There isn't much good news coming out of India. Here's some: Mumbai's data is showing what looks like vaccines reducing fatalities. First bit of data: the fraction of deaths amongst the over-60s in Mumbai has been steadily dropping. Let's explore a bit. (1/7)

The effect is now very significant. Deaths in the over-60s this week are around 40% less than expected based on trends upto late March and deaths in under-60s. Also, the timing is about right: vax starts for the over-60s on March 1 - first hints of an effect by April 8.

In theory the shift could reflect:
- more recorded fatalities amongst younger people
- fewer recorded fatalities amongst older people
- some mixture of the two.
Which is it? Could more young people be getting severe disease for some reason?

You could get more young people with severe disease because of wider spread in the slums. Or some unknown property of a new variant. But: during April, delayed CFR (delay of 18 days) has been falling: fatalities today are ~30% *fewer* than we expect from cases 18 days ago.

The fall in delayed CFR has occurred without much fall in test positivity - it seems like a genuine drop in fatality rate.

So: the age-shift seems to be more about
fewer fatalities amongst older people,
and less about
more fatalities amongst younger people.

Could deaths amongst older people be going uncounted? Again, it is possible.

Ultimately, we can't be completely sure that the relative drop in deaths amongst the elderly is a vaccine effect. But we can be hopeful and keep looking at the data.

Finally, the virologists have some good news too.

We don't yet know for sure what variants are circulating in Mumbai. But whatever they are, I think the vaccines (mainly AZ in Mumbai) are effective against them. (7/7) https://twitter.com/3RakeshMishra/status/1385245798544207875