The Real Question is How to Restart

When??

When can we finally go back to something resembling a “normal” life, go back to work and restart our economies, embrace again long-lost family, friends, co-workers? And, in too many cases, when can we get together to properly mourn those we have lost?

I cannot tell you “when.” Everyone wants the answer to that question. I do, too. But that is not the first question to ask.

The first question is “how”?

So, let’s discuss “how”.

First, just as much as “when” depends on “how,” “how” depends on “where”.

We are still in the “first wave” of the pandemic. There are substantial differences in rates of localized spreading as well as containment measures: differences between countries and also within different regions/states of the same country.

For example, Lombardia (Milan region in Italy), London (UK), and New York City are all at a very different stage in their pandemic than the vast majority of the rest of the countries they are in. There are all sorts of very messy complications in assuming what a “second wave” could look like and when it could hit. A LOT of these complications arise from having different regions not maintaining contemporaneous strict lockdowns (because the virus spreads there later, or because of much lower population density, or because of different local health policies). I will not go into that as my head hurts just by thinking about it (I am maintaining a strict regimen of responsible alcohol consumption during the lockdown, so that is NOT the reason for the headache, thanks very much for asking).

I will add, however, one comment from Rhode Island Governor Gina Raimondo (another fellow Italian, and a former medtech VC) on Mon. Apr 13. Her remarks came during the launch of a regional coordination initiative between governors of NY, NJ, CT, RI, DE, PA, MA: “The reality is this virus doesn’t care about state borders, and our response shouldn’t either.”

I will focus below mostly on the US, since this is where I now live. I will also try to translate some lessons hopefully learned by other countries who entered the peak of their regional clusters before the US. For some background, please refer to the previous articles in the Timmerman Report.

• Lessons from Italy. Mar. 10
• Eight Days Later, Italy vs. US. Mar.18
• Let it Rip or Shelter at Home? Mar. 23
• What’s the Real Risk of Death? Apr. 1

Before I start, a (not so quick, it turns out) reminder. Lesson No. 1 with exponential infection spreading: you might think you are running ahead of the viral spread, but it is probably the virus that is way ahead of you.

There is now substantial evidence, as if things were not bad enough, that this virus has an R-naught (R0) probably well in excess of 2, which makes it very infectious indeed. If you want a primer on this, and do not want to read my older Timmerman Report posts, you should watch a movie. 

“Contagion”, the eerily prescient film from 2011 on a fictitious pandemic originating from Hong Kong, in my opinion, should be shown in every house all around the world for the next 6 months. It should become practically required school viewing, and get a “posthumous” Oscar. Watch, in particular, the TV interview Jude Law’s character gives roughly mid-film. Though a great actor (one of my wife’s very favorites), he is not a good guy in the movie (and I am still confused by how he manages an Australian accent at times). BUT he is spot-on about exponential spreading. The scientific consultant to the movie, Larry Brilliant, helped eradicate smallpox. Apart from having the most apt last name in the universe (can you tell I am jealous? No? Did you forget my first name? Go check your Shakespeare), Larry is a GREAT epidemiologist: read more about him here.

Apart from a few, required dramatic Hollywood flourishes (a vaccine created, manufactured and distributed in only 133 days against a totally new virus is practically science fiction; viruses do not “mutate” dramatically that quickly, etc.), the movie covers incredibly well virus provenance (also from bats originally), infectivity from touch (do not touch your face, wash hands), and contact tracing, as well as many other important aspects (who gets vaccine first etc.). I watch it almost once a week now. It does also have a happy ending, I think. This tragedy we are living through will, too.

Now, back to “when”.

I am assuming by now that you are all more or less familiar with the three gating “phases” to re-open US states’ economies in the coming months: after all, the audience of this report is quite knowledgeable about Phase I, II and III trials. And this is very much a trial we are all going through.

For those who are not as avid a consumer of news (or not US-based), the US administration has given state governors guidance and criteria on how to reopen state economies. Note that states (from the very little I know of the US Constitution) very much have discretion on how to execute these guidelines. The full outline from the administration is here.

There are a number of gating criteria which need to be satisfied before entering Phase I:

• Downward trajectory of influenza-like illnesses (ILI) AND COVID-like syndromic cases reported within a 14-day period;
• Downward trajectory of documented positive cases within a 14-day period AND downward trajectory of positive tests as a % of total tests within a 14-day period (with flat or increasing volume of tests)
• Hospitals need to be able to treat all patients without resort to crisis care (i.e., I guess, without having to ration critical care like ICU beds or ventilators); AND they need to have a robust / at scale testing program in place for healthcare workers (including serological testing, not yet available).

Subsequently to satisfying these criteria, each of the phases should last, at a minimum, 14 days. Phase I includes many of the current lockdown measures (avoid non-essential travel; do not gather in groups). BUT it says venues such as restaurants, churches and sport arenas “can operate under strict physical distancing protocols”. In Phase II, non-essential travel can resume, schools can reopen and bars can operate “with diminished standing-room occupancy”. Phase III is almost back to our old definition of “normal”, in which states that are seeing a continued downward trend of symptoms and confirmed cases could allow “public interactions” with physical distancing and unrestricted staffing at worksites. Visits to care homes and hospitals can resume and bars can increase their standing room capacity. The at-risk demographic should still avoid public, crowded areas.

So, back to “how”. What do we need to achieve to get even to Phase I?

• Testing: the guidelines rightly focus on tests as a key gating criterium for starting to think about re-opening. We need two types, and probably many different formats of test, since there is a tradeoff between speed vs volume to think about: a) RNA tests, to check if you have virus genetic material in you (which, presumably, means you are / have been infected); b) serological tests, unfortunately not yet really available at scale in the US, to check if you have developed antibodies against the virus (which, presumably, means you have been infected and have developed some sort / level of immunity, at least for some time). What we need is BOTH types of testing, in massive scale, deployed as soon as possible BEFORE we can think of re-opening. Ideally you should be able to test ~1% of the US population per week with RNA tests, and very, very roughly half of that in serological tests. We are, unfortunately, very far from achieving that level of scale in the US. Just to give some statistics, as I write this morning Eastern Time April 17, the US (population ~330M) has performed ~a total of 3.4M RNA tests (~1% of the US population, or ~10k tests/1M people), of which ~678k came back positive (~20%). In comparison, Germany and Italy (population ~80M & 60M respectively), have performed ~1.7M and ~1.2M tests, or roughly 2x as many as the US per 1M population. So, it is absolutely essential to increase the volume of RNA tests by a factor of at least 5-10x in the US in the next 4-5 weeks. The speed of obtaining a test result also needs to dramatically increase: developing rapid testing formats that can provide a response within 15-30 minutes is essential to allow contact tracing and to isolate infected people immediately (more on contact tracing below). Serological testing is also absolutely necessary. There have been, and continue to be, a series of issues on these tests but a number of diagnostics companies (from Abbott to Roche) are announcing rapid plans to scale up their platforms to address these. That said, there is unlikely to be any serious availability of serological testing until June at the earliest. Without serological testing, it is very hard to identify immune people and therefore to “release” them back into the community, with the obvious positive implications for the economy, healthcare infrastructure, etc. Also, and very important: as I write, NY State + NJ represent ~44% of all confirmed cases in the US, and have performed ~21% of ALL tests in the country (worldometers.info/coronavirus). So, we have a long way to go. If you want to really geek out on what does it mean to “scale testing”, I have written a few words in a little appendix at the bottom of this post. (See “Addendum”)
• Contact Tracing: this piece of the puzzle is extremely important, as ramping up testing capacity is necessary but not sufficient to contain the pandemic. What should you do once somebody has tested positive to the RNA test? They should be isolated immediately, ideally away from their family, and everybody they have been in contact with should also receive the test: all their “contacts” need to be “traced” and tested. This is all the more important as there is now substantial evidence that: a) there is a substantial (probably ~50%) number of people who are infected and contagious but asymptomatic, BUT ALSO b) that people who will go on to develop symptoms start spreading the virus starting ~2 days AHEAD of showing symptoms. Contact tracing is equivalent to old shoe-leather detective work, and involves extensive interviews with the people who tested positive, their families, their co-workers. Therefore, it requires a large number of trained individuals to be performed correctly. It is very labor intensive. There are initiatives for automating some of that hard task by using phone apps to supplement human operators (S. Korea and Singapore have used similar apps with great success): Google and Apple are joining forces to use their combined dominance of mobile phone operating systems to develop such an app. Salesforce is also joining the fray. There are obvious privacy concerns associated with such an app and its use of location and contact data: those are not for me to address, here or elsewhere, and they are important concerns. That said, I do not personally believe, seeing the level of spread of the contagion in the US, that there will be a sufficient number of qualified trained human operators that could perform this task without some technology support.
• PPE / Masks: we need to continue (start??) ramping up manufacturing of protective equipment and masks for healthcare workers (and eventually for the general population). Data reported from Italy are staggering and scary: roughly 10-20% of all healthcare workers in Lombardia tested positive to the virus. These selfless warriors are literally sacrificing their lives to care for patients. No effort should be spared to provide them and the hospitals they work in with equipment and relief. We are talking multiple hundreds of millions of masks needed per month: healthcare workers need to be able to use multiple (3-5) N95 masks / day to take a break, drink and eat, and not have to recycle the same mask, which is sadly the case right now. It is also important, since there is evidence of aerosol transmission with this virus, that people use masks in general if/when leaving their homes. Since we do not know if a person is spreading the virus or not, masks help “contain” the radius of infectious particle spread, as well as provide some protection for at-risk demographics (especially if masks are N95 or N99 type).

Until we see ALL these fundamental steps in place, NATION WIDE, it is very hard to start thinking about “when.” If only New York were to follow these steps, then it’s only too easy for people to spread the virus elsewhere as people travel around the country by planes, trains, buses and cars. What we are doing right now with social distancing measures, and at incredible, horrific economic costs, is lowering R-naught (R0), the rate of infection of the virus, below 1. This is essential if we are going to catch up.

There is evidence it’s working. Every day there is a lower number of new people who become infected. This is a prerequisite step to be able to reduce the number of infected people to manageable numbers.  The painful work of social distancing to bring down the rate of transmission has to be done first. Without that step, testing and contact tracing are impossible: there are just too many people infected to trace all their connections.

Think of New York City, with roughly ~50% tests coming back positive: how do you trace contacts in a densely populated urban area, which necessarily requires the use of public transportation to function, when there are so many cases? Also do not forget, New York was very much part of the first wave: the same will apply to any other large metro area as soon as we relax restrictions, without having a commensurable capacity of performing tests/contact tracing (and provided we have rapid assay formats).

So, it is sensible, in the government’s guidelines above, to stress a “reduction” in the number of positive tests over 14 days (roughly the period required for a person to not infect others any more). This should allow the current pool of people infected to shrink, perhaps even very substantially, to numbers low enough to be traced if tested positive. I also encourage you (if you have not gotten to watch the movie yet) to do a mental exercise of how many people we interacted with in close proximity, and how many surfaces we used to touch during the “old normal” days we used to live in. It is literally hundreds, at least in my case (Italians are eminently social and we are also quite tactile in our social interactions). Again, watch “Contagion” (Disclosure: I do not receive royalties on the movie streams…). 

So, now to the (blind) guessing part: “when” do we think we might have figured out the “how” well enough to hazard some projections for when Phase I will start?

Spoilers: I do not think it is going to be before June, at the earliest. This is not a pessimistic projection, I believe. PLEASE take that with a “couple of mines” worth of salt. I would really like to see some random RNA and serological testing in a couple of places (outside of NY or NJ or Michigan) to understand more how diffused the virus is in the country. As of this writing on Apr. 17, we are still flying blind. I cannot wait to see the results of the serological random survey of 100,000 people that Germany is performing.

Some other random data points to support the very rickety framework I used to come to that “June” guess:

On Apr. 14, Anthony Fauci (in the impossibility that anybody who reads this does not know who he is, “Tony” is director of the National Institute of Allergy and Infectious Diseases (NIAID) and member of the White House Coronavirus Task Force), told the Associated Press that the May 1 reopening projection was “a bit overly optimistic” for many areas of the country. That’s because a strong testing and tracing system would be needed before social distancing measures were lifted. I guess we all agree with the good doctor here, especially since Fauci is an Italian name (which, hilariously enough, means “Jaws” in Italian: the comic implications / associations with the movie of the same title and the various decisions made by elected officials in said movie are too funny to mention). “Overly optimistic” feels a bit “overly kind” to that prediction statement, but, you know, some people have to be mindful of their audience.

Yesterday, Apr. 16, New York Gov. Andrew Cuomo said that his state would remain under stay-at-home orders until May 15. New York is likely beginning to descend the slope of the infectious spread. Note that a number of other countries (from Italy to the UK) have repeatedly pushed off the timing of a potential “re-opening” due to the scale of the infection spread being detected as more testing has gone online. I am guessing the same thing will happen here, especially since New York is detecting a positive testing rate of ~50% of tests – an incredibly high rate. The statistic can be biased since they are only testing people at high probability of being infected, but still.

Germany will reopen schools on May 3, with Austria even earlier than Germany. Both countries did a great job of containing the infection early on, have first-class and abundant hospital critical care units (Germany has the highest ICU bed concentration probably in the world) and both scaled up testing very early. So, even states and cities which were very much hit in the first wave might need to wait longer to get there. Not every city / state hit in the first wave is in the same spot: The San Francisco Bay Area and all of California are way ahead of the curve. So is Washington state: both acted early and ramped up social distancing / testing early, and have also much lower population density than New York City.

The issue is then, again, the “where” to decide the “how” that gets us to “when”: you can see here when different states issued stay-at-home orders. The differences are very wide, and frankly, puzzling. How do we solve for interstate travel/traffic to avoid a large second wave, at the same time as summer kicks in and people want to escape their home confinements and rejoin families/ friends as soon as restrictions are lifted?

This is a huge problem, and has massive implications for how we behave. I do not think we can simply rely on individual self-policing of behavior. I know this does not jibe with the ruggedly individualistic, “frontier” culture of the country. But, guess what: the virus does not care. We should.

Follow Otello Stampacchia on Twitter: @OtelloVC

This article expresses the personal views and perspectives of the author. The views and perspectives expressed here do not necessarily represent the views or perspectives of Omega Fund Management, LLC or any officer, director, partner, member, manager or employee of Omega Fund Management, LLC or any of its affiliated entities.

Addendum: More on Testing

“Testing capacity”, as well as “contact tracing” are very vague terms. As I am not sure everybody is familiar with what they actually mean, here are some thoughts: to test and contact trace effectively, at the scale we are talking about here (literally hundreds of millions of test), you need to think and coordinate a huge number of factors (each of which is very likely a rate-limiting factor). Some of these factors are: type of testing (batch testing in a central laboratory, which can process at high throughput a large number tests but has slow turnaround, vs localized testing in city hospitals?); number of locations where tests are available (dense cities vs rural areas: you do not want people piling up in a queue to get tested); number of total test kits; reagents needed to produce the tests (which are in very scarce supply right now); personnel trained to extract the material needed to perform the test; personnel trained to perform the test (they are not necessarily the same); protective equipment needed for all this personnel. Same thing for contact tracing: it is a process that needs to be taught, requires a certain amount of psychological expertise and understanding of what is going through the head of the person in front of us, who might well be under a certain amount of distress. So it’s hard to train thousands of people to perform this without some technical tools to aid in contact tracing. So this is not simple. And every US state should not have to go through a steep learning curve to develop all this infrastructure, in all its complexity and inevitable mistakes, on their own. This is where it pays to adopt best practices in a coordinated fashion across the country.