Coronavirus Sars-CoV-2/Covid-19 Megathread

[The Brain]

I'll send some care packages, J.

[Zanza]

Anecdotally, India is really bad now. Two of my Indian team members are Covid positive right now. 

[Jacob]

Anecdotally, India is really bad now. Two of my Indian team members are Covid positive right now. 

I think it's more than anecdotally at this point. The data says ~1.6 million new corona cases in the last week.

[Jacob]

I'll send some care packages, J.

Right now the Chinese government cannot arrest me, I don't think 

[Zoupa]

This covid wave in India is really bad and I hope the world can help India respond. But it's also a demonstration of how incredibly effective these vaccines are - see this table below:

Unless I'm missing something, how does that table tell you that the vaccines are effective?

You would need the number of positives in the placebo group to determine efficacy.

[DGuller]

For that matter, how can you tell anything from the table where digit separators are used haphazardly?

[Sheilbh]

For that matter, how can you tell anything from the table where digit separators are used haphazardly?

Aren't those just Indian numbers - lakhs and crores etc?

Unless I'm missing something, how does that table tell you that the vaccines are effective?

You would need the number of positives in the placebo group to determine efficacy.

That was the trial though there's no placebo group. This is just in the context of a huge second wave  - I think now at the highest number of reported new daily cases by any country during the pandemic. And there seems to be very limited effects for people who've been vaccinated - and far fewer serious ones such as Singh.

[Jacob]

Aren't those just Indian numbers - lakhs and crores etc?

Sure the convention is different, but that doesn't justify the absence of dividers in the Covishield row.

[Zoupa]

For that matter, how can you tell anything from the table where digit separators are used haphazardly?

Aren't those just Indian numbers - lakhs and crores etc?

Unless I'm missing something, how does that table tell you that the vaccines are effective?

You would need the number of positives in the placebo group to determine efficacy.

That was the trial though there's no placebo group. This is just in the context of a huge second wave  - I think now at the highest number of reported new daily cases by any country during the pandemic. And there seems to be very limited effects for people who've been vaccinated - and far fewer serious ones such as Singh.

There's no placebo group?  Sorry Sheilbh, but you cannot infer anything from those numbers then. No matter the context (huge second wave or not).

[Sheilbh]

I mean this is the context though:


And India from gene sampling have a lot of diferent variants. The ICMR's point was despite this there are very few cases among the vaccinated and those that exist do not seem to be developing into serious cases (including for the Indian developed covaxin which they've authorised for emegency use during its phase 3 trial).

I get this is mainly helpful for countering points around not getting vaccinated because there have been cases of people getting infected after both doses by showing that it's very rare.

More generally outside of clinical trials when you're trying to get everyone to get a vaccine and when it might lead to people's deaths to set up a placebo group - how do you study their effectiveness? Do you just look at the non-vaccinated?

Sure the convention is different, but that doesn't justify the absence of dividers in the Covishield row.

No - but it's what the Indian authorities have published - it's from their slide pack on a recent briefing

[Valmy]

You would think a country famous for its bureaucracy would know how to make a good slide.

[Admiral Yi]

There's no placebo group?  Sorry Sheilbh, but you cannot infer anything from those numbers then. No matter the context (huge second wave or not).

You can't definitively conclude anything, but you can tentatively conclude that the vaccines are ossum.  It would take some bizarre omitted variable bias to conclude these vaccines were useless.

[Sheilbh]

You would think a country famous for its bureaucracy would know how to make a good slide.

But I don't think it's famous for an overly efficient/competent bureaucracy

[alfred russel]

There's no placebo group?  Sorry Sheilbh, but you cannot infer anything from those numbers then. No matter the context (huge second wave or not).

You can't definitively conclude anything, but you can tentatively conclude that the vaccines are ossum.  It would take some bizarre omitted variable bias to conclude these vaccines were useless.

I don't think so...

Realclearpolitics says that India has 16,257,164 confirmed cases. The first result I see on google says they have 1,369,699,147 people, meaning 1.19% have tested positive. That means about 0.09% per month we've been going through this...

If they just started really ramping up vaccine numbers in India, and you assume a several week lag in testing reporting, it isn't obvious that 0.04% positive is really better than the overall population. It probably is - especially since covid is currently exploding in India, but the only thing to really conclude is the info provided is stupid.

[Admiral Yi]

I don't think so...

Realclearpolitics says that India has 16,257,164 confirmed cases. The first result I see on google says they have 1,369,699,147 people, meaning 1.19% have tested positive. That means about 0.09% per month we've been going through this...

If they just started really ramping up vaccine numbers in India, and you assume a several week lag in testing reporting, it isn't obvious that 0.04% positive is really better than the overall population. It probably is - especially since covid is currently exploding in India, but the only thing to really conclude is the info provided is stupid.

You need to compare like with like.

100% of the vaccine recipients have been tested.  100% of the general population has not been tested.  The comparison should be between vaccine recipients who have been tested and the general population who have been tested.