Coronavirus SARS-Cov2 (COVID-19)

[jayakris]

Apparently the State Bank of India also did a Covid data study and made a report. High Covid fatality in states linked to elderly population, peak a month away — SBI report (But experts refute this assertion).

An article that says a few reasonable things. The comments from Giridhar R. Babu (professor and head, lifecourse epidemiology, at the Public Health Foundation of India) at the end of the article are quite acceptable for us who think through this stuff. This was interesting:

“However, in a declining epidemic, test positivity rates will steadily decrease, followed by daily case counts, followed by daily death counts. We have to track all of these indicators, though the last of these three will be the most reliable trend indicator,” he said

Very nice way to put it and it makes perfect logical sense.

We have a YES on the first criterion. We have had a decreasing of positivity happening for 3-4 weeks though it leveled a bit lately (but not going up or anything).

The daily case count is still going up, so it's a NO on the second criterion.

The death count is not decreasing as a total (but it is decreasing in AP, TN, WB and maybe UP too). But the total also has been fairly flat for almost 3-4 weeks now, as you can see from the 7-day average death graph for India at worldometers. There is even a hint of a drop, though today we showed our max number from several days ago, around 1080 deaths. So it is a MAYBE on the third (and most reliable) criterion.

Keep the hope up that something good will slowly happen in the near future.

[jayakris]

And a short one-page world report in the Lancet magazine - Is India missing COVID-19 deaths? (Patralekha Chatterjee).

This also quotes Giridhar Babu who mentions villages where death registration is pretty low, but I also liked the Kerala virologist Rijo John's comment that over 65% of the deaths are anyway from 4 states where the death registration is 100% (MAH, TN, KAR and DEL).

---------------------

ICMR Update: 46,679,145 total tests... Tests on Thursday: 1,169,765... So we should expect around 1120K or so tests' results on Friday too. Lab count: +8 to 1631

[suresh]

Selected Numbers for Sept 4, 2020

AP +10776 from 59919 tests @18%
Kerala +2479
Delhi +2914 from 36218 tests @8%
TN +5976 from 81588 tests (79 deaths) 7.3%
J&K +1047
UP +6179 from 146601 tests @4.2%
Bihar +1978 from 150195 tests @1.3%
Karnataka ( +9280 (116 deaths)

[suresh]

This also quotes Giridhar Babu who mentions villages where death registration is pretty low, but I also liked the Kerala virologist Rijo John's comment that over 65% of the deaths are anyway from 4 states where the death registration is 100% (MAH, TN, KAR and DEL).

There was an article in the HIndu by two guys who claimed that the number of deaths is under reported by a factor 5. The main basis of that conclusion was that only 22% of deaths are registered in India. No one denies that there is under reporting. But it is totally wrong to use such a simplistic approach to a huge and heterogenous country., Kerala and UP are so different. Here is a nice discussion on why that factor of 5 is horribly wrong.

[jayakris]

Haha... 19218 cases in Maharashtra. Now I guess this is the unlock cases in MAH. What a joke.

[sameerph]

And looks like Mumbai is getting a second waive, 1929 cases today. It has gone up a lot in last 1 week from below 1000 where it was I think just about a week ago. And the way I am begining to see increase in crowds on roads and buses and all, the cases are going to increase.

[jayakris]

And looks like Mumbai is getting a second waive, 1929 cases today. It has gone up a lot in last 1 week from below 1000 where it was I think just about a week ago. And the way I am begining to see increase in crowds on roads and buses and all, the cases are going to increase.

Looks like. A month ago we were below +1000 a day and the positivity had dropped to somewhere in the 10-14% range. Now we are are consistently above 1500 and the positivity is in the 16-19% range. You cannot get the curve to stay low if the positivity does not come down below 8% or so because somewhere or other it is much higher and you are letting the disease spread. After the good numbers started coming from late July, Mumbai's testing actually went down a bit 3 weeks ago to 6K to 8K tests a day. Rather than take the foot of the pedal, they should have been testing 12 and 13K people a day (and the AG tests should've been for the areas with less disease prevalence where any hot-spot-identification or building-complex-based testing would be more difficult).

Just 5000 extra AG tests per day in Mumbai was all that we are talking about. Absolutely doable by ordering kits and making them available to the hospitals and clinics. You need no personnel to do it. People would go get tested if they felt anything. At least a portion of them would, and that's enough to quickly see where the disease is coming up and then to do some RT-PCR at building complexes, or a few street-side AG test centers. Not sure which parts of Mumbai has the growth now, but that is all fully allowed to develop by BMC and people like Ashwini Bhide who should've been on top of it.

Many parts of Maharashtra have had 2nd and 3rd waves. Just totally messed up in the whole state.

At least they have increased the testing this week from 70K max to 95K now in Maharashtra. It should be 200K but I will take anything for now! It's mind-boggling that Bihar and UP have been doing 125K+ per day for weeks now, but Maharahstra was testing 60K-75K and getting 3 to 8 times as many cases. Nobody is doing great contact tracing or anything in UP and Bihar. They just make AG tests available and monitor hot spot areas. That's it. Just do it. Stupid officials, and a hapless CM who has only stupidity all around.

[jayakris]

There was an article in the HIndu by two guys who claimed that the number of deaths is under reported by a factor 5. The main basis of that conclusion was that only 22% of deaths are registered in India. No one denies that there is under reporting. But it is totally wrong to use such a simplistic approach to a huge and heterogenous country., Kerala and UP are so different. Here is a nice discussion on why that factor of 5 is horribly wrong.

Thanks Suresh. The article by OmmenC. Kurien in reply to that junk in The Hindu, was fantastic! Repeating the link in the tweet you gave - From COVID19 pandemic to epidemic of spurious analysis
I need to do a to him. Right on point, and so well-written.

---------------

Meanwhile, for the record, we had +87.1K today. The tests were more too, as the states reported about 1148K... Positivity 7.6%... After dropping from near 12% to around 7.2% (three-day average) over a 5 week period, we have now inched up to around 7.6% in the last 8 days, as Chhattisgarh, Jharkhand and Maharashtra keep adding serious plus numbers every day.

[Atithee]

I had posted this yesterday or so I thought. Not sure what happened. Some facts on people not getting tested that I believe is keeping the test numbers down. Note the excerpt about Mumbai.

Don’t hide Covid symptoms: 31% die in 24 hrs of admission, reveals Mumbai civic corporation data
https://www.hindustantimes.com/cities/d ... GPxpM.html

When HT spoke to health volunteers, they complained that often people hide information about their symptoms during inspection. Fearing quarantine, people also take fever-reducing medicines. “When we try to do door-to-door inspections, especially in buildings with infected patients, people even refuse to come out for thermal scanning. If they have aged people, they don’t even tell us. Then we have to check the voters’ list to find them,” said a health volunteer from R-North ward that covers Dahisar. Dr Supe laid emphasis on people’s cooperation.

[jayakris]

^^^ I assume this happens everywhere, not just in Mumbai. BMC should be ashamed of themselves for blaming people for this. If they had made AG tests available, a whole bunch of these people would take them before things are too late. 31% dying within 24 hours. It is unacceptable, and I bet it is no more than 10 or 15% in other places. We knew all along that the reason for the much higher death rate in Mumbai was the late admission of patients and not anything lacking in the doctors and hospitals compared to those in Malappuram or Dhanbad or Udaipur. Patient admission is late entirely because they got tested too late too.

--------

ICMR update: 47,738,491 total tests... Friday tests: 1,059,346... Less than the last couple of days. The states may still report something near 1.1M, I guess... Lab count: +13 to 1643

[suresh]

Oommen C Kurian raises the issue that I have brought up in this forum. Are TN's numbers cooked?

[jayakris]

Oommen C Kurian raises the issue that I have brought up in this forum. Are TN's numbers cooked?

I don't know if TN was lying, but they were clearly withholding data and releasing it slowly over a period or something like that. It has been in the 5900s for way too many days. And Oommen's statement that the testing went up is also not fully correct. It did not go up much. His graph itself shows it. But that actually adds strength to his conclusion that there may be cooking going on. It is possible to withhold data on one of the two things (either testing numbers or the case numbers) and show it to be flat, but I don't know how they can make both of them flat!

[jayakris]

Note the excerpt about Mumbai.
When HT spoke to health volunteers, they complained that often people hide information about their symptoms during inspection. Fearing quarantine, people also take fever-reducing medicines.

Just saw this tweet
@ANI. 25m
To encourage testing for #COVID19, Punjab Chief Minister Captain Amarinder Singh today announced distribution of free food packets to poor families who do not want to get themselves tested for fear of isolation impacting their meager earnings: Punjab Chief Minister's Office

Maybe this will help. Get tested, get free food!

[suresh]

I don't know if TN was lying, but they were clearly withholding data and releasing it slowly over a period or something like that. It has been in the 5900s for way too many days. And Oommen's statement that the testing went up is also not fully correct. It did not go up much. His graph itself shows it. But that actually adds strength to his conclusion that there may be cooking going on. It is possible to withhold data on one of the two things (either testing numbers or the case numbers) and show it to be flat, but I don't know how they can make both of them flat!

I think that they are playing around with when positive results are announced. That one day spike when TN crossed was some kind of correction. So the fudging can't be huge. I don't quite understand the logic. The good news is that positivity rates are down and the number of deaths (can't be fudged that easily) are down.

@jay Did you see that ICMR has recommended that states do RAT's in containment areas. I hope this forces TN to do a rethink and add RA tests to their arsenal.

Link to current ICMR recommendation

A.Routine surveillance in containment zonesandscreening at points of entry:

Choice of Test(in order of priority):
i.Rapid Antigen Test (RAT) [as per attached algorithm]
ii.RT-PCR or TrueNator CBNAAT

1.All symptomatic (ILI symptoms) cases including health care workers and frontline workers.
2.All asymptomatic direct and high-risk contacts (in family and workplace, elderly ≥ 65 years of age, immunocompromised, those with co-morbidities etc.)of a laboratory confirmed case to be tested once between day 5 and day 10 of coming into contact.
3.All asymptomatic high-risk individuals (elderly ≥ 65 years of age, those with co-morbidities etc.)in containment zones.

*RAT for containment zone: Ideally,it is suggested that 100% people living in containment zones should be tested by RAT particularly in cities where there has been widespread transmission of infection

[jayakris]

@jay Did you see that ICMR has recommended that states do RAT's in containment areas. I hope this forces TN to do a rethink and add RA tests to their arsenal.

I saw that. ICMR was way too late on sending this out. Who knows if TN would listen!

Meanwhile, Maharashtra had 20.5K cases today. From 90K tests at 22.7% positivity. No comments. I'm out of words to complain. Not sure whom to yell at, anyway.