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 |
Some concerns |
Quote: "Randomization was generated using a computer random number generator. We used sequentially numbered sealed envelopes of identical appearance containing either hydroxychloroquine or placebo, ensuring allocation concealment." Comment: Allocation sequence random. No information on whether the envelopes were opaque, ensuring concealment. Any baseline imbalances appear compatible with chance |
| Deviations from intervention |
Low |
Quote: “Participants, investigators assessing participant eligibility and recruitment, assessing outcomes and follow-up, and/or dealing with data management and analysis were all blinded to arm allocation. Only one person unrelated to participant recruitment and follow-up, clinical assistance, data management and analysis had access to this information. Participants in the control group followed the same treatment schedule with placebo tablets that were indistinguishable from hydroxychloroquine tablets. Active surveillance of each participant was conducted monthly by blinded physicians unaware of the trial arm assignments.”
Comment: Blinded study (participants and personnel/carers) Data for the outcome were analyzed using intention-to-treat analysis. This method was considered appropriate to estimate the effect of assignment to intervention. Risk assessed to be low for outcomes: Adverse events. Serious adverse events. |
| Missing outcome data |
High |
Comment: 269 participants randomized, 269 participants analyzed for safety outcomes but "a total of 253 (94.1%) participants had completed the first month of follow up".
Data not available for all or nearly all participants. Reasons: HCQ arm - withdrawal of consent (n=1), adverse events (n=1), lost to follow-up (n=2), participant lost study drug (n=1). Reasons: Placebo arm - withdrawal of consent (n=3), adverse events (n=5), participant unable to receive drug as per protocol (n=1), SARS-CoV-2 infection (n=1), other (n=1). No evidence that the result is not biased (no sensitivity analyses, different reasons for missingness and imbalance of missing data). Consent withdrawal and inability to receive drug as per protocol are documented reasons unrelated to the outcome; SARS-CoV-2 infection is exclusion of post-baseline ineligible participants, hence not missing data. Due to the other reasons, missingness could and is likely to depend on the true value. Risk assessed to be high for the outcomes: Adverse events. Serious adverse events. |
| Measurement of the outcome |
Low |
Comment: Method of measuring the outcome probably appropriate.
Measurement or ascertainment of outcome probably does not differ between groups. Blinded study (outcome assessors) Risk assessed to be low for the outcomes: Adverse events. Serious adverse events. |
| Selection of the reported results |
Low |
Comment: The prospective registry, protocol and statistical analysis plan were available. The current protocol version1.5 is dated June 2, 2020 hence it is not prospective, thus the prospective registry was the main reference for this assessment.
Results were not selected from multiple outcome measurements or analyses of the data. Timepoints were different in the registry but this was due to early stoppage of the trail on the basis of a very low incidence rate among study participants. Trial analyzed as pre-specified. Risk assessed to be low for the outcomes: Adverse events. Serious adverse events. |
| Overall risk of bias |
High |