Dietary index for gut microbiota score and incident inflammatory bowel disease: a prospective cohort study in the UK Biobank

Purpose Dietary patterns may be associated with inflammatory bowel disease (IBD) development. These patterns have been proposed to affect gut microbial and immune-related processes. However, prospective evidence on the correlation between the dietary index for gut microbiota (DI-GM) and incident IBD remains limited. The associations of DI-GM with incident IBD, Crohn’s disease (CD), and ulcerative colitis (UC) were investigated. Methods This prospective cohort study included 198,204 participants from the UK Biobank. The DI-GM was constructed from dietary components selected on the basis of their reported associations with gut microbial features. Multivariable Cox proportional hazards models were used to estimate associations of DI-GM with incident IBD, CD, and UC. We performed restricted cubic splines, subgroup analyses, sensitivity analyses, as well as competing-risk analyses accounting for death. Results During a median follow-up of 9.46 years, 1,170 incident IBD cases were identified. In the fully adjusted model, each 1-point increment in DI-GM was associated with a lower hazard of IBD (hazard ratio [HR] = 0.948, 95% confidence interval [CI]: 0.914–0.983; p = 0.004). Analyses using quartiles and predefined DI-GM categories showed consistent inverse associations with significant dose–response trends. Restricted cubic splines suggested an approximately linear inverse association. In general, associations were consistent across subgroup and competing-risk analyses and most sensitivity analyses. Subtype-specific analyses displayed inverse associations with both CD and UC when DI-GM was modelled continuously, with more consistent categorical findings for UC. Conclusion Higher DI-GM scores were correlated with a lower hazard of incident IBD in this large prospective cohort. The above-mentioned findings provide prospective epidemiological evidence for an association between DI-GM and IBD development but do not confirm causality or microbiota-mediated mechanisms.