Re:AstraZeneca may have provided incomplete efficacy data NIAID
by freebuju on Tuesday March 23, 2021 @11:00AM
(#26555837)
100% effective against hospitalization is 100% obvious BS. I don't need to be a statistician to make this conclusion.
Re:AstraZeneca may have provided incomplete efficacy data NIAID
by thehappypm on Tuesday March 23, 2021 @11:27AM
(#26556133)
The real-world data they have shows 100% protection compared to placebo in the trial. That's all the number means. Would you prefer they round down to 99% for you?
Re:AstraZeneca may have provided incomplete efficacy data NIAID
by freebuju on Tuesday March 23, 2021 @12:21PM
(#26556805)
I'd prefer they focus on doing research with actual data & figures that actually matter. And getting it right while at it.
IMO the 100% effective xxx was absolutely unnecessary. It becomes easy for anyone to tear down this report. AZ aren't helping themselves with such claims.
Also prefer they round off nothing. Report your data as-is. Majority of the people aren't stupid. They don't need your loose interpretations nor any nudges. Certainly not from the vax manufacturer.
Re:AstraZeneca may have provided incomplete efficacy data NIAID
by ceejayoz on Tuesday March 23, 2021 @12:21PM
(#26556810)
100% effective against hospitalization in the trial, because no one wound up hospitalized in the experimental group. You'd prefer they lie?
Re:AstraZeneca may have provided incomplete efficacy data NIAID
by freebuju on Tuesday March 23, 2021 @01:05PM
(#26557320)
I think they've already done so (worst case scenario) or they're just incompetent (best case scenario).
And if am not wrong, they should be publishing the final report based on this research.
Re:AstraZeneca may have provided incomplete efficacy data NIAID
by ceejayoz on Tuesday March 23, 2021 @02:23PM
(#26558370)
They submitted the analysis of the Phase III trials for publication in February. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3777268
> There were no hospitalisations in the ChAdOx1 nCoV-19 group after the initial 21 day exclusion period, and 15 in the control group.
You can disbelieve it, if you like, but you can hardly accuse them of not publishing it.
This is an illustration of why lay people applying "common sense" to data sets and statistics can be a problem.
The efficacy stats are based on small numbers, but with sound stats behind them. For example, from https://www.nytimes.com/interactive/2021/03/03/science/vacci...:
> What’s more, all the vaccines look as if they have a high efficacy against more serious outcomes like hospitalization and death. For example, no one who got Johnson & Johnson’s vaccine had to go to the hospital for a Covid-19 infection 28 days or more after getting an injection. Sixteen people who got the placebo did. That translates to 100 percent efficacy, with a confidence interval of 74.3 percent to 100 percent.
Expanding from the trials (with tens of thousands of people) to billions of people will mean that number can't stay at 100% forever in real-world use. It's still a great number to see.