102 comments

[ 3.1 ms ] story [ 54.7 ms ] thread
I have a lot of friends and relatives in NYC. At a certain point, volunteer medics started discouraging people in their neighborhoods from going to the hospital because nobody was coming back alive. Because family members were barred from visiting and staying with their loved one, there were people literally dressing up as hospital staff and “sneaking” their loved one out of the hospital because they were hearing terrifying reports of the high mortality rates there. Really crazy.
Something is odd with either the patients or the methods. In Sweden the mortality rate is around 20% [1] (in Swedish, unfortunately).

[1] https://sverigesradio.se/sida/artikel.aspx?programid=83&arti...

Is there an english translation of this?
translate.google.com

"Region Stockholm Eight out of ten corona patients survive intensive care"

Given the importance, I didn’t want to rely on google translate.
I am relying here on Google translate, so I might miss something. That mortality in the article is about patients being in ICU, but not every patient in intensive care is intubated. Actually, to my knowledge, doctors are now avoiding intubation as long as possible, which explains the "increased" mortality rate.

You can see it the other way around, of those people, where they didn't see any other possibility, they still manage to save 1 in 10 with intubation.

Google translate is correct. I was going to make the same comment, the article is talking about ICU patients and does not mention if they were intubated or not. The 175 patients in the study might not have been put on ventilators, the article does not specify!
In Denmark most patient in intensive care are in respirators. Sweden is probably similar.
I will try to find a source, but AFAIK (and checked with a doctor) only patients that need intubation currently get ICU treatment in Sweden.
I think the question was not how many patients that need intubation are in the ICU but how many patients in the ICU need intubation. I think the suggestion is that the first number is 100% while the second is much lower until you hit a point where ICU beds are so full that only to most critical cases are put into the ICU.
Yes, agree that there is a difference. Regarding Sweden, it seems like all patients in ICU are intubated. An explanation might be that because of the low number of ICU beds in Sweden only people deemed to have a fair chance of survival is admitted.

[1] is the ICU register that gives some statistics, e.g. that median age is 60 years and 25% of patients don't have any risk factor.

[1] https://www.icuregswe.org/data--resultat/covid-19-i-svensk-i...

As another data point/anecdote, I heard the number 90% from someone working in ICU in Switzerland a few weeks ago. It certainly wasn't a scientific study, but their point was that basically most people intubated were dying in the end, which was really disheartening.
Here's my rough translation (not Google Translate):

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

Eight of ten corona patients survive intensive care

Published Sunday 12 april 17.33

More covid-19 patients are now being discharged from hospital in Stockholm region, and eight out of ten patients survive intensive care.

After having examined the first 175 patients admitted to intensive care with covid-19 at Karolinska University hospital, the conclusion is that 48 out of the 62 patients that were discharged did survive intensive care. The other 14 patients died.

- I am cautiously optimistic as the immediate survival rate seems to be better than what we were expecting when we started doing this, says David Konrad, head doctor at the ICU at Karolinska University hospital in Stockholm.

- We had read accounts and reports from other parts of the world that said perhaps 10-25 percent of the patients admitted to ICU would survive.

- This is a very preliminary result and we don't know what will happen in the longer term. We don't yet know if anything will happen to the discharged patients, or in what state of health they will be. We have treated too few patients so far to be certain of any of our conclusions, says David Konrad.

The ICU at Karolinska University hospital currently has 126 covid-19 patients and the hospital has a total of 177 ICU beds.

There are 78 remaining ICU beds in all of region Stockholm, and there are about six to twelve covid-19 patients being admitted to ICUs in the region per day. Despite this, David Konrad thinks there are signs of a slowdown in the influx of new patients.

- We have also started to discharge more patients. This indicates that we've reached a plateauing phase where there's a better balance between the number of incoming patients and patients who can leave the ICU, says David Konrad, though he does point out that the situation may change.

The Health Authority has also updated the nationwide statistics today. 899 people have died in association with covid-19 in Sweden, which is an increase of 12 people compared to yesterday.

The government has pointed out that one should be careful to not draw premature conclusions from what's being reported, due to time lag.

In total, 10 483 people have been confirmed infected by the virus in Sweden thus far.

Marcus Admund Funck marcus.admund_funck@sverigesradio.se

> Among all patients who received mechanical ventilation, the mortality rate was 76.4% for adults ages 18 to 65, and nearly all patients over age 65 died (97.2%), reported Karina Davidson, PhD, of Feinstein Institutes for Medical Research at Northwell Health in Manhasset, New York, and colleagues, writing in JAMA.

If that is the death rate for people who need ventilators then a lot of the arguments for "flattening the curve" are less compelling.

The whole idea of "flattening the curve" was to preserve hospital capacity in order to save lives. If most of the people who need ventilators are likely to die whether or not hospital capacity is available, then the trade-off between saving lives and saving the economy has not turned out the way we expected.

Not everyone who needs medical help for Covid-19 needs a ventilator. Some just need supplemental oxygen.

And people go to hospital with other conditions. If hospitals are 100% overwhelmed with Covid-19 cases, then they may struggle to treat strokes, heart attacks, car crash victims, cancer patients, etc...

Flattening the curve will save lives.

> Flattening the curve will save lives.

Yes, but how many lives will it save? The answer is unclear but it seems to be lower than what most people would expect.

The basic assumptions were not well supported when the decisions were made. All the models we had were based on a very, very low level of testing and unreliable information from other countries.

What is your source on this? This opinion seems to be against what experts continue saying.
What's an acceptable death rate look like for you? Anecdotally does that number change if we're talking about your life, or the lives of people you are responsible for?
How many jobs lost per lives saved is worth it to you? 100? 1000? And how long are the lives saved being prolonged?

In California it looks like more than 200 jobs could have been lost per life saved and that's a lot of human suffering. https://marginalrevolution.com/marginalrevolution/2020/04/ca...

Jobs come back. Dead people do not.
A lack of jobs can create dead people, whether through crime or through an inability to afford healthcare or shelter or food.

I'm not convinced that such deaths outweigh the lives saved through these measures, but they're still worth mentioning if we want a full picture of the impact.

This might sound good as a sound bite but offers no thought behind it. One thing that is clear is that we cannot go to extremes on either end.

On one side - we cannot stop the economy to contain the virus, because that would turn into every person having to grow their own food and carry their own water with a bucket from the river. Many many people would also die.

On the other side - we also cannot ignore the pandemic and carry as if nothing is happening because that would over-burden the hospitals to the point that they will not be able to treat people with other issues. Many many people would die as well.

So there indeed has to be a balance. And the question "what price are we willing to pay for saving a life" seems to be on point.

A lot of the jobs that were lost won't come back for years. It's not like we can flip a switch and all the businesses that closed will be open again. The small business economy has been destroyed.
Have you accounted for the human suffering caused by the loss of those lives on the survivors? How many jobs is that worth?
True, but that outlook depends a lot on age, life experience and other factors; a lot of people will agree with you unless it's their own life/family.

Also; por qué no los dos? The USA has enough money and resources (and billionaires) to save the lives + not have the suffering (basic income/welfare). That they choose not to might not be the concern of the people who voted for those things but got Trump; they don't want their lives or loved ones to be chosen over 'jobs/economy' just because the people in power are overly greedy and short-sighted.

Why do you assume those jobs could be saved? Do you think they would have carried on as normal if the pandemic had been allowed to spread unchecked?
Given the number of people who were working until the lockdowns were announced by the government, and how many customers were still buying nonessential services, I think the majority of those jobs would not have been lost.
This is nowhere near as bad as things would be if it was left unchecked.
I don't think the data supports that assertion. I don't have the link handy, but OpenTable published data showing that in Atlanta at least, restaurant bookings were down ~90% before the stay-at-home order was issued. The situation was similar in many other areas as well.
> What's an acceptable death rate look like for you?

We make arbitrage like this when setting pollution laws, or speed limits. The issue here is that we don't know the cost of flattening the curve. We know that GDP is strongly correlated with life expectancy, and the loss of GDP worldwide will have important sanitary impact, but it's hard to quantify this.

That being said, I'm not saying we shouldn't try to flatten the curve. Considering we don't have the data to make the right decision, we use a greedy algorithm. But I'm quite convinced that we'll be worse off in the long term. Incidentally, we choosing to save old people over poor people.

Most economists say we should listen to health professionals and not re-open to early. And that if we do re-open too early its likely to cause more long term damage to the economy

http://www.igmchicago.org/surveys/policy-for-the-covid-19-cr...

Most economists have stayed as far as humanely possible from evaluating lives vs GDP.

Which is a shame.

It’s like the one example where some back of the envelope yet informed numbers might improve public policy vs the status quo..blind panic and a livelihood crippling total lockdown.

I’m not sure if it’s possible to have an earnest conversation about that when one party can easily claim the moral high ground by expressing outrage that the other party might not be willing to go to literally all costs to save a single life.
I do see some of that argument being made, however, I see a lot more false information coming from one side, as well as ignoring the advice of experts, and actively quieting people in the government who dare to publicly disagree with the president's statements. So, yeah it is pretty hard to have an unbiased discussion on this. I obviously have my own biases here :).
> the other party might not be willing to go to literally all costs to save a single life.

That's an easy thing to say if the cost you pay is working from home and not going out on the town.

Not so easy when the cost is the end of a small business you worked years to build.

To be clear, I fully support Strict social distancing and think the idea of their being a trade-off between government policy on the matter and economic performance is off-base. People will social distance whether the government tells them to or not, and if the pandemic worsens substantially then the economic impact will be even worse. That is to say that the best thing to do from both a lives saved AND an economic perspective is to contain the pandemic to a manageable level. </br> All that throat clearing out of the way... this is a silly argument and one that is not applied in any other domain. Should we ban cars? How can you possibly say that the many thousands of people who die in automobile accidents each year are an acceptable loss of life?
> Should we ban cars?

From cities? You'll find that many will agree with that.

I do agree with your general point though.

> Should we ban cars?

To put it in perspective, approximately 37,000 people die per YEAR from automobile accidents in the USA

With all the measures already in place, and strong possibilities of under counting COVID19 has already killed more than 50,000 people in less than 4 months.

A better argument would be banning obesity. ~300,000 people die from obesity related diseases every year

Obesity is actually the top chronic condition risk factor for COVID-19.
We definitely need to ban all cars, airplanes, busses. But more importantly close all factories and borders. Forbid sodas. We need to safe every single life.

Bonkers.

You can save the life of a child for like $3k so the life of a diseased old person should probably be worth either that or lower.
I read the study and nowhere does it mention when or how ventilators were applied.

If ventilators are a scarce resource/bottleneck and ventilation is applied by prioritisation then I could see the mortality rate skewing up as a function of treatment being applied too late.

It is worth bearing in mind the mortality rate of non-swamped health care systems.

For every NYC/Paris/Lombardy/Wuhan there is a South Korea/Taiwan/Hong Kong with very different data profiles.

I would love to see the mortality rate for ventilation support in these countries/cities...

Can we expand the capacity of the healthcare system for those treatments? Oxygen is not without danger, but it can be safely handled by non-professionals, and it's relatively easy to manufacture.

I've never actually heard anybody discuss any treatments besides ventilators, so I have no idea what's feasible.

Remdesivir seems promising, also variolation, also plasma therapy (IVIG).

Currently the lockdown makes sense for one reason, to get something that at least makes it seasonal flu-deadly (about 5-10 times less than currently).

Accidental release of data for remdesivir showed no benefit.

https://www.statnews.com/2020/04/23/data-on-gileads-remdesiv...

Maybe the data wasn't ready for release, and the final version will show some benefit, but it's likely that remdesivir has little benefit.

(Genuinely confused about why this got downvotes.)

(comment deleted)
People don’t like bad news.
Thanks! Too bad :/ Last week it seemed promising...

(I have no idea who would downvote your comment.)

If it was a case of just seeing if they need oxygen then it'd be trivial.

The issue is when they deteriorate, what do you do then. There are hundreds of ways for people in hospital to "go wrong" knowing when where and how to intervene requires both nurses and doctors. All of which take 3-9 years to train

Besides close monitoring of people on O2 to adjust amounts and intubate if that time comes, there are potentially other components of supportive care that might alter outcomes, like IV anti-virals (e.g. remdesivir) or anti-coagulants, e.g. heparin. Some anti-coagulants can be given at home, but the strong doses being discussed also come with high risk of bleeding, benefitting from bed rest and close observation. For a good slice of the sickest COVID patients, hospitalization is helpful and hard to replicate at home, even with our limited therapies.

That said, oxygen and heparin at home could be good for a lot of patients. Improvements in prognostication may help distinguish that group in the future.

But could oxygen be given outside of hospital environment?
Oxygen is very regularly given outside of the hospital for a range of ailments. There is a thriving industry selling home and portable oxygen delivery units.
1. Non-ventilator hospital resources are also stretched to the limits in NYC.

2. The high death rate might mean that they are at the point of only intubating hopeless cases, who otherwise have a 0% chance of survival.

3. If you don't flatten the curve, you're going to be killing doctors.

> The whole idea of "flattening the curve" was to preserve hospital capacity in order to save lives

The lives being saved by flattening are not limited to covid-19 patients requiring intubation.

It's also about saving the lives of people who aren't covid-19 infected at all, but require emergency hospital care, which there's a constant supply of on any given day in a major city.

Frankly I'm surprised we're even having to clarify this at this late stage, sigh.

People seem to forget that people are still getting ill as usual, in addition to the covid cases. Hospitals still have to treat everyone they used to, in addition to all the new covid patients
In europe at least, all hospital switched to treating only essential procedures, basically almost anything non life threatening is postponed.

It got into ridiculous situations where a friend who is an urology doctor was on paid leave for last few weeks, simply no work for them under these rules. Another friend, radiologist in a hospital heavily hit by covid, worked only 4 hours/day shifts, instead of usual 10-12 ones.

So no, hospitals definitely don't treat everybody they used to, only those that would literally die if postponed. And even there sometimes they don't, ie they stopped resuscitating severe heart attacks here few weeks ago, because its too equipment-intense and uncertain. Equipment is +-same as that used for covid patients.

Most of the normal workload for hospital isn't emergency, but all those non-severe treatments like dermatology, arthritis, broken bones etc. That's maybe 80% of the staff.

> It got into ridiculous situations where a friend who is an urology doctor was on paid leave for last few weeks, simply no work for them under these rules. Another friend, radiologist in a hospital heavily hit by covid, worked only 4 hours/day shifts, instead of usual 10-12 ones.

I'm assuming specialists ain't allowed to work as generalists, then? I'd imagine that specialized doctors (like urologists) should be able to double as PCPs should the need arise (like, you know, a global pandemic). I ain't a doctor, though, so I'm perfectly willing to accept "specialist and generalist practice are too different from one another" as the answer here.

> The lives being saved by flattening are not limited to covid-19 patients requiring intubation.

It's not even about just saving lives if the thing I saw recently about the effects of COVID-19 on survivors is accurate - diminished lung capacity, impaired liver function, possible cardiac issues, possible neurological issues, blood clots that may lead to strokes later, etc.

Currently 750k survivors - health systems could possibly be looking at a generation or more of strain coping with them.

Or we can flatten the curve and reduce the number of infections and thence the number of infected survivors.

I think what you mention is completely stopping infections, which is certainly a great goal - albeit a very costly one for many countries which started the right actions too late, too little.

Flattening the curve is only about spreading the same number of infections over time, so that at a given time, the health system is not overwhelmed. It will not reduce the number of infections, and will likely augment the number of survivors, who might have to deal with hard long term consequences.

It will most definitely reduce the number of infections. It allows us to reach the point of where measures like test-and-trace become possible after a smaller number of infections, and those measures can further suppress and even eradicate the disease before it manages to infect everyone.
Testing and tracing would have been a great idea in February when the outbreak could still be contained. Now it's gone too far for that to be practical. So we're likely to end up with ~70% infected eventually (approximate herd immunity limit), it's just a question of when.
That is the whole point of lockdowns, to bring the outbreak down enough that these methods are reasonable again.
You missed the point. Now that infections have spread so far, the level of lockdowns that would be necessary to bring the outbreak down to that level aren't even possible in an open society. But it will be a good approach to try during the early stage of the next pandemic.
I admire your optimism that the people in charge might learn from this when they clearly haven't[1] from the last few.

[1] Excepting places like Taiwan, etc., who clearly did learn.

There is not a constant supply. ER departments in most hospitals are underutilized right now. Doctors and nurses are being laid off or having their hours cut due to lack of work.
I don't hear that for the first time. But question i always ask is: are the ventilators only for the worst of the worst cases? (because there arent enough) so you end giving them to people who probably die anyway.

BUT:we know there are long term effects on your organs and lungs from corona. Would giving ventilators to not critical or semi critical people reduce the long term effects and damage?

So, a hospital at, lets say 70 percent might still be netter longterm for infected people than close to the edge. (not to mention heart strokes etc)

Tbf we dont know that there are long term effects on organs and lungs from covid-19. Its only existed for 6 months. Its impossible to know what will happen long term at this point.
True, but we can make some educated guesses based on the long term outcomes of SARS and MERS patients.
> are the ventilators only for the worst of the worst cases? (because there arent enough)

If there aren't enough, you give them to the best of the worst.

If there are two sick people and one needs vent. One almost dies. One would have long term effects, but the the other one would have reduced long term effects with a vent. The vent would go to the one almost dying. Then you would have two survivers. But one with worse long term effects.

But as i said, i have no idea if a vent would reduce long term effects.

> are the ventilators only for the worst of the worst cases? (because there arent enough) so you end giving them to people who probably die anyway.

If a patient can't breathe independently, they are by definition a "worst of the worst case". Regardless of whether they contracted corona or something else.

Putting someone on a ventilator is not a cure. It's a last ditch aid that buys the patient a bit of extra time for their body to pull through. However, that's in no way or shape a guarantee that a patient will actually pull through.

When doctors decide to hook a patient to a ventilator, they already accept that that their patient might not make it.

However, that's not an argument to not deploy ventilators or let patients die.

Ventilators are used to give any patient a maximal chance to pull through. The upshot is that ventilators gives a percentage of patients a second chance.

The moral question then is to what lengths society is willing to go to provide that chance to as many patients who are hit hardest by this virus. Understanding that some patients are less likely to pull through then others. That's where you end up with ethical questions about who gets to live and who doesn't and what criteria medical staff is using to make that decision.

> we know there are long term effects on your organs and lungs from corona.

We know that there is damage to organs and lungs. However, this is a novel virus. It's unclear what the actual impact of that damage will be in a general population over a longer period of time.

> Would giving ventilators to not critical or semi critical people reduce the long term effects and damage?

Putting someone on a ventilator is a Hail Mary and it's a delicate operation. It's literally shoving a tube down a patient's windpipe and hooking up an advanced air compressor. Do it wrong, and the patient dies or is left with debilitating complications for life (i.e. damaged vocal cords, overextending the lungs when pushing air,...).

It's also a traumatizing event: patients who are put on a ventilator are put in an induced coma for an extended amount of time which comes with it's own trade offs and potential complications.

> So, a hospital at, lets say 70 percent might still be netter longterm for infected people than close to the edge. (not to mention heart strokes etc)

It's important to understand that maintaining an ICU in normal times is extremely cost intensive. An ICU unit is, after all, intended for patients who are at the edge between life and death. It's a place where life is generally counted in minutes or hours. And above all, despite the fancy technology and the expertise, it's a place where fate, luck and chance determine in which way the scales tips.

The 'shortage' of ICU beds all over the world isn't just a reflection of how ill-prepared we were for a pandemic, it's above all a testament of the deadliness of this pathogen, of it's ability to ravage a human body.

While increasing the ICU capacity of hospitals is absolutely tantamount to save as many lives as possible, I think humanity also might want to show a due sense of humility to a complex force of nature that we don't really control. For all our technological capabilities, we tend to forget in our daily existence that life is very much fragile and that living to see another year is less self granted then we may have come to assume. For many, that might come as a very stark and uncomfortable truth.

> If that is the death rate for people who need ventilators then a lot of the arguments for "flattening the curve" are less compelling.

edit - to the downvoters who refuse to read anything before downvoting, here you go:

Boston 25 News: "According to a story in The Washington Post, autopsies are showing that many people who died from the COVID-19 virus have hundreds of microclots present in their lungs at the time they die. The autopsies were not revealing pneumonia as was expected, but lung damage from blood clots and the damage was appearing in the kidneys, liver, heart, intestines and brain, the Post story reported."

This is the death rate of people being incorrectly treated. If you treat patients thinking that pneumonia is their primary problem, putting them on ventilators, when the primary problem is that they have microclots in their lungs and you fail to treat the clots (which has been the case until very recently), you're going to kill a large percentage of those patients.

Tens of thousands of patients have died because the medical system somehow didn't notice the large volume of blood clots in patients.

It's the blood clots causing the high mortality rates, sudden deaths that are difficult to account for, and the oxygen problems re the lungs. They're finally beginning to address this by immediately treating patients with blood thinners. The virus causes an extreme increase in inflammation, which is causing blood clots including in the lungs which is causing oxygen deprivation.

A horrific mistake has been made over the past few months of just ventilating everyone with breathing problems, when it was the blood clots in the lungs causing the problem. They tried to solve the wrong problem.

https://www.reuters.com/article/us-health-coronavirus-usa-bl...

https://www.livescience.com/coronavirus-blood-clots.html

https://www.boston25news.com/news/trending/coronavirus-are-b...

About a thousand articles are out there over the last week about this.

Going off the article alone, it would be hard to notice this, but the conclusions of the actual study are based on patients with a final results (either death or discharge), but that only includes about 46% of the cases. Specifically, as noted by ivanonymous, that's only including 320 patients on ventilators who died or were discharged versus 861 still on ventilators. If all 861 patients recover, that 90% would drop massively (it's hard to say how much without breaking the numbers down more completely).
(comment deleted)
Lockdown protestors need to understand this. If you're unhealthy or unlucky enough to require intubation, you will likely die.

I don't understand why people aren't taking this seriously.

Because the overall mortality rate is less than 0.5%.
That's not necessarily true. The article on MedPageToday barely notes it, but the actual study clearly spells out that only about 46% of patients were included in the study as having a final results (either death or discharge). As noted by ivanonymous, that's 320 patients requiring intubation (282 died) vs 831 still on ventilators, hopefully recovering.

TL;DR: looking at the headline and not the study itself is deeply misleading.

As always with these links, lots of data, very little information...
If you want to survive to COVID-19, don't eat too much ! High blood sugar is very bad for this desease. Time to make a diet.
Notably, "Similar to prior data, hypertension, obesity, and diabetes were the most common comorbidities."

State of health is clearly essential in this fight. And more actionable than magic bullets or vaccines.

Really who cares about science when the US President comes up with sheer utter garbage like bleach and UV light inside your body and nobody bats an eyelid.
Niels Ryberg Finsen was awarded the Nobel Prize..."in recognition of his contribution to the treatment of diseases, especially lupus vulgaris, with concentrated light radiation, whereby he has opened a new avenue for medical science."

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

Lupus Vulgaris is primarily a skin disease.

I think your comment here is disingenuous at best.

Even under the most generous reading of Trump's statements, how in the world would it be remotely possible to "inject disinfectant" in any way?

His speculation, particularly in an area where he has no expertise (but of course he still thinks he knows what he's talking about), is wildly dangerous; but of course, he can't shut up and let experts talk.

Worth noting, he never said the exact words "inject disinfectant", but did allude to putting disinfectant chemicals and methods inside the body. Still dumb, but be careful quoting those words like that, Trump supporters will (possibly rightly) jump down your throat for that, and promptly discard anything you might say
We already know for over two weeks that ventilators do more harm than help. The proper treatment are ECMO's, putting oxygen directly into the blood. The hemoglobin/iron problem.

The high mortality rate is also trivial. Everybody who doesn't need oxygen already helped themselves. Those who do need it are the severe cases with wrong treatment. But even with correct treatment the rate would be very high. Plus the followup damages to all organs because if he oxygen deficiency.

That ~90% figure is wildly misleading.

It's based on 282 deaths among the 320 ventilated patients who either died or were discharged.

But 831 patients were still on ventilators!

It's a snapshot taken too soon.

Estimates of mortality on vents very widely still, partly because of real underlying variation in practices or population, partly because the data is really messy.

This email series from Mass General is a good orientation (discusses this study, links to discussions of previous studies): https://mailchi.mp/db30d9d2cb24/tz4idnzryr-4406129?e=acf498e...

Any vent strategy being debated on Twitter (early! late! APRV!) is also being discussed by working pulmonary critical care docs. Judging those discussions or the variations in practice as an outsider is hard. But relative silence on Twitter doesn't equal mindless orthodoxy.

Just to back up your statement, from the actual study, under Limitations:

> Fifth, clinical outcome data were available for only 46.2% of admitted patients. The absence of data on patients who remained hospitalized at the final study date may have biased the findings, including the high mortality rate of patients who received mechanical ventilation older than age 65 years.

MedPageToday notes this, but unlike the study, they start with the headline of 90% mortality, and don't mention until the very last sentence "that clinical outcome data were only available for less than half of admitted patients." This just seems wildly irresponsible reporting; I don't know if the reporter didn't really understand the limitations, or what, but at the very least, that should not have been the headline, not without a major proviso included at the very beginning.

There is a lot of variance in icu outcomes. My wife is a pulmonologist in Houston and just finished a week of virus icu. She had only 1 death, and even that person she could have kept going but patient was ready to quit and didn’t want any more heroic measures being 85.
I must be incredibly lucky to know of two friends' family members with covid19, both above 65, both went on ventilators, both now off ventilators and recovering. Neither one in NYC, fwiw.