I last wrote about ivermectin here, but I’m getting so many question about it that I need to revisit the topic. Although (I’ve said this before), I believe that I will regret doing it, because I expect the signal/noise in the comments section to degenerate to mid-pandemic levels in response.
The mechanistic story here has always been confused, but to be honest, that doesn’t bother me too much. There are a lot of effective drugs whose exact mechanisms we’re unclear about. But keep in mind that if you argue in favor of ivermectin because of its antiviral activity in cell assays, that these levels are far off of what is reached in the reported clinical effects (when there are any – see below). You can’t have both of those arguments working at once: if you build your case on the in vitro results, then you need to regard most of the clinical data as having been dosed at far too low a level to be relevant. I’m not interested in fighting about the mechanism of action, though – the real question is, does it work? If it does, we can figure out how it happens later.
My current opinion is pretty much exactly that of the WHO guidance: I do not think that the current evidence is strong enough to say that ivermectin is a useful therapy for coronavirus patients. I know that there are quite a few studies out there in the literature, but they suffer from various combinations of small sample size, poor trial design, not enough data reported, and (in many cases) inconclusive statistics. I think that WHO page does a solid job of evaluating the literature to that point, and overall, the better the quality of the evidence, the more it tends to show little or no effect of ivermectin.
Since that recommendation in March, nothing has appeared that changes my mind about that. This study from Egypt compared two groups of about 80 patients in an open-label design, which is certainly not ideal. But it failed to show any statistically significant differences between the treatment group and the controls. This study from Lebanon was more positive: it looked at fifty asymptomatic SARS-CoV-2 positive patients who received a single dose of ivermectin versus fifty asymptomatic age-matched positive controls. The treatment group showed a statistically significant change in cycle threshold when tested by PCR for viral load, indicating a lower viral load. But that goes against the earlier evidence (see the WHO page) that ivermectin treatment had either no effect on viral clearance or (in some cases) lengthened the time needed. This study from Iran was double-blinded, but had only 35 patients in each group. The authors report a shorter duration of symptoms and shorter hospitalization in the ivermectin-treated group, but the statistics for the two groups still overlap, from what I can see.
Update: one of the more positive studies now appears to have fallen apart, and very badly, with strong evidence of faked data and poor controls. It appears that if this one is taken out, that some meta-analyses of ivermectin trials that have pointed towards possible benefit may no longer even do so. . .
And this paper, which appears to have come out in time to be included in the WHO guidance, is one of the larger studies. A team in Colombia looked at 200 ivermectin-treated patients and 200 controls with mild coronavirus infections, and found no statistical differences between the two. Objections have been raised to that trial’s use of an oral suspension formulation, I should note.
All in all, though, the most compelling reports of ivermectin’s effects seem to come from the smallest and least controlled samples (all the way down to anecdotal results) while the larger and more well-controlled trials tend to produce equivocal evidence at best. This very much reminds me of the hydroxychloroquine situation, which topic I have no desire whatsoever to revisit. A similar landscape of “the harder you look, the less you see” obtained there, too. And I have to say, there is a passionate constituency for ivermectin treatment, as there was for hydroxychloroquine. I hear from people who are convinced that this is the cure for the pandemic, and they are (variously) baffled that others don’t see it, zealous about spreading the word, or even ready to accuse the vaccine manufacturers and others of actively suppressing this treatment.
But as I said above, I look at the data and I’m not convinced, or certainly not yet (and neither are the WHO reviewers). The only way I can see the reaction of some of my correspondents is if they have been looking at all the most positive reports, accepting them completely and ignoring everything else, and that’s no way to treat the medical literature. This was the case for HCQ as well, I’m afraid. If you haven’t had to mess with drug discovery for a living, it’s understandable that you hear that Some Person Somewhere was very sick, took New Therapy X, and suddenly got better, and then assume that there it is, the cure has been found. But that’s not how it works. Real results stand up when you run larger, better-controlled trials, but most early results don’t turn out to be all that real. Even when this is your job, it’s frustrating to watch this happen, so I can only imagine how baffling it is if you haven’t seen this kind of evaporation before.
And as for the further bunch that are ready to go the conspiracy-theory route, well, as you’d imagine, I’m not having it. As usual, letting that stuff into your head simplifies everything enormously. Things get way too simple, actually. Everything bad is Their Fault, and you’re on the side of the good guys, the angels, struggling against the dark evil forces. It’s a bit like the pre-modern habit of thinking that made someone, every time they come down with some illness or problem, immediately wonder what witch or evil spirit did this to them. Someone has to be to blame, because nothing “just happens”. For the advanced conspiracy theorist, there are no accidents and there are no coincidences: things either directly support the all-encompassing theory, or they just show how the conspiracy is even bigger than it first appeared. It’s non-falsifiable – you can like your favorite conspiracy framework or you can love it, and those are your only choices.
So let me finish up by saying that my mind is not yet made up about ivermectin. I can be convinced by good data; if I couldn’t be, I shouldn’t be doing my day job at all. But I am not too optimistic – the data so far are consistent with a lot of other sorta-kinda-maybe-maybe not things I’ve seen over the years, where if you climb up on the right chair and hold your hand up to your face to block out the exact right stuff then things might look OK, but otherwise not so much. The good data that would dispel this are going to have to be really good, and the longer this goes on the less likely they seem.