Ivermectin, the latest supposed treatment for COVID-19 being touted by anti-vaccination groups, had “no effect whatsoever” on the disease, according to a large patient study.
That’s the conclusion of the Together Trial, which has subjected several purported nonvaccine treatments for COVID-19 to carefully designed clinical testing. The trial is supervised by McMaster University in Hamilton, Canada, and conducted in Brazil.
These are the opening two paragraphs of Michael Hiltzik, “Major study of Ivermectin, the anti-vaccine crowd’s latest COVID drug, finds ‘no effect whatsoever’,” Los Angeles Times, August 11, 2021.
This is good reporting but shoddy writing. I’m sure he’s right that ivermectin is “being touted by anti-vaccination groups.” Something he leaves out is that it’s also touted by some pro-vaccination people who want treatments for people who, whether or not they are vaccinated, get Covid.
He also writes:
The findings on Ivermectin are yet another blow for advocates promoting the drug as a magic bullet against COVID-19.
True, if the results hold up. They’re also a blow for advocates who promise the drug as an effective treatment against COVID-19 even if not close to “a silver bullet.”
Hiltzik also writes:
The study’s results on Ivermectin haven’t been formally published or peer-reviewed.
As I said, it’s good reporting.
UPDATE: Notice the comment by Charley Hooper below.
READER COMMENTS
Charley Hooper
Aug 12 2021 at 2:34pm
(1) I disliked the article in the LA Times. It’s very much “Trump’s people are wrong again.” Can’t we just discuss the scientific results without having to refer to tribalism?
(2) Notice that the results show that ivermectin did help patients. Relative risk: 0.91 (0.69-1.19). Mortality relative risk: 0.82 (0.44-1.52). In other words, ivermectin patients were 9% less likely to progress and 18% less likely to die. I know what comes next: “but it’s not statistically significant.” Yes, but the results are in the correct direction and a bigger trial might have shown statistically significant results. Plus, these results show a probability of 75% that patients on ivermectin did better than patients on placebo. The headline (“no effect whatsoever”) is just plain wrong. If I were sick and had to decide between ivermectin and placebo today, just based on this one trial, I would still go with ivermectin.
(3) This trial is valid only if it was well-designed and run. Was it powered at the right level? I’m not in a position to judge that. Plus, it would be nice to see it replicated.
(4) One of the tricks in medicine is knowing who should get a therapeutic and when. Perhaps some of these patient should have gotten ivermectin (they did well on it) and others shouldn’t have (they didn’t do well on it). Further analysis may show who benefited and who didn’t. Perhaps patients should have gotten ivermectin sooner or later than they did in this trial.
(5) Ivermectin holds promise in prophylaxis. Even if it fails in treatment, good reporting should point out that that’s not the end of the story for ivermectin.
RES
Aug 25 2021 at 1:28pm
Your point (2) is not right. If it is not found statistically significantly different, it cannot be claimed to be beneficial. Imagine you toss a coin 100 times, you are unlikely to get 50:50, but something in a range close to that because of random distribution of results. Therefore if you got 53 heads and 47 tails, this is not adequate grounds to claim that the coin is more likely to land heads. If you tossed that same coin another 100 times, you might very well end up with 45 heads and 55 tails.
Your points 3-5 are all valid criticisms. However, I feel that there is often a lopsided analysis applied: the unrealistic expectations for ivermectin (and HCQ before it) come from a failure to apply the same standards of skepticism to early, positive studies that were unfortunately very low quality (and in one significant case, probably fraudulent). Once the idea, maybe even “hype” is established, it is much harder to shift from our thoughts and risk that hope transcends reason.
From my scientific perspective, the overall picture for ivermectin is looking very poor. If the drug were going to have a substantial effect, it should be much clearer by now. Even despite the limitations of studies in number and quality, there are just too many studies finding it is ineffective to believe it can be a substantial benefit. Marginal, maybe; but my personal suspicion is that it will be a total bust, like HCQ. In fact, I suspect it was never a credible drug in the first place: that’s precisely why there was so little interest to investigate it.
Rebecca Menes
Aug 31 2021 at 8:41am
No — statistical significance is NOT the end of the discussion. This is a point that is maddeningly mis-understood.
In the absence of other information the result is still your new “best estimate,” it’s just a weakly supported estimate.
The correct (but little used) model is the “loss function.” No result is ever completely certain or uncertain, and the “significance” is only a shorthand, a rule of thumb, that assumes that the payoffs of type 1 and type 2 errors are symmetric. If they are asymmetric, then that changes the interpretation.
In this case, the payoffs are NOT symmetric. Ivermectin is one of the safest drugs we have, with a vanishingly small number of negative effects, and it is so cheap as to be essentially free. So the cost of giving it even if it doesn’t work is essentially zero.
On the other hand, the upside is significant — possible lives saved.
In this instance, a “significance” of 75% likelihood of positive effect is a good reason to administer the drug.
Charley Hooper
Aug 12 2021 at 2:46pm
Also, did the author of the LA Times piece think the success shown by Luvox (fluvoxamine) in the study didn’t merit mention?
Alan Goldhammer
Aug 12 2021 at 5:20pm
The ivermectin story is a total s**t storm and has been from the get go. The in vitro data that “supported” its use required 10 times the concentration that is achievable when given to humans. I’ve looked at all the trial data and they are worthless, underpowered, and studies done under questionable conditions. Alas, the search for a magic bullet against Covid will continue and those who have no clue about how difficult drug discovery really is will continue to tout the ‘cure of the day.’ There are still die-hard hydroxychloroquine people around who are convinced that the correct levels of zinc and azithromycin were not used in any of the trials. They refuse to give up.
I documented all of this last year in real time. It’s too sad to rehash fully here. Time to move on.
Ken P
Aug 12 2021 at 5:38pm
Good points Charley. If I was making a decision on similar data I would probably move forward with improved population targeting and and timing (pt 4) and shoot for improved power. Ivermectin was not disproved here.
MikeP
Aug 12 2021 at 9:07pm
Can’t we just discuss the scientific results without having to refer to tribalism?
No. No we can’t.
If there is one thing this whole disastrous overreaction to coronavirus has shown, it’s that the mainstream media has exactly zero interest in scientific results unless they are the results of their favorite scientists — actual scientific accuracy notwithstanding.
Tribalism sells. Science presented as religion suffices for the media’s needs.
Alan Goldhammer
Aug 13 2021 at 8:10am
I posted yesterday on this but it was either delayed or rejected, don’t know which. I will only note that ivermectin’s utility is not supported by the in vitro data that was used to support its use as an anti-viral for Covid. The serum concentration of the drug required to treat Covid can not be reached in the body. this was known over a year ago and is why most responsible clinical trialists did not pursue it. All of the studies that I reviewed were under-powered or of suspect quality.
[Your previous version of this comment was removed for using a four letter word. Using foul language violates our comment policies. You are a long-standing commenter here, and we should not have to spell that out for you. –Econlib Ed.]
Michael
Aug 13 2021 at 1:28pm
The underlying data can be found here:
https://dcricollab.dcri.duke.edu/sites/NIHKR/KR/GR-Slides-08-06-21.pdf
This is an adaptive clinical trial that is designed to allow the evaluation of many treatments. Patients entering the trial are assigned to one of several treatments (or placebo), and as some treatments are found to be effective or not, they are replaced in the trial so that additional therapies can be assessed.
The presentation looks at 3 therapies: metformin (approved for type 2 diabetes), ivermectin, and fluvoxamine (an SSRI antidepressant).
Metformin was dropped as worsening of Covid-19 was more common in the metformin group than in the placebo group. 32/217 (17.2%) of metformin group had extended ER obsefvation or hospitalization (the outcome in all the trials) versus 27/206 (14.5%) of the placebo group.
Ivermectin was looked at in a bigger patient group. 86/677 patients in the ivermectin group vs 95/678 patients in the placebo group experienced the outcome. So the difference between groups here, non-significant as discussed above and in the comments, was 9 patients out of 678.
Then there was fluvoxamine. 74/742 patients who got fluvoxamine, compared with 107/738 who got placebo, experienced the worsening outcome. That’s a difference of 33 patients of 742, or about 4.5%, nearly 1 in 20. That was a significant result: relative risk 0.69 (0.52 to 0.91).
I’m not sure why anyone would be interested in ivermectin and not fluvoxamine based on these trial results.
Michael B. Goodkin MD, FACC
Aug 13 2021 at 5:22pm
I don’t know the specifics of the protocol. With the delta virus, at least 0.3 mg/kg daily is needed. There is an obsession with randomized trial data. They frequently don’t predict what happens in real life. Keytruda looked great for non small cell lung cancer trials but when 20,000 real life medicare patients were evaluated it was barely better than chemotherapy. Remdesivir wasn’t very good in 3000 patients prior to EUA but in real life in 100,000 patients it was much better.
Uttar Pradesh, India, 210 million people started widespread use of ivermectin last July 2020. By December their mortality rate was 11 times less than the US. Mexico City, 22 million people, had profound benefit from ivermectin which its mayor publicly announced. The problem is that the press will not report on anything positive about repurposed drugs.
There is substantial data on ivermectin prophylaxis. Not letting unvaccinated people know they can be protected with 0.2 mg/kg once a week is criminal.
Insisting patients get no treatment instead of a cheap, safe drug that shows some efficacy is crazy.
NIH has done nothing with any repurposed drugs. It has actively hidden ivermectin. Now fluvoxamine will probably not get an EUA. It was turned down 1/22/21. They did nothing with famotidine and certrizine, over the counter, which stabilize mast cells, had several positive nonrandomized studies and prevent serious COVID infection in patients with mast cell disorders.
Had any of these drugs been shown to work, the vaccines would not have gotten an EUA. It was all intentional on the part of NIH to push vaccines and help drug companies.
Joe Munson
Aug 17 2021 at 2:56am
I’m seeing some studies that say its good, I’m seeing some that say its bad. I’m seeing Brett Weinstein interview experts that say its probably good (and he says its good).
At this point I dont know what to believe, though I guess I err on the majority.
Mike Maloney
Sep 11 2021 at 7:18am
Perhaps you should read this paper on nih.gov
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8248252/
Conclusions:
Moderate-certainty evidence finds that large reductions in COVID-19 deaths are possible using ivermectin. Using ivermectin early in the clinical course may reduce numbers progressing to severe disease. The apparent safety and low cost suggest that ivermectin is likely to have a significant impact on the SARS-CoV-2 pandemic globally.
Comments are closed.