et al., Clinical Microbiology and Infection, doi:10.1016/j.cmi.2020.10.004 (preprint 8/15) (Peer Reviewed)
Retrospective study of HCQ use in 9 hospitals in the Netherlands, showing no significant difference in mortality with HCQ/CQ or dexamethasone. Late stage (admitted to hospital with positive test or CT scan abnormalities). 4 of 7 hospitals started treatment only after further deterioration. Short cutoff (21 days) - other studies have shown treated patient cases resolved faster and more control patients remaining in hospital at this time.
In the preprint, 58 of 341 control patients died. In the journal version, 53 of 353 control patients died.
Significant differences between hospitals - HCQ hospitals had significantly older patients with significantly more comorbidities. Non-HCQ hospitals were "tertiary academic centres" whereas HCQ hospitals were "secondary care hospitals". Residual confounding likely. This study compares overcrowded regular hospitals with undercrowded academic hospitals.
A subset of patients were excluded due to transfer to other hospitals. This introduces bias because patients in critical condition are not transferred. For examples, patients benefiting from HCQ treatment may have been transferred to the tertiary centres and excluded from analysis, increasing the percentage of critical cases in the secondary hospitals.
Among the seven (H)CQ-hospitals, the timing of start of (H)CQ treatment differed; three hospitals started at the moment of COVID-19 diagnosis, four started after diagnosis but only when patients clinically deteriorated e.g., when there was an increase in respiratory rate or increase in use of supplemental oxygen.
Most patients received CQ instead of the safer HCQ, receiving late treatment with CQ. Patients were given an initial dose of 600mg CQ then every 12 hours, for 5 days a dose of 300 mg, for a total of 3600mg CQ. This dose is likely to be toxic, see for example .
Authors mention a subset of hospitals started treatment relatively earlier, which seems like the most important area to analyze, but no results are provided.
Peters et al., 8/15/2020, retrospective, Netherlands, Europe, peer-reviewed, 21 authors.
risk of death, 9.0% higher, RR 1.09, p = 0.57, treatment 419 of 1596 (26.3%), control 53 of 353 (15.0%), adjusted per study.
Effect extraction follows pre-specified rules
prioritizing more serious outcomes. For an individual study the most serious
outcome may have a smaller number of events and lower statistical signficance,
however this provides the strongest evidence for the most serious outcomes
when combining the results of many trials.