— January 13, 2026 · Fasting-Mimicking Diet Eases Crohn’s Disease in Trial

Table of Contents

1. At a Glance
2. What Was Done
3. What Was Found, Number by Number
4. Side Effects, Dropouts and Adherence
5. How a Five-Day Diet Could Work
6. How It Fits With Earlier Research
7. What It Cannot Show
8. What the Researchers and Coverage Said
9. What Remains Unknown
10. Sources and Primary Documents
11. Key Research Papers
12. Connections
13. Featured Videos

At a Glance

• Paper: Kulkarni, Fardeen and colleagues, senior author Sidhartha R. Sinha, Stanford University School of Medicine. Nature Medicine, published online January 13, 2026 (print issue March 2026, volume 32, pages 1023–1033).
• Who took part: 97 adults aged 18 to 70 with mild-to-moderate Crohn’s disease, median age 45, enrolled between 2019 and 2023; 38 came through Stanford and the rest from 59 other institutions around the country.
• Design: open-label randomised controlled trial, two diet participants for every control (65 versus 32), registered as NCT04147585.
• The diet: five consecutive days a month for three months, about 1,090 kcal on day 1 and 725 kcal on days 2 to 5, from a plant-based, commercially prepared meal kit; usual eating the other 25 or so days.
• Main result: clinical response 69.2% (45 of 65) versus 43.8% (14 of 32), P = 0.03; clinical remission 64.6% (42 of 65) versus 37.5% (12 of 32), P = 0.02.
• Inflammation markers: fecal calprotectin −22.0% versus +8.0% (P = 0.03); C-reactive protein (CRP) −1.0% versus +36.9% (P = 0.06, not statistically significant).
• Durability: three months after the last cycle, response was 56.9% versus 53.1% (P = 0.83), no longer a significant difference.
• Safety: no severe adverse events in the diet group; fatigue (52.3%) and headache (50.8%) were common and mostly mild; 21.5% of the diet group withdrew, against 9.4% of controls, none because of side effects.
• Funding and conflicts: a private charitable trust and the US National Institutes of Health, among others; one co-author holds equity in the company that sells the meal kit.

Back to Table of Contents

What Was Done

Crohn’s disease is one of the two main forms of inflammatory bowel disease (IBD), the other being ulcerative colitis. The paper notes that about 20–30% of people with Crohn’s have a milder course, and that in the United States no medicine other than corticosteroids is approved by the FDA specifically for mild disease. It also notes that the most common question people with IBD ask their gastroenterologist is “What should I eat?”, while most diet evidence in adult Crohn’s comes from small or non-randomised studies.

The team screened 279 people through a national recruiting campaign. Of 119 recruited, 22 decided not to take part before randomisation, leaving 97 who were randomised in blocks of six, two to the diet for every one to the control group. To join, people had to be 18 to 70 years old with a Crohn’s Disease Activity Index (CDAI) above 150 and no higher than 450. The CDAI is a standard symptom score widely used in Crohn’s trials; it draws on factors such as the number of bowel movements, abdominal pain, body weight and the haematocrit (a blood count), and the trial defined remission as a score of 150 or below. People were excluded if they were pregnant or nursing, allergic to nuts, had a body-mass index below 18, took diabetes drugs that can cause low blood sugar, had serious heart disease or a history of fainting, had liver or kidney problems that a low-glucose, low-protein diet could worsen, or had short bowel syndrome, an ostomy, a total colectomy or an ileoanal pouch.

For five consecutive days each month, for three months, the diet group was asked to eat only, and all of, the contents of a pre-packed box: soups, nutrition bars, snacks and supplements, described in the paper as plant-based, low in calories, sugars and protein and high in unsaturated fats. Day 1 provided 1,090 kcal (10% protein, 56% fat, 34% carbohydrate) and days 2 to 5 provided 725 kcal each (9% protein, 44% fat, 47% carbohydrate). A research dietitian could swap an item that was poorly tolerated for a recipe with matching nutrients. For the rest of each month the diet group ate as usual. The control group kept its usual diet throughout. Both groups stayed on whatever treatment their own gastroenterologist had prescribed.

The primary outcome was “clinical response 70”: a fall in CDAI of at least 70 points, or a CDAI of 150 or below, measured within a week after the third diet cycle. Secondary outcomes included remission (CDAI 150 or below), a stricter 100-point response, changes in CRP, the erythrocyte sedimentation rate (ESR) and fecal calprotectin, patient-reported symptoms, quality of life and colonoscopy findings, checked after the first cycle, after the third and after a three-month washout. Anyone who dropped out was counted as a treatment failure. Participants received $200 for their time, and $400 if they also had a colonoscopy.

The groups were broadly similar at the start, with some imbalances: the diet group had more women (80.0% versus 56.3%) and more stricturing disease (30.8% versus 9.4%), while the control group had more obesity (31.3% versus 15.4%) and more use of advanced therapy such as biologics (71.8% versus 60.0%). About two-thirds of participants had mild disease, and the median CDAI was 196. A regression analysis that adjusted for the imbalanced factors found that none of them significantly changed the odds of remission.

Back to Table of Contents

What Was Found, Number by Number

All figures below are from the intention-to-treat analysis, which keeps every randomised participant in the group they were assigned to.

• Clinical response (primary outcome): 69.2% with the diet versus 43.8% with the usual diet (P = 0.03), a difference of about 25 percentage points.
• Clinical remission: 64.6% versus 37.5% (P = 0.02).
• Stricter 100-point response: 66.2% versus 40.6% (P = 0.02).
• CDAI change: a median fall of 105 points with the diet versus 76 points without it (P = 0.02).
• After only one cycle: response 66.2% versus 43.8% (P < 0.05) and remission 60.0% versus 37.5% (P = 0.04).
• Fecal calprotectin: mean change −22.0% versus +8.0% (P = 0.03). In an analysis added after the trial was designed, 37.0% of the diet group had a fall of 50% or more, against 6.3% of controls (P = 0.01).
• Blood markers: CRP −1.0% versus +36.9% (P = 0.06) and ESR +10.7% versus +15.2% (P = 0.87); neither difference was statistically significant.
• Patient-reported remission (fewer than four loose stools a day and little or no abdominal pain): 47.7% versus 25.0% (P < 0.05).
• Quality of life: a score above 50 on the short IBD questionnaire (SIBDQ) in 46.2% versus 25.0% (P < 0.05); participants who felt they were in remission, 24.6% versus 6.3% (P = 0.03).
• Need for stronger treatment: steroids, a new advanced therapy or a dose increase in 27.7% versus 25.0% (P = 0.99), no difference.
• Three months after the last cycle: response 56.9% versus 53.1% (P = 0.83) and remission 55.4% versus 43.8% (P = 0.39), no significant difference.

Subgroups. Response was higher with the diet in mild disease (75.0% versus 47.8%) and in moderate disease (57.1% versus 11.1%, the latter among only nine controls), in Crohn’s of the colon (82.4% versus 33.3%) and of the ileum and colon together (71.4% versus 30.0%), but not in disease limited to the ileum, the last part of the small intestine (55.5% versus 60.0%, P = 0.99). Among people taking no Crohn’s medicine at baseline, response was 76.9% versus 33.3%, an analysis the authors added after seeing how many participants were not on advanced therapy. Subgroups this small give wide uncertainty.

Colonoscopy. Only six participants agreed to a colonoscopy, five in the diet group and one control; the authors attribute this to pandemic-era disruption and patients’ hesitation to undergo an elective procedure. A reader who did not know the group assignments scored four of the five diet participants as in endoscopic remission after the third cycle, including one who went from severe activity (a Simple Endoscopic Score of 19) to none (0). The one control stayed mildly active. Six people cannot show whether the diet heals the bowel lining, and the authors say so.

Back to Table of Contents

Side Effects, Dropouts and Adherence

No participant in the diet group had a severe (grade 3) adverse event. Seven controls did; in the paper’s adverse-event table these were mainly severe abdominal pain (six people) and diarrhoea (two), which are symptoms of the disease itself. Median body weight did not change in either group.

The diet had its own side effects. Fatigue was reported by 52.3% of the diet group and headache by 50.8%, mostly mild; the paper notes both are commonly reported in fasting. Mild (grade 1) symptoms reported more often with the diet than without it (not every difference was statistically significant) included fatigue (40.0% versus 9.4%), headache (41.5% versus 0%), bloating (24.6% versus 3.1%), dizziness (18.5% versus 3.1%) and difficulty concentrating (12.3% versus 0%). In the control group, diarrhoea (50.0%) and abdominal pain (37.5%) were the most common complaints.

Fourteen diet participants (21.5%) withdrew, against three controls (9.4%), 17 in all (17.5%). The paper states that none of the withdrawals were due to adverse effects and lists the reasons in a supplementary table. Because dropouts were counted as failures, the withdrawals worked against the diet in the main analysis. Of the diet group, 76.9% followed the diet in all three cycles (no more than 400 extra calories on any diet day and all five days completed); 87.5% of controls kept to their usual diet. Among participants who stuck to their assigned diet, response was 82.0% versus 50.0% and remission 76.0% versus 39.3%, and in that group CRP also fell more with the diet (−15.7% versus +36.9%).

Back to Table of Contents

How a Five-Day Diet Could Work

The idea behind a fasting-mimicking diet is to give the body some of the signals of a fast while still providing a small amount of food, so that people can stay on it for five days and then go back to their normal meals. The authors list several possible routes by which it could quiet Crohn’s inflammation, most of them drawn from other studies rather than tested here:

• Lower inflammatory signalling. The authors cite studies in which calorie restriction suppressed inflammatory pathways such as NF-κB, the NLRP3 inflammasome and TNF, which drive much of the gut damage in Crohn’s.
• Repair of the gut lining. In mice, the diet and the refeeding that follows activated intestinal stem cells that replace damaged lining. That is an animal result.
• Gut bacteria. The kit is rich in prebiotic fibres, and in mouse colitis the diet increased Lactobacillaceae, a bacterial family linked to calmer colitis. This trial did not measure the gut microbiome.
• Cell clean-up. Fasting triggers autophagy, the recycling of damaged parts inside cells.
• Ketones. Fasting raises blood ketones such as β-hydroxybutyrate, which have been shown to influence inflammatory cells in the gut in other studies.

The trial itself ran blood tests on a small subset: plasma from 13 diet participants and immune cells from 14, drawn after one diet cycle and compared with each person’s own baseline, not with the control group. Fourteen of the 20 most reduced chemical pathways in plasma belonged to the metabolism of arachidonic and linoleic acid, the fatty acids from which the body makes inflammatory signalling fats. Breakdown products of two leukotrienes (LTB4 and LTE4), which are raised in IBD, fell, and an anti-inflammatory lipoxin product rose. In white blood cells, activity of the genes for TNF, IL-1β, IL-18, NLRP3 and two chemokines (CXCL10 and CCL20) went down, while NF-κB and IL-10 did not change significantly. The authors call parts of this “hypothesis-generating”: the samples were not specially stabilised and some fragile molecules may have been missed.

Why the diet seemed to help colonic but not ileal disease is unexplained. The authors suggest that the immune environment of the colon reacts more to food and bacterial material passing through it, and that the small-intestine lining turns over faster, but they present this as a hypothesis.

Back to Table of Contents

How It Fits With Earlier Research

Healthy volunteers. The diet was first tested in people in a 2017 randomised trial of 100 generally healthy US adults. Three monthly five-day cycles reduced body weight, trunk and total body fat, blood pressure and the growth factor IGF-1 compared with an unrestricted diet, with no serious adverse effects. A post hoc analysis found that CRP and other risk markers improved more in participants who were at risk to begin with. The Crohn’s trial based its sample-size estimate on how often CRP returned to normal in that study.

Animal colitis. A 2019 study in mice with chemically induced colitis found that four-day fasting-mimicking cycles reduced gut inflammation, increased intestinal stem cells and protective bacteria and reversed the tissue damage, whereas water-only fasting improved some markers without reversing the damage. Transferring Lactobacillus or stool bacteria from treated mice also helped. The same paper reported that in a human trial, three diet cycles lowered markers of whole-body inflammation. The mouse findings are animal results; they suggested the Crohn’s trial but do not show what happens in people.

Ulcerative colitis. A small 2025 pilot trial in Miami gave two five-day cycles of a low-calorie, plant-based, fasting-mimicking diet (the same 1,090 then 725 kcal pattern) to adults with active ulcerative colitis who were starting advanced medicines. It ended early with 32 enrolled and 23 completing. Response was 57% versus 35% (P = .11), so the primary end point was not met, although a continuous symptom score improved more with the diet (P = .039). There were no serious diet-related adverse events and no weight loss.

Other Crohn’s diets. The best-established diet treatment is exclusive enteral nutrition, a liquid formula in place of all food for several weeks. A 2018 Cochrane review of 27 trials found, with evidence it graded as very low quality, that in adults it achieved remission less often than corticosteroids (45% versus 73%; risk ratio 0.65, 95% CI 0.52 to 0.82), while in children remission was 83% versus 61% (risk ratio 1.35, 95% CI 0.92 to 1.97, not statistically significant); the most common reason for stopping was being unable to tolerate the formula. Whole-food alternatives have followed. In children, the Crohn’s Disease Exclusion Diet with partial formula was better tolerated than exclusive formula (97.5% versus 73.6%) and kept more children in steroid-free remission at week 12 (75.6% versus 45.1%) than exclusive formula followed by partial formula. CD-TREAT, an ordinary-food diet built to match the formula, produced a response in 4 of 5 children, and remission in 3, in an early study. In adults, the 2021 DINE-CD trial randomised 194 people with mild-to-moderate symptoms to the Specific Carbohydrate Diet or a Mediterranean diet; symptomatic remission at week 6 was similar (46.5% versus 43.5%). DINE-CD compared two diets with no usual-diet control, which is the gap the Stanford team says its trial addresses. The site’s overview of these approaches is on the IBD diet page.

Treatment targets. The 2021 STRIDE-II consensus of the International Organization for the Study of IBD lists symptom relief and normalisation of blood and stool markers such as calprotectin as short-term treatment targets, with clinical remission and endoscopic healing as long-term targets. The Crohn’s trial reported symptom improvement and a fall in calprotectin, the kind of short-term measures STRIDE-II names; it could not assess endoscopic healing.

Back to Table of Contents

What It Cannot Show

• No blinding. Participants knew whether they were on the diet, and the main outcome is a symptom score that partly depends on how people feel. The authors note that people with mild disease “may be more prone to assessor bias when reporting symptoms particularly in diet studies.” Calprotectin, a laboratory measure, is less open to that bias and also moved in the diet’s favour.
• A large control-group response. About 40% of controls improved with no diet change. The authors attribute this to continued medicines and the natural ups and downs of mild disease. The size of the diet’s effect rests on the gap between two moving groups.
• Short and not lasting on its own. The diet was given for three months. Three months later the difference was no longer statistically significant, so the trial says nothing about long-term control, and the authors’ suggestion that monthly cycles might need to continue has not been tested.
• Healing not measured. With colonoscopy in six people, the trial cannot show whether the bowel lining healed.
• Who it applies to. Most participants had mild disease and were predominantly White, and many groups were excluded, including people who are underweight, pregnant, on low-blood-sugar-risk diabetes drugs or with serious heart, liver or kidney disease. The authors say the results cannot speak for moderate-to-severe disease.
• Small subgroups and added analyses. The results for moderate disease, colonic disease and people on no medicine come from small groups, and some analyses (the 50% calprotectin fall, the no-medicine group) were added after the trial was designed.
• Mechanism data are thin. The blood tests came from 13 or 14 diet participants after one cycle, without a control comparison. In the lipid figure, the multiplicity-adjusted value for one pair of oxidised linoleic-acid markers (the HpODEs) is q = 0.31, above the paper’s 0.10 threshold, although the text groups them with the significant decreases.
• One product. The trial tested a specific commercial meal kit. Whether a home-prepared diet with the same calories and nutrients would do the same was not studied.
• Registry detail. The ClinicalTrials.gov record lists an actual enrolment of 86, while the paper reports 97 randomised; the paper does not explain the difference.
• Conflict of interest. The meal kits were bought by the research team from the company that sells them. One co-author “has equity interest” in that company “and has filed patents related to the fasting-mimicking diet”; the paper states that he receives no consulting fees from the company and “has committed his shares of the company to charitable organizations.” The other authors declare no competing interests.

The authors themselves sum up that, given “the small and open label nature of this study and limited endoscopic data,” the findings need to be “interpreted with care.”

Back to Table of Contents

What the Researchers and Coverage Said

Stanford Medicine’s news release of January 13, 2026, “A new diet option for mild-to-moderate Crohn’s disease,” quoted the senior author, Sidhartha R. Sinha: “We were very pleasantly surprised that the majority of patients seemed to benefit from this diet.” He added, “We noticed that even after just one FMD cycle, there were clinical benefits,” and, on the state of diet advice, “We have been very limited in what kind of dietary information we can provide patients.” He also said: “There’s still a lot more to be done to understand the biology behind how this and other diets work in patients with Crohn’s disease.” The release states that the diet “significantly improved both physical symptoms and biological indicators”; of the inflammation markers in the paper, fecal calprotectin changed significantly, while CRP and ESR did not. The release says the researchers are now exploring whether changes in the gut microbiome help explain the results; that is future work, as this paper reports no microbiome data.

HealthDay’s January 14 story was headlined “Two-Thirds Of Crohn’s Disease Patients Benefit From Fasting Diet, Clinical Trial Shows” and opened by saying that fasting “a handful of days each month can significantly improve GI symptoms.” Its figures (69% versus 44% response; 65% versus 38% remission) match the paper. Neither the release nor that story mentioned that the difference had faded three months after the last cycle, that the trial was unblinded, that only six people had a colonoscopy, or that more diet participants than controls withdrew. Neither quoted an independent expert.

The paper itself describes the result as showing that the diet “is superior to a baseline diet for inducing clinical response, clinical remission and biochemical improvement in mild-to-moderate CD” and as support for “further investigation of FMD as an adjunctive therapy for chronic inflammatory diseases.” The authors describe the trial as, to their knowledge, “the largest randomized, controlled trial of a solid, non-exclusion diet in adults with CD.”

Back to Table of Contents

What Remains Unknown

• Whether repeating the five-day cycle every month keeps people in remission for a year or more, and whether that is safe and tolerable over time.
• Whether the diet heals the bowel lining, which needs colonoscopy in many more people.
• Whether the effect holds in moderate-to-severe Crohn’s, in disease limited to the ileum, and in more diverse populations.
• Which part does the work: the calorie cut, the low protein and sugar, the fats, the fibres in the kit, or simply eating differently for five days.
• Whether a home-prepared version matching the kit’s calories and nutrients gives the same result.
• Whether gut bacteria change in people with Crohn’s as they did in mice, and whether that matters.
• How the diet compares head to head with established diet treatments such as enteral nutrition or the Crohn’s Disease Exclusion Diet, and how it combines with biologic and other advanced medicines.

Back to Table of Contents

Sources and Primary Documents

Figures attributed to the trial above come from the full text of the paper’s NIH author manuscript in PubMed Central (including its tables and figure legends), its PubMed record and its trial registration, not from the press coverage.

1. Kulkarni C, Fardeen T, Gubatan J, Ye J, Jarr K, Dickson E, et al. (2026). A fasting-mimicking diet in patients with mild-to-moderate Crohn’s disease: a randomized controlled trial. Nature Medicine 32(3):1023-1033. Epub January 13, 2026. — doi:10.1038/s41591-025-04173-w · PubMed PMID: 41530382
2. Author manuscript of the same paper, PubMed Central PMC13293041 (NIHMS2169882). — PMC13293041
3. ClinicalTrials.gov record NCT04147585, “Effects of an Intermittent Reduced Calorie Diet on Crohn’s Disease” (lead sponsor Stanford University; started December 18, 2019; status completed; enrolment listed as 86). — ClinicalTrials.gov NCT04147585
4. Stanford Medicine news release, January 13, 2026: “A new diet option for mild-to-moderate Crohn’s disease.” — med.stanford.edu release

Coverage read for this report: HealthDay, January 14, 2026 (“Two-Thirds Of Crohn’s Disease Patients Benefit From Fasting Diet”).

Back to Table of Contents

Key Research Papers

Background papers for this story. Each entry was checked against PubMed and Crossref before it was written down.

1. Wei M, Brandhorst S, Shelehchi M, Mirzaei H, Cheng CW, Budniak J, et al. (2017). Fasting-mimicking diet and markers/risk factors for aging, diabetes, cancer, and cardiovascular disease. Science Translational Medicine 9(377):eaai8700. — PubMed PMID: 28202779
2. Rangan P, Choi I, Wei M, Navarrete G, Guen E, Brandhorst S, et al. (2019). Fasting-Mimicking Diet Modulates Microbiota and Promotes Intestinal Regeneration to Reduce Inflammatory Bowel Disease Pathology. Cell Reports 26(10):2704-2719.e6. — PubMed PMID: 30840892
3. Damas OM, Leo M, Colina N, Mijares M, Fernandez I, McALeavy J, et al. (2025). A Pilot Randomized Control Trial to Assess the Adjunctive Effect of Diet on Response to Advanced Therapies in Patients With Ulcerative Colitis. Clinical Gastroenterology and Hepatology 23(13):2579-2587.e2. — PubMed PMID: 40378983
4. Narula N, Dhillon A, Zhang D, Sherlock ME, Tondeur M, Zachos M (2018). Enteral nutritional therapy for induction of remission in Crohn’s disease. Cochrane Database of Systematic Reviews 4(4):CD000542. — PubMed PMID: 29607496
5. Levine A, Wine E, Assa A, Sigall Boneh R, Shaoul R, Kori M, et al. (2019). Crohn’s Disease Exclusion Diet Plus Partial Enteral Nutrition Induces Sustained Remission in a Randomized Controlled Trial. Gastroenterology 157(2):440-450.e8. — PubMed PMID: 31170412
6. Svolos V, Hansen R, Nichols B, Quince C, Ijaz UZ, Papadopoulou RT, et al. (2019). Treatment of Active Crohn’s Disease With an Ordinary Food-based Diet That Replicates Exclusive Enteral Nutrition. Gastroenterology 156(5):1354-1367.e6. — PubMed PMID: 30550821
7. Lewis JD, Sandler RS, Brotherton C, Brensinger C, Li H, Kappelman MD, et al. (2021). A Randomized Trial Comparing the Specific Carbohydrate Diet to a Mediterranean Diet in Adults With Crohn’s Disease. Gastroenterology 161(3):837-852.e9. — PubMed PMID: 34052278
8. Turner D, Ricciuto A, Lewis A, D’Amico F, Dhaliwal J, Griffiths AM, et al. (2021). STRIDE-II: An Update on the Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE) Initiative of the International Organization for the Study of IBD (IOIBD): Determining Therapeutic Goals for Treat-to-Target strategies in IBD. Gastroenterology 160(5):1570-1583. — PubMed PMID: 33359090

PubMed Topic Searches

1. PubMed: Fasting-mimicking diets and inflammatory bowel disease
2. PubMed: Randomised diet trials in Crohn’s disease
3. PubMed: Exclusive enteral nutrition in Crohn’s disease
4. PubMed: Fecal calprotectin and treatment response in Crohn’s
5. PubMed: Calorie restriction and intestinal inflammation

Back to Table of Contents

Connections

• News Archive
• Crohn’s Disease — causes, symptoms and treatment research
• Inflammatory Bowel Disease (IBD) — the hub for Crohn’s and ulcerative colitis
• Ulcerative Colitis — the other main form of IBD
• Crohn’s vs Ulcerative Colitis — how the two diseases differ
• Diet for IBD — Specific Carbohydrate, Mediterranean, CDED and enteral nutrition
• IBD Diagnostic Workup — calprotectin, colonoscopy and MR enterography
• IBD Biologics — TNF, IL-23 and integrin inhibitors
• Fasting — intermittent and extended fasting research
• Caloric Restriction — the science of eating less
• Autophagy and mTOR — the cell biology of fasting
• Your Microbiome — the bacteria of the gut
• The Gut Barrier and Your Microbiome — interactive animation
• C-Reactive Protein (hs-CRP) Test — the blood marker of inflammation
• Mediterranean Diet — the comparison diet in the DINE-CD trial

Back to Table of Contents

Featured Videos