3.6 | Univariable meta-regression and subgroup analysis
To identify the source of heterogeneity, we performed meta-regression analyses. Three study characteristics were used as covariates for univariable meta-regression analysis, including whether patients with a history of H. pylori infection were excluded, whether enrollment was consecutive, and whether a single reference standard for H. pylori infection was used (Supplementary Figure 5 ). The results showed that whether patients with a history of H. pylori infection were excluded significantly affected the sensitivity of RAC for the diagnosis of H. pylori infection (p<0.01). Likewise, whether enrollment was consecutive significantly affected the sensitivity of RAC (p<0.05). Thus, the heterogeneity could partly be explained by meta-regression analysis.