Some of the data in this paper by Wu and colleagues (BMJ 2025;388:e082583, doi:10.1136/bmj-2024-082583, published 15 January 2025) have been updated. Disruptions to clinical service during the covid-19 pandemic affected the data verification process at some of the trial centres leading to inaccurate data being entered in the trial database. Although these errors were subsequently corrected, the results in the published paper were based on the outdated data. The article and PDF have since been corrected.
Survival data
Nine participants (9/1106) were considered as losses to follow-up in the database during the covid-19 pandemic. After the pandemic, survival data were obtained for these participants, and all analyses except the survival analyses were updated. After correction, the hazard ratio for the primary outcome (28 day mortality) changed to 0.97 (0.76 to 1.24) [not 0.99 (0.77 to 1.27)] and for 90 day mortality to 0.95 (0.77 to 1.17) [not 0.94 (0.76 to 1.16)].
Immunological data
The original dataset included errors for lymphocyte counts in 35 participants (35/1106; 3%). As a result, minor amendments have been made to supplementary table S8 and figure S2.
Data sharing
The data sharing statement has been revised.