced diet and healthy eating patterns, and about 40% of patients respond to this diet. 

    -- those who do not respond then proceed to the FODMAP diet (Fermentable Oligosaccharides, Disaccharides, Monosaccharides, and Polyols) which has been found to have a success rate in 50% to 70% of patients

        --  components of  this diet target avoiding short-chain carbohydrates that are poorly absorbed in the small intestine and can cause digestive stress including gas, bloating, and pain especially in those with IBS, and restriction of sugars found in wheat, diary dairy, certain fruits/veggies, alcohol. This diet could be modified as potential triggers are identified and is not necessarily a long-term diet

            -- however,  FODMAP is quite complex, restrictive, costly, and socially inconvenient; adherence to this diet is suboptimal (https://pubmed.ncbi.nlm.nih.gov/30543574/ ), and tends to decrease over time

            -- of concern, this FODMAP diet can be detrimental on the microbiome

    -- the unstated assumption in this current study is that patients in the TDA did not move onto FODMAP if insufficient TDA response (at least, there was no indication they did in the article or supplementary material)

 -- the American Gastroenterological Association published their AGA Clinical Practice Update on the Role of Diet in Irritable Bowel Syndrome: Expert Review in 2022, noting that: "The low-FODMAP diet is currently the most evidence-based diet intervention for IBS. Healthy eating advice as described by the National Institute of Health and Care Excellence Guidelines, among others, also offers benefit to a subset of patients with IBS." in their full review they do note that the Mediterranean Diet could be considered based on a small study but needed more supportive data: https://www.gastrojournal.org/action/showPdf?pii=S0016-5085%2821%2904084-1

-- this study, however, found that MD was not only noninferior to TDA but was actually superior in the primary endpoint of achieving at least a 50 point change in the IBS-SSS, and a nonsignificantly higher proportion achieving the 100-point reduction

-- and there was a statistically significant greater reduction in the mean IBS-SSS and in the frequency of abdominal pain, though no significant difference in dietary satisfaction and adherence to both diets

 

 

Limitations:

-- this study required participants to have online access and speak English. it was okay if they were on antidepressants but not if they were being titrated up on them. Also patients were not included if they were already on a FODMAP diet. these exclusions do limit generalizability of the intervention to other groups

    -- the age limits precluded pediatric patients, non-English speakers, and non-tech savvy people, limiting generalization to these groups.

    -- though the researchers found that the MD was superior to the TDA, they did not fully test the traditional approach in that they did not move TDA-nonresponders to FODMAP as per the usual approach

-- this study had a pragmatic design and was virtually administered, which in this study may have led to inaccurate responses to the questionnaires. But, a pragmatic study also leads to a more generalizable outcome to actual patients (vs patients selected from specialty clinics, for example)

-- the assignment to different diets was randomized, but the participants were aware of their assigned diets. this could lead to a placebo effect if they were predisposed to the content of their assigned diet. but, without a tightly controlled randomized study with people getting similar appearing foods with similar caloric values etc and being delivered to them, the choice of an open-label study is about the best that could be done

-- the outcome after 6 weeks was impressive, but that does not necessarily represent a long-term outcome in a long-term disease

-- the study was also biased to a large percentage of white women, again not reflecting many other sites/cultures where different foods are eaten... also, potentially limiting generalizability of the results

-- we do not have access to granular data on the specific foods. for example, some of the vegetables consumed can lead to gas-producing effects and more IBS symptoms

-- there is the gut/brain axis of the microbiome, and the Mediterranean Diet is known to be beneficial, with its antioxidant and anti-inflammatory effects

    -- the gut/brain axis (see microbiome gut brain axis JClinInvest2015 in dropbox, or doi:10.1172/JCI76304) has been pretty well described showing that the gut microbiome can affect emotional behavior, stress- and pain-modulation, and brain neurotransmitters (the gut microbiota elaborates many neurotransmitters; animal studies have found major effects on the expression of brain signaling systems; some of these changes could affect appetite/diet/weight as well as the placebo effect (see https://www.frontiersin.org/articles/10.3389/fpsyt.2022.824468/full )

so, 

-- pretty impressive study that the Mediterranean Diet seemed to be the best diet in those with irritable bowel syndrome, in a group skewed to more severe symptoms

-- and, the Mediterranean diet has been shown to have many global benefit in terms of cardiovascular disease, breast cancer risk, knee osteoarthritis, osteopenia,  atrial fibrillation, diabetes, cognitive health, NAFLD (now MAFLD), systemic inflammation, overall mortality (for many blogs on this, see https://gmodestmedblogs.blogspot.com/search?q=Mediterranean+diet&updated-max=2025-05-27T11:20:00-07:00&max-results=20&start=13&by-date=false )

-- the Mediterranean Diet seems to me to be the best diet to try initially, especially in light of its pretty global benefit to health...

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them