Coronavirus SARS-Cov2 (COVID-19)

[rajitghosh]

There may be some other reason for false negatives. In Mumbai and Pune labs are taking 4-5 days the get the reports out. By then the virus in the swabs may have got killed and hence not got detected.
By the way Maharashtra deaths are under reported. Dr. Bhagwan Pawar reported some 99 deaths in Pune district yesterday and his death count is close to 11000 against 8600 in the state report.
Other districts may be under reporting also.

[jayakris]

^^ Yeah, I am sure MAH will have a reconciliation of at least 2000 deaths one of these days. CFR furing this second wave has been really low (at around 0.6%). Things coludn't have improved this much suddenly in Maharashtra where it was like 3 to 6 percent earlier.

[jai_in_canada]

Quick answer: Stay away from WhatsApp, and do not believe even a word of anything that comes on WhatsApp.

But seriously... it is true that mutations can cause RT-PCR to become less accurate. USFDA had said this too earlier this year. Not terribly bad but there could be a bit more of false negatives (Not sure how much, but instead of around 4% or so, initially, it may have gone to 10% or something).

Thanks, Jay.

Btw, as I mentioned when I came back to life on this forum about a month ago, I have been off of all social media for over a year. The misinformation and disinformation was making me sick. So I social media distanced and washed my hands of WhatsApp.

But once in a while now I've been succumbing to curiosity and checking in on my high school group. In there today there was a back and forth between a guy who is a scientist (well, a food scientist) and a medical doctor on the topic of false negatives. I shake my head over the vague statements that such highly qualified people (who should know better and do better) send out.

No wonder the pandemic of misinformation and fear is worse than the virus itself. Although the fear factor seems to be gone now.

Maybe natural selection is at work.

[jayakris]

In there today there was a back and forth between a guy who is a scientist (well, a food scientist) and a medical doctor on the topic of false negatives. I shake my head over the vague statements that such highly qualified people (who should know better and do better) send out.

Doctors are the least informed people on epidemiology. They know how this can be treated and all that, but don't take anything they say seriously when it comes to this virus' and pandemic's dynamics. This is mostly engineers' and economists' arena. They understand it better than the doctors.

[jai_in_canada]

But how is a given test result recognized to be a false negative? If the US CDC is claiming 4% originally and 10% now are false negatives, how are they counting that? Or is it just a WAG?

[jai_in_canada]

Doctors are the least informed people on epidemiology. They know how this can be treated and all that, but don't take anything they say seriously when it comes to this virus' and pandemic's dynamics. This is mostly engineers' and economists' arena. They understand it better than the doctors.

Thank you, Jay. Please come talk to certain members of my family. They think doctors are automatically experts in virology, epidemiology and vaccines. And I become a Trump supporting redneck from the backwoods of Arkansas if I said otherwise.

[jayakris]

But how is a given test result recognized to be a false negative? If the US CDC is claiming 4% originally and 10% now are false negatives, how are they counting that? Or is it just a WAG?

I suppose they have observed that a lot more people are turning symptomatic right after testing negative on RT-PCR than before. A false negative in the case of a totally asymptomatic infection will be impossible to establish. The 10% number was not from CDC. It was just something I pulled out of thin air, to say that RT-PCR is still pretty good in catching most of the cases even at 2 or 3 times inaccuracy. I think the inaccuracy will depend on the particular mutant, and they may not have good enough data, as a genome analysis for mutants is done only on a small fraction of the tested cases (and if the fraction of asymptomatic cases are more/less in the case of any mutant, any data based on how many turn false negative would be tougher to say).

[jayakris]

There seems to be fair fraction of UK and the double-mutant strains floating around now. Just saw a release from Jharkhand, for instance. They sent 52 positive cases for genome analysis and found 9 of them to have the UK strain and 4 to have the double mutant strain. But that is just one sample set from JH.

[jai_in_canada]

But how is a given test result recognized to be a false negative? If the US CDC is claiming 4% originally and 10% now are false negatives, how are they counting that? Or is it just a WAG?

I suppose they have observed that a lot more people are turning symptomatic right after testing negative on RT-PCR than before. ... I think the inaccuracy will depend on the particular mutant, and they may not have good enough data, as a genome analysis for mutants is done only on a small fraction of the tested cases (and if the fraction of asymptomatic cases are more/less in the case of any mutant, any data based on how many turn false negative would be tougher to say).

But couldn't the symptoms be due to the flu or sinus infection rather than Covid-19?
And regarding not having proper genome sequencing done for the mutants, wouldn't that potentially also result in false positives?
Which then makes me wonder, how can they tell a false positive from an asymptomatic positive?
Sorry if I'm being annoying like a 2 year old with my incessant questions.

[jayakris]

I don't know Hindi, so I actually didn't know what the heck our Govt's latest "mela" name from Delhi meant - "Tika Utsav". Apparently we have been doing a Tika Utsav for the last few days. I find that tika means injection. Okay...

So, if they do both Covishield and Covaxin, does that become a "Vaccine Tika Masala"?

[jai_in_canada]

There seems to be fair fraction of UK and the double-mutant strains floating around now.

India being colonized again by double mutants from the UK???!!! Jharkhand, huh? I guess it's the East India Covid Company!! History repeats!

[jayakris]

But couldn't the symptoms be due to the flu or sinus infection rather than Covid-19?
And regarding not having proper genome sequencing done for the mutants, wouldn't that potentially also result in false positives?
Which then makes me wonder, how can they tell a false positive from an asymptomatic positive?

I think in the proper studies on this, they have tested against other flu, sinus etc, because easy blood tests are there for those. As for telling the false positive from an asymptomatic positive, they can do properly selected samples where multiple RT-PCR's are done, and of course genome sequencing can give full information on the prevalanece of the specific virus in the body. Still there will be statistical questions, and they would rely on standard hypothesis testing and 90% confidence intervals and all that. Basically statisticians' work. But they have done some studies of these kinds in various places around the world duringthe past year.

[jai_in_canada]

Thanks, Jay. So obviously a lot more to tracking the pandemic than just testing and reporting the positives, negatives, recoveries and deaths. No doubt different places employ different quality standards to how they track positive results, negatives, false positives, false negatives, asymptomatic positives, recoveries, cause of death etc.

[jayakris]

And we almost hit 200K today.... It was +199.6K... Testing was not great either, as we did only 1660K (may be affected by yet another pointless celebration - Ugadi/Yugadi/Bihu/Bisu/Vishu today?)... So the positivity was terrible at 12.02% .... It is all pretty much out of hand everywhere. I do have a sneaky suspicion that we are not going to rise like this for more than another week or two. Maybe even as soon as in another 2-3 days, we may start slowing towards a peak. That is, we may hit an inflection point. But the rise will continue for some more time, and even the most optimistic scenario should possibly take us at least to 250K or 300K per day, IMHO. But what do I know! I am also generally only praying, and latching on to Punjab (and maybe Maharashtra) reaching inflection points.

The vaccinations were okay today, at 2.88M first doses. But generally we are running about 15% lower than the rate last week. Expecting only around 20M people to be done this week, as opposed to 23M last week. But the vaccinated people's total went over 100M today (114M doses).

-----------
ICMR Update: 262,003,415 total tests... Wednesday tests: 1,384,549... This need to go up. It is stuck right now at 1.4M by ICMR and about 1.7M reported by the states. Not enough.

[jayakris]

Lots of details on viral mutations and the Indian strains, in this article - Is the 'DESI' coronavirus variant behind India's Covid-19 boom? (India Today)