An Epidemiologist Answers COVID-19 FAQs

Please submit your existential questions to "Ask a philosopher."

Source: Emily Morter/unsplash

Since mid-March I have, like most working parents in the US, had to navigate simultaneously working a full-time job while also being a full-time parent. When my university moved classes online, the message to faculty was: You must keep the quality of instruction high, but you must be flexible to students’ situations and needs. I did my best to do both, but it is also true that last week when I submitted grades for the 85 students I had across two classes and graduated five Master’s of Public Health students, I was...what’s the word?... elated to be done. My students likely felt a similar sense of relief. Online teaching consumed all of my mental energy, and like many working parents I found myself creating a “Must-do” list that got shorter and shorter as the realities of being under stay-home orders became more concrete. And this blog was not on that list. Neither was showering on a daily basis, just to give some perspective.

But, now that I’ve turned that corner, I want to get back to the mission of this blog, which is about making epidemiologic concepts and principles more transparent to help readers understand health research and what it means (or doesn’t) for their lives. Several readers have sent me questions about our overstaying guest COVID-19 over the past month, and I want to use this post to respond to them. I think these are thoughtful questions, and I thank you for your patience. I greatly appreciate your engagement with these posts and I hope you find them useful.

Question 1: Given that COVID-19 has more severe health complications for older adults, who are also more likely to get annual influenza and pneumonia vaccinations, could it be that these vaccinations are contributing to the risk of severe complications for these groups through over-stimulation of the immune system?

There is an important idea in epidemiology called “confounding by indication.” What it means is that sometimes it looks like the treatment for a disease is associated with some poor health outcome, but that association is just reflecting the fact that the people who get the treatment have some health indication that makes it more likely for them to experience the outcome in the first place. In this case, the *reason* the CDC wants older adults to get vaccinated against influenza because they are more likely to have health conditions that make complications from respiratory infections more likely to occur - that is, they are most likely to experience death from influenza because of their underlying health status. But because they are also more likely to get the vaccine, it can look like the vaccine is causing the respiratory complications.

So to answer the question, no, there is no evidence that influenza vaccination increases the likelihood of infection with or complications from COVID. Here’s a well-resourced article on the issue that I won’t reiterate, but here's the take-away: This question has been investigated — not with COVID-19 specifically, but with other coronaviruses — and the answer for adults is “No.” From the article: “Serious COVID-19 disease occurs primarily in adults, and we do not have evidence of flu vaccine causing virus interference in adult age groups.”

Finally, even if you don’t like the article, the question itself is misleading. There are many groups that are especially advised to get the flu vaccine, not just older adults (about 68% of adults age 65+ got the influenza vaccine during the 2018/19 season). These include children 6+ months to 17 of age (~63% got the influenza shot in 2018/19) and there’s no evidence that those groups are at higher risk of complications from COVID (in fact, they seem to be protected from it). Moreover, African Americans are disproportionately experiencing serious complications from COVID, but their influenza vaccination rate is lower than most other racial/ethnic groups (39.4% among adults versus 49.7% among non-Hispanic whites). So even at an ecological level, the data just don’t support this conjecture.

[Slide aside: I also want to say that this idea — that vaccinations harm your ability to mount appropriate immune responses — is pushed by slick, fear-mongering “advocacy” groups that have a political agenda. When I say “Vaccines are safe and they are effective,” what I mean is that “The overwhelming majority of people will have better health outcomes with vaccination versus without it.” Because of the truth of that statement, I ensure that everyone in my family, from my young son to my elderly in-laws, is vaccinated. That statement doesn’t mean there aren’t sometimes adverse effects of vaccines, or that they prevent 100% of infections. There is no magic bullet. But these “advocacy” groups hold up those exceptions to the rule as evidence of bad science, etc. as a way to prey on people’s fears. And if you’ve ever had a health scare — for yourself or someone you love — you know how hard it is to think rationally when you are in a state of terror. These groups depend on that fear.]

Question 2: Do coronaviruses (as a general family of virus types) have latency (that is, could they become latent infections)? Or, could the apparent “reinfections/recurrences” of the virus be the result of faulty testing?

This question is difficult to answer right now because we haven’t had COVID for long enough and we are still learning about the virus. But today, May 6, 2020, I feel confident saying the following: People who are symptomatic with COVID do develop some immunity to the virus, and the vast majority of “reinfections” that have been reported are the result of testing errors. This statement has the following footnotes:

1. We don’t know how long the developed immunity will last.
2. We don’t know if asymptomatic people develop immunity, or if they do, if it is as protective as it is for those who were symptomatic.

We don’t know (1) because, well, our inability to travel into the future, and we don’t know (2) because we generally know very little about asymptomatic people, including how many there even are (or were).

Coronavirus Disease 2019 Essential Reads

Update on COVID-19 and How It Changes the Brain

How Political Partisanship Can Be Deadly

Finally, I want to note that even the WHO has walked back its (radically poorly-worded, IMO) statement about there being "no evidence" of immunity post-infection to clarify that it just meant "don't use history of COVID-19 to indicate the person has complete immunity from this virus for all time."

I have a question. Will there be treats?

Source: Camylla Battani/Unsplash

Question 3: True or False

3a: COVID-19 is more infectious than Influenza and, because we have inadequate testing capability, more people than we can count are hypothetically believed to be infected.

True.

3b; Staying isolated, maintaining 6 feet distance, and wearing masks will slow down the spread, but these actions will not eradicate the virus.

True.

3c: Slowing down the spread will reduce the burden on healthcare system and slow down the number of people who need ventilators and slow down the number of people who die.

Partially true. It will not just slow down the spread, it will reduce the total number of cases that occur. It will do this because it gives our population time to develop herd immunity and to institute policies that reduce the risk of spread, which will ultimately reduce the likelihood that any given person gets infected when we lift stay-home orders. So it isn’t just changing the shape of the curve (flat vs. steep), it is also reducing the area under the curve (the actual number of cases). Also because infection does not equal death, we can have an increasing number of cases without a proportional increase in the number of deaths if we protect people who are most at risk of dying from infection once we reduce community spread (that is, the spread we have now).

3d: As we gradually return to a fully operational economy, the virus will have access to those people who were never infected and, as reports from China suggest, it may be able to reinfect the same person or it may have a biphasic lifespan in some people.

Partially true. See my response to Question 2.

3e: The burden on the healthcare system will continue as the economy ramps back up, the ventilators will be used and people will continue to die, although possibly not as many assuming there are enough ventilators and medications.

Partially true. See my response to question 3c.

3f: The virus is going to cause the cytokine storms in those with the comorbidities — heart disease, obesity, COPD, diabetes, asthma, cancer, immune system disorders, and those over 65. And this inflammatory response is more likely to kill these people.

Partially true. It can cause these complications in anyone, of any age, with any health status. It is more likely to occur in people with these characteristics. And this immune response contributes to the risk of death, but it isn’t the only thing that kills people.

3g: Going forward, protecting these individuals [people with existing health conditions] could be made a higher priority than isolation for everyone.

Yes, and we are already moving in that direction, if you look at the latest CDC testing priorities, which were updated on May 3, 2020.

3h: Covid19 is not going away until there is a vaccine or a cure.

Depends on what you mean by “going away.” It isn’t going to be eradicated like smallpox anytime soon, without question.

But if what you mean is “not causing thousands of deaths a day,” we will get there before a vaccine is developed as long as we a) keep some social distancing efforts in place, and b) are able to do widespread testing (particularly of asymptomatic or mildly-symptomatic people) which will allow us to contact tracing to quarantine confirmed or suspected cases to prevent widespread community transmission.

And finally, when we get a vaccine, the role that vaccine will play in controlling the virus will depend on three things: that a) the vaccine is safe enough that a sufficient number of people in the population can get it, b) a sufficient number of people actually get it (recall my earlier post about how only about 60% of the U.S. population gets an annual influenza vaccine, despite the fact that this vaccine is very safe and offers moderate protection), and c) the vaccine provides sufficient protection from infection (not all vaccines are equally effective — the influenza vaccine being an example of one that is relatively less effective, and the MMR vaccine being an example of one that is relatively more effective).

In short, vaccine development is just one part of a multi-pronged approach to controlling the impact this virus has on our health and broader society.

3i: If the number of total of infected persons is ever known, the death rate will more closely resemble that of Influenza, even though many more people were infected.*

Unknown (and unknowable) at the current time. Currently, it looks like it will be worse than influenza, but we just don’t know. I highly recommend this post by Nate Silver at 538 that helps explain why the data we have can’t answer the questions we want it to.

*Updated 5/8/2020: New study by John Ioannidis (a pretty smart guy - I recommend his readings in my classes regularly - study pre-print is available here) suggests (hopefully) that the case fatality rate for people <65 years old and without existing medical conditions is, in absolute terms, pretty uncommon - but still not zero. I am still trying to decide if I like his strategy of expressing COVID mortality risk in terms of the likelihood of dying in a car accident during a relatively short trip (again, uncommon, but still happens)... on the plus side, it helps anchor it to something we may have some familiarity with, but on the other hand, it feels a bit dismissive in that it doesn't reflect the tremendous mental, social, and economic costs that go into trying to care for a person with COVID (who ultimately dies from its complications), either for the person's family (who generally cannot visit with them) or the healthcare workers.