There is no practical way to "isolate the sick and the elderly" when there are no supplies of PPE or testing kits for care workers. The whole point is that the sick and elderly need care, and when there's a pandemic that care requires extra measures to prevent the spread of infection.

In reality the government did the exact opposite - knowingly sending elderly people with Covid infections back to care homes, and then saying "Oh that's nothing to do with us - every move was signed off by a clinician."

And that in turn makes even less sense when you realise that tens of millions were spent on Covid-ready Nightingale Hospitals, which were opened to great fanfare - and then literally left empty, while old people were being sent to nursing homes to infect others.

Evil, or "simply" stupid and incompetent? It doesn't even matter any more A lot of people are dead for no good reason, and far more are going to die in the UK before this is over.

This is just completely wrong. It's a useful exercise to work out how many people would need to die before herd immunity could be achieved. Use mortality numbers based on what we've learned about the virus and estimates of what percentage of the population would need to be infected to achieve herd immunity based on the behavior of any number of other viruses.

And that's assuming infection confers immunity with this virus, which we don't know yet.

As for the death rate, the rate to the non vulnerable population is immaterial. With appropriate isolation we could absolutely achieve herd immunity with minimal losses. Instead we have sent covid infected patients back in nursing homes in both the US and the UK with catastrophic results.

The numbers don't look very immaterial to me. [1] Granted there's a weird amount of variance between regions and there's a lot that isn't certain.

> If you are still questioning whether infection confers immunity, then what other strategy than herd immunity do you suggest? If we don’t develop immunity naturally, there won’t be a vaccine either. Permanent lockdown?

(Let's hope some reasonable amount of immunity is conferred! It seems likely, but I think it's useful to consider the worst case.)

I'd be happy with people doing what they ought to be doing anyway: exercising some caution and doing the obvious things to limit the spreading of the bug. I'm in a constant state of amazement at how bad we are at all of this.

In the crazy worst case where there isn't much immunity conferred by the infection, we could hopefully rely on Darwin to eventually provide us with a less dangerous version of the virus.

[1] https://en.wikipedia.org/wiki/Mortality_due_to_COVID-19#Mort...

You're looking at the numbers for the case fatality rates (CFR). The CFR is not the same as the infection fatality rates (IFR) which represent "the fatality rate in all those with infection: the detected disease (cases) and those with an undetected disease (asymptomatic and not tested group)... The IFR will always be lower than the CFR..."[1]

Spanish government ran a seroprevalence study and got the following estimates of COVID-19's IFR[2]:

For the age group under 50, the estimated IFR is about 0.003%

For the age group 50 - 69 yo, it's 0.04%

And for those over 70, it's 4.1%

[1] https://en.wikipedia.org/wiki/Case_fatality_rate#Terminology

[2] https://www.mscbs.gob.es/profesionales/saludPublica/ccayes/a... Look for the table on page 15

Here is 14th May report https://www.mscbs.gob.es/profesionales/saludPublica/ccayes/a... that is referenced below the table.

For some reason they chose to use total from the page 3, 19.155 total by age not 27.321, total by region from page 2. I presume that they do not know the details of about 8000 diseased, but it does not matter when it is the total number.

Again while number of >= 70 year old matches the table on the page 3 - 16559, so do age groups <10 and 10-19 but other numbers do not.

The table on the page 15 contains only 263 from the age group 50-69 but the table on the 14th May report contains 2303. The IFR for this age group would be then 0.32%. Not very big but not negligible either.

Group 20-49 contains only 23 people from the 20-29 but omits 62 people from the group 30-39 and 201 people from the group 40-49. The IFR for this age group would be then 0.03%.

Does the document break out IFR-S? Some news came out in recent days indicating IFR-S in the United States was around 1.3%.

I agree, 4% IFR of people over 70 is a scary number. In my opinion, governments around the world should really focus on protecting these people.

The good news is for people under 70 personal danger from the virus is close to zero.

"Does the document break out IFR-S?"

Don't know, sorry. Deciphering that table is really easy as the column names are almost the same as the English terms. And Google Translate does a really good job on the footnotes.

But reading the whole paper I'll have to leave to someone who actually reads Spanish.

I feel like I'm missing some key context that would lead a person to declare that with any sense of certainty. It's on a chart and everything! But it's not in agreement with everybody else's charts, and it's early days when it comes to the epidemiology.

(and at least in the US, our treatment capabilities are vast, but our testing regime is cartoonishly bad)

I haven't seen any other IFR estimate that included a breakdown by age group from a government or another authoritative source. Do you have anything particular in mind?

Is it a certain thing the risk for people under 70 is close to zero? Strictly speaking, no, it isn't.

It's unlikely we're going to see precise IFR numbers any time soon. But countries around the world are lifting lockdowns. Where I am, they're lifting it too. So I'll have to evaluate my personal risks without full information.

Why do I tend to rely on Spanish numbers?

- They are one of the hardest hit countries. If they had relatively few fatalities in the age group under 70, it must be reasonable to assume other countries aren't going to have significantly more.

- Don't have a link handy, but numbers published by Italian government had similar numbers of fatalities per each age group. (They didn't try to estimate IFR, though.)

- I don't think Spanish govt manipulates their numbers. But -- if anything -- they have an incentive to overestimate the IFR to justify one of the strictest lockdowns and a high number of dead.

I'm privileged enough, to be working from home. And staying as safe possible. I just think bad economy can ultimately take lives (not only livelihoods) of more people than COVID-19.

But one difference with the Spanish documents is they don't estimate their IFRs.

And if I'm not missing anything, doing it yourself is not trivial if possible at all. It's not enough to know that 5% (an arbitrary number, just to give an example) of a country's population developed antibodies. You'd also need to know the age structure of those 5%, which may or may not match the country's age structure.

Yes, modeling prevalence is complex but once you have the total estimated infected then finding IFR becomes trivial. Calculating SMR would be much more difficult.

Many (most?) countries do not produce reports in English but only in their native language.

When you search in the native language then it is easier to find relevant data. Especially when you catch the right keywords they use.

Here is 14th May report https://www.mscbs.gob.es/profesionales/saludPublica/ccayes/a... that is referenced below the table.

For some reason they chose to use total from the page 3, 19.155 total by age not 27.321, total by region from page 2. I presume that they do not know the details of about 8000 diseased, but it does not matter when it is the total number.

Again while number of >= 70 year old matches the table on the page 3 - 16559, so do age groups <10 and 10-19 but other numbers do not.

The table on the page 15 contains only 263 from the age group 50-69 but the table on the 14th May report contains 2303. The IFR for this age group would be then 0.32%. Not very big but not negligible either.

Group 20-49 contains only 23 people from the 20-29 but omits 62 people from the group 30-39 and 201 people from the group 40-49. The IFR for this age group would be then 0.03%.

I don't really understand why they are using numbers from one table and not the other.

I guess, it may be related to the dates they conducted the seroprevalence study, but it's only a wild guess.

I'll try to look through the documents more carefully later, maybe there's an explanation.