We thank Drs. Xu and Giang for their letter. 1 We performed a metaanalysis of randomised controlled trials (RCTs) of peppermint oil in irritable bowel syndrome (IBS) 2 ; hence, all eligible studies were RCTs.The risk of bias in studies was assessed according to the Cochrane risk of bias tool, 3 as recommended, and as we have used in multiple other meta-analyses in IBS. [4][5][6][7][8] We agree that there were high levels of heterogeneity in two of our analyses, a fact that we had acknowledged. We incorporated this heterogeneity, as well as possible publication bias, the risk of bias of included RCTs, and uncertainty around the effect, into our assessment of the quality of evidence by GRADE criteria. 9 Hence, we judged this as very low and stated that a future large negative trial, when pooled with the existing studies, may demonstrate that peppermint oil is not efficacious in IBS.Since we did not have access to individual patient data, examining the influence of factors such as age or IBS severity, which Xu and Giang suggest, would not be possible. We reported IBS subtypes of recruited patients in the table of included studies. High levels of heterogeneity persisted in our subgroup analysis of RCTs of small intestinal-release peppermint oil for global IBS symptoms (I 2 = 91%), and moderate heterogeneity persisted in the subgroup analysis of trials of small intestinal-release peppermint oil for abdominal pain (I 2 = 60%). We now report data examining efficacy for global symptoms or abdominal pain according to trial duration (≤4 weeks versus >4 weeks) in Table 1 below. Finally, we were uncertain why Xu and Giang felt it was inappropriate to pool adverse events data, which is surely of interest to clinicians and patients.