Elucidating the non-genetic risk factors for celiac disease: an umbrella review of meta-analyses

Abstract

The breadth and validity of the associations of nongenetic risk factors with celiac disease (CeD) are elusive in the literature. We aimed to evaluate which of these associations have strong epidemiological credibility and assessed presence and extent of potential literature biases. We systematically searched PubMed until April 2024 for systematic reviews and meta‐analyses of studies examining associations between putative risk factors and CeD. Each association was categorized in five evidence grades (convincing, highly suggestive, suggestive, weak, and not statistically significant) based on broadly used criteria for evaluating quality of evidence in observational studies. Five eligible publications were included, describing 15 meta-analytic associations on seven nongenetic risk factors, three of which were nominally significant (P < 0.05). None of the associations received a strοng or highly suggestive evidence. One meta-analytic association received suggestive evidence, namely any infections during childhood and adulthood for a higher risk of CeD (OR, 1.37; 95% CI, 1.2–1.56; P=3.77 × 10−6). Two meta-analyses reported weak evidence, pertaining to current smoking for a lower risk of CeD (OR, 0.52; 95% CI, 0.32–0.84; P=7.84 × 10−3) and use of antibiotics for a higher risk (OR, 1.2; 95% CI, 1.04–1.38; P 14.8 × 10−3). The rest of the meta-analyses did not report statistically significant results, and pertained to breastfeeding, time of gluten introduction, rotavirus vaccination, and cesarean section. No association of nongenetic risk factors for CeD received high levels of evidence. The evidence was suggestive for the association of any infections during childhood and adulthood with higher risk of CeD. More and prospective future research is warranted.