Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

India coronavirus: The 'mystery' of low Covid-19 death rates

bbc.com

Post reply on HN
Print view View original

361–370 of 381 posts

Re: India coronavirus: The 'mystery' of low Covid-19 death rates

#361

Post by klmadfejno »

somewhereoutth wrote: post #301
klmadfejno wrote: post #238

Earlier quoted context omitted.

It's simply not realistic to keep a highly infectious disease forever out of a country like India (unless the virus just doesn't thrive there for environmental or demographic reasons). What are they going to do, shut down entire cities every time there's a lapse in security? If the world did somehow divide into green and red zones, you would have a worse problem on your hands. The red zones are the wealthier western…

Land borders are of course harder to seal than air - but not impossible. Having to shut down entire cities if there is a breach will make their vigilance that much higher. The green zone will prosper, perhaps those wealthier western countries finding themselves in the red zone will no longer look so desirable. Of course I'm hoping it won't come to this, but it does seem to be the direction we are heading.

You've written green and red, and somehow assumed that green is good. Green means your population is vulnerable to an infectious disease that, in this universe, is highly prevalent in all of the wealthy countries. Long term its a weakness.

Re: India coronavirus: The 'mystery' of low Covid-19 death rates

#362

Post by klmadfejno »

ikiris wrote: post #278
klmadfejno wrote: post #260

Earlier quoted context omitted.

You can do this in a closed system. India is not a closed system. At some point, someone is going to screw up. You can't shut down a city every time that happens.

Evidence shows otherwise. You're even complaining about it.

You're right, that's fair, technically India can do this if it wants to forever isolate itself from the rest of the global economy and into greater levels of poverty than ever before.

I had assumed India didn't want to be a closed system.

Re: India coronavirus: The 'mystery' of low Covid-19 death rates

#363

Post by ksk »

kranner wrote: post #330
ksk wrote: post #314

Earlier quoted context omitted.

Where did you read that? Was it a reputable source that cited peer-reviewed data? We use RT-PCR every single day at our company (vaccine development) and we couldn't do our jobs if it performed so abysmally. The RT-PCR is the gold standard when it comes to detection.

https://pubs.rsna.org/doi/10.1148/radiol.2020200432 Sorry, I should have qualified that fully. I meant clinical sensitivity for RT-PCR for SARS-CoV-2 for samples drawn from throat swabs, specifically. I know that my local hospital is only taking throat swabs, not nasal swabs; and all the press photos I've seen for testing in India is of throat swabs.

Well, then that's not really the RT-PCR that's at fault, its the way the disease presents itself, that makes it difficult to find a reliable point of sampling. Although, I can see how in casual conversation one might say that the RT-PCR test isn't reliable.

Re: India coronavirus: The 'mystery' of low Covid-19 death rates

#364

Post by tptacek »

makomk wrote: post #245
jcranmer wrote: post #229

Earlier quoted context omitted.

No. Per capita, our testing is absolute crap--South Korea and Singapore are good examples of countries that actually have done thorough jobs in testing everybody. South Korea's positive test rates are somewhere in the single digits, the US is around 20-30%. That's not a sign we're "testing far more than most countries."

The US has actually been testing more people per capita than South Korea (or, indeed, most countries) for a while now - if I remember rightly, for about a month in terms of testing rate and a week or so in terms of total tests done. South Korea is running at about 6,000 tests per day right now with a peak capacity of about 20,000 last used a month or two ago. If scaled up to the US population that would correspond to…

That's become true recently (we're now neck and neck with South Korea), as Korea's case rate has plummeted; South Korea has essentially weathered this storm, or at least the first wave of it. But it's nowhere close to true compared to Italy; our per-capita test rate is half that of Italy's. And Italy's test positive rate is something like half that of ours; the point 'jcranmer makes --- that you can work out whether we're undertesting from our test positive rate --- stands.

Re: India coronavirus: The 'mystery' of low Covid-19 death rates

#365

Post by solarkraft »

segmondy wrote: post #338
CivBase wrote: post #124

Earlier quoted context omitted.

A lockdown is not supposed to reduce the total number of cases; it is supposed to "flatten the curve" so those cases take a lot longer to develop. The intent is to drag the virus out for as long as possible so we can prepare our hospitals, develop more effective treatments, and get closer to a vaccine. A lockdown does not make the virus go away sooner; it prolongs it . According to the Johns Hopkins dashboard[0], Ind…

Nah, a stay at home order is to flatten the curve. A lockdown is to burnout the virus. If you lockdown for about 2x the time the virus incubates and people follow the lockdown. You can snuff the virus out. At the beginning of the lockdown, assume someone just got it and lives at home with others no symptoms but they spread it to other's that live at home, assume those don't have symptoms and they stay in for the reco…

Snuffing out the virus would be a pretty impressive feat and it might just work. It might even be awesome for a while - but wouldn't you have to completely eliminate any potential vectors for the virus to get into the country again? Does that really mean virtually no immigration? Even with rigorous testing, if there's a single false negative, wouldn't that completely eliminate the gains of having gotten rid of the virus?

Re: India coronavirus: The 'mystery' of low Covid-19 death rates

#366

Post by pkphilip »

arihant wrote: post #150

This should not be a mystery at all. India locked down entire country (no state rights in Epidemics laws), despite federal structure. Very few things can be bought, even Amazon is only delivering "essentials." But more importantly, India has been contact tracing like a greedy algorithm on steroids since case #1. There was a positive case in a condo nearby mine. The entire condominium building and adjacent locality is…

Unfortunately, from what I have heard from my doctor friends, the government does lie about the numbers. Not all the state govts are doing this to the same extent but I have heard first hand information for doctors on the ground that the govt is hiding and misreporting numbers.

Re: India coronavirus: The 'mystery' of low Covid-19 death rates

#367

Post by rovolo »

legolas2412 wrote: post #351
rovolo wrote: post #298

Earlier quoted context omitted.

NYC certainly got hit bad, but I wouldn't say it's the only badly hit city. Here is deaths/total pop for some other cities >1mil population which I would also call "hit hard". NYC: 1.07‰ New Orleans: 0.65‰ Detroit: 0.60‰ Boston: 0.44‰ Source: https://www.nytimes.com/interactive/2020/04/03/upshot/corona...

I think you're off by a factor of 10 in your numbers

I'm using the per-mille (‰) rather than per-cent (%) symbol. Sorry for the confusion, but it's a more convenient unit for these quantities. This is how it'd look in percentages:

NYC: 0.107%

New Orleans: 0.065%

Detroit: 0.060%

Boston: 0.044%

https://en.wikipedia.org/wiki/Per_mille

Re: India coronavirus: The 'mystery' of low Covid-19 death rates

#368

Post by eloff »

sershe wrote: post #341
eloff wrote: post #312

Earlier quoted context omitted.

A lot of Asian countries are indeed showing that this is possible. That's good news! Good luck trying to contain the flu that way. SARS-COV2 seems to be very virulent, but not so virulent like flu or measles that it can't be contained.

Isn't R0 for COVID supposed to be much higher than flu? That was one of the major parts of the arguments for lockdown that I've seen.

It is, but paradoxically it also seems easier to control. Perhaps because while it is infectious through the air, it is maybe not infectious over the same distance - but not as infectious inside a close distance.

I'm just guessing really. People say flu can't be controlled in the same manner. But we also haven't done that experiment like we're doing now. So maybe the people who say that are just wrong.

Re: India coronavirus: The 'mystery' of low Covid-19 death rates

#369

Post by pkphilip »

ramraj07 wrote: post #18

I can give some ground level information. I live in a very densely populated street (total length is just two hundred meters, my guess is at least a thousand people live here). No one is dying here without others noticing. Further, there's a great amount of scrutiny into whether anyone's sick (see the reports of doctors getting beaten up even by the cops). As far as I know, no one has died in this street in a month.…

There are some numbers which simply don't make any sense.

Take today's data:

Delhi - 3,314 cases. (likely to go up as data is updated for tonight). Total population of Delhi is 19 Million.

Rajasthan - 2,438 cases. Total population: 68.9 Million

Haryana - 311 cases. Total population: 25.4 Million

Haryana has less than 10% of Delhi's covid case numbers while being a much larger state and while sharing a big border with both Delhi and Rajasthan where the covid cases are much higher. Also, huge numbers of people travel between Haryana and Delhi on a daily basis. Haryana's numbers make no sense.

Uttar Pradesh - 2,134 cases. Total population: 204 Million

About 30% less than Delhi while being much, much larger than Delhi and sharing a huge border with Delhi with vast numbers of people travelling to and from Delhi! Uttar Pradesh numbers too seem heavily under reported.

Bihar - 403 cases. Total population: 99 Million

Bihar has much, much lower covid cases than even Uttar Pradesh when Uttar Pradesh's numbers were themselves already suspiciously low. Bihar shares a huge border with Uttar Pradesh and there is simply no way Bihar with a much poorer healthcare infrastructure can have such low covid numbers.

Both UP and Bihar has huge numbers of migrants working in Delhi, Mumbai etc which makes these numbers even more dubious.

So on the face of it, there are big issues with these numbers being put out by some of the biggest states in India. If these numbers were more accurate, the overall figures for covid cases from India will be much higher.

India shares a massive border with China and is right next to China while having almost the same number of covid cases as Peru - which is halfway across the world from China and which is relatively shielded from the rest of the world in comparison to India - in terms of number of travellers coming in etc.

Re: India coronavirus: The 'mystery' of low Covid-19 death rates

#370

Post by legolas2412 »

rovolo wrote: post #367
legolas2412 wrote: post #351

Earlier quoted context omitted.

I think you're off by a factor of 10 in your numbers

I'm using the per-mille (‰) rather than per-cent (%) symbol. Sorry for the confusion, but it's a more convenient unit for these quantities. This is how it'd look in percentages: NYC: 0.107% New Orleans: 0.065% Detroit: 0.060% Boston: 0.044% https://en.wikipedia.org/wiki/Per_mille

Ahh, today i learned
Post reply on HN