Most of these are retrospective and observational, and they fail to answer the question posed by the very controversial (and flawed) Raoult paper: that is, that

Most of these are retrospective and observational, and they fail to answer the question posed by the very controversial (and flawed) Raoult paper: that is, that early administration during the course of the disease provides a clinical benefit. Also in many the control is standard of care, which can mean everything and nothing. The positive studies suffer from the same issues, by the way. Especially controls are pretty bad. These studies can't give the answers that a proper randomized controlled trial (against placebo), like the U of Minnesota one (but there are a few others) can. Some studies focus on mortality, but I doubt HCQ helps there: like remedisivir, the measure should be time to symptom clearance / hospital discharge. Time of administration also matters a lot (see again the NEJM paper on lopinavir / ritonavir, a failure, then the Lancet paper that gives more insight at earlier time points). One study I've read gave HCQ at 16 days post symptom onset (median): that time is definitely too late for any antiviral therapy. I reiterate: I have no personal interest in having HCQ succeed. What I want to see is good science.