Extended Data Fig. 4: Per-protocol analysis of clinical, patient-reported, and laboratory outcomes after the third FMD cycle.

Data from participants who completed the study were assessed after the third FMD cycle; for participants in the control group, this corresponded to approximately 12 weeks after baseline. a–c, Significantly more participants met criteria for clinical response 70 (a, 82.0% versus 50.0%, P < 0.01), clinical response 100 (b, 78.0% versus 42.9%, P < 0.01), and clinical remission (c, 76.0% versus 39.3%, P < 0.01) after completing the three FMD cycles compared to participants who made no dietary changes. d,e, Compared to baseline, a significantly higher percentage of participants reported remission by PRO (d, 58.0% versus 25.0%, P < 0.01) and improvement in SIBDQ score (e, 56.0% versus 28.6%, P = 0.02) after the third FMD cycle compared to participants in the control arm. f,g, There were significant differences in mean percentage change in CRP (f, −15.7% versus 36.9%, P < 0.01) and fecal calprotectin (g, −36.5% versus 8.9%, P < 0.01) between the FMD and control arms. a–e, Percentages of participants meeting the criteria for response. f,g, Mean percentage change measured from baseline to after the third FMD cycle; error bars represent standard error of the mean (s.e.m.). P values were calculated by Fisher’s exact test or chi-square test (a–e) or t-test (f,g); *P < 0.05, **P < 0.01. FMD, fasting-mimicking diet; PRO, patient-reported outcomes; SIBDQ, short inflammatory bowel disease questionnaire; CRP, C-reactive protein.