Reporting Quality Assessment of COVID-19 Clinical Trials in China at the beginning of the Pandemic: An Empirical Study based on the CONSORT Statement

Objectives: This study aimed to evaluate the reporting quality and risk of bias of Chinese COVID-19 RCTs. Methods: The Chinese clinical trial registry (ChiCTR) was searched for registered COVID-19 clinical trials before April 2, 2020. The reporting quality and risk of bias of included literatures were assessed using the Consolidating Standards of Reporting Trials 2010 (CONSORT 2010) and Cochrane risk of bias tool, respectively. Results: Of the 283 clinical trials identified, 172 RCTs were included. Some items including objectives, trial design, participants, interventions, outcomes and registration were presented in all RCTs, other items in limited RCTs, including blinding (11.63%), allocation concealment mechanism (14.53%), implementation (47.09%), title/abstract (54.07%), protocol (94.77%), sequence generation (97.09%), and funding (98.84%). The top three score among 25-item were background and objectives (10.00), registration (10.00) and funding (9.90), while the lowest four score all belonged to randomization part, including blinding (0.59), allocation concealment mechanism (0.88), implementation (1.43) and sequence generation (2.71). Among the RCTs, 75.58% showed low risk of bias in random sequence generation, while only 9.30% low risk of bias in allocation concealment. Besides, 39.53% presented with high performance bias and detection bias. Conclusions: The reporting quality of RCTs registered on the ChiCTR was significantly lower than expected, especially in randomization and blinding. Researchers are advised to refer to the CONSORT statement when designing a RCT.