| Bias | Author's judgement | Support for judgement |
| Randomization |
Some concerns |
Quote:"Double blind randomization was achieved by on-site physician choosing identically loaded syringes with saline or reconstituted BCG"
Comment: Allocation sequence probably random. No information on allocation concealment. Method of randomization unconventional, with no information on who prepared the syringes |
| Deviations from intervention |
Some concerns |
Quote: “Double blind randomization was achieved by on-site physician choosing identically loaded syringes with saline or reconstituted BCG”
Comment: There was no clear information on whether participants and providers were aware or unaware of the assigned intervention. Data were analyzed using available case analysis. All patients received co-interventions of interest such as corticosteroids, antimalarials and anticoagulants. 1 patient (in experimental group) was excluded from analyses due to receipt of Remdesevir upon ICU admission, while 4 patients in the control group who also received Remdesevir were included. |
| Missing outcome data |
Low |
Comment: 60 patients randomized; 59 patients analyzed.
4 patients only partially included in analyses because they withdrew or were lost to follow up. Risk assessed to be low for the outcomes: Mortality. |
| Measurement of the outcome |
Low |
Comment: Blinding uncertain.
Risk assessed to be low for the outcomes: Mortality. |
| Selection of the reported results |
Some concerns |
Comment: The study registry and protocol were available, but no statistical analysis plan.
Timepoint for the outcome Mortality in the registry and protocol is described as "Time From enrollment" while there is no clear timepoint reported in the manuscript. Risk assessed to be some concerns for the outcomes: Mortality. |
| Overall risk of bias |
Some concerns |