Comparison of Intervention Effects in Randomized Controlled Trials Conducted in China and Other Countries: Evidence from a Meta-Epidemiological Study

<div>Background Randomized controlled trials (RCTs) conducted in China have increased rapidly in recent years, but concerns remain regarding their methodological quality. We aimed to investigate whether intervention effect estimates differ between RCTs conducted in China and those conducted in other countries using a meta-epidemiological approach. </div> <div>Methods We conducted a meta-epidemiological study using data from Cochrane systematic reviews published between 2011 and 2014. We selected meta-analyses of binary outcomes including at least three RCTs and at least one trial conducted in China. For each meta-analysis, we estimated the ratio of odds ratios (ROR) comparing intervention effects between Chinese and non-Chinese trials. Combined RORs were calculated using random-effects models. Subgroup analyses were performed according to intervention type and outcome type. Sensitivity analyses adjusted for sample size, risk of bias domains, and trial location. </div> <div>Results We included 68 meta-analyses comprising 897 RCTs, including 153 conducted in China and 744 conducted in other countries. Overall, there was no statistically significant difference in intervention effects between Chinese and non-Chinese trials (combined ROR 0.88; 95% CI 0.77-1.01). However, when trials conducted in Hong Kong and Taiwan were excluded, Chinese mainland RCTs showed significantly larger intervention effects compared with non-Chinese trials (combined ROR 0.80; 95% CI 0.66-0.97). </div> <div>Conclusion Overall intervention effect estimates were not significantly different between Chinese and non-Chinese RCTs. However, trials conducted in mainland China tended to report larger intervention effects than trials conducted elsewhere. Further research is needed to understand the methodological and contextual factors that may explain these differences. </div>