Challenges in Results Robustness of Trials with Missing Data for the Primary Endpoint: Insights from Coronary...

To assess the consequence of different degrees of missing primary endpoint data for randomized controlled trials and to find the influence factors.

PubMed, Cochrane Library, EMBASE and ClinicalTrials.gov were searched up to Nov 30, 2023. We included trials of the drug-coated balloon/drug-eluted stent with angiographic outcomes as the primary endpoint. The tipping-point analysis was used to deal with the missing data for the primary endpoint. The inconsistency rate, tipping-point standardized effect size (SES) and tipping-point ratio were used to assess the result robustness.

A total of 101 trials were included, which had 109 trial comparisons. Among them, 89 (81.7%) comparisons had superior/non-inferior conclusions (H₀ rejected); 85 (78.0%) comparisons had a missing rate of ≥10%, and 30 (27.5%) comparisons had a missing rate of ≥20%. For H₀ rejected comparisons with a missing rate of ≥10%, the median of inconsistency rate, tipping-point SES and tipping-point ratio was 32.2% (IQR 19.7%, 45.4%), 0.90 (IQR 0.17, 1.79) and -1.53 (IQR -2.43, -0.39). A higher missing rate and a larger (worse) observed-target SES were associated with a more unreliable result.

A high dropout rate and inflated target effect size could cause an unreliable result. We emphasize a robust evaluation of the results for clinical trials with missing data for the primary endpoint.