Concordance between randomized controlled trials and target trial emulation: A meta-analysis

Abstract Background Randomized controlled trials (RCTs) are the gold standard for causal inference but are often impractical. Trial emulation offers an alternative using observational data, yet concerns about its validity persist. This study assesses the concordance between Target trial emulation (TTE) and their corresponding RCTs and identifies design features influencing replication success. Methods We conducted a systematic review and meta-analysis of studies published from 2010 to 2025 that explicitly aimed to emulate a target RCT, and 82 TTE-RCT pairs were identified. Concordance was assessed through Pearson correlation coefficients and relative odds ratios (ROR). Factors associated with replication success were evaluated through regression and subgroup analyses, focusing on emulation design, population characteristics, and outcome definitions. Results Among the 82 TTE-RCT pairs, the Pearson correlation coefficient was 0.55 (95% CI: 0.38 to 0.69) and the summary ROR was 0.99 (95% CI: 0.94 to 1.03, I2 = 66%). In 38 pairs with closer emulation designs, higher agreement was observed (Pearson = 0.86, 95% CI, 0.77 to 0.92). Factors impairing concordance included differences in baseline age and sex, treatment initiation during hospitalization, and poor outcome emulation quality (p < 0.05). Subgroup analysis showed systematic underestimation of treatment effects for venous thromboembolism (ROR = 0.76, 95% CI: 0.59 to 0.98, I2=36%) and major adverse cardiovascular events (ROR = 0.93, 95% CI: 0.89 to 0.97, I2=19%). Conclusions Current real-world trial emulations demonstrate moderate concordance with corresponding RCTs. Improving emulation design, particularly in baseline population alignment, outcome ascertainment, and leveraging multi-source linked databases, can enhance replication success. Key messages • Current trial emulations moderately replicate RCTs, but better design significantly improves concordance. • Enhancing emulation methods can strengthen real-world evidence, supporting more reliable healthcare decisions when RCTs are not feasible.