95 comments

[ 3.4 ms ] story [ 169 ms ] thread
It's unfortunately not yet translated but I think the graphs and tables are clear enough and can easily be translated by Google Translate.

Personally my surprise is that it seems half of those deceased people did not receive any kind of antiviral therapy.

Possible explanations for the absence of antiviral therapy: lack of availability, reluctance to prescribe untested treatment (all antiviral treatments are currently experimental), some of the antivirals actually work really well and a lot of the people who got them are missing from this study because they recovered. I hope it's the latter.
Some excerpts (courtesy of Google translate):

The mean age of deceased and positive COVID-19 patients is 79.5 years (median 80.5, range 31-103, RangeInterQuartile - IQR 74.3-85.9). There are 601 women (30.0%). The figure 1 shows that the median age of patients COVID-19 positive deaths is more than 15 years higher than that of patients who contracted the infection (median age: patients who died 80.5 years - patients with infection 63 years). The Figure 2 shows the number of deaths by age group. Women who died after contracting COVID-19 infection they are older than men (median ages: women 83.7 - men 79.5)

The Figure 4 shows the therapies administered in patients who died COVID-19 positive during hospitalization. ThereAntibiotic therapy was the most widely used (83% of cases), least used antiviral (52%), most rarely steroid therapy (27%). The common use of antibiotic therapy can be explained by presence of super infections or is compatible with initiation of empirical therapy in patients with pneumonia, pending laboratory confirmation of COVID-19. In 25 cases (14.9%) all 3 therapies were used.

The figure 5 shows, for the patients who died positive COVID-19, the median time in days, who spend from onset of symptoms to death (8 days), from onset of symptoms to hospitalization (4 days) and from hospitalization to death (4 days). The time elapsed from hospitalization to death was 1 day longer in those who were transferred to resuscitation than those who did not they were transferred (5 days against 4 days).

To date (17 March), 17 COVID-19 positive patients have died under the age of 50. In particular,5 of these had fewer than 40 and were all male people aged between 31 and i39 years with serious pre-existing pathologies (cardiovascular, renal, psychiatric pathologies, diabetes, obesity).

Thank you for your summary. It was very informative.
huh, no one else has mentioned this, am I interpreting this wrong? If 30% of deceased positives are female, then 70% are male? That's quite a big difference!
this is highly anecdotal, but my family is from a small town in italy, population of ~1500, and the men are way less healthy from a lifestyle perspective - lots of drinking, smoking, hard manual labor without lung protection, etc etc, while the women mostly stay home handling home duties and socializing
I think prior lung damage is going to shake out as the number one prior factor in the likelihood of severe infection and probability of death.
(comment deleted)
Will make an unsubstantiated conjecture that men are dying younger and at a higher rate because they've experienced more prior lung damage during their work and smoking habits.

Particularly in northern Italy, which is heavily industrialized.

Doesn't the median age of 60 years for positively diagnosed patients suggest a huge number of non-detected cases?
Indeed. They only test who is really really ill
Depends on the median age of the population. Italy skews older.
Probably tens of thousands of undiagnosed, asymptomatic.

Aka super carriers.

The virus is smart and has evolved from an evolutionary standpoint. Infect younger and asymptomatic patients with the goal of spreading and infecting as many as possible. Eventually, the more infected, the more that will die. The young are just carrier pigeons and don’t even know it.

> The virus is smart and has evolved from an evolutionary standpoint. Infect younger and asymptomatic patients with the goal of spreading and infecting as many as possible.

I think you are giving a bit too much agency to the virus. It has evolved in such a way that optimises for further spreading for sure, but I do not think it has a sneaky strategy to use young people as attack vectors.

What's more likely is that it has evolved to replicate and spread within the human population, while not acting so fast as to kill most people. The problem is that replicating at a rate that doesn't kill most people will kill some of the oldest and sickest within a population.

Another important insight is that Viruses, and germs in general, try to evolve and adapt inside the host. They don’t gain if they end up killing the host because that ends them too. Not to sound apocalyptic but I think their holy grail is something like what the machines do to humans in the movie Matrix
> Another important insight is that Viruses, and germs in general, try to evolve and adapt inside the host.

That assumption has no basis at all. Evolutionary pressures don't happen due to a intentional guidance from the evolving species. Individuals mutate by chance and replicate as they can. That's it.

(comment deleted)
> The virus is smart a

You're making a mountain out of a mole hill. The virus spreads as it's possible, and those who carry the virus experience the infection differently depending on some traits.

Assuming this was true, and there's not tons of dead young people, it also would seem to back up that people under 60 are at far less risk.
Right. One issue we’re having globally is that we simply cannot get an accurate fatality rate.

We’re only seeing the sickest, most critical patients and the most overwhelmed hospital systems in cities.

I think MOST patients being tested in hotspots like Lombardy are those who are already at the hospital, and if you’re at the hospital you’re not in good shape.

Probably tons of positive cases at home with a fever, or mild symptoms or no symptoms.

And of course, tons of people who are not tested and probably never will be. Totally asymptomatic.

We can’t get an accurate fatality rate without these “infected but asymptomatic” cases.

Didn’t an Italian doctor do a small study showing 98% asymptomatic positive cases?

Germany currently looks to be an interesting case. As of CNN this morning, they have 13,957 cases confirmed, and 31 deaths. The deaths may of course crank up to a "normal" level, but I wonder if it's possible that they are testing many more people and getting a more accurate view.

(of course, this comment will probably age poorly)

Oh. This article says it way better than I could: https://www.ft.com/content/c0755b30-69bb-11ea-800d-da70cff6e...

Sorry about the paywall. If you search for it via google and click it from there that's how I found it (and why I didn't realize it was paywall)

Basic summary: Too Early to tell. But Germany is doing 160,000 tests/week, which is more than lots of countries. They may be getting a more complete picture of who is all infected.

We have enough data from closed experiments (Diamond Princess) to say that COVID sends 5% or more to the ICU. That’s the number that matters right now, not CFR. Fatality rate itself is much more dependent on whether the ICU beds are full (Italy, Iran) than anything else. Unfortunately, beds are about to be full in several US cities.
But that's not reliable either. Average age on the Diamond Princess was 58. With 1/3 being over 70. Average age in the US is 38, with 15.2% being over 65.

Given that age is a very large component here, I don't think you can extrapolate data from the Diamond Princess to the general population.

I found page 3 most remarkable: 48% of those who die of covid-19 in Italy also suffer from three or more other diseases (or did recently), and 99.2% suffer from one or more.
Right. 75% suffer from 2 or more. And 99% suffer from 1.

Not good for the US given the high obesity / diabetes rates here.

What counts as a co-morbidity? Obesity seems fairly ... well, trivial. I have high (total) cholesterol. Does that count? Constipation? Itchy spot on the back of my knee?
There's a table in the PDF, you can read it.
The co-morbidities reported here are mentioned in the report (in Italian). Translated by Google:

  diseases N %
  Ischemic heart disease 117 33.0
  Atrial fibrillation 87 24.5
  Stroke 34 9.6
  Hypertension 270 76.1
  Diabetes mellitus 126 35.5
  Dementia 24 6.8 COPD 47 13.2
  Active cancer in the past 5 years 72 20.3
  Chronic liver disease 11 3.1
  Chronic renal failure 64 18.0
Hypertension and Diabetes seems very prevalent in older people. Especially in western countries like the US.
Maybe it is just correlation between old age and hypertension, but I find it interesting how ACE2 which the coronavirus uses as its entry point into cells is also part of regulating blood pressure.
A very high fraction of folks past age 75 have one or more of these diseases.

Before concluding that they contribute, the next question should be, how many people in that population at that age have ZERO comorbid diseases? Likely, it's very few. So these comorbidities may actually have little causal role leading to mortality. We can't know from just this data.

Until we learn more, we can't conclude 1) that comorbidities were causal, or 2) WHICH comorbidity really matters.

there is also the possibility that the comorbidities are indeed causal but we can't do anything about it since most elderly people would have them anyway, and hence besides being an interesting fact it's effectively indistinguishable from a pure statistical age correlation.
Most people might have some conditions that just haven't been diagnosed yet. Some people are obese, diabetic smokers with high blood pressure that still have a more robust cardiovascular system to someone living perfectly healthy. Certainly not the rule, but not impossible. My physics and sports teacher dropped dead with 30-something from an aneurysm. We know the probable risk factors, but you could probably just say they were in poor general health.
The risk factors they count are named in the PDF. They're not "all diseases known to man", they're "cancer in the past five years" etc.

Which might not be an optimal selection of things to count. It's not obvious that a small cancer three years ago ought to be counted and a bigger one six years ago not? But it's what they counted.

"My physics and sports teacher dropped dead with 30-something from an aneurysm"

Just for clarification: did he have covid19?

Some interesting data here.

- Mean age of 79.5 sheds light on Italy’s extremely high fatality rates; in essence, it’s the (very) elderly that are dying due to complications from viral pneumonia. Which begs the next question...

Why are they overwhelmingly treating patients with antibiotics in cases of viral pneumonia and not antivirals (Remdesivir)/ chloroquine?

Sure, these are “experimental” therapies but decent data out of China/South Korea shows these therapies work. Perhaps they found out too late?

- The younger fatalities (17) show multiple, serious co-morbidities and smoking is not listed; an assumption can be made a fair amount of these younger patients smoke. But again, an assumption.

- Almost 50% of patients showed 3 or more co-morbidities - this is high and important to note. 25% of patients showed 2 co-morbidities. Roughly 75% of patients had 2 or more co-morbidities (!).

- Sample size (2003) is good given their current 3,500 fatality numbers.

Not a medical doctor but a few things I’m struggling to figure out:

- How did so many elderly get infected? Did the disease simply spread in close quarters where many elderly live? Elderly folks aren’t necessarily out and about drinking espresso and touching surfaces yet alone having younger asymptomatic carriers cough on them.

I wonder if Italy is similar to a Kirkland, Washington situation. High density of elderly folks spreading infection.

It’s obvious that SARS-Cov-2 is highly, highly contagious but it’s interesting how we’re seeing these somewhat “bomb” explosions of infection: Wuhan > Daegu > Kirkland > Lombardy > NYC next.

Sure, quarantine works but the rate of new infection stays rather localized and then just annihilates everyone around it.

Perhaps it’s a viral load issue; viral load increases exponentially the more we have infected. Why you see doctors and nurses infected / critical and dying even with full PPE.

Let’s hope the Italians figure out a way to get this curve to fall of ASAP. Hoping they have a similar effect to Wuhan’s curve and just drop down rather than flatten.

> Elderly folks aren’t necessarily out and about drinking espresso and touching surfaces yet alone having younger asymptomatic carriers cough on them.

From what I understand about Italian culture, they kind of are. Older Italians seem to have much stronger social lives than in the US.

There’s truth to this and perhaps I jumped to an invalid American based assumption, stupid given that I’m European.

In Europe, the elderly are out and about, walking, sitting at parks, drinking coffee, the works. They’re actually still doing this as we speak in countries, from what friends and family tell me. Go figure.

However, even if they are in cafes, grocery stores, parks - the numbers are still very high.

This virus is either incredibly efficient and contagious (spread by totally asymptotic carrier pigeon patients > elderly) and/or it’s been there for a long time. Months.

If my experience with 'southern' cultures is anything to go by, the amount of interaction between young and old people probably plays a huge role.

For comparison, I live in NL and for me and the vast majority of my 20-30-something friends, visiting grandparents is relatively rare. For many of us even visiting parents is a 'once every x months' kind of thing.

On the other hand, when I lived around the mediterranean, not only was it expected to regularly interact with parents and grandparents, but it was often the case that they lived together, or at least close by.

Perhaps the 'quarantining'/neglect of the elderly in Northern-Europe that I've often criticised as inhumane is actually saving them in this particular situation.

> and/or it’s been there for a long time. Months

I believe this to be accurate and have argued for it elsewhere.

> This virus is either incredibly efficient and contagious (spread by totally asymptotic carrier pigeon patients > elderly) and/or it’s been there for a long time. Months.

It's incredibly contagious. Infection rate doubles every 2.5 days.

On the elderly infections: https://www.wired.com/story/why-the-coronavirus-hit-italy-so...

On antibiotics, aside from secondary infections, there’s a paper out talking about azithromycin: “Our preliminary results also suggest a synergistic effect of the combination of hydroxychloroquine and azithromycin. Azithromycin has been shown to be active in vitro against Zika and Ebola viruses [20-22] and to prevent severe respiratory tract infections when administrated to patients suffering viral infection [23].”

Not sure of they used azithromycin but would’ve been good to see that verified.

Yeah, I remember reading this as well.

MedCram did a good video showing how Zinc, for example, actually impairs viral replication but Zinc cannot enter the cell... without chloroquine :)

Seems chloroquine has the key to the cell, otherwise nothing goes in.

> Why are they overwhelmingly treating patients with antibiotics in cases of viral pneumonia and not antivirals (Remdesivir)/ chloroquine?

My guess is that they try to avoid an additional bacterial super infection with antibiotics. Also these antivirals can be dangerous to (very) old people. I think because of liver malfunctions. But take all I say with a grain of salt.. this is all stuff in the back off my head that I remember vaguely from my father (a doctor).

I wonder if intubating these patients is actually killing them. I have ZERO evidence but is there an off chance they’re actually treating them incorrectly?

There’s the cytokine storm problem although I just wonder, for some weird and maybe even wrong reason, that the standard of care, the protocol for this particular viral infection is not intubation for most?

If you look at a video of the ICU or ward in Lombardy, everyone is either intimated or negative pressure oxygen. Yes, their Sp02 levels are bad and they have breathing difficulties. So I get why they are following standard protocol - but I just wonder if this particular infection, at its peak, requires a different approach.

But enough Doogie Howser from me.

Intubation can lead to bacterial pneumonia, which would be treated with antibiotics.
> I have ZERO evidence
My understanding is that the virus greatly hinders your immune response in the lungs and that opens the door to microbial infections. This is what causes the severe pneumonias that kill people. Kurzgesagt did a nice video explaining this: https://www.youtube.com/watch?v=BtN-goy9VOY
We all want to live long, and people now often live till 80+, but when a person 80+ years old dies does it make sense to search for a cause of the death? My grandmother died last year from pneumonia. She was 89, which is way above the average life expectancy. Any small infection can kill a person at this age. Is there a statistics that shows the mortality rate in Italy before the pandemic.
> but when a person 80+ years old dies does it make sense to search for a cause of the death?

Of course! We want to know how to prevent the deaths of others in the same condition.

Something else to keep in mind is this part:

> The figure 5 shows, for the patients who died positive COVID-19, the median time in days, who spend from onset of symptoms to death (8 days), from onset of symptoms to hospitalization (4 days) and from hospitalization to death (4 days).

At this point, we know the elderly and people with pre-existing conditions are dying fast; it is also not unreasonable to think they're more vulnerable overall.

But not enough time has passed to say yet whether younger people without pre-existing conditions are not particularly susceptible or are just a little less susceptible but take longer to succumb. If the statistics are a week for someone over 65, from onset to death, and a month for someone under 40 w/o pre-existing conditions, it's going to skew the hell out of the statistics this early in the outbreak.

> Why are they overwhelmingly treating patients with antibiotics in cases of viral pneumonia and not antivirals (Remdesivir)/ chloroquine?

Because a bunch of these patients are people with suspected, not confirmed, covid-19 and so they're treating pneumonia which they think is probably bacterial but possibly viral.

Also, they were thinking that covid-19 in already vulnerable people was damaging the lungs and making bacterial pneumonia much more likely, so they were treating preventatively.

> yet alone having younger asymptomatic carriers cough on them.

Droplets in the air are one route of transmission, but there are others. It's likely that most people are not infected via this route, but via fomites. Infected people cough onto a surface, and later someone touches that surface and then their face.

> Infected people cough onto a surface, and later someone touches that surface and then their face.

And then what? How does the virus get from their face into the lungs in enough quantity to infect them?

> And then what?

They touch their mouth, or eyes, or nose.

> How does the virus get from their face into the lungs in enough quantity to infect them?

They're already infected. I don't understand the question. Are you asking how the virus replicates? It's a strand of RNA that takes over human cell replication. It binds to ACE2, which is how it targets lungs. https://blogs.plos.org/dnascience/2020/02/20/covid-19-vaccin...

People touch their face all day everyday, even when they've been told not to. Even when they're telling other people to stop doing it. https://twitter.com/Kojoanan/status/1235275598697771011

It's unusual for people in public places to be coughed on. I can't think of it happening to me in the past 5 years.

Healthcare professionals wear masks (and goggles, and protective clothing, and sometimes gloves) because their work involves close contact with ill people who are coughing over them.

> How did so many elderly get infected? Did the disease simply spread in close quarters where many elderly live? Elderly folks aren’t necessarily out and about drinking espresso and touching surfaces yet alone having younger asymptomatic carriers cough on them.

Ha. Italians in N. Italy often living multigeneration in close quarters. Spanish and italians, men and women, ritually kiss on the cheeks as a greeting. Brits and Germans don't so much.

Italians are very social to an old age and live longer than many nations.

https://www.sciencedirect.com/science/article/pii/S120197121...

> We estimated excess deaths of 7,027, 20,259, 15,801 and 24,981 attributable to influenza epidemics in the 2013/14, 2014/15, 2015/16 and 2016/17

Italy always has this problem

> How did so many elderly get infected? Did the disease simply spread in close quarters where many elderly live? Elderly folks aren’t necessarily out and about drinking espresso and touching surfaces yet alone having younger asymptomatic carriers cough on them.

I was wondering the same thing, but judging by my own country crowds of elderly people is a thing. For example they do not use technology as much as others and tend to concentrate in places that provide alternative means to function without technology.

Another assumption is that touching surfaces is not how the virus spreads. Being within a few meters of carriers, not necessarily coughing, just talking and breathing, is how people catch it. And since masks are not available and dumb propaganda discourages people from using even a piece of clothing to protect them from inhaling droplets, people catch the virus so easily.

Perhaps it doesn't matter that much how many people smoke right now, but how much lung damage they accrued during their lifetime.

The median age of the deceased is around 80 years.

According to a random source I found on the internets, in the 1950s and 1970s between 70% and 50% of the male population was smoking:

https://www.niussp.org/wp-content/uploads/2016/05/Schermata-...

It is everywhere. The 'pockets' you think you are seeing are an artifact of exponential growth. When infection rate doubles every 2.5 days, the incubation time is 5-14 days and it take 3-4 weeks to die the first 4 weeks look harmless. But by then it is already well out of control.
This is encouraging for healthy youngsters BUT age and health are also the criteria for receiving treatment in Italy at the moment. As the epidemic grows the categories of people that get ventilation decrease and the deaths will get younger and healthier.
Even if a youngster as a higher change of survival, there are serious effects reported. Fibrosis (damage of the lung tissue) lasts for much longer (over 4 weeks) than it is known in any typical flu. And what happens after 4 weeks is still totally unclear.

As we don't know yet of the long term effects, please don't take that lightly!

Source for this comment?

All pneumonia’s have decreased lung function post-infection and require therapy to retain full or close to full lung function.

Curious where you’re getting this info from

(comment deleted)
(comment deleted)
How many were so sick that they would have died anyway?

I read that they were testing dead people and the theory was, their count was so high because they would add people to it that would have died anyway but just happen to be infected right before their death.

Edit: I think this is an interesting question. Why do Italy, France and Spain have such high death ratios? Are they measuring more than needed or are the others measuring not enough?

Right, but you need to establish the cause of death and it's kind of important. Like for example my father was fighting cancer for 8 years before he died, but when he passed away the hospital wrote that he died of "cardiac arrest", since his heart just couldn't go anymore. So in some statistic somewhere he will exist as a person who died of heart problems, not someone who died because of cancer. That's not right either I don't think.
This is a problem with counting diabetes deaths as well. We have evidence to believe that people with diabetes are much more likely to get heart disease or cancer, which is what they usually die from. There's lots of debate on if mortality due to diabetes is underestimated for the same reasons.
I can only speak for France, where I live, but the situation is possibly the same in Spain and Italy.

Over here they only test people who are "very sick", in other words people who need medical attention. They are not testing people with no or "mild" symptoms. Basically this means if you don't feel like you're about to die, you're not tested and therefore you're not counted.

So to answer your question, they're actually not measuring enough.

In Germany they also only test people who have specific symptoms AND contact to high risk people/places.
Summarizing the deaths, by age distribution is quite telling (page 4):

- no deaths under 30

- Less than 1% of deaths under 50

- less than 4% of deaths under 60

- 87% of deaths above 70

This disease is incredibly dangerous for our elderly, and isolating them to prevent infection should be our top priority.

That's because Italy's hospitals are overcrowded and very old patients are simply not treated at all. It's a vicious circle because they're thought being less likely to survive, are denied treatment, which makes them even less likely to survive.
(comment deleted)
Triage is unfortunately applied in some hospitals, especially around Bergamo.

But you would need to provide some data to make the claim that the stats are skewed this much purely because all elderly are just cast away. Given the vast majority of hospital cases are actually the elderly.

Finally, proper ethical triage never looks at how long a patient is expected to live after their recovery as a condition.

> Finally, proper ethical triage never looks at how long a patient is expected to live after their recovery as a condition.

But it does look at chances of survival, which for this particular disease looks very closely correlated with how long a patient is expected to live after their recovery.

Confirms that this is a disease of the elderly and unwell, and everyone else should be getting on as normal.
This is about deaths, not cases. If everyone else tries to get on as normal then the virus will spread like wildfire among them and the elderly and unwell will quickly become infected.
Isolate the elderly and unwell, let the virus spread among the rest.
How could you possibly isolate 10s of millions of people who are thoroughly mixed in with the rest of society? Many of them live with or are dependent on the younger healthier people. Even if the isolation were possible, letting the virus spread like that would overwhelm the health system and push the death rate way up.
The same way they're doing it now.