Cheap antidepressant shows promise treating early Covid-19

Post by SkittyDog » Thu, Oct 28, 2021, 4:37 AM UTC

darkerside wrote: post #19

SkittyDog wrote: post #12

Earlier quoted context omitted.

16% > 11% represents a 31% decrease in the absolute number of hospitalizations. That's nearly a ONE THIRD improvement in the hospitalization rate. In places where hospitals are at or near capacity, that would make a huge difference. Or were you trying to make a statement about the statistical validity of the result, because the outcomes differ by 5%? Because that's not how statistics work. A well constructed study of…

N = 1500 One sample had 79-86 out of 750 hospitalized, and the other had 117-123 out of 750. Does that seem sufficient to demonstrate statistical significance? Edit: found the actual numbers in the study 79 [11%] of 741 vs 119 [16%] of 756

Post by darkerside » Thu, Oct 28, 2021, 4:50 AM UTC

pertymcpert wrote: post #9

blinkingled wrote: post #5

> In the group that took the drug, 11% needed hospitalization or an extended ER stay, compared to 16% of those on dummy pills. Is that really promising? Very close to margin of error if not within.

> The results, published Wednesday in the journal Lancet Global Health, were so strong that independent experts monitoring the study recommended stopping it early because the results were clear.

Post by rscho » Thu, Oct 28, 2021, 4:51 AM UTC

chmod600 wrote: post #20

SkittyDog wrote: post #12

Earlier quoted context omitted.

16% > 11% represents a 31% decrease in the absolute number of hospitalizations. That's nearly a ONE THIRD improvement in the hospitalization rate. In places where hospitals are at or near capacity, that would make a huge difference. Or were you trying to make a statement about the statistical validity of the result, because the outcomes differ by 5%? Because that's not how statistics work. A well constructed study of…

I don't know what the other post had in mind, but does seem like smaller effect sizes have a way of evaporating under scrutiny. One third is pretty big when looking at it from a group perspective, but not really that big from a personal perspective. If someone told me that taking a pill would reduce my chances of hospitalization by 5%, I would certainly be looking closely at the side effects.

Post by ekianjo » Thu, Oct 28, 2021, 4:53 AM UTC

darkerside wrote: post #19

SkittyDog wrote: post #12

Earlier quoted context omitted.

16% > 11% represents a 31% decrease in the absolute number of hospitalizations. That's nearly a ONE THIRD improvement in the hospitalization rate. In places where hospitals are at or near capacity, that would make a huge difference. Or were you trying to make a statement about the statistical validity of the result, because the outcomes differ by 5%? Because that's not how statistics work. A well constructed study of…

N = 1500 One sample had 79-86 out of 750 hospitalized, and the other had 117-123 out of 750. Does that seem sufficient to demonstrate statistical significance? Edit: found the actual numbers in the study 79 [11%] of 741 vs 119 [16%] of 756

Post by darkerside » Thu, Oct 28, 2021, 4:58 AM UTC

ekianjo wrote: post #28

darkerside wrote: post #19

Earlier quoted context omitted.

N = 1500 One sample had 79-86 out of 750 hospitalized, and the other had 117-123 out of 750. Does that seem sufficient to demonstrate statistical significance? Edit: found the actual numbers in the study 79 [11%] of 741 vs 119 [16%] of 756

Most drugs get approved with less strong effect than this