Note: The risk of bias by domain corresponds to the highest risk of bias among outcomes by domain.
The overall risk of bias corresponds to the overall highest risk of bias assessed among outcomes.
| Bias | Author's judgement | Support for judgement |
| Randomization |
Low |
Quote: "Randomisation was conducted as a manual randomisation scheme using SAS 9·4 PROC PLAN to generate random allocation schedules. The randomisation schedule was uploaded through an online web-based system, in a 1:1 ratio to receive standard care (control group) or standard care plus the favipiravir and hydroxychloroquine combination, with a randomisation block size of eight. It was stratified by enrolling centre and by being admitted to ICU or on mechanical ventilation. The sites access the randomisation system and could not know the sequence of the treatments. The investigator disclosed the arm only after all information regarding patient enrolment was recorded in the online system." Comment: Allocation sequence random. Allocation sequence concealed |
| Deviations from intervention |
Some concerns |
Quote: “The trial was open-label”
Comment: Unblinded study (participants and personnel/carers) Deviations from intended intervention arising because of the study context: No participant cross-over. Administration of co-interventions of interest: corticosteroids were reported and balanced between arms (111 Vs 114). Biologics were reported and not balanced (8 Vs 15). Antivirals were mentioned but no figures reported. Administration of biologics was not balanced and could affect the outcome. Nevertheless, this domain was rated as some concern as it is impossible to distinguish deviation because of trial context and deviation because of intervention effect. MORTALITY, CLINICAL IMPROVEMENT, WHO SCORE 7 AND ABOVE. Our analysis for the binary outcome is an intention-to-treat analysis. This method was considered appropriate to estimate the effect of assignment to intervention. Risk assessed to be some concerns for the outcomes: Mortality (D28). Mortality (D60 or more). WHO score 7 and above (D28). Clinical improvement (D28). TIME TO DEATH, TIME TO CLINICAL IMPROVEMENT Participants were analyzed according to their randomized groups for the outcome. Of note, 7 vs 7 participants were excluded from the analysis post-randomization. 5 in the intervention arm Vs 2 in the control arm withdrew consent (missing data accounted for in domain 3). 2 in the intervention arm Vs 5 in the control arm were randomized by error (not eligible). This method was considered inappropriate to estimate the effect of assignment to intervention for this outcome. There was probably no substantial impact of failure to analyze participants according to their randomized groups. Risk assessed to be some concerns for the outcomes: Time to death. Time to clinical improvement. |
| Missing outcome data |
Some concerns |
Comment: 268 participants randomized; 254 participants analyzed for clinical improvement and WHO score 7 and above; 245 analyzed for mortality (D28); 229 analyzed for mortality (D60 or more).
CLINICAL IMPROVEMENT, TIME TO CLINICAL IMPROVEMENT, WHO SCORE 7 AND ABOVE Data available for all or nearly all participants randomized. Of note, withdrew consent (5 vs 2), randomized in error (2 vs 5). Risk assessed to be low for the outcomes: Clinical improvement (D28). Time to clinical improvement. WHO score 7 and above (D28) MORTALITY, TIME TO DEATH Data not available for all or nearly all participants randomized. No evidence that the result is not biased. Reasons: withdrew consent (5 vs 2), randomized in error (2 vs 5), lost to follow up for day 28 results (6 vs 3), lost to follow up for day 28 results (8 vs 8). Missingness could depend on the true value of the outcome. Not likely that missingness depended on the true value of the outcome (similar reasons and proportion of missingness between arms). Risk assessed to be some concerns for the outcomes: Mortality (D28). Mortality (D60 or more). Time to death. |
| Measurement of the outcome |
Some concerns |
Comment: Method of measuring the outcome probably appropriate.
Measurement or ascertainment of outcome probably does not differ between groups. Unblinded study (outcome assessor) MORTALITY, TIME TO DEATH Mortality is an observer-reported outcome not involving judgement. Risk assessed to be low for the outcomes: Mortality (D28). Mortality (D60 or more). Time to death. WHO SCORE 7 AND ABOVE For WHO score 7 and above, we consider that the assessment cannot possibly be influenced by knowledge of intervention assignment. Risk assessed to be low for the outcomes: WHO score 7 and above (D28). TIME TO CLINICAL IMPROVEMENT, CLINICAL IMPROVEMENT Clinical improvement (time from randomisation to an improvement of two points (from the status at randomisation) on a seven-category ordinal scale) requires clinical judgement and could be affected by knowledge of intervention receipt, but it not considered likely to in the context of a pandemic. Risk assessed to be some concerns for the outcomes: Clinical improvement (D28). Time to clinical improvement. |
| Selection of the reported results |
Some concerns |
Comment: The protocol and statistical analysis plan (retrospective, 11 Aug 2020) and registry were available (prospective, May 19 2020).
Mortality (D28), Mortality (D60 or more), Clinical improvement (D28) and WHO SCORE 7 and above were not pre-specified. Time to Clinical improvement specified in registry with different time frame (Day 28). No information on whether the result was selected from multiple outcome measurements or analyses of the data. Trial probably not analyzed as pre-specified. Risk assessed to be some concerns for the outcome: Mortality (D28). Mortality (D60 or more). Clinical improvement (D28). Time to clinical improvement. WHO score 7 and above (D28). |
| Overall risk of bias |
Some concerns |