Part 1: Introduction
Times being what they are, in which challenging events abound and good information is hard to come by, I am delving back into writing about infectious disease (ID). While I’ve not been posting here about the intersection of ID, preparedness, and biosecurity, I have continued to work on these problems as a consultant for corporations, the US government, and the WHO. More on that in a bit, because my experience on the ground at the WHO definitely colors my perception of what the organization has said about events in China.
These posts will primarily be a summary of what we do, and do not, know about the current outbreak of the disease named COVID-19, and its causative agent, a coronavirus known officially as SARS-CoV-2 (for “SARS coronavirus-2”). I am interested in 1) what the ground truth is as best we can get to it in the form of data (with error bars), and I am interested in 2) claims that are made that are not supported by that data. You will have read definitive claims that COVID-19 will be no worse than a bad flu, and you will have read definitive claims that the sheer number of severe cases will overwhelm healthcare systems around the world, potentially leading to shocking numbers of fatalities. The problem with any definitive claim at this point is that we still have insufficient concrete information about the basic molecular biology of the virus and the etiology of this disease to have a good idea of what is going to happen. Our primary disadvantage right now is that uncertainty, because uncertainty necessarily complicates both our understanding of the present and our planning for the future.
Good sources of information: If you want to track raw numbers and geographical distribution, the Johns Hopkins Coronavirus COVID-19 Global Cases dashboard is a good place to start, with the caveat that “cases” here means those officially reported by national governments, which data are not necessarily representative of what is happening out in the real world. The ongoing coverage at The Atlantic about testing (here, and here, for starters) is an excellent place to read up on the shortcomings of the current US approach, as well as to develop perspective on what has happened as a result of comprehensive testing in South Korea. Our World In Data has a nice page, updated often, that provides a list of basic facts about the virus and its spread (again with a caveat about “case count”). Nextstrain is a great tool to visualize how the various mutations of SARS-CoV-2 are moving around the world, and changing as they go. That we can sequence the virus so quickly is a welcome improvement in our response, as it allows sorting out how infection is spreading from one person to another, and one country to another. This is a huge advance in human capability to deal with pathogen outbreaks. However, and unfortunately, this is still retrospective information, and means we are chasing the virus, not getting ahead of it.
How did we get here?
My 2006 post, “Nature is Full of Surprises, and We Are Totally Unprepared”, summarizes some of my early work with Bio-era on pandemic preparedness and response planning, which involved looking back at SARS and various influenza epidemics in order to understand future events. One of the immediate observations you make from even a cursory analysis of outbreaks is that pathogen surveillance in both animals and humans needs to be an ongoing priority. Bio-era concluded that humanity would continue to be surprised by zoonotic events in the absence of a concerted effort to build up global surveillance capacity. We recommended to several governments that they address this gap by aggressively rolling out sampling and sequencing of wildlife pathogens. And then not much happened to develop and real surveillance capacity until — guess what — we were surprised again by the 2009 H1N1 (aka Mexican, aka Swine) flu outbreak, which nobody saw coming because nobody was looking in the right place.
In the interval since, particularly in the wake of the “West Africa” Ebola outbreak that started in 2013, global ID surveillance has improved. The following years also saw lots of news about the rise of the Zika virus and the resurgence of Dengue, about which I am certain we have not heard the last. In the US, epidemic planning and response was finally taken seriously at the highest levels of power, and a Global Health and Security team was established within the National Security Council. That office operated until 2018, when the current White House defunded the NSC capability as well as a parallel effort at DHS (read this Foreign Policy article by Laurie Garrett for perspective: “Trump Has Sabotaged America’s Coronavirus Response”). I am unable to be adequately politic about these events just yet, even when swearing like a sailor, so I will mostly leave them aside for now. I will try to write something about US government attitudes about preparing to deal with lethal infectious diseases under separate cover; in the meantime you might get some sense of my thinking from my memorial to virologist Mark Buller.
Surprise? Again?
Outside the US government, surveillance work has continued. The EcoHealth Alliance has been on the ground in China for many years now, sequencing animal viruses, particularly from bats, in the hopes of getting a jump on the next zoonosis. I was fortunate to work with several of the founders of the EcoHealth Alliance, Drs. Peter Daszak and Billy Karesh, during my time with Bio-era. They are good blokes. Colorful, to be sure — which you sort of have to be to get out of bed with the intention of chasing viruses into bat caves and jumping out of helicopters to take blood samples from large predators. The EcoHealth programs have catalogued a great many potential zoonotic viruses over the years, including several that are close relatives of both SARS-CoV (the causative agent of SARS) and SARS-CoV-2. And then there is Ralph Baric, at UNC, who with colleagues in China has published multiple papers over the years pointing to the existence of a cluster of SARS-like viruses circulating in animals in Hubei. See, in particular, “A SARS-like cluster of circulating bat coronaviruses shows potential for human emergence”, which called out in 2015 a worrisome group of viruses to which SARS-CoV-2 belongs. This work almost certainly could not have picked out that specific virus before it jumped to humans, because that would require substantially more field surveillance and more dedicated laboratory testing than has been possible with existing funding. But Baric and colleagues gave a clear heads up that something was brewing. And yet we were “surprised”, again. (Post publication note: For more on what has so far been learned about the origin of the virus, see this absolutely fantastic article in Scientific American that came out today: How China’s “Bat Woman” Hunted Down Viruses from SARS to the New Coronavirus, by Jane Qiu. I will come back to it in later installments of this series. It is really, really good.)
Not only were we warned, we have considerable historical experience that (wildlife consumption + coronavirus + humans) leads to zoonosis, or a disease that jumps from animals to humans. This particular virus still caught us unawares; it snuck up on us because we need to do a much better job of understanding how viruses jump from animal hosts to humans. Unless we start paying closer attention, it won’t be the last time. The pace of zoonotic events among viruses related to SARS-CoV has accelerated over the last 25 years, as I will explore in a forthcoming post. The primary reason for this acceleration, according to the wildlife veterinarians and virus hunters I talk to, is that humans continue to both encroach on natural habitats and to bring animals from those habitats home to serve for dinner. So in addition to better surveillance, humans could reduce the chance of zoonosis by eating fewer wild animals. Either way, the lesson of being surprised by SARS-CoV-2 is that we must work much harder to stay ahead of nature.
Why is the US, in particular, so unprepared to deal with this virus?
The US government has a long history of giving biological threats and health security inadequate respect. Yes, there have always been individuals and small groups inside various agencies and departments who worked hard to increase our preparedness and response efforts. But people at the top have never fully grasped what is at stake and what needs to be done.
Particularly alarming, we have recently experienced a unilateral disarming in the face of known and obvious threats. See the Laurie Garrett article cited above for details. As reported by The New York Times,
“Mr. Trump had no explanation for why his White House shut down the Directorate for Global Health Security and Biodefense established at the National Security Council in 2016 by President Barack Obama after the 2014 Ebola outbreak.”
Yet this is more complicated than is apparent or is described in the reporting, as I commented on Twitter earlier this week. National security policy in the US has been dominated for many decades by people who grew up intellectually in the Cold War, or were taught by people who fought the Cold War. Cold War security was about nation states and, most importantly, nuclear weapons. When the Iron Curtain fell, the concern about large nations (i.e., the USSR) slipped away for a while, eventually to be replaced by small states, terrorism, and WMDs. But WMD policy, which in principle includes chemical and biological threats, has continued to be dominated by the nuclear security crowd. The argument is always that nuclear (and radiological) weapons are more of a threat and can cause more damage than a mere microbe, whether natural or artificial. And then there is the spending associated with countering the more kinetic threats: the big, shiny, splody objects get all the attention. So biosecurity and pandemic preparedness and response, which often are lumped together as "health security", get short shrift because the people setting priorities have other priorities. This has been a problem for both Democrat and Republican administrations, and demonstrates a history of bipartisan blindness.
Then, after decades of effort, and an increasing number of emergent microbial/health threats, finally a position and office were created within the National Security Council. While far from a panacea, because the USG needs to do much more than have policy in place, this was progress.
And then a new Administration came in, which not only has different overall security priorities but also is dominated by old school security people who are focussed on the intersection of a small number of nation states and nuclear weapons. John Bolton, in particular, is a hardline neocon whose intellectual roots are in Cold War security policy; so he is focussed on nukes. His ascendence at the NSC was coincident not just with the NSC preparedness office being shut down, but also a parallel DHS office responsible for implementing policy. And then, beyond the specific mania driving a focus on nation states and nukes as the primary threats to US national security, there is the oft reported war on expertise in the current exec branch and EOP. Add it all up: The USG is now severely understaffed for the current crisis.
Even the knowledgeable professionals still serving in the government have been hamstrung by bad policy in their ability to organize a response. To be blunt: patients are dying because the FDA & CDC could not get out of the way or — imagine it — help in accelerating the availability of testing at a critical time in a crisis. There will be a reckoning. And then public health in the US will need to be rebuilt, and earn trust again. There is a long road ahead. But first we have to deal with SARS-CoV-2.
Who is this beastie, SARS-CoV-2?
Just to get the introductions out of the way, the new virus is classified within order Nidovirales, family Coronaviridae, subfamily Orthocoronaviridae. You may also see it referred to as a betacoronavirus. To give you some sense of the diversity of coronaviruses, here is a nice, clean visual representation of their phylogenetic relationships. It contains names of many familiar human pathogens. If you are wondering why we don’t have a better understanding of this family of viruses given their obvious importance to human health and to economic and physical security, good for you — you should wonder about this. For the cost of a single marginally functional F-35, let alone a white elephant new aircraft carrier, we could fund viral surveillance and basic molecular biology for all of these pathogens for years.
The diversity of pathogenic coronaviruses. Source: Xyzology.
Betacoronaviruses (BCVs) are RNA viruses that are surrounded by a lipid membrane. The membrane is damaged by soap and by ethyl or isopropyl alcohol; without the membrane the virus falls apart. BCVs differ from influenza viruses in both their genome structure and in the way they evolve. Influenza viruses have segmented genomes — the genes are, in effect, organized into chromosomes — and the virus can evolve either through swapping chromosomes with other flu strains or through mutations that happen when the viral polymerase, which copies RNA, makes a mistake. The influenza polymerase makes lots of mistakes, which means that many different sequences are produced during replication. This is a primary driver of the evolution of influenza viruses, and largely explains why new flu strains show up every year. While the core of the copying machinery in Betacoronaviruses is similar to that of influenza viruses, it also contains an additional component called Nsp-14 that corrects copying mistakes. Disable or remove Nsp-14 and you get influenza-like mutation rates in Betacoronaviruses. (For some reason I find that observation particularly fascinating, though I can’t really explain why.)
There is another important feature of the BCV polymerase in that it facilitates recombination between RNA strands that happen to be floating around nearby. This means that if a host cell happens to be infected with more than one BCV strain at the same time, you can get a relatively high rate of new genomes being assembled out of all the parts floating around. This is one reason why BCV genome sequences can look like they are pasted together from strains that infect different species — they are often assembled exactly that way at the molecular level.
Before digging into the uncertainties around this virus and what is happening in the world, we need to understand how it is detected and diagnosed. There are three primary means of diagnosis. The first is by display of symptoms, which can span a long list of cold-like runny nose, fever, sore throat, upper respiratory features, to much less pleasant, and in some cases deadly, lower respiratory impairment. (I recently heard an expert on the virus say that there are two primary ways that SARS-like viruses can kill you: “Either your lungs fill up with fluid, limiting your access to oxygen, and you drown, or all the epithelial cells in your lungs slough off, limiting your access to oxygen, and you suffocate.” Secondary infections are also more lethal for people experiencing COVID-19 symptoms.) The second method of diagnosis is imaging of lungs, which includes x-ray and CT scans; SARS-CoV-2 causes particular pathologies in the lungs that can be identified on images and that distinguish it from other respiratory viruses. Finally, the virus can be diagnosed via two molecular assays, the first of which uses antibodies to directly look for viral proteins in tissue or fluid samples, while the other looks for whether genetic material is present; sophisticated versions can quantify how many copies of viral RNA are present in a sample.
Each of these diagnostic methods is usually described as being “accurate” or “sensitive” to some degree, when instead they should be described as having some error rate, a rate than might be dependent on when or where the method was applied, or might vary with who was applying it. And every time you read how “accurate” or “sensitive” a method is, you should ask: compared to what? And this is where we get into uncertainty.
Part 2 of this series will dig into specific sources of uncertainty spanning measurement and diagnosis to recommendations.