Ready
a few things
first Read what you quoted: HOSPITALIZATIONS were significantly less. Unfortunately DEATHS were NOT
quote from the study: “was associated with significantly less hospitalizations and 5 times less all-cause deaths.”
Note the change, while they WERE able to show that hospitalizations were significantly less (odds ratio 0.16, 95% CI 0.06-0.5). they WERE NOT able to show statistical significance (odds ratio 0.2, 95% CI 0.03-1.5; p=0.16)
This is important. Say you have 1000 Covid patients divide them in two groups group A and Group B. You dont do anything different to the groups. You’d expect similar outcomes in both right? Is it IMPOSSIBLE that in 1 group all 500 live and the other group half die? of course not. It is very very unlikely but not impossible. Would you dream in that unlikely scenario that merely label a covid pt as “Group A” would lead o a better outcome? Of course not. you would correctly conclude that this was a fluke (assuming the 2 groups were in fact similar) .
with me so far?
So with any study. If the POSSIBILTY of “flukes” exist, how do you know that Group A who got the treatment did better in a statistically significant way? maybe it was a fluke?
S othats where P values come in. A p value calculates the odds that the findings are due to chance alone. A P value of < 0.05 is taken to be “statisticly significant”.
so with regard to hospitalizations the treated group had less hospitalizations thatn the untreated group. The p value was <0.001 this means that there is a less than 0.1% chance that thsi was due to chance alone. It is statistically significant (Though we will next have to examine if in fact it is just the HCQ that differentiates the groups).
However., with regard to DEATH. The P value was 0.16. this is greater than 0.05 that means the study does NOT show that anything other than chance alone casues any difference between the 2 groups.
Or to quote the studies discussion “Treatment with the triple therapy resulted in a numerically lower
rate of all-cause deaths. In the absence of clinical details about the untreated patient
group, THE LOWER RATE OF ALL-CAUSE- MORTALITY IN THE TREATED GROUP WAS NOT STATISTICALLY SIGNIFICANT” (emphasis added, I mean caps for emphasis not yelling)
Let me know if you are with me so far
I want to keep this in manageable chunks. The next step would be to look at the 2 groups, but first its important to understand the conclusion, the study did NOT show reduction in death among the treated group.