Peter's COVID Consolidated Brief for 29 March

...See the latest in my 2 Apr Brief!

COVID-19 is a rapidly changing situation and it is hard to keep up to date. However, right now I am following COVID-19 full time and I read widely and I read a lot. I’m going to experiment with providing a public consolidated brief that tries to consolidate everything I read into one short, actionable list so other people don’t have to re-create my work. This way I can save time and fight research debt.

This brief assumes you are up to date on most things that have happened since around the 25th of March and will aim to keep you up to date on the latest over the past three days or so.

Note that I am not a domain expert and I urge some caution in over-relying on my selection and interpretation of these links.

This brief follows my research agenda. I am going to keep that up to date as well. I will also keep eyes on the LessWrong Coronavirus Agenda and submit to the LessWrong links database. Further discussion will be in the EA Coronavirus Facebook Group.

Doing Your Part! How You Can Stay Safe and Help the Fight!

Rob Besinger offers some advice for staying safe that I have not had the time or expertise to verify, but will reprint uncritically:

definitely, definitely self-quarantine
Avoid people
If you do need to be around people, wear something over your mouth and nose
Don’t touch your face (duh)
Wash your hands (duh)
Eat well, sleep well, get exercise
Consider stockpiling a month of food (...if you still can at this point, IMO good to have a at least a week or so above your usual amount of food)
Consider printing out copies of your health records
Regularly disinfect commonly touched surfaces like door handles and light switches
Consider covering commonly touched surfaces with copper tape
Probably stop taking NSAIDs like ibuprofen
Probably even-better-advice-than-normal to consume 2000-6000 IU of Vitamin D daily, in the morning
Consider running an air purifier
Understand how COVID-19 usually presents and progresses, so you can make an informed guess about how likely you are to have it
Take zinc immediately if you start feeling any cold-, flu-, or COVID-19-like symptoms
Start monitoring your oxygen immediately if you develop a fever or experience significant chest tightness or difficulty breathing

80,000 Hours puts out a list of things people can do to help with COVID:

Research to understand the disease and to develop new treatments and a vaccine.
Determine the right policies, both for public health and the economic response.
Increase healthcare capacity, especially for testing, ventilators, personal protective equipment, and critical care.
Slow the spread through testing and isolating cases, as well as mass advocacy to promote social distancing and other key behaviours, buying us more time to do the above.
We also need to keep society functioning through the progression of the pandemic.

80,000 Hours thinks it is people should switch to working on COVID-related projects if they’re roughly in the top 4% of people best suited to work on it—typically people who:

have highly relevant skills and/or useful connections—especially those who have medical training, can help with urgent hardware or software engineering efforts, or have knowledge of vaccines, public health, and government institutions
are not otherwise doing really important work
are highly motivated and informed on COVID
can switch into COVID-related work and switch back after without derailing one’s long-term career

Here’s my personal list of things you can do:

Find your equilibrium, prepare yourself, and make sure you are okay first before trying to help. Make sure you have what you need to continue to be healthy and successful. Find a way to have a happy quarantine. Here’s a bunch more ideas. There are also a million articles on this topic (these are my favorite four out of the 40+ I’ve seen).
If you are already working in an essential industry, are a valiant healthcare worker, etc., definitely keep doing that.
If you have the skills to contribute to vaccines, antivirals, etc… obviously do that.
Rest a lot if you feel sick. Do what you need to do to self-care and look after your mental health.
Contact your government decision-makers and let them know you support the shutdown and value public health. Now is an unusually important time to make your voices heard and convince others to do the same!
If you are a publicist, social media influencer, or have celebrity contacts, consider getting them onboard with maintaining public support.
If you have social media experience and/or online advertising experience, consider helping out with some social media campaigns.
Research one of my research ideas for coronavirus and publish your findings.
If you have expertise in data science or forecasting, besides trying to work on these research questions, it seems worth throwing significant time to various forecasting efforts like Metaculus’s Pandemic Questions, the Good Judgment Open, and/or Kaggle. This could potentially scale to consume a significant amount of EA talent, though it may not be that neglected.
If you have deep learning and image recognition experience, you could try to join https://www.covid19challenge.eu/
Find a project on “Help with Covid”, which also lets you filter by skill. Read through LessWrong and the “Effective Altruism Coronavirus Discussion” FB group. Look through this list of EA approaches. However, be wary of low neglectedness and widespread duplication of work.
Spend time helping aggregate and organize information, maybe by making the Coronavirus Tech Handbook nicer and updating Wikipedia.
Reach out to your local community, friends, family, and neighbors and make sure they feel supported and are doing okay in this trying time.
With the pause in normal work now could be a great time for some personal and organizational reflection. Self-evaluation can pay big dividends in the long-term. Perhaps now you have time to re-evaluate long-term strategy, evaluate hiring practices, management style, employee morale, team culture, etc.

80,000 Hours also suggests donations for COVD-19 relief: Center for Health Security at John Hopkins, Gates Foundation COVID-19 Funds, and the Center for Global Development.

You can now help fight COVID using your laptop’s spare cycles via Folding@Home.

A Glance at The Latest Situation

Things are still getting bad quickly. See FT’s latest graph reprinted below. …As Justin Wolfers puts it: “Project the U.S. line forward just 7 days, and we’ll be at 10,000 deaths in total. Project it forward a week after that, and we’ll be at 10,000 per day.” (Hopefully we’ll have flattened the curve a bit since then.)

The case numbers look bad, but at least the second derivative (growth in growth) shows some good news. The New York Times reports:

However, the situation still is dire: “The rate of increase in cases [in New York City] is far higher for the number of cases than it was in Wuhan or Lombardy, once they had reached similar numbers of cases. Other metropolitan areas, like Detroit and New Orleans, stand out as places where a coronavirus outbreak might escalate quickly without preventive measures. The Seattle and San Francisco areas, in contrast, seem to have made serious progress in flattening the curve.”

Here’s a very good visualization of percent of people infected by region in the world.

Note that case numbers are related to testing numbers, differences in who gets tested, and how cases, tests, and deaths are reported—and these all can differ country-to-country. This might be why Germany has such a low death-per-case rate. Italian data may also be underreported. Same with Spanish data. We should be prepared for the data to be a bit wonky and for comparisons to not be entirely apples-to-apples.

For example, I’ve been watching China-related COVID-19 reporting with a lot of anticipation… they seem to be doing very well at containing the virus and could become a model for the rest of the world. However, there’s also a lot of disinformation and misinformation around China. There’s good reason to distrust their numbers. There’s also good reason to distrust the distrust of their numbers. Some Chinese doctors who tested negative for coronavirus have later tested positive. …One thing that is easy to verify: China re-closes all cinemas over fear of a second coronavirus outbreak.

Japan should be having a terrible time right now. They were one of the first countries to get coronavirus cases, around the same time as South Korea and Italy. And their response has been somewhere between terrible and nonexistent. A friend living in Japan says that “Japan has the worst coronavirus response in the world (the USA is second worst)”, and gets backup from commenters, including a photo of still-packed rush hour trains. [...]

But actually their case number has barely budged over the past month. It was 200 a month ago. Now it’s 1300. This is the most successful coronavirus containment by any major country’s, much better than even South Korea’s, and it was all done with zero effort.

The obvious conclusion is that Japan just isn’t testing anyone. This turns out to be true – they were hoping that if they made themselves look virus-free, the world would still let them hold the Tokyo Olympics this summer.

But at this point, it should be beyond their ability to cover up. We should be getting the same horrifying stories of overflowing hospitals and convoys of coffins that we hear out of Italy. Japanese cities should be defying the national government’s orders and going into total lockdowns. Since none of this is happening, it looks like Japan really is almost virus-free. The Japan Times is as confused about this as I am. [...]

Also, what about Iran? The reports sounded basically apocalyptic a few weeks ago. They stubbornly refused to institute any lockdowns or stop kissing their sacred shrines. Now they have fewer cases than Spain, Germany, or the US. A quick look at the data confirms that their doubling time is now 11 days, compared to six days in Italy and four in the US. Again, I have no explanation. [...]

The third world …is in really deep trouble, isn’t it?

The numbers say it isn’t. Less developed countries are doing fine. Nigeria only has 65 cases. Ethiopia, 12 cases. Sudan only has three!

But they probably just aren’t testing enough. San Diego has 337 diagnosed cases right now. The equally-sized Mexican city of Tijuana, so close by that San Diegans and Tijuanans play volleyball over the border fence, has 10. If we assume that the real numbers are more similar (can we assume this?), then Mexico is undercounting by a factor of 30 relative to the US, which is itself undercounting by a factor of 10 or so. This would suggest Mexico has the same number of cases as eg Britain, which doesn’t seem so far off to me (Mexico has twice as many people). [...]

Nigeria and Mexico and so on make me confused in the same way as Japan – why aren’t they already so bad that they can’t hide it? If the very poorest countries in sub-Saharan Africa were suffering a full-scale coronavirus epidemic, would we definitely know? In Liberia, only 3% of people are aged above 65 (in the US, it’s 16%). It only has one doctor per 100,000 people (in the US, it’s one per 400) – what does “hospital overcrowding” even mean in a situation like that? I don’t think a full-scale epidemic could stay completely hidden forever, but maybe it could be harder to notice we would naively expect.

Tyler Cowen also asks “Where does all the heterogeneity come from?”: “Can anyone shed light on why the death rate is not higher in Iceland? Is it that the death rate is about to burst a week from now? [...] Similarly, Sweden hasn’t restricted public life very much and they do not seem to be falling apart? [...] It is possible that Cambodia, Thailand, and Vietnam still will be hit hard, but so far the signs do not indicate as such. Warm weather may play a positive role, though that remains speculative. The latest weather paper appears credible and indicates some modestly positive results. Of course weather won’t explain the relative Icelandic and Swedish success, if indeed those are truly successes.”

Maybe it’s still just a matter of time? Vox argues that Mexico’s coronavirus-skeptical president is setting up his country for a health crisis and Japan’s coronavirus crisis may be just beginning.

...Also, beware Goodhart’s Law in these metrics—originally US states were strongly incentivized to underreport, but now that relief is tied to caseload, they are strongly incentivized to overreport.

There is now data on ICU beds by US county.

...So Just How Bad Could This All Get?

“From Spain to Germany, Farmers Warn of Fresh Food Shortages” warns a Bloomberg article, mainly due to fewer workers being available to pick fruit. I’m pretty skeptical of COVID-related food shortages, but I still think it is important to monitor and be on the lookout. The NYTimes reports that there is still plenty of food in storage and supply chains are currently getting replenished just fine.

Gaze into the Crystal—The Latest Modeling and Forecasting

A Stanford team produces a stochastic model to forecast a “lightswitch approach” to lockdowns, where we alternatively lockdown and un-lockdown to continually keep the case load manageable enough to not overwhelm hospitals.

Vipul Naik estimates when we will get out of lockdown. He thinks the strict “shelter in place / go out for emergencies only” might get relaxed back to “most things closed”-level lockdown by mid-June and to the “most things open except large groups still banned”-level lockdown by summer 2021 and back to business as usual by summer 2022.

A survey of 18 epidemiologists say COVID-19 will cause approximately 195,000 deaths in the US and that a “second wave” of the virus is likely to occur between August and December.

Dr. Fauci predicts over 100,000 Covid-19 deaths in the United States.

A hospital-specific model tries to use more specific information about shortages. And another model, not tied to specific hospitals, but still modeling the impact of PPE shortages.

Now Let’s Talk Policy Response

Harvard Business Review outlines some key lessons from Italy:

“Recognize your cognitive biases. In its early stages, the Covid-19 crisis in Italy looked nothing like a crisis. [...] Threats such as pandemics that evolve in a nonlinear fashion (i.e., they start small but exponentially intensify) are especially tricky to confront because of the challenges of rapidly interpreting what is happening in real time.”
“The most effective time to take strong action is extremely early, when the threat appears to be small — or even before there are any cases. But if the intervention actually works, it will appear in retrospect as if the strong actions were an overreaction. This is a game many politicians don’t want to play.”
“The systematic inability to listen to experts highlights the trouble that leaders — and people in general — have figuring out how to act in dire, highly complex situations where there’s no easy solution.”
“Avoid partial solutions. A second lesson that can be drawn from the Italian experience is the importance of systematic approaches and the perils of partial solutions. The Italian government dealt with the Covid-19 pandemic by issuing a series of decrees that gradually increased restrictions within lockdown areas[… I]t backfired for two reasons. First, it was inconsistent with the rapid exponential spread of the virus. [...] Second, the selective approach might have inadvertently facilitated the spread of the virus.”
“Consider the decision to initially lock down some regions but not others. When the decree announcing the closing of northern Italy became public, it touched off a massive exodus to southern Italy, undoubtedly spreading the virus to regions where it had not been present.”
“An effective response to the virus needs to be orchestrated as a coherent system of actions taken simultaneously.”
“These rules also apply to the organization of the health care system itself. Wholesale reorganizations are needed within hospitals (for example, the creation of Covid-19 and non Covid-19 streams of care).”
“Finding the right implementation approach requires the ability to quickly learn from both successes and failures and the willingness to change actions accordingly. Certainly, there are valuable lessons to be learned from the approaches of China, South Korea, Taiwan, and Singapore, which were able to contain the contagion fairly early. But sometimes the best practices can be found just next door. Because the Italian health care system is highly decentralized, different regions tried different policy responses. The most notable example is the contrast between the approaches taken by Lombardy and Veneto, two neighboring regions with similar socioeconomic profiles.”
Good policies: “Extensive testing of symptomatic and asymptomatic cases early on. Proactive tracing of potential positives. If someone tested positive, everyone in that patient’s home as well as their neighbors were tested. If testing kits were unavailable, they were self-quarantined. A strong emphasis on home diagnosis and care. Whenever possible, samples were collected directly from a patient’s home and then processed in regional and local university labs. Specific efforts to monitor and protect health care and other essential workers.”
“It is especially important to understand what does not work. While successes easily surface thanks to leaders eager to publicize progress, problems often are hidden due to fear of retribution, or, when they do emerge, they are interpreted as individual — rather than systemic — failures.”
“Collecting and disseminating data is important. Italy seems to have suffered from two data-related problems. In the early onset of the pandemic, the problem was data paucity. More specifically, it has been suggested that the widespread and unnoticed diffusion of the virus in the early months of 2020 may have been facilitated by the lack of epidemiological capabilities and the inability to systematically record anomalous infection peaks in some hospitals.”
“More recently, the problem appears to be one of data precision. In particular, in spite of the remarkable effort that the Italian government has shown in regularly updating statistics relative to the pandemic on a publicly available website, some commentators have advanced the hypothesis that the striking discrepancy in mortality rates between Italy and other countries and within Italian regions may (at least in part) be driven by different testing approaches.”
“In an ideal scenario, data documenting the spread and effects of the virus should be as standardized as possible across regions and countries and follow the progression of the virus and its containment at both a macro (state) and micro (hospital) level.”

Rhode Island essentially declares war on New York City:

Rhode Island police began stopping cars with New York plates Friday. On Saturday, the National Guard will help them conduct house-to-house searches to find people who traveled from New York and demand 14 days of self-quarantine.

“Right now we have a pinpointed risk,” Governor Gina Raimondo said. “That risk is called New York City.”

WHO is very cagey / weird about Taiwan. WHO Director General, Bruce Aylward is asked about Taiwan’s membership in WHO… he responds by hanging up and then pretending to have not heard the question. Taiwan is not a member of WHO due to China’s insistence that Taiwan is a part of China.

A Bit About Life Under Quarantine

Americans have been changing their plans a lot:

Wild… “Our estimate is that about 12-13% [of churches in the US] were still open to in-person worship as of this week.”

The Psychologists Also Have Something to Say About This

For those studying the social psychology of the pandemic, there is now a research tracker.

The EA Australia Research Collaboration is running surveys to help policymakers with decision making about how to allocate resources to tackle COVID-19:

We doing something quite different from other ongoing surveys. Most of these are about country level comparisons and not about understanding behavioural drivers. For example, we can report things by location and demographic, but also why people are not doing the behaviours, e.g., 75% of males between 30 and 40 are social distancing but only 50% of males between 20-30. At only 55% adherence, the inner west region reports the lowest amount of social distancing. The main capability barriers are commuting and desire to see friends. 97% of those surveyed are always washing their hands, suggesting that this need no longer be a key communication target. [...]

Please consider joining us in collaboration. You can contribute by:

Helping to disseminate the survey via social networks or panel data.
Reaching and helping policy makers in your country with the data we collect
Helping to modify our report template to provide useful and interesting information to policy makers.
Developing reports and doing analysis for policy makers
Providing us with feedback based on your discussion data collection
Helping with write up and dissemination when we seek to publish this work

If you contribute to this project in any significant way then you will be recognised on all outputs and be an author on any subsequent paper. The bar for recognition will be relatively low (perhaps ~5 hours of work).

If You Still Own Envelopes, Check Their Backs—Here’s the Latest Cost-Benefit Analysis

The IGM Forum regularly polls economists about US economic policy. Economists rarely agree. But this time, economists are unanimous that the large contraction in economic activity is worth fighting coronavirus, we should not abandon the severe lockdowns, and the government should invest more in policy response.

“A comprehensive policy response to the coronavirus will involve tolerating a very large contraction in economic activity until the spread of infections has dropped significantly.”

“Abandoning severe lockdowns at a time when the likelihood of a resurgence in infections remains high will lead to greater total economic damage than sustaining the lockdowns to eliminate the resurgence risk.”

“Optimally, the government would invest more than it is currently doing in expanding treatment capacity through steps such as building temporary hospitals, accelerating testing, making more masks and ventilators, and providing financial incentives for the production of a successful vaccine.”

A study using data from the 1918 Flu found that greater economic growth was connected with lockdowns as opposed to the opposite. Faster social distancing also saved a lot of lives during the 1918 Flu.

Now Just What are the Tech Overlords up to?

Apple launches a COVID-19 screening tool. According to TechCrunch: “The site is pretty simple, with basic information about best practices and safety tips alongside a basic screening tool which should give you a fairly solid idea on whether or not you need to be tested for COVID-19.”

The New York Times calls for “Big tech needs to rapidly build and scale a cloud-based national ventilator surveillance platform which will track individual hospital I.C.U. capacity and ventilator supply across the nation in real-time. Such a platform — which Silicon Valley could build and FEMA could utilize — would allow hospitals nationwide to report their I.C.U. bed status and their ventilator supply daily, in an unprecedented data-sharing initiative.” Probably already being done by 18 different groups now, but might still be worth exploring?

Zuckerberg donates $25M to the Gates Foundation to fight coronavirus. It’s an accelerator to find new possible antiviral drugs and total funding for the accelerator is now at $125M.

Not to be outdone, Google pledges to donate $800 million and 3 million face masks in an effort to combat the coronavirus… however, over 75% of this fund comes in the form of Google ad grants and cloud computing credits.

Lastly, Mayo Clinic and Amazon launched a collaboration to increase COVID-19 testing and vaccine development: “The private industry effort, spearhead by Mayo Clinic’s John Halamka, M.D. and other industry leaders, plans to leverage the strengths of healthcare organizations, technology companies, non-profits, academia, and startups to provide a focused response to the coronavirus outbreak.” It’s not super clear what this means but I’m glad it’s happening.

~

Maybe we shouldn’t be using Zoom? Consumer Reports reports “Zoom Calls Aren’t as Private as You May Think”. Engadget reports “Zoom happens to be a privacy nightmare with a terrible security track record” and that “Zoom collects your physical address, phone number, your job title, credit and debit card information, your Facebook account, your IP address, your OS and device details, and more and traffics that data with whomever it’s doing business with”.

~

Predictably, Edward Snowden warns about the dramatic increase of tech surveillance.

And How Do We Get Out of this Mess? Vaccines, Treatments, Testing, Tracing, etc.

US testing is no longer increasing exponentially:

...But I’m told this is a game changer: Abbot launches molecular point-of-care test to detect novel coronavirus in as little as five minutes. They’ve already received FDA emergency use authorization.

Dr. Fauci outlines ambitious plan to scale up COVID-19 vaccine. Looks like the proposal is to start ramping up production of a vaccine while the candidate is still in Phase II clinical trials. This risks spending a ton of money producing a vaccine that ultimately might not get approved by the FDA. Dr. Fauci has suggested hundreds of millions in incentives to make this work.

The Washington Post reports that “hVIVO, a clinical research group in London, has attracted more than 20,000 volunteers willing to be infected with tamer relatives of the virus that causes Covid-19 in exchange for a fee of £3,500 ($4,480).”

To stop COVID-19, test everyone, repeatedly: “We propose an additional intervention that would contribute to the control of the COVID-19 pandemic and facilitate reopening of society, based on: (1) testing every individual (2) repeatedly, and (3) self-quarantine of infected individuals. By identification and isolation of the majority of infectious individuals, including the estimated 86% who are asymptomatic or undocumented, the reproduction number R0 of SARS-CoV-2 would be reduced well below 1.0, and the epidemic would collapse. This testing regime would be additive to other interventions, and allow individuals who have respiratory symptoms due to other causes to return to work, but would have to be maintained until a vaccine becomes available. Unlike sampling-based tests, population-scale testing does not need to be very accurate: false negative rates up to 15% could be tolerated if 80% comply with testing, and false positives can be almost arbitrarily high when a high fraction of the population is already effectively quarantined.”

And Now a Word From the Lamestream Media

Kelsey Piper is the best. Vox, and other media outlets have not apologized for continually downplaying the coronavirus and calling out people as fearmongering. For example, on 13 February, Vox made fun of the prescient tech industry, deriding them with “Although public officials in the area say the virus is contained for now, some in the tech industry fear the virus will spread out of control.” Kelsey Piper, Vox Future Perfect author, is at least willing to admit she made a mistake. No one else seems to be… yet Kelsey is getting all the hate for it. :(

(I also made a mistake by privately stockpiling on 8 February, but not telling anyone of my fears out of a meta-fear of looking like some dorky prepper. I also apologize for this.)

The Non-Profit Impacts

American non-profits that (a) existed before 1 March 2020, (b) have fewer than 500 employees, and (c) keep staff on payroll can get what might amount to free money from the government. This deserves urgent further investigation.