| Bias | Author's judgement | Support for judgement |
| Randomization |
High |
Quote: “Randomisation was done using an odd-even methodology applied to registration numbers, in a consecutive fashion of 1:1 ratio”
Comment: Quasi-randomization. A consecutive odd-even allocation suggests probably no allocation concealment. |
| Deviations from intervention |
Some concerns |
Quote: “open-label”
Comment: Unblinded study. No participant cross-over. No information on administration of co-interventions of interest: antivirals, anticoagulants, biologics and corticosteroids. Data were analyzed using intention-to-treat analysis |
| Missing outcome data |
High |
Comment: 82 patients randomized; 62 patients analyzed; 40 patients analyzed for outcome of interest.
Data unavailable for >5% of population. 18/82 patients randomized but not included because of prior symptom duration. 2/82 patients randomized not included because of insufficient data. The result is probably not biased (same reasons for missingness and balance of missing data). Only 20 patients in each arm tested for viral negative conversion with no information on how they were selected and reason for missingness, hence no evidence that the result is not biased. No information on whether missingness could or is likely to depend on its true value. Risk assessed to be high for the outcome: Incidence of viral negative conversion. |
| Measurement of the outcome |
Low |
Comment: Unblinded study
Viral negative conversion is an observer-reported outcome not involving judgement. Risk assessed to be low for the outcome: Incidence of viral negative conversion. |
| Selection of the reported results |
Some concerns |
Comment: The protocol, statistical analysis plan and registry were not available.
Risk assessed to be some concerns for the outcome: Incidence of viral negative conversion. |
| Overall risk of bias |
High |