Hacker News Reader

Comment by roelschroeven | Hacker News Reader

Even if that is true, it is missing the point. Avoiding infection is not just for your own good; more importantly, keeping the speed of the virus spreading down is for protecting everybody.

And what you say isn't even true. Plenty of healthy young people get very very sick. The percentages are lower than for older and/or less healthy people, but the numbers are still significant.

Don't think you're invulnerable, cause you're not. Don't think that if you get no or only mild symptoms, no harm is done. Think about the other people you have unknowingly infected, and how you've contributed to the spread of the virus. (Where 'you' = the generic 'you', not you personally).

View on HN

>> Don't think you're invulnerable, cause you're not. Don't think that if you get no or only mild symptoms, no harm is done.

Such statements were true a few months ago. Now 1 in 6 Londoners have had the disease. So those 1 in 6 know exactly how it will impact them because it already has. There are now more than a few people out there who can confidently say "I'll be fine" because they are fine. In coming days/months many people will be getting tests and learning that they have suffered and recovered from the disease at least once. Others will learn that they have been living with someone who has had the disease but it didn't spread even within the household. The days of shouting "you don't know, you might die" at every naysayer are behind us.

"With further statistical adjustments, they found about 17% of London adults had been infected; this compares to only 4.3% in the north-east of England."

https://www.bbc.co.uk/news/live/world-52749186

Your reply would be better if it bothered to address the other sentences in the paragraph you quote from.

It's incredibly tedious that you didn't.

It's fine if the reply would be about the same, the problem is that you don't address the meaning of the words you quote!

> Now 1 in 6 Londoners have had the disease.

Please do not spread misinformation. England has had 132k reported cases. London has a population of 8.9M.

"The top stories on Thursday: UK government data suggests around one in six people in London and one in 20 elsewhere in England have had coronavirus"

"With further statistical adjustments, they found about 17% of London adults had been infected; this compares to only 4.3% in the north-east of England."

https://www.bbc.co.uk/news/live/world-52749186

That's the _BBC_, perhaps the most respected news organization on the planet, and unarguably an expert on all things London. They are citing information released by the UK government: 1 in 6 Londoners. Misinformation my a--.

The BBC has a fairly dire reputation for it's scientific reporting

One of the things we have learned during this pandemic is that reputable sources are capable of publishing false information. So we can’t just take their assertions as facts. We have to ask them to explain how they arrived at them.

The claim that one in 6 Londoners have had the virus is particularly suspicious because Sweden made similar claims about infections in Stockholm based on rough estimates and they turned out to be wrong when actual studies were carried out.

It was the government figures, and it raised far more questions - London has an IFR of half that of the rest of the UK, does that mean it’s spread more widely amongst healthy people? Does the antibody year require a higher viral load (which you’d get spending a long time on the tube compared with driving to work in north east).

Need larger antibody tests and data broken down by demographic, globally, but we’ll have them by summer.

The BBC an expert in all things London? Maybe if you define away all the people in the city who aren't like BBC employees! Your view of the BBC is quite out of date. Most of the British population say now in polls they don't trust it. It's an open question for how long it'll stick around.

Thank you for the source. I only checked Google data and wasn't aware of the report.

> "England has had 132k reported cases. London has a population of 8.9M."

I personally know about a dozen people in London who have had probable Covid-19, including myself. None of those cases were ever actually confirmed by testing or "reported" to anyone.

Same here, and that's ignoring the asymptomatic cases, so you can at least triple/quadruple that number.

Can you? Citation needed, not heresay.

Helpful neighboring comments from hours before yours: https://news.ycombinator.com/item?id=23287505 https://news.ycombinator.com/item?id=23288767

Unless you want a source that many cases are asymptomatic, in which case you can easily google that yourself.

Until you actually have an antibody test result, be careful to equate "probably COVID-19" with "COVID-19". I have a friend who lives in the NYC metro area who was sure he had COVID-19 in March. All the symptoms matched, felt like he was dying for 6 weeks. Never tested, because tests were in short supply then, but his doctor said it was very likely COVID. He just took an antibody test and got the results yesterday. Negative for COVID-19.

Swine flu was going around this winter, which also will knock you out for a month. If you didn't specifically get a coronavirus PCR test positive, there's about a 10:1 chance that what you actually had was the flu. Outside of major hotspots like NYC and Chelsea MA, I believe the serological studies that suggest about 2-4% of the population has had coronavirus.

I can't believe that people are quick to tell me that the pneumonia I had in January couldn't possibly have been COVID-19, but they say that every death since January is COVID caused.

Who, exactly, says every death since January is caused by COVID-19? I cannot find any statistics which suggest that about 20 million people have died to coronavirus this year.

> but they say that every death since January is COVID caused.

I have no idea where you came up with this nonsense, but if anything there are academic studies demonstrating that, even in countries that do a honest job reporting covid19 deaths, Covid19 deaths have been systematically underreported.

Is that "deaths with COVID-19" or "deaths by COVID-19"?

You may be confusing this with excess deaths count, which is tracking the difference between deaths in a given month of 2020 and deaths in the same month in previous year (or years, averaged). Assuming no other unknown disease or phenomenon other than COVID-19 is silently ravaging our societies, this is the most reliable metric we have - as government after government has been caught on underreporting COVID-19 deaths.

But that's clearly a false assumption. It assumes hospitals can be emptied and society put under house arrest with no excess deaths as a consequence, which is not only a very strange assumption but contradicted by huge amounts of evidence.

There's by now a whole lot of people, including many experts in many domains, who are concluding the lockdown will kill more people than the disease itself. Far more if it turns out they had little impact, which there is also quite some evidence of.

You can filter out deaths by suicide and similar, of which there isn't many so far (but yes, maybe the lockdown will kill more; it hasn't now). You kind of already have to filter out things like deaths by traffic accidents, of which there's much less now.

Only 8% fewer traffic deaths in March: https://www.npr.org/sections/coronavirus-live-updates/2020/0...

And consider that heart attack and stroke centers are seeing precipitous drops in patient referrals, almost certainly meaning that the fear of catching COVID-19 is causing many people to avoid treatment and die at home: https://www.washingtonpost.com/health/patients-with-heart-at...

It’s not clear yet how reliable the antibody tests are. Antibodies fade over time, and we don’t know how long Covid-19 antibodies remain detectable in your blood. It’s quite possible that after 6 weeks or so, you’ll no longer have easily detectable levels of antibodies.

We also do not know the seroconversion rate for mild or asymptomatic cases (every study measuring this seems to use an entirely-hospitalized or mostly-hospitalized cohort).

EDIT: Hey hey it looks like some information is finally coming out, and what do you know, it supports the idea that milder cases don't develop IgG antibodies as often: https://www.biorxiv.org/content/10.1101/2020.05.21.108308v1

But among a cohort of mildly-symptomatic hospital workers, almost all developed detectable antibodies: https://www.medrxiv.org/content/10.1101/2020.05.19.20101832v...

Well, did your symptoms match the disease?

Have you had high fever and anosmia (lack of sense of smell) or maybe difficulty breathing?

If yes it probably was Cv19, if not, probably no.

In my case, fever which lasted less than 24 hours. Felt like I was coming down with a bad flu, but the fever was over much more quickly than any flu would be.

Lingering symptoms for a week or so afterwards: “shortness of breath” feeling with slightly sore/tingling chest (but no congestion, snot or phlegm), mild dry cough, all-over muscle aches, feeling much more tired than usual, and sleeping longer than usual. I did not notice any loss of sense of smell.

Hard to tell... if it was a high fever (> 39C or whatever that is in your favourite measurement) that might have been it.

It might be worth getting an antibody test if it's suspected.

https://en.m.wikipedia.org/wiki/Sampling_(statistics)

Reported vs actual are very different numbers.

In Toronto Canada:

- Number of tests available are limited still (and were severely so until recently) - People who are showing symptoms but aren't high risk have been turned away after pre-screening - People who are showing symptoms but aren't front line workers have been turned away after pre-screening - People who have had severe exposure but aren't showing symptoms have been turned away after pre-screening

(Heck, understanding it's an anecdote, but my neighbour showed all the symptoms, after waiting hours and talking to all the pre-screeners, they told him, literally "Yup, you have it; we don't need to bother testing". )

Anecdata notwithstanding, it is crucial for people to understand that the number of tested positive cases is by definition, always, under all circumstances, always going to be some subset of people actually sick. And sometimes significantly so... sometimes by order(s) of magnitude.

I am always rolling my eyes at breathless news reports along the lines of "the actual number of deaths may be higher than the official count!"

Of course they are going to be higher.

Even the official counts can be fuzzy. For example, people dying of cancer are often carried off by pneumonia due to their weakened state. Was the cause of death the cancer or the pneumonia?

Root cause analysis is a tricky thing

Died of a heart attack

Caused by blood loss

From being impaled on a fence

After a car crash

While not wearing a seatbelt

And driving too fast

On an icy road

The data (and you have to scroll down to find the details) is based on antibody tests carried out on a sample of 1000 people living in London and elsewhere and extrapolated from that, and not on people who have tested positive for the virus. The selection criteria are not mentioned, but a random sample of 1000 would have an error margin of 3%. False positive rate for the test is 0.2%: https://www.bbc.co.uk/news/health-52656808

> So those 1 in 6 know exactly how it will impact them because it already has.

It could impact them differently if they contract it again.

How common is that? I know there have been a few cases where it's believed to happen, but similarly there are some cases of people getting chicken pox more than once. With chickenpox, that's rare enough that chicken pox is generally considered something that you only get once, but it does happen.

For chicken pox, it is not as rare as you might think. There are routine vaccinations for children and elderly alike.

I frankly would be unsurprised if an eventual covid vaccine is similarly administered to at-risk age groups, if not everyone.

That doesn't answer my question though. What does the data actually say about the covid reinfection rate? I've heard of anecdotes, but haven't seen any numbers.

The common cold is a coronavirus. There’s reason to believe immunity to this might last only a handful of months.

Have there been any confirmed cases of people contracting it after having antibodies and having symptoms?

That is as true for any flu - or indeed any disease - as it is for COVID-19. There are many diseases out there and most people don't know very much about any of them regarding what a 2nd infection might do. It isn't a reason for them to behave any differently to how they behaved in 2019.

See the thread on coronavirus passports. While unpopular in some quarters, it's the beginning of a way out.

Only if immunity is long lasting, which it may not be.

So what are the numbers if they are significant?

Avoiding infection is not just for your own good; more importantly, keeping the speed of the virus spreading down is for protecting everybody.

Twas ever thus. Norovirus kills 200,000 people every year, and yet we don't close entire cities every time there's an outbreak.

IIRC, noro is too infectious for a lockdown to work, even though you can only catch it from the… emissions… when the patient goes off at both ends.

> Norovirus kills 200,000 people every year

That's true, but it kills very few people in the developed world. Deaths from norovirus are mostly a result of inadequate treatment, and/or lack of detection.

And, frankly, if covid was only going to kill 200k people, we probably wouldn't be closing anything much. It has already killed almost 350k people, and that's _with_ most countries imposing variously draconian measures.

I don't understand this thinking. Do you think it's fine if new endemic diseases are added that will kill multiple of 10k or 100k people a year, every year?

Nobody freaks out over automobile deaths.

The entire car safety industry is a result of people freaking out over automobile deaths. Car safety is reactive. People die, friends /relatives /society freak out (justifiably) and the car industry responds. These responses are also debated and argued by the population (seatbelts, safety ratings, speed limits etc.)

This is preposterous. The measures we are taking will kill far more than the disease itself when all is said and done - look at the referral numbers for stroke centers alone. If this was a test of our collective ability to meet an emerging threat, then we failed about as badly as it is possible to fail. What’s worse is that we failed not so much because of an initial honest mistake, but because we’ve turned out to be superstitious cowards. Shame on us all.

We shouldn't compare it with the number of people who eventually die from covid. We should compare it with the number of people who would have died if there were no measures taken.

I’m not OK with that option. We can make up nonsense on computers as many times as we like, but it will never mean that reality “would have” been anything. Science doesn’t work on “would have” questions, and anyone who claims otherwise is a charlatan.

So we may not be able to resolve this disagreement. Just understand that I’m at least as pissed off as the people on the other side.

> The measures we are taking will kill far more than the disease itself

You're comparing (cost of measures) to (cost of disease after taking those measures), which is a nonsensical comparison. It's like saying "parachutes are too expensive; almost no one dies from skydiving!"

How many people would die from skydiving if skydivers stopped using parachutes but continued to skydive? Am I a charlatan if I give a rough number?

Surely there is a continuum of uncertainty between the skydivers and "it's impossible to know." Surely diseases can be modeled, and those models used to predict outcomes with various degrees of uncertainty. Surely those predictions have meaning.

We can test the prediction that parachutes improve skydiver survival with a simple experiment. It would be an inhumane experiment, so in this case it is fair to assume the theory would be proven correct (but we can always test a similar prediction that parachutes prevent human-like objects from being destroyed, if you prefer).

With a global pandemic, it is in fact impossible to know, and anyone who claims otherwise is in fact a charlatan. That’s the key detail you’re missing here: The skydiver theory can be phrased as a “will” question, but this is always going to be a “would have” question. We can’t run an experiment to test any prediction - where’s the control Earth? If we can’t run an experiment, it is not science. Period.

You are allowed to believe whatever you like, but this shameful episode will be the final nail in the coffin for all pop-bullshit “science”: https://guscost.com/2020/05/12/pandemic-woo/

Reply on news.ycombinator.com