It’s likely that a considerable time will pass before we have access to a vaccine, too much time for us to remain hunkered down in an airlock, waiting. We’ll need to absorb a more nuanced understanding of real risk factors in order to make personal decisions and influence reasonable policy.
What follows is a layman’s summary of the very spotty available research as of May 2020. Almost all of the available information is preliminary, based on small samples with little if any experimental validation. It’s marginally better than anecdote, yet it’s all we have to work with for now. Use with caution.
Why should anyone who isn’t an epidemiologist even attempt to summarize this data? Thanks to the abdication of leadership by our government, we all have to process this information and make decisions on it, with or without any competence. Feedback and corrections are welcome, but we all have to start somewhere.
Discover Magazine offered an excellent explanation of the uncertainty around these questions. It forms a fine disclaimer:
Since SARS-CoV-2 is a new virus, the research community is only just beginning to do controlled experiments. Therefore, all the information we have comes from observing patients who were all infected in different ways, have different underlying health conditions, and are of different ages and both sexes. This diversity makes it difficult to make strong conclusions that will apply to everyone from only observational data.
How do you contract C19? At a summary level, to become ill with C19 a patient must get a threshold load of C19 virus particles into vulnerable respiratory membranes. Transmission seems to occur mostly by particles reaching the nose or throat via airflow, but sometimes through the eyes. There is no evidence of transmission via food. Contaminated surfaces could theoretically be a vector for C19 spread, but no contract tracing has yet identified a case. Transmission via touching a surface would require an intersection of many unlikely coincidences and might not be viable on any mass scale.
When discussing transmission vectors, it’s important to remember a key difference between viruses and bacteria. Viruses cannot replicate outside a host. There is a vibrant debate about whether viruses are living things at all. We are used to scrubbing down counters and guarding against surface contamination because bacterial pathogens can grow and thrive in almost any environment. A virus on a surface is merely waiting to die. It needs a lot of help to survive. A lot of the hygiene practices we’ve absorbed for other purposes don’t have the same relevance with C19.
How much of the virus is necessary to cause an infection? We’re not sure, but we have some clues.
“Viral load” is the quantity of virus particles present in a sample of bodily fluid. There is a threshold amount of exposure to particles of a pathogen necessary to initiate an illness. Patients’ viral load will vary over the course of their illness, and by certain personal traits. How much of the virus a patient “sheds” via different emissions also varies.
For comparison, exposure to one particle of the Norovirus is associated with a .5% chance of developing the illness. It’s estimated that exposure to as few as 18 particles can reliably launch the norovirus illness in most subjects. With most pathogens, the viral load to which a patient is exposed in their initial infection can influence the severity of the illness. Research on C19 is producing contradictory results on this question.
What viral load is necessary to spread C19? This is not settled, but based on mouse studies on earlier SARS viruses, it seems likely that the necessary viral load is around 1000 particles. For comparison, the infectious dose of Influenza A is about 200 particles.
That might seem like good news, except for one of the unique characteristics of C19. With the flu, patients shed the highest volume of the virus as their symptoms peak. C19 patients seem to shed peak volumes of the virus just before the onset of symptoms. Viral shedding appears to begin about 2-3 days before onset of symptoms. One study found that 44% of the cases traced had been transmitted by a pre-symptomatic patient. This sucks, because it makes it harder to contain the disease by screening for symptoms. This pattern of presymptomatic transmission helps explain why C19 has become such a dangerous pandemic despite in many other ways being less virulent than other common illnesses.
What is the viral load expelled by a cough, speaking or sneezing? How much C19 is contained in, for example, a patient’s sneeze? We don’t have an experimental answer, and any answer would be influenced by the patient’s viral load at that time. We do, however, have some idea of the density of droplets expelled in different interactions.
A cough releases about 3,000 droplets. A sneeze, 30,000. A single exhale anywhere from 50-5000. Droplets exhaled during speech can vary exponentially based on volume. How much virus might be contained in a single droplet could vary based on the virus, the patient’s viral load, and the stage of that patient’s illness, but for comparison, a single sneeze could expel hundreds of millions of virus particles from a flu patient at the peak of illness. Most falls to the ground quickly, but some could remain suspended in the air for some time.
Speaking increases the release of respiratory droplets about 10 fold; ~200 copies of virus per minute. Again, assuming every virus is inhaled, it would take ~5 minutes of speaking face-to-face to receive the required dose.
The exposure to virus x time formulae is the basis of contact tracing. Anyone you spend greater than 10 minutes with in a face-to-face situation is potentially infected. Anyone who shares a space with you (say an office) for an extended period is potentially infected.
How are you most likely to contract the disease? In China the top risk factors were “living with the case (13.26%), taking the same transportation (11.91%), and dining together (7.18%)” with those percentages representing the odds of spread, not the prevalence of that particular vector. Risk of cross-infection within a hospital was low, less than 2%.
Certain environments that combine close quarters, long personal contact, or consistent exposure to carriers have developed into “super-spreaders,” accounting for large multiples of spread. Prisons, nursing homes, meat-packing plants, churches, restaurants and call centers have been identified as particularly virulent environments.
A study of a call center in South Korea highlights the potential risks of almost any closed environment, especially the open-plan offices that have become common in recent years. Mere speaking releases far fewer virus particles than sneezing or coughing, but an environment in which people are speaking consistently, in relatively close quarters, for hours on end in a closed indoor ventilation system can quickly build up the necessary airborne viral load for transmission. The call center research suggests that concerts, sporting events and nightclubs might also be hyper-spreading environments.
Other high-contagion environments have been funerals and church services. An explanation may come from a study on a choir practice that became a super-spreader event. In Mt. Vernon, WA. A choir practice on March 10 with 60 participants, none of whom were symptomatic, led to 45 cases and 2 deaths. Members were careful to maintain social distancing and used hand sanitizer, yet 2 hours of shared air was deadly. A single breath may contain very few virus particles, but multiply those breaths by minutes and then hours, and a small exposure can grow exponentially.
What kinds of interactions appear less likely to spread the virus? There is almost no available evidence of outdoor exposure, minus direct personal contact. A Chinese study of 318 outbreaks found only one based on outdoor exposure. Brief, low-contact interactions are very unlikely to spread the disease without a cough or sneeze.
There is some evidence that even in households with an infected member, proper sanitary and mitigation methods may successfully limit spread. Studies are finding that the “secondary attack rate” of the disease within households ranges between 16% and 10%, much lower than one might expect. In China, close personal contacts of infected medical staff very rarely contracted the disease, suggesting that the disciplines around hygiene and mask-wearing well-engrained among medical professionals can limit spread even within households.
Masks, even relatively simple ones, can be helpful by limiting the range of spread of aerosol particles from breath, coughs and sneezes. Masks do not prevent the spread of the disease absolutely. Few of the masks in common use successfully filter out particles from the air we breathe. They act as a damper, limiting how far breathed particles spread, thereby suppressing risk factors. Masks aren’t a fix. They are a suppressant. Combined with social distancing and limits in sustained contact, they can be a vital key to reducing transmission.
What does this mean for the shape of any potential end to a lockdown? We can begin to resume something like normal life once we’ve implemented testing and contact tracing. At the end of January this would have been a manageable process. Had the Administration consulted and followed its own pandemic playbook when the first case was detected in January, lockdowns might have been limited to the Seattle and NYC areas, we might not have needed more than a few thousand tests per day, and contact tracing might have allowed life to proceed more or less as normal across the rest of the country. Too late for that now.
For normal life to resume, even after initial availability of a vaccine, testing and contract tracing would have to happen on a historically unprecedented scale. It still has to happen, regardless the cost or complexity. Even after a vaccine is available, testing and tracing will remain a necessary gateway to normalcy.
Since our leaders are incapable of meeting this challenge, what do we do in the meantime? Until a vaccine and a testing/tracing scheme is enacted, there are certain environments which will remain lethal. Common environments like churches, nightclubs and dine-in restaurants probably cannot resume anything like normal function, safely. Sporting events and concerts are out of the question. Anyone calling workers back to an office, especially a call center, or the close environment of a meat-packing plant is guilty of negligence per se. Meat packing plants could perhaps operate safely with workers protected by safety respirators and eye protection, but this would also require shorter shifts and impact safety and effectiveness.
Grocery stores and other retail might be able to operate safely if their indoor spaces are sufficiently large, workers and patrons wear masks, employees are consistently tested, and careful crowding controls are maintained. Commercial air travel meets all the high risk factors for transmission, though airlines have been highly secretive about cases.
A month ago we learned that over a hundred American Airlines attendants and 41 pilots had tested positive. The LA Times reported three weeks ago that 15 flight attendants had died, but the industry is not posting up to date stats. We have no information about testing rates or infection rates among attendants.
Mass transit, by contrast, appears relatively safe. Big, dense cities like Hong Kong, Taipei, Tokyo and Seoul have been very successful containing the disease without ever closing mass transit. Compared to air travel, mass transit involves brief interactions, with rarely any talking, and we’ve learned that surface contamination is a rare vector for transmission, if a vector at all. How often does anyone spend 3 hours in a subway car? And if they did, they wouldn’t be packed in next to the same person that whole time. Urban mass transit is orders of magnitude safer than air travel.
Meanwhile, we probably don’t need to worry about our mail, your Amazon packages, or our food. Hand washing would be an effective response to any surface-based transmission and this ritual of washing all my groceries is more religious than preventative.
Short-timeframe, outdoor interactions, such as during walking or jogging, probably cannot transmit the disease without someone sneezing on you. And even in a home with a C19 patient, reasonable sanitary precautions can limit the odds of further transmission.
We are in for the long-haul. No helping is coming before the election. Even with a vaccine, the disease will remain a factor in our lives for months or years to follow. Maintaining an OCD level OCD paranoia is too much to ask. As data becomes available to separate the real threats from the unreal, we should make reasonable adjustments.
Go outside. Get used to the mask. Relax about the groceries and the deliveries. Stay away from crowded indoor spaces. Be kind to people. If you’re fortunate enough to have an income during this time, be generous. In fact, find ways to be generous even if it hurts. It’s a way to maintain connection. Brace for years of inconvenience and start thinking about ways we can use this disruption to build a better world.