Association between diabetes mellitus and diverticulosis: a cross-sectional analysis

Diverticulosis is a multifactorial condition influenced by genetic, age-related and cardiometabolic factors. Given overlapping risk profiles, its association with diabetes mellitus remains of interest, although evidence is inconsistent.

In a retrospective cohort study of 5924 asymptomatic individuals undergoing colorectal cancer screening in Austria, participants were stratified by glycemic status and assessed for diverticulosis via endoscopy. Incidence rate ratios (IRRs) were estimated using Poisson regression with robust standard errors, sequentially adjusting for demographic and cardiometabolic variables. Interaction and sensitivity analyses evaluated effect modification by age, sex, and metabolic parameters.

The prevalence of diverticulosis increased with diabetes status: lowest in participants without diabetes (29%), intermediate in prediabetes (41%), and highest in diabetes (46%; p < 0.001). In fully adjusted models, prediabetes remained moderately associated with diverticulosis (IRR: 1.10; 95% confidence interval, CI: 1.01-1.21; p = 0.034), while the association with diabetes was attenuated (IRR: 0.98; 95% CI: 0.86-1.11; p = 0.743). Right-sided diverticulosis showed a weak residual association with prediabetes (RRR: 1.62; p = 0.012) but not with diabetes. Associations were modified by age, sex, body mass index (BMI) and hypertension, with stronger effects observed in younger, female, non-obese, normotensive individuals.

The association between (pre)diabetes and diverticulosis is largely attributable to shared metabolic risk factors. Observed signals for right-sided diverticulosis should be considered exploratory and hypothesis-generating, warranting further studies. Given the cross-sectional design, causality cannot be inferred and our findings should be interpreted as associations only.