| Bias | Author's judgement | Support for judgement |
| Randomization |
Some concerns |
Quote from the protocol: "An electronically generated randomization list was prepared by an independent
statistician, with 110 blocks of 4 participants per block. This randomization list was
generated on R software (Package ‘blockrand’). The list will be accessible only to nonblinded
pharmacists in the study, in an attempt to minimize observation bias. Participants will
be randomized by the study pharmacist to their designated treatment regimen at the time of
inclusion and will be subsequently identified throughout the study only by their allocated
study number, always assigned following chronological order." Quote from the pre-print: "An electronically generated randomization list was prepared by an independent statistician, with four blocks of 110 participants per block. This randomization list associated each patient's study number with an opaque surface hiding the treatment group designation. The list was accessible only to non-blinded pharmacists in the study, in an attempt to minimize observation bias. Participants were randomized by the study pharmacist to their designated treatment regimen at the time of inclusion and were subsequently identified throughout the study only by their allocated study number, always assigned following chronological order. Unmasking was available to DSMB members in case of severe adverse events." Comment: significant baseline differences between groups in mean age, non-significant but possibly clinically substantial differences in diabetes and hypertension (see Table 1). |
| Deviations from intervention |
Low |
Comment: Unblinded study. There is no evidence of deviation from the intended intervention that arose because of the trial context’ (i.e., no cross-over; co-interventions administered are balanced between arms). A comparable proportion of participants in the two arms received glucocorticoids (2,5% in the high dosage arm and 7% in the low dosage arm). Participants were analyzed according to their assigned intervention. There was no imputation for missing outcome data (the trial was terminated early). |
| Missing outcome data |
Low |
Comment: 81 randomized/81 analyzed (see Table 2). Risk assessed to be low for the outcome: mortality. |
| Measurement of the outcome |
Low |
Comment: objectively measured outcome. Risk assessed to be low for the outcome: mortality. |
| Selection of the reported results |
Low |
Comment: The protocol was avaialble. The statistical analysis plan was not available. Data for the outcome Mortality are presented in the Results for day 13, but not for day 7 (pre-specified time point). Risk assessed to be "low" for the outcome: mortality. |
| Overall risk of bias |
Some concerns |