A recent systematic review and meta-analysis by Shiha et al, synthesising 25 studies and nearly 50 000 participants, estimated that approximately 1 in 10 individuals self-reported gluten or wheat-related symptoms.1 The associations with female sex, psychological distress and irritable bowel syndrome (IBS) were clinically significant. However, the limitations acknowledged by the authors are essential for interpretation and should inform future research.

First, pronounced heterogeneity should be interpreted as a clinical signal rather than solely as statistical variation. The pooled self-reported non-coeliac gluten/wheat sensitivity (NCGWS) prevalence of 10.3%, accompanied by substantial between-study heterogeneity, likely reflects residual confounding, diagnostic variability, geographical and cultural differences and genuine biological diversity.1 Consequently, self-reported NCGWS is unlikely to represent a single epidemiological entity. Differences in definitions, survey tools, exclusion criteria and rates of medical evaluation make it difficult to directly compare countries or draw firm causal conclusions from subgroup analyses. NCGWS constitutes a composite…