Update on the coronapanic fueled by Imperial College

by philg

Previously from the U.K.:

Yesterday: “UK has enough intensive care units for coronavirus, expert predicts” (New Scientist)

[the expert[ said that expected increases in National Health Service capacity and ongoing restrictions to people’s movements make him “reasonably confident” the health service can cope when the predicted peak of the epidemic arrives in two or three weeks. UK deaths from the disease are now unlikely to exceed 20,000, he said, and could be much lower.

New data from the rest of Europe suggests that the outbreak is running faster than expected, said [expert]. As a result, epidemiologists have revised their estimate of the reproduction number (R0) of the virus. This measure of how many other people a carrier usually infects is now believed to be just over three, he said, up from 2.5. “That adds more evidence to support the more intensive social distancing measures,” he said.

Who is this “expert” telling the Brits to chill out? Neil Ferguson, the lead author of the March 16 Imperial College paper that told the British to freak out.

(Professor Ferguson doesn’t explicitly say “my Oxford competitors are right and I was wrong”, but upping R0 to “just over three” supports the Oxford theory. Let’s assume “just over 3” is about 3.4. He previously was thinking 2.5. How do these compare? If we just do the simple exponential and don’t bother to consider people meeting the already-infected, after 8 exchanges of germs, the disease has spread 12X more widely under this quietly buried tweak. That’s the difference between 5% of the population and 60% that are infected and, by inference (since 60% of Britons aren’t dead), a huge difference in the lethality of COVID-19.)

How about the peak? The Imperial College model has a chart showing mid-May, without mitigation attempts and June with even the most basic actions, such as isolating sick people. Most of the Britons who need “critical care” won’t be able to get it.

Now it seems that the peak will arrive in “two or three weeks,” i.e., no later than Easter(!) and, as noted above, nearly everyone who needs critical care will get it (but, unfortunately, 20,000 will still die).

The text of the March 16 paper is just as inconsistent with the latest statement as the graphs. On page 16 of that paper, he says that “even if all patients were able to be treated, we predict there would still be in the order of 250,000 deaths in GB”. So in a perfect world where the UK never ran out of critical care capacity and therefore every patient had access to all useful medical interventions, 250,000 would die just in Great Britain (not including Northern Ireland). He is saying now that, in an imperfect world where the hospitals are hanging on the hairy edge of figuring it out, 20,000 will die in the entire UK (including Northern Ireland). [see comments below for some discussion]

‘I am surprised that there has been such unqualified acceptance of the Imperial model,’ Professor Gupta [of Oxford] told the Financial Times.

Perhaps Imperial College professors don’t accept the Imperial model!

(And what does 20,000 deaths actually mean? Are they people who would have lived another 50 years in great health? Or mostly people who would have died within a year or two from an underlying condition?)

What does this mean for the U.S.? Unclear. The “chill out” opinion assumes a competently run health care system that already has policies in place for rationing care (i.e., super old/sick people that were in the ICU in the U.S. in 2019 would have been unplugged and buried in the U.K.). “In the UK’s health system, rationing isn’t a dirty word” says that the U.K. won’t spend more than $40,000 to keep someone alive for another year in good health. That’s the cost of a handful of ED visits here in the U.S., without even being admitted! If we used U.K. standards, half of the older people who are admitted to a hospital would be given three ibuprofen and sent home in an Uber Comfort.

[March 28, 2020 update, from “As the rest of Europe lives under lockdown, Sweden keeps calm and carries on” (Guardian):

Anders Tegnell, Sweden’s state epidemiologist, believes it is counterproductive to bring in the tightest restrictions at too early a stage. … His team at the Public Health Agency of Sweden is critical of the Imperial College paper that warned this month that 250,000 people in the UK would die if the government failed to introduce more draconian measures. A week later Johnson ordered the police to implement a partial lockdown to combat the virus, telling people they “must stay at home”.

“We have had a fair amount of people looking at it and they are sceptical,” says Tegnell. “They think Imperial chose a number of variables that gave a prognosis that was quite pessimistic, and that you could just as easily have chosen other variables that gave you another outcome. It’s not a peer-reviewed paper. It might be right, but it might also be terribly wrong. In Sweden, we are a bit surprised that it’s had such an impact.”

]

Full post, including comments

Why do we care about COVID-19 deaths more than driving-related deaths?

by philg

Working with third-year medical students involves much struggling with SQL, R, and data, but also chatting about the topics of the day. This year it is coronavirus, of course. Of the nine M3s that I work with most commonly, at least one believes that he has already had COVID-19 and recovered. Absent significant testing capability for infection or antibodies, however, these bright young people are as much in the dark as anyone else.

Leaving aside the true alarmists, estimates of likely U.S. deaths from COVID-19 seem to range from 50,000 to 500,000 in a “life goes on” scenario. The prospect of this many deaths has motivated us to shut down society and mobilize for what people say is a “war” (let’s hope it isn’t like any of the wars that the U.S. has fought and lost since 1960, e.g., Vietnam War, War on Cancer, War on Poverty, Iraq War, Afghanistan War, etc.).

This is as it should be, right? Let’s take a mid-range estimate. The prospect of 275,000 people dying is terrible and should motivate us to bold action. Yet roughly 36,500 Americans die every year in motor vehicle-related accidents (NHTSA data from 2018, the latest available).

This led to a discussion regarding human psychology. We are pretty confident that there will be more than 275,000 car-related deaths over the next 8 years in the U.S. Maybe this should motivate us to bold action, but it actually does motivate us to do almost nothing.

In the 24th year of the smartphone, we don’t bother with a car-to-smartphone communication system, for example, that would reduce pedestrian fatalities (since the car would know where all of the pedestrians were; I wrote about this in 2016; ordinary Bluetooth range seems to be roughly 100 meters outdoors). Considering the nation as a whole, we don’t invest much in separated (e.g., with a curb) bike lanes like they do in Denmark and Holland. We don’t cut the speed limit on the Interstate back to 55 or lower. We don’t say that cars have to have electronic governors so that it simply is impossible to speed (“I’m sorry, Dave, I feel you pressing the accelerator, but I can’t go faster than 35 mph on this stretch of road”). We don’t re-engineer the road network to eliminate traffic lights in favor of (a) traffic circles, and (b) overpasses. We don’t put in a car-to-traffic light communication system so that the car knows when the light is red and will hit the brakes before we inadvertently drive through an intersection (imagine a traffic light that broadcasts in Bluetooth “I am the light at Massachusetts Avenue and Vassar St. and am currently green for Vassar St.”). We don’t ask America’s nerds to stop working on clever Internet ad technology and try to come up with innovative ideas for reducing the carnage on our roads. We’re willing to invest $trillions to reduce the death toll from coronaplague, but hardly a dime to build centerline dividers on more of our two-lane roads so as to eliminate head-on collisions.

As with most discussions about psychology, we came to no conclusion!

Readers: What is the answer? Why do we accept that hundreds of thousands of Americans will die in the next 10-20 years because of our failure to invest in engineering and infrastructure today, but we can’t accept that up to hundreds of thousands of Americans will die in the next year because we didn’t do a sufficiently thorough shutdown?

Related:

  • Sweden’s Vision Zero, kicked off in 1997, which worked to reduce fatalities until it stopped working in 2013.
Full post, including comments

Move to avoid estate tax before coronavirus kills us?

by philg

Now that the grim reaper seems to be among us, is it time to move away from the 12 states that assess estate taxes? Massachusetts, for example, deprives heirs of 10-16 percent of the value of their inheritance, for estates valued at over $1 million (i.e., for anyone who dies while owning a decent apartment or house in the Boston area). The highest state tax rate is reached even for those whose estates aren’t worth enough to be taxed at all by the Federales.

What about income tax? A lot of us will have to work from home for the next two years. Why not do this from the Ritz Dorado Beach in Puerto Rico and cut income tax to 4 percent via Act 22? Puerto Rico seems to have eliminated its estate and gift taxes in 2017 so even if 183 days per year of heat and humidity don’t protect you from coronavirus your savings will be protected.

Full post, including comments

About 2.5 percent of Boston hospital beds occupied by COVID-19 patients

by philg

Friends who work at Boston’s biggest hospitals (MGH and Brigham and Women’s) get the information that the public would surely love to have, i.e., what percentage of hospital beds are occupied by patients stricken with COVID-19. The answer as of this morning? About 2.5 percent, with perhaps an additional 2-4 percent “under investigation” (can’t get test results even for inpatients?).

Is that good news for those worried about contracting the coronavirus? Maybe not. The LA Times says that ICU beds at their hospitals were already full with non-COVID-19 patients. As you might expect from the three-month wait that was required for a non-elite to see a doctor in the U.S. in 2019, the health care system as experienced by the 99.8% did not have any headroom.

And, of course, if we believe that hospitalization follows infection by at least two weeks (but does it, on average?), the surge may be yet to come. The folks who heeded New York City Mayor Bill de Blasio’s March 2 call to join him in a crowded movie theater “despite Coronavirus”, and their new Tinder friends, and the new Tinder friends of those Tinder friends, may yet arrive in the ED.

Readers: Has anyone found a good source for mean time from infection to hospitalization and/or mean time from infection to requiring a ventilator? Dyson will begin cranking out ventilators by “early April” (CNN).

[Update: I talked to a physician at Emerson Hospital in Concord, Massachusetts. This serves Middlesex County, the hardest hit county in Massachusetts (a profitable place to be a family court plaintiff though; having sex with an already-married medical specialist will pay about as well as going to medical school and working as a primary care doctor). There are 179 beds, 6 of which are occupied by victims of the coronaplague (3.34 percent).]

[Update, evening 3/27, text message from friend: “I asked [physician wife] how her nurse friend is doing at the hospital. She said she has not worked in a week because the ER is so dead she can’t get hours. The hospitals are empty.”]

Full post, including comments

Are celebrities jumping the coronavirus testing queues doing us a favor?

by philg

From yesterday: “Elite versus Non-Elite access to COVID-19 testing”.

On further reflection, however, maybe celebrities jumping the queue are doing us a favor. We’ve learned that Harvey Weinstein (68), Jackson Browne (71), and Prince Charles (also 71) have tested positive for coronavirus and are recovering at home from their flu-type symptoms.

These guys are not examples of youth, vibrant health, or genetic diversity, right? They probably wouldn’t have been tested if they’d been ordinary peasants. Can we relax a bit knowing that (a) their positive diagnoses suggest that a reasonably high percentage of the population has already been exposed (consistent with the Oxford epidemiologists who may never be interviewed by America’s hysteria-dependent media), and (b) their mild symptoms suggest a lower-than-expected percentage of the infected need medical intervention?

[Ironic: Jackson Browne was infected in Wuhan on the Hudson at a concert to benefit victims of a different virus: HIV (New York Post). He came out of an “anti-Trump” concert in 2017 unscathed, fortunately, and also a crowded event in Los Angeles: “Trump loathing unifies the diverse crowd at the massive L.A. women’s march”. If Jackson Browne’s case had become serious, his life might have been saved by the president he loathes, since Trump sent one of the Navy’s two hospital ships to Browne’s home city of Los Angeles (the other one is going to New York City). Trump could certainly have sent the ships to Florida and Texas, right? Or addressed the inequality that New Yorkers and Californians decry by sending the ships to lower-income states such as Mississippi, Louisiana, Alabama, and South Carolina. Will Democrats in LA and NYC at least send a thank-you note to Trump even if they will never vote for him?]

Meanwhile, in Massachusetts we have limited information on the non-celebrities. The March 25 update says that there are 1838 “confirmed cases”. Are they watching Netflix at home? Or near death on a ventilator in an ICU? For 1385 the answer is “Under investigation”. 103 are known to be hospitalized and 350 are known to be “not hospitalized”. Is it truly impossible for our state public health officials to call up the hospitals and ask “How many COVID-19 inpatients do you have right now?” (Contrast to Texas, where the governor has ordered daily reports on bed availability. (with hundreds of billions of dollars invested in health care IT, why aren’t these data already streamed up to each state’s department of health?).)

Related:

Full post, including comments

Full range of coronaplague opinions

by philg

From Sunetra Gupta, professor of theoretical epidemiology at Oxford, via the Daily Mail:

Coronavirus could already have infected HALF the British population and been spreading in the UK since JANUARY, Oxford University study claims

‘We need immediately to begin large-scale serological surveys – antibody testing – to assess what stage of the epidemic we are in now,’ she said.

The Oxford university research offers a contrasting view on the disease to the study that is informing government policy. It was carried out by experts at Imperial College London.

‘I am surprised that there has been such unqualified acceptance of the Imperial model,’ Professor Gupta told the Financial Times.

The Imperial study has led to the Government imposing the extraordinary shutdown on the basis that, without such rules, the disease could claim up to 250,000 lives.

Who wants to bet whether these “Oxford experts” get interviewed by the folks at the New York Times who say that Donald Trump should let the “experts” speak?

From Wuhan on the Hudson, on the other hand, “Coronavirus: New York Infection Rate Is ‘Doubling About Every 3 Days,’ Cuomo Says” (NPR):

The rate of new coronavirus cases in New York is “doubling about every three days” and is speeding up even more, Gov. Andrew Cuomo said Tuesday. “That is a dramatic increase in the rate of infection.”

The new estimates are “troubling and astronomical numbers,” the governor said. He added that the apex of the curve of rising coronavirus cases in New York is still 14 to 21 days away, according to the latest projections. The governor also said New York is in urgent need of ventilators and other vital resources.

“We need the federal help, and we need the federal help now,” Cuomo said.

(If Federal help is critical and Trump has the discretion to allocate that help among states, perhaps Cuomo now regrets telling Donald Trump “good riddance” when he moved to Florida, thus escaping New York State and City taxes? When one quarter of the economy is the Federal government, is it helpful to state residents to pick a fight with the guy who has some amount of control over the $5 trillion?)

Facebook friends who voted for Hillary love Governor Cuomo, by the way. A sampling of perspectives:

Gov. Andrew Cuomo. For all his state’s problems, and they are life and death problems….he still inspires, like a true leader does. No childish name calling. No race baiting. No ignorant rants. Just inspiration

Sick of the administration’s sideshow “press conferences?” This is what leadership looks like: honest, tough, authoritative, reassuring.

I am falling in love with Andrew Cuomo🥰🥰🥰 [The previous governor got/purchased some love from female Democrats as well, no?]

[Did Cuomo do anything tangible in terms of preparing hospitals, obtaining masks, ventilators, and PPE, starting social distancing before New York’s caseload blew up, etc., or has he been able to win hearts and minds with fine words only?]

COVID Denier Aaron Ginn needed to be kicked off Medium partly for spreading the heresy that coronavirus did not spread well in heat and humidity. Now there are heretics at M.I.T. New York Post:

Scientists at the Massachusetts Institute of Technology found that 90 percent of COVID-19 transmissions that occurred until Sunday happened in regions with low temperatures — between 37.4 and 62.6 degrees Fahrenheit.

“Wherever the temperatures were colder, the number of the cases started increasing quickly,” Qasim Bukhari, a computational scientist at the university who co-authored the study, told the New York Times. “You see this in Europe, even though the health care there is among the world’s best.”

That pattern applies in the US too, Bukhari told the paper.

The outbreak has developed more slowly in Southern states like Arizona, Florida and Texas compared to New York, Washington state and Colorado, for example, according to Bukhari.

In California, the rate is somewhere in the middle.

Epidemiologists have seen a similar pattern with other viruses — including the flu, which generally follows a November-to-April trend in the Northern Hemisphere, Dr. Deborah Birx, a member of the White House’s coronavirus task force, said during a recent briefing, according to the Times.

The four types of coronavirus that cause the common cold also pick up in the winter and drop off in the summer.

At least two other studies have come to similar conclusions — including one conducted by Chinese researchers at Beihang University and Tsinghua University.

There you have it! Maybe it will kill most of us (Professor Cuomo). Maybe it has already done most of whatever it is going to do (Professors at Oxford). Maybe it will taper down to a manageable burn by June (MIT).

(Meantime, if we believe MIT, move to Florida, Texas, or Las Vegas and then leave the air conditioning off? Check the family law situation in those states before agreeing to any move, though! Florida offers “permanent alimony” while Texas offers “no alimony”. Texas and Nevada cap child support profits, unlike most states in the frigid COVID belt.)

Related: A February 9 tweet from New York City’s top public health official…

Full post, including comments

When do car makers cut prices?

by philg

We’d like to buy or lease a new car, but right now we have two cars and almost nowhere to go. I expected to fight through traffic to Harvard Medical School every day this month. Instead, I’ve been on Webex and Zoom. There’s no point going to the supermarket, which has been stripped bare by the hoarding neighbors who, just a few months ago, were touting their superior virtue and empathy level compared to the out-of-state Deplorables who voted for Trump. We can’t go to the art museums, which are closed. We can’t go to restaurants, which are closed.

At the current rate of usage, our current fleet will last for at least 40 years (an aquarium maintenance service operator told us that she now has more than 300,000 miles on her Honda Odyssey!).

Shouldn’t the car manufacturers be having coronavirus special deals? A car plainly isn’t worth as much as it was two months ago. Why are the prices mostly the same? Are people actually buying cars at a similar rate to what it was a year ago? Is the shutdown of car factories roughly balancing the collapse of demand? Even with the U.S. and Europe paralyzed, Japan and China are open for business, right? They can produce a ton of cars, can’t they?

Full post, including comments

Elite versus Non-Elite access to COVID-19 testing

by philg

Email received from the president of Harvard, Larry Bacow, today at 1:15 pm:

Earlier today, Adele and I learned that we tested positive for COVID-19. We started experiencing symptoms on Sunday—first coughs then fevers, chills, and muscle aches—and contacted our doctors on Monday. We were tested yesterday and just received the results a few minutes ago. We wanted to share this news with all of you as soon as possible.

Two days from first symptom to test result.

Text message from an M.I.T. Ph.D. in engineering, today a little earlier:

Yes, a friend of mine in Boston had to wait 6 days to get tested, then another 4 days for the result

Ten days from symptoms to result (positive, unfortunately, and then the rest of the family caught it too, casting doubt on the 10% household transmission stat that has appeared in some articles; everyone is recovering without hospitalization).

Conversation this morning with some Harvard Medical Students:

At Partners [the Harvard-affiliated goliath of Boston-area hospital systems] we can’t order a coronavirus test unless the patient is admitted.

[Separately, the “emergency” is not so urgent as to have ruffled the feathers of the Massachusetts state government’s license raj. One of the Harvard students is 4th year and will soon be eligible for a medical license here in Massachusetts. “I don’t know how I’m going to get licensed,” he said. “There are a ton of forms that I need to give to the state and they all have to be notarized. Where will I find a notary?” I.e., the emergency is not so dire that they’re willing to give a provisional license to anyone whom Harvard Medical School verifies is a recent graduate, then sort out the rest of the paperwork after the plague has abated.]

As of September 2019, President Bacow was a cheerleader for more low-skill immigration to the U.S.. Email to the Harvard community:

Not just as a university president, but as the son of refugees and as a citizen who deeply believes in the American dream, I am disheartened by aspects of the proposed new criteria for people seeking to enter our country. They privilege those who are already educated, who already speak English, and who already have demonstrable skills. They fail to recognize others who yearn for a better future and who are willing to sacrifice and work hard to achieve it. Had these same rules been in place when my parents each immigrated, I doubt they would have been admitted, and I would not be writing this message today.

My parents, like most immigrants, loved this country in part because they had the experience of growing up someplace else. They appreciated its aspirations of freedom and opportunity for all, and never took these ideals for granted. But they were also not uncritical of their new home. They wanted it to be the very best place it could be, a goal to which we all should aspire. Indeed, it is the role of great universities to foster an environment that encourages loving criticism of our country and our world. Through our scholarship and education, through our encouragement of free inquiry and debate, we ask not just why things are as they are, but how they might be better. To be a patriot is also to be a critic and not to accept the status quo as inevitable.

The new academic year is a chance for all of us to commit ourselves to creating a community that welcomes and embraces people from across the nation and around the world, people whose distinctive voices and varied experiences are essential to our common endeavor.

At the time the email was sent, every Harvard building that I needed to access was locked down with 100% ID checks at the door by security guards assisted by RFID readers. None of the new migrants would be welcomed into a Harvard building to use the restroom or eat in the cafeteria. The University provided Bacow with a mansion in one of America’s most desirable neighborhoods; he wouldn’t be competing with the new arrivals for housing. So maybe the U.S. could grow to 400 million and Bacow’s day-to-day quality of life wouldn’t suffer.

But why would President Bacow want to see a vastly -expanded-through-low-skill-immigration United States given that it was already taking 2-4 months in the Boston area to get a non-emergency appointment with a physician, a sign of a health care system that would snap during the next breeze of demand? I guess we now know the answer: he never had to wait.

Full post, including comments

Public transit and viral plagues in the U.S.

by philg

As soon as the coronaplague hit Massachusetts, Boston’s government-run public transit system (MBTA) cut its schedule. Despite the reduced number of people trying to move around, the result was packed trains (March 17 photos), perfect for spreading more plague.

I wonder if this is a good argument for why we need completely automated trains. As the U.S. gets more and more packed with humans (see “Modern Immigration Wave Brings 59 Million to U.S., Driving Population Growth and Change Through 2065” (Pew)), coronavirus-style plagues are inevitable. If mass transit systems have to cut schedules, perhaps due to human workers choosing not to show up and expose themselves (see “With unlimited paid sick leave for coronavirus symptoms, why will anyone work?”), the result will be a government-run service spreading the plague that other government agencies, e.g., public health, are trying to contain.

New York City (“Wuhan on the Hudson”?) seems to be the worst-hit region of the United States and it is also the place where Americans are most likely to take subway trains.

(Anecdotally, U.S. trains are far more packed than the ones I rode on Chinese metro systems in Shanghai (metro population pushing 35 million) and Suzhou (comparable to Boston). The Chinese trains run so frequently, often every 1-3 minutes, that they are less crowded than Boston and New York City trains. It might also help that there are a lot of lines and that the trains are fast.)

Readers: What do you think? To reduce the impact of the next plague in the U.S., should we invest in automated transit systems that can run with good amounts of social distance during plague periods?

Related:

Full post, including comments