I’d have to look them up again, as I don’t recall the percentages, but I do remember that the combination of chloroquine and azithromycin did result in a full cure in six days. Chloroquine alone took longer...again I don’t recall numbers, but definitely more than six days.
A Freeper supplied a link so I read through the paper.
You are correct that inclusion of azithromycin seemed to make a dramatic difference.
A few comments:
The authors of the paper noted that the experiment was not completely randomized. The control group came from places other than the treated group.
Also, six patients in the treated group, I think, were dropped after the trial began. Three of those went to an ICU.
The report that "100%" were cured evidently represents the six patients who received both the hydroxychloroquine and the azithromycin. Those with only the hydroxychloroquine did not recover as quickly as six days after "inclusion" which I assume means inclusion in the trial.
Probability values are provided in Table 3 in the paper. I'm no expert in calculating these values for controlled experiments. I believe they indicate the probability that the observed differences only occurred by chance. If p is 0.002 then that would indicate that only 2 times out of a thousand trials would one expect that the difference noted was the result of chance.
What's important to realize is that the low p value does not indicate that repetition of the study would show the same impressive results. Also, repeating the study with the same limited sample size won't necesarily give us a better understanding of how good the treatment might be.
Another way to look at the significance of such a study is to recognize that we have thousands of drugs on the market. If every study had a p value of 0.002, then that means one study out of every 500 would be incorrect. If we have 10,000 drugs on the market, then we would risk having 20 drugs that might have no benefit whatever, despite having a small study supporting them.