Monday, April 6, 2020

In Veneto, Keeping People out of the Hospital

From the Financial Times, a comparison of coronavirus treatment in two of Italy's most populous regions, Lombardy and the Veneto:
Struck by a human catastrophe unseen in Europe outside of war, with military trucks taking away corpses from the city of Bergamo, Lombardy has a death rate of 17.6 per cent.

Nearby Veneto’s stands at 5.6 per cent. While virologists caution that the percentage death rate is closely tied to the level of testing, they also attribute the gap to other factors, such as Veneto’s reluctance to hospitalise compared with its neighbour.

“Veneto has a very low mortality compared to the rest of Italy,” said Professor Andrea Crisanti, a leading virologist from the University of Padua, in charge of a mass testing programme across Veneto. “This shows that our approach has worked well so far.”

As of Saturday Lombardy, which has a population of 10m people, accounts for 8,656, or 56.3 per cent, of Italy’s total declared deaths from the virus of 15,362. Meanwhile Veneto, which has a population of 4.9m, has suffered 607 official deaths out of 10,824 diagnosed cases.

Higher levels of testing and tracing in Veneto is the most widely cited explanation for why the region has managed to control its outbreak more effectively than its neighbours.  . . .  Yet experts say testing is not the only reason for the lower death rate.

Venetian doctors also cite the region’s expertise in infectious disease, something they trace back to its pioneering history dealing with viruses arriving in its port from the east. The word quarantine derives from quarantena, the Venetian word for “40 days”, or the amount of time ships arriving from plague-ridden destinations were isolated.

For Giorgio Palù, one of Europe’s leading virologists, and scientific adviser to the governor of Veneto, a critical factor has been the number of diagnosed patients taken into hospital.

The number of diagnosed patients who were taken in hospitals for clinical treatment at the start of the outbreak was about 65 per cent in Lombardy, Prof Palù said.
This compares with 20 per cent in Veneto, where the majority were told to stay at home unless urgent care was required.

“There were different instructions given to the sick by the different regional health authorities,” he said. “Yes, there has been more testing in Veneto but people were kept at home and not taken into hospitals. The more patients you admit to the hospital, the more cases you get. You create the outbreak as at the start nobody was taking care of sampling the doctors or nurses, [so] you were taking home the infection.”

His observation comes as more than 60 Italian doctors and health workers have died, the majority of these in Lombardy. A group of doctors from the Papa Giovanni XXIII hospital in Bergamo warned last month that hospitals had become the main source of transmission of Covid-19 infections, and urged more patients to be treated at home.

“We are learning that hospitals might be the main Covid-19 carriers,” they wrote in the New England Journal of Medicine. “They are rapidly populated by infected patients, facilitating transmission to uninfected patients.”


David said...
This comment has been removed by the author.
G. Verloren said...

"This compares with 20 per cent in Veneto, where the majority were told to stay at home unless urgent care was required."

...but that's how hospitals work. You don't go to the hospital when you don't need urgent care. Why are the Italians going to the hospital when they don't need things like IVs, supplemental oxygen, and respirators?

G. Verloren said...

Or put another way, why are uninfected individuals (aside from medical workers themselves, for whom it is an inevitable occupational hazard) getting infected at hospitals? Why are they going to hospitals if they aren't infected?

The only thing I can think of is that people are getting sick with something else (I've heard talk that this is a double flu season?), and the people who have access to testing can confirm if they just have the flu and go home, while the people who don't have access to testing can't be certain what they have, and thus the only safe assumption is that it's COVID and they need to be hospitalized?

John said...

@G - they go to hospitals to have babies or because they broke their legs or whyever else people go to the hospital.

People can be given supplemental oxygen and IVs at home. And they are in Veneto.

David said...


I'm sympathetic to the idea that those with the flu-like symptoms could be kept at home, with periodic visits, IVs, etc. This is the German approach, and apparently the Venetian one as well.

But what about those with the pulmonary symptoms? In the last comment thread, you seemed to suggest that such people should stay at home, not infect hospital staff, stay off ventilators, and, well, wait for death. Did you really mean that? I for one found it rather terrifying.

G. Verloren said...


Yes, obviously people who are having babies or who break their legs go to the hospital. But you're once again ignoring the context of your own posts.

We're talking about people in Veneto being told to stay home - but it's not as if women delivering children and victims of major accidental trauma can realistically just not go to the hospital! We're talking about people who don't require urgent care, and mothers in labor and people with shattered femurs absolutely do require urgent care, and thus aren't relevant to what we're talking about!

John said...

Well, I am against a lot of end of life care. I myself want no extraordinary measures taken to keep me alive. But I don't think that's the issue here.

I am not exactly opposed to putting coronavirus patients on ventilators, I'm just saying that we lack the data showing that it does any good. Until we know I guess we should keep trying.

But what if it turned out that putting people on ventilators in hospitals ended up killing more doctors and nurses than it saved patients? This is not impossible. As I said, it happened just a decade or so ago during an Ebola outbreak, and the people I quoted in Veneto are clearly worried about exactly that scenario. In that case I would say, we should close the hospitals. We should never practice medicine that leads to more people dying rather than fewer.

I am baffled that y'all are so committed to giving more medical care without even pausing to ask if it helps. I am all for giving care that helps, but shouldn't we try to figure out whether what we are doing is helping or not? Or exactly what it is that helps and what might be harmful? Aren't these important questions?

You cannot ever assume that treating people in hospitals helps them, and you cannot trust front-line doctors to know whether their actions are helpful or not. After all Galenic doctors were sure for about 1,600 years that bleeding people helped them, and they were just as smart and observant as modern doctors. It happens all the time that doctors think they know how to help patients but then some statistician does a good study and finds that these interventions do no good at all, or actively cause harm. This happened recently with using grafts to bypass clogged arteries in the neck, which turned out to increase the risk of stroke. Those surgeons thought they were saving lives but really they were taking them. More medicine is not always better.

I want to save the most lives possible. That does not equate to giving the most hospital care. The examples of Germany and and Venice show that care can be given to most patients at home, and that this results in lower death rates. And G, a lot of those patients being treated at home in Germany and Venice are people who would be hospitalized in the US or Britain. Who is an "urgent" case is often a matter of opinion, or local custom. (Consider childbirth, which we see as a medical emergency requiring hospital care but most women can do perfectly well at home.)

The goal in fighting a pandemic has to be keeping people alive. That is my goal. What worked in Korea was relentlessly tracking the contacts of everyone who tested positive for the virus. That saved far more lives than could ever be saved by ventilators. Social distancing is working for us and in Europe, saving, again far more people than we could ever imagine putting in hospitals.

Those are the measures that matter, and that is where we should be putting our emphasis. The number of lives lost to any shortage of ventilators will be a rounding error compared to those that would be lost by relaxing the social shutdown even one day too soon.

David said...

I think for a lot of folks, it's not about statistics. It's about their and their loved ones' individual fates. If I or anyone in my family gets the pulmonary kind of COVID, they're getting a ventilator, and I don't give a fuck what the numbers say.

It's also about the social compact. I really believe this. Fundamental to our society is the idea that, if you're sick, you can get help. Someone will give you care and try to do something that might actually work, as the ventilators do sometimes. I've seen no stories indicating that they always, or almost always, fail. Looking at statistics and then deny that help because you or some other authority judges that help "ineffective" breaks a fundamental trust that people value.

Yes, that part of the compact puts some people in greater danger. But it's still the compact.

It's similar to the fundamental trust that's broken when a business fires people because their quants decide that's the thing to do at this moment. I'm not talking about the current crisis. And such quantitative business approaches may be better in the long run for quantitative measures of the economy. But people fundamentally aren't quantitative beings who are happy to just let whatever the numbers say decide their fate. They want to trust that they have a place to work and that they can get paid there and not worry if some quant gets a more accurate set of data that means they're out. Relationships are based on a moral compact, not statistics.

Imagine if people were told that, if they have a home invasion, they shouldn't call 911, and in any case the police won't come, because we've determined that in home invasion cases, calling the police is 20% likely to get the police killed, and will result in no improvement in the chances of the people living in the home surviving. To do things that way would be horrible. I don't want to live in a society like that.

Perhaps it is these sorts of approaches that has led to some of the distrust in our society, and some of the hostility to "experts."

For a lot of us, these approaches are deeply alienating.

Perhaps a way to think of it is that, to the rest of us, letting a person wield data the way you do feels like giving them a nuclear weapon, in Scott Alexander's sense. It's too powerful, and it's not a weapon we have or want.

G. Verloren said...


All very well said.

Numbers and rationality are invaluable tools that we would be utterly lost without, but they alone don't solve our problems. We always have to weigh the rational against the moral; the realistic against the idealistic. And sometimes, that means choosing an option that is measurably sub-optimal or not wholly rational, simply because it is the more humane or just choice.

"But is that logical? Surely, we must-"
"Logic, logic, logic... Logic is the beginning of wisdom, Valeris, not the end."

"Don't grieve, Admiral. It is logical. The needs of the many... outweigh..."
"The needs of the few."
"Or the one."

"My father says that you have been my friend. You came back for me."
"You would have done the same for me."
"Why would you do this?"
"Because the needs of the one... outweigh the needs of the many."