A Doctor On How To Make Covid-19 Decisions When So Much Is Unknown

The pandemic forces all of us to weigh risks and benefits in the face of uncertainty.



A person wearing a mask entering a train in Queens, New York, on April 29.


This month, two shoppers at a Target in Los Angeles fought a security guard, breaking his arm, when he tried to tell them to follow the store policy and wear masks. The American Covid-19 response has been befuddling to say the least; at times, it seems like we are facing two battles — one against the virus, and one against each other.

A central part of this frustration between fellow Americans has been a failure of leadership to unite us against a common enemy. With confusion and inconsistency around everything from masks to medications, it’s become harder to navigate information and misinformation, find direction, or even know if we are going in the wrong one.

And as much as there is uncertainty among the general public, so too is there across levels of leadership, including at the federal government, the Centers for Disease Control, and the World Health Organization.

What is unequivocal is this: We are operating under extreme uncertainty. Most of us have never been so directly impacted by a pandemic; the best comparisons we have are to the flu, yet it is quite clear we are dealing with something very different.

As a doctor, I have always had to embrace uncertainty in the hospital when I am caring for patients, and to understand my clinical decisions less as absolute truths and more accurately as careful calculations of risks and benefits. As such, uncertainty has not felt to me as foreign as it has to many other Americans in this moment. Decision-making under uncertainty demands that we are careful enough that our hoped-for benefits outweigh our feared risks, yet not so careful that it stops us from taking action altogether.

In a March 13 press conference, the WHO’s executive director of the Health Emergencies Program, Dr. Mike Ryan, remarked on the pandemic response, “If you have to be right before you move, you will never win.” Yet what we’ve seen is the failure to move on a number of critically important decisions by those who are paralyzed by the lack of perfect information.

Universal mask-wearing is one example: Opponents, including the CDC early on, have said that universal masking has not yet been proven to reduce transmission of SARS-CoV-2, the virus that causes Covid-19. While it is true that we aren’t sure what the effect of universal masking at a population level will be, that is because we have not had time to study this effect with Covid-19 yet. This doesn’t mean masks won’t help us greatly — it means we aren’t completely sure if and how much they will.


Here, the practice of weighing risks and benefits can help us realize that masks are worth using. The potential benefits of consistent universal masking, particularly if using a high-filtrate equivalent mask, like N95, are huge. We could potentially prevent thousands of deaths, slow the spread significantly, and eventually drive R0 (a measure of how quickly the disease spreads) consistently below 1, forcing the epidemic to burn out.

If masks are not all that they were cut out to be, we really don’t lose much. Some have also argued that there is a risk that masks may reduce how much people physically distance (thinking they are safe), but there is no data to support this “risk-compensation” argument. The same was assumed for seat belts — the possibility they would increase reckless driving — and that was shown to be untrue. The CDC has endorsed population-level masking, and many have felt they should have done this much sooner.

The decision to close schools was another such exercise in the challenge of decision making under uncertainty. There were a number of debates over the benefits of closing schools, with some prominent academicians arguing against it because we lacked the data to know whether it was worth doing. Now, reports suggest that children likely play a notable role in transmission, and that closing schools may in fact have reduced the epidemic’s surge by 40 to 60 percent.

This is a prime example of the old saying: Better the devil you know than the devil you don’t. If you’re not sure, close the schools. You can reopen them (or address the ripple effects of closing them), but you can’t rewind a runaway outbreak.

The same logic could have been applied on a broader level, as it relates to instituting city-wide lockdowns. Data now suggests that the New York City outbreak alone may have accounted for most of the other outbreaks throughout the country so far. And even early in the pandemic, we had reason to believe that a city-wide lockdown could potentially have a notable effect — with Wuhan, China, as our case study. (Even now, it is hard to know how much of an effect each intervention had, the lockdown being one of many.) But the authorities hesitated to close down travel into and out of our country’s most populous city, while quick action here may have changed the US Covid-19 trajectory.

It is also important to consider the consequences of moving too fast and being wrong — because being wrong can also cause unnecessary harm.

President Trump’s tweets about hydroxychloroquine-azithromycin come to mind as an example. Acting on the preliminary findings of a small French study with a number of methodological limitations, the president jumped to promoting the medications. Ultimately, they have yet to be proven to have significant benefit for Covid-19 patients, but are well-known to increase the risk of deadly heart rhythms and have already resulted in harm in a number of cases.

Acting without robust data can be reckless and harmful when the risks are just as great, if not greater, than the possible benefits. Medicine’s mantra of “First, do no harm” must temper the predisposition of leadership to act too fast and rely on unproven treatments. With masks, the risk of being wrong — that masks don’t have a huge benefit at the population level — leaves us still with our remaining strategies of testing, tracing, and isolating, and likely does not set us back significantly. But with unproven medicines, there is a much quicker path to making life and death calculations. Here, taking a chance is a much higher-stake move.

The preprint publishing industry has also perpetuated this rush-to-conclusion tendency. Preprints are research that is made publicly available before being formally vetted for publication in a printed scientific journal. These early reports allow for rapid dissemination of research, but in a form that has not undergone peer review by other experts in the field and which can be easily misinterpreted by nonexperts.

It is fueled, in part, by the insatiable appetite for quick scientific results, which is understandable during a rapidly evolving situation like the one we are in now. But they also must be read with extreme caution.

There have been a number of examples of this system failing. One prominent one was a preprint about early serological data from Santa Clara County, California, which epidemiologists and public health experts swiftly called out as problematic.

To add an additional layer of complexity to this, we have to overcome the challenge of countering misinformation, a gargantuan task in itself. As if it weren’t enough to be simultaneously dealing with uncertainty and complex trade-offs, misinformation is truly the “parallel” epidemic in almost all infectious disease outbreaks; it happened with Ebola and Covid-19 is no exception.

Misinformation erodes trust, and in turn limits the possibility of coordination or collaboration in an outbreak response. It also creates even more uncertainty and makes people less likely to believe robust scientific data when we do have it.

The antidote to misinformation is a reliable countermeasure from trustable leaders. This may come at a national level. But when it does not, it must come from trusted professionals who hold technical expertise teamed up with journalists and communicators who can help translate information to the masses.

As a doctor working closely on the Covid-19 public health response, I can appreciate the challenges of getting responses perfectly right — avoiding being too slow or too quick, either of which could cause unintended harm, and properly weighing the risks and benefits.

We must recognize that we are operating under uncertainty now — but that we always were, even in the pre-Covid-19 world. We just didn’t need to think about it quite as often, as the consequences of our decisions were less far-reaching.

It’s not easy, but the world we live in has always been dictated by trade-offs, with both successes and failures. We also have to realize, perhaps most importantly and humbly, that we could be wrong, and that we may need to correct course. In the end, the only thing that we know with certainty is that we need to keep moving forward, together.

Source