The study found a 61% lower rate of hospitalization in patients treated with Ivermectin compared to placebo -- but the authors didn't report that in their paper. You have to request the data from the authors' to see that.

The paper reports "Emergency department visit, hospitalization, or death" for ivermectin. The active group had 23/406 = 5.7%. The control had 16/394 = 4.1% the control was safer than the ivermectin group.

And it reports "Hospitalization or death". The active group had 4/406 = 1.0%. The control had 15/394 = 1.3% This was a reduction in hospitalization and death, but only a 23% reduction in rate, and a reduction of 0.3% which for the numbers in the study is about one person.

No result is signficant, and trying to double it by excluding the death as if that's better than hospitalization is disingenuous of you at best.

the authors consider only metformin results to be positive however the symptom results hidden in the appendix (Figure S4) show the opposite: the other two drugs, including Ivermectin, show faster improvement but no difference is seen for metformin.

I'm looking as figure S4, and it shows none of the expected symptom scores differ from the control by more than the error bars.

The subjects were instructed to take Ivermectin on an empty stomach, which is known to reduce the potency of the drug by nearly a factor of three.

That's not the recommendation I find with google. Can you link me to that information?

Ivermectin (Oral Route)

It is best to take this medicine on an empty stomach.

Late treatment: Ivermectin is known to work best with early treatment, as soon as symptoms start.

Is it? Can you link me to the research that finds that too?

The treatment arm for Ivermectin was three days of active medication followed by eleven days of placebo pills, compared to 14 days of active medication for the other two drugs being trialed.

That gives a dose over the three days of 1.2-1.4 mg/kg of weight. Poison control suggests individuals who have received more than 2 mg/kg be seen in a hospital..

So even one more day would be dangerous, and two would recommend hospitalisation for ivermectin overdose.

No specific antidote for ivermectin exists, so prehospital should be supportive. Coma is common in large overdoses and the airway should be appropriately managed.

I'm surprised that they got that dose through the ethics committee. It's cutting it pretty close to dangerous.

The source used in this study was rated 7th out 11 for quality in a blinded test.

Where the test was one dish with 6 worms in it, and the end point washow long it took to kill each worm. At least run it a dozen times in case the starting health of the worms or inconsistencies in per worm dosage causes due to the flow in the initial pipetting was the source of some of the variation.

Editorial bias: the study's authors found that Metformin was a promising treatment for Covid, but the journal editors changed their conclusion.

Getting the authors not to p-hack isn't "bias". The primary composite end point was defined. Metformin didn't achieve 95% confidence on that end point.

The three placebo groups had very different outcomes, which suggests that they weren't properly randomized, or some other confounding factor.

Which outcomes had the most statistically significant differences?

One of the primary objectives of the trial was to look at whether the three drugs improved viral load

I can't see that. Where do you get that from?

Confused conclusions in the paper: the authors report odds ratios which are notoriously difficult to interpret, even for scientists.

The confidence interval is the best thing to look at. If it spans an odds ratio of 1, the study doesn't show a significant effect.

This study is junk science.

I'm not convinced, but I look forward to your response to my queries.