5 April 2020
Just wear it
The topic of masks seems relatively simple; here’s the emerging consensus (but note some dissenters):
- Everyone in a significantly impacted area should wear one in public.
- The main reason is to reduce the risk of infecting other people.
- You should do it even if you don’t have symptoms, because it seems like many more people may be both asymptomatic and contagious than we thought.
- A surgical mask, a homemade cloth mask, or even a scarf helps for this purpose.
- These simple masks might or might not protect you, but it doesn’t see like they could hurt.
- They could have a side benefit of keeping you from touching your face (not clear that there is evidence for this).
- In terms of social signalling, until recently wearing a mask in the west (not Asia) might have been stigmatised as signalling that you were sick; but quickly it will come to signal that you are a responsible citizen.
- P95 masks definitely offer more protection, and if there were unlimited supplies I would recommend wearing one in public. But for as long as there are PPE shortages, you should instead see if you can donate them to a local hospital or nursing homes.
What do experts say?
The case for: The New York Times’ roundup is helpful. Here’s the CDC’s advice: wear a mask, even a homemade one. Austria now requires them in grocery stores. These experts say you should.
This roundup of recommendations from health authorities around the world in the Lancet tips cautiously in favour of wearing masks. (Dated 20 March, somewhat out of date.)
The case against?
David Heyman, professor of infectious disease epidemiology at the London School of Hygiene & Tropical Medicine (that’s way up there in terms of credentials), is sceptical. His arguments for caution: (a) We don’t have strong evidence to support a recommendation one way or another. (b) It’s possible (again, little evidence) that in removing a mask you could increase your risk of infection. (c) Encouraging mass adoption of masks could reduce the supply available to hospital workers. (d) Wearing a mask could give a false sense of security and cause us to take less care elsewhere (e.g., in washing hands).
Also note that the WHO has not updated their advice as of today, 5 April 2020: “If you are healthy, you only need to wear a mask if you are taking care of a person with suspected 2019-nCoV infection.”
(As an aside–though I come out in favor of wearing masks in public for those in impacted areas, I love arguments like Heyman’s, which are brave enough to question the conventional wisdom and, with cool rationality, consider the potential counterarguments.)
I also appreciated this post by Ed Baker: https://medium.com/@esbaker/homemade-face-masks-could-be-the-silver-bullet-for-defeating-covid-19-5ed96d61544b
Asymptomatic transmission
I’ve heard increasing numbers of claims about asymptomatic transmission recently, which is a really important topic.
As a matter of definition, I’m using “asymptomatic transmission” to include two types of situations: transmission during a pre-symptomatic period of a case that later goes on to develop symptoms; and transmission from a case that never develops symptoms.
It’s also worth noting, as this New York Times article does, that there is no standard definition of “asymptomatic.” It often turns out to mean “pre-symptomatic.“
Before considering the current evidence, let’s also distinguish between two questions.
- Are a significant proportion of those infected with COVID-19 mostly or completely asymptomatic (to the point that they might not even know that they were infected)?
- Are a significant proportion of instances of transmission coming from infected individuals during a period in which they are asymptomatic?
The second is the important question in the near term. Lots of asymptomatic transmission would be bad news for our ability to control the spread; and in turn, that makes it harder to return to normality. We would need stricter control measures for longer. And subsequent outbreaks (e.g., from imported cases) would be harder to detect and contain.
On the other hand, if little asymptomatic transmission is happening, but there are significant number of asymptomatic cases, that would (strangely enough) be really good news. It would mean that there are likely many more cases than we know about, which would mean that the IFR (the true fatality rate) is much lower, since the denominator of the CFR (the measured fatality rate) would be understated–i.e., that the virus is less deadly than we currently think. And it would mean getting to herd immunity faster.
What does the evidence say so far?
In this NPR interview, the head of the CDC, Robert Redfield, says that as much as 25 percent of people infected with the new coronavirus don’t present any symptoms but can still transmit the illness to others. And even those who do go on to present symptoms can be infectious 48 hours prior to symptoms.
This Business Insider article lists many examples of studies (some small-scale) consistent with either asymptomatic cases, asymptomatic periods of a case, or asymptomatic transmission (they don’t always distinguish among these).
This New England Journal of Medicine paper says, “The viral load that was detected in the asymptomatic patient was similar to that in the symptomatic patients, which suggests the transmission potential of asymptomatic or minimally symptomatic patients.” Note that this was one patient out of 17, so a tiny sample.
This paper looks at the population of a nursing home in Washington State, and finds that roughly half of the 76 who tested positive were asymptomatic on the day they were tested.
At the same time, the WHO’s situation reports continue to say (quoting 2 April 2020 sitrep), “There are few reports of laboratory-confirmed cases who are truly asymptomatic, and to date, there has been no documented asymptomatic transmission. This does not exclude the possibility that it may occur. Asymptomatic cases have been reported as part of contact tracing efforts in some countries.”
Because of the vagueness about what “asymptomatic” means, one academic argues that the distinction between symptomatic and asymptomatic transmission is unhelpful, and should be replaced by “documented” vs “undocumented.” Using this distinction, he and collaborators published a paper that estimated that 86% of infections in China were from undocumented cases.
My conclusions are as follows:
- There are almost certainly asymptomatic periods of cases that go on to be symptomatic.
- There are probably fully asymptomatic cases (which could just mean: really mild symptoms).
- There’s increasingly strong evidence that transmission can occur in the absence of symptoms.
- I haven’t found any good sources to enable us to estimate the proportion of transmission that is asymptomatic.
- There are surely many instances of transmission from undocumented cases–perhaps even the majority of all transmission.
Why is Germany’s fatality rate so low?
A good New York Times article on this topic. The key points:
- Germany’s CFR is 1.4% vs 14% in Italy, 10% in Spain, France and Britain, 4% in China, 2.5% in the US, and 1.7% in South Korea.
- The main reason is probably a statistical one: Germany’s exceptionally high rate of testing likely means that the denominator of the CFR calculation is less understated in Germany than elsewhere. (A good reminder that what we care about is IFR, not CFR, and that CFR’s can’t be compared across regions with different testing rates.)
- Germany’s infected population (at least the population we know about) is unusually young:
- “But the average age of contracting the disease [in Germany] remains relatively low, at 49. In France, it is 62.5 and in Italy 62.”
- Why? “Many of the early patients caught the virus in Austrian and Italian ski resorts and were relatively young and healthy.”
- (But note that these average ages are among detected cases; probably places like France and Italy, which test a far lower proportion of the population than Germany and focus testing on serious cases, have many more cases among young people than they have detected or reported.”
- Germany may be doing a better job treated patients which would reduce the fatality rate. “But there are also significant medical factors that have kept the number of deaths in Germany relatively low, epidemiologists and virologists say, chief among them early and widespread testing and treatment, plenty of intensive care beds and a trusted government whose social distancing guidelines are widely observed.”
- “The reason why we in Germany have so few deaths at the moment compared to the number of infected can be largely explained by the fact that we are doing an extremely large number of lab diagnoses,” said Dr. Christian Drosten, chief virologist at Charité, whose team developed the first test.
- “When I have an early diagnosis and can treat patients early — for example put them on a ventilator before they deteriorate — the chance of survival is much higher,” Professor Kräusslich said.
- Germany started with a higher ICU capacity and built more capacity quickly. “All across Germany, hospitals have expanded their intensive care capacities. And they started from a high level. In January, Germany had some 28,000 intensive care beds equipped with ventilators, or 34 per 100,000 people. By comparison, that rate is 12 in Italy and 7 in the Netherlands. By now, there are 40,000 intensive care beds available in Germany.”
Overall, it appears that the German approach is a role model that should be widely followed.
Robert Schiller on economic impact
Nobel laureate Robert Schiller is a hero of mine; a great, long-term, big-picture thinker. SO I was excited to see his opinion piece in the New York Times on the stock market.
Key excerpts:
“[I]s the market cheap today? Is it expensive? Here’s my equivocal answer: It’s not as highly priced as it was just months ago.” […]
I worry that the present anxious situation may stay in the collective memory for decades, much as the stock market crash of 1929 did. That could make people more risk-averse, possibly portending lower valuations on the stock market. […]
Based on history we would expect to see it to be a reasonable long-term investment, attractive at a time when interest rates are low. […]
As a practical matter, my advice is to look at your portfolio to make sure that it is not so heavily weighted to stocks that further losses would be unbearable. Otherwise, I’d try not to worry too much about the stock market.”
Modeling humour
Subjective Probability 