Coronavirus SARS-Cov2 (COVID-19)

[suresh]

Yes, tons of information on that thread by Chinmay Tumbe. I think similar methodology was followed in all estimates (including those by Rukmini whose numbers he uses). That is why posted the link.

[suresh]

via @TNCoronaUpdate
#TamilNadu 15 June

• TN - 11,805
• Today's Deaths - 267
• Today's Tests - 1,70,961
* TPR - 6.9%

• Coimbatore - 1,563
• Erode - 1,270
• Chennai - 793

Big States:

KL - 12246 (TPR of 11.8% from 104,120 tests -- so testing has marginally gone up but this is not enough)
TN - 11805
AP - 5741
KA - 5041
OR - 3405
WV - 3268

Total: 41506 (MH to come)

Notes: 1. Kerala overtakes TN to take top position in number of cases. The good news is that TPR is coming down slowly. A week of 140K+ tests should do wonders. Just do random 40K RATs -- who cares if you catch a 1000 cases that way.
2. We will be around 64-65K for today -- maybe lower if MH does well.

[depleter]

5041 from Karnataka out of 132600 tests with a positivity of 3.8%. Now Maharashtra will be the key.

[suresh]

Big States: (I cannot edit my previous entry and putting in a new one)

KL - 12246 (TPR of 11.8% from 104,120 tests -- so testing has marginally gone up but this is not enough)
TN - 11805
MH - 9350
AP - 5741
KA - 5041
OR - 3405
WV - 3268

Total: 50856 (MH to come)

Notes: 1. Kerala overtakes TN to take top position in number of cases. The good news is that TPR is coming down slowly. A week of 140K+ tests should do wonders. Just do random 40K RATs -- who cares if you catch a 1000 cases that way.
2. We will be around 65K for today.

[jai_in_canada]

Basically, his compilation of numbers from multiple sources for the 4 states are pointing to something closer to what JIC was thinking (around 8 to 14 times the deaths) than what I was thinking (4 to 7 times). And some populous states like MP may have 30 times the deaths and all that. It is hard to believe, but as more and more such data come out, we have to believe it.

Jay, I think you are giving me credit for something that I did not do.

I did not even attempt to estimate the actual fatalities that have happened due to Covid-19 (compared to what has been reported) based on the excess deaths stats. The reason I refrained from doing that is because when I thought about the excess deaths during the pandemic, they could have happened due to one of the following several causes:
1) Deaths due to Covid-19 that were correctly recorded as such.
2) Deaths due to Covid-19 but were ascribed to a different illness, either due to the existence of multiple other morbidities that made it difficult to ascertain the true cause of death and was therefore not assigned to Covid-19, or due to negligence/laziness/incompetence on the part of the hospital, or because the person had died at home and the relatives did not want to report it as a Covid death due to the stigma.
3) Deaths not directly due to Covid-19 but due to unavailability of treatment (for cancer, heart disease, diabetes etc.) due to medical facilities being overwhelmed with Covid-19 cases.
4) Deaths not directly due to Covid-19 but due to public health measures to control the disease -- mainly suicides.
5) Deaths due to other epidemics occurring at the same time e.g. dengue fever.

I could not figure out how to parse out the above categories to arrive at a more accurate estimate of excess fatalities directly due to Covid-19 i.e. items 2, 3 and 4. So I did not try to estimate that.

The only thing I did was try to estimate what the reported cases would have been if India's testing regimen had been as penetrating as those of countries that have done a better job of testing. My estimate was that India would have reported about 4 times more cases. Again, this is NOT the number of people who have had the disease as in seroprevalence. This is just what would have been reported if India had tested on par with a France (at 1.1 million tests per million) or the UK (at 2.8 million tests/million). Since India has reported 29 million cases to date, I feel that greater test penetration would/should have resulted in the reported number being 120 million. This is of course, a huge absolute number that would have scared the world. But looking at it as cases/million, it would still have meant that India's cases/million was comparable to the UK (66,000 cases/mil), France (87,000 cases/mil) and Brazil (81,000 cases/mil); and lower than the US (103k cases/million).

The only difference of opinion that I had with you was on the reported cases. You thought India had reported only about 50% of what it could have if it had tested like the advanced countries. I thought India had reported only 25% of what it could have if it had tested like the advanced countries.

I had no commentary on estimating deaths.

[Atithee]

The rule of thumb in India even in certain times is 10x for number of deaths. This has to be higher than that. We don’t need a lot of research. I think it’s close to 15-20 in Corona related deaths.

[jayakris]

Basically, his compilation of numbers from multiple sources for the 4 states are pointing to something closer to what JIC was thinking (around 8 to 14 times the deaths) than what I was thinking (4 to 7 times). And some populous states like MP may have 30 times the deaths and all that. It is hard to believe, but as more and more such data come out, we have to believe it.

Jay, I think you are giving me credit for something that I did not do.
...
I had no commentary on estimating deaths.

I know, Jai... The estimates I was attributing to your judgment is really only a corollary of your case estimates, with some reasonable CFR estimates for hospital and non-hospital deaths that I used at my end! But your case estimates are not that far off-the-chart on the implied death count, as I initially thought. The reality may be up to 8 or 10 times the deaths we have seen reported, not as low as 4 or 5 like I thought. If so, three times the case numbers you estimated is possible too. It just would be hard to believe for me, with the TPR we have (no country that has our TPR seems to have that kind of death rates). But we are both within the realm of possibilities, when it comes to "true case total" estimates.

[jayakris]

I have been stuck in some meetings and have not done the state-wise case update for today. Will do it a bit later...

[jayakris]

So we had +62.2K today (Ladakh outstanding)... From about 1991K tests at 3.12% ... The seven-day average dropped by 4.3K to 77.7K... The deaths total dropped to 1469 (excluding the 1070 earlier deaths in MAH)

Vaccinations went fine today, at around the same rate as last Tuesday... 2.6M by 7 pm (2.2M first doses).

---------------
ICMR Update: 383,306,971 total tests... Tuesday tests: 1,930,987

[Kumar]

That is a much smaller rise on Tuesday compared to last week.

[jayakris]

^^^ I was also a bit surprised at the drop in the case total today. I thought it would be some 5K more.

[suresh]

^^^ I was also a bit surprised at the drop in the case total today. I thought it would be some 5K more.

This was a pleasant surprise for me to wake up to. The numbers from the non "Big states" were not going down fast enough. It turned out to be under 12K yesterday. Assam overtook Odisha in the number of cases -- Assam is testing a lot and TPR is under 3%. It is more a case of Odisha numbers going down.

[jayakris]

I had forgotten to go back and update the weekly vaccination stats at the end of last Saturday. Just did that. Here is the link.

[suresh]

One more on the death count but on a positive note. Way to go Jharkhand.

[jayakris]

^^ Now, if only they would do something on vaccinations.... I guess UP, Bihar, Jharkhand just do not have the system to do that at a high rate.