Yesterday, I wrote a story examining how much COVID-19 treatment protocols have improved since the early days of the pandemic—a thorny question that’s difficult to tease out of the data. One of the people I interviewed for that story was Dr. Megan Ranney, an emergency physician and professor at Brown University in Rhode Island. Our conversation was fascinating enough that I was unhappy not to have been able to get more of it into the piece, so we decided it would be worthwhile to publish the interview in its entirety. A few snippets of our conversation that appear here also appeared there. The following has been slightly edited and condensed.

It seems as though we’re getting better at treating COVID-19, but it’s hard to know exactly how much better we’re getting, exactly what things are helping. For doctors making these treatment decisions, what goes into your decision-making for what treatments and therapies you’re using?

So there’s kind of two parts to the decision-making. Part one, and the part that we all are most comfortable with, is relying on peer-reviewed evidence. So for [antiviral drug] remdesivir, for [anti-inflammatory drug] dexamethasone, even at this point for proning patients who are short of breath and for trying to avoid intubation as long as possible—those are all practices that at this point have semi-reliable, published evidence behind them. And that’s really the gold standard; that’s what we’re almost comfortable with.

The other side is the anecdote, the case series, the shared experiences of groups of physicians. This was particularly prominent early in the pandemic when we really had nothing, and no guidance whatsoever. We saw Facebook and WhatsApp groups spring up, as well as email listservs, to try to allow real-time sharing of best practices and to allow people to compare what worked in one setting to what worked in another. There were lots of case studies; some of them were published. It led to some small things, like those hydroxychloroquine studies, which have since been shown to not be accurate. But it also led to people saying, this is my practice pattern, and this is what I see helps my patients. 

That’s how we’ve developed practices like, before we discharge patients, we have them walk around on a pulse oximeter and see if they desaturate. That’s not something that’s in the peer-reviewed literature, but that enough physicians said that they were doing and seeing that that helped predict a downward course that it became part of many of our practice. And that was purely through kind of sharing anecdotal experience.

On top of that you have medical societies and hospitals who are trying to collate these various levels of evidence and create standard protocols. And those are shifting as the evidence changes. In medicine, there’s actually levels of evidence, like grades that we use, for how much can we trust this? And right now, with COVID, those levels of evidence for the most part are super, super low. So you’re going to find that there’s a lot of variation in practice patterns and in hospital or clinic protocols between one institution and another, because there’s so much out there still that is kind of unknown.

And then for stuff like convalescent serum, or other drugs, then you get into questions that are, do we have a study that’s ongoing? Is it even available? How do we ration the limited resources? Remdesivir being a great example: We have some evidence that it works for the really sick patients, but we don’t have enough for everyone. So then you get into these complex ethical questions as well.

As far as clinical practice is concerned, one thing I’ve been trying to brush up on over the last couple days is some of these newer terms that are creeping into the COVID research discourse—things like monoclonal antibodies, medicinal signaling cells, some of the stem cell research that’s been going on. Clearly these are all small-scale trials that are happening at research institutions. Is that just a completely different track from the kind of stuff clinical physicians are looking at right now? How do those tracks relate to one another?

Yeah, so that’s a really great question and it’s a complicated answer. I’ll do my best to answer, with the caveat that I am at an academic institution, and I am a researcher myself.

So you sort of straddle the two.

Exactly—I’m at a place that’s going to have more access to trials, and I deeply believe in the value of trials. So it’s kind of two things. One is, most of those therapeutics that you just mentioned, that I would really consider to be experimental therapeutics, are available only in the context of a clinical trial right now. And those clinical trials are occurring at both big and large hospitals and big and large clinics. But most of them are available only through enrollment in a trial. And that’s how it should be, because we shouldn’t be giving people stuff if we don’t know if it’s gonna work or not. So that’s the should answer.

But the reality answer is that, even though many of these treatments are not theoretically approved for use outside of clinical trials, there are some places, particularly with convalescent serum, where it is being offered outside of trials.

And that is both because of financial means—it’s more likely to be available for people who are, you know, willing to pay out of pocket sometimes, or for better funded hospital systems—but also out of desperation, right? As a physician, the worst thing is to see your patient getting sick and dying in front of you and not being able to do anything. And as a patient or family member, you want hope.

And so, if you have access, I can understand why people are offering some of these truly experimental treatments outside of clinical trials, because if you have access to it and you think it’s going to help your patient, of course you want to help your patients.

Is it the case that sometimes these things can work at cross purposes with each other—that there’s treatment happening outside a clinical trial environment? Does it ever make it more difficult to get the clinical results?

Yeah, no, it absolutely does. There are some studies, particularly with convalescent plasma, that we’ve been having trouble reaching target enrollment goals on, because of the fact that treatment is available outside of the trials. Because why would a patient—you can understand that if a patient feels that they’re guaranteed to receive it and they don’t understand that there may be potential risk involved, that why would they agree to be randomized?

They wouldn’t want to enter a placebo-controlled environment.

Exactly. Particularly if your doctor is telling you, this thing is going to help you—you’re not going to go search out a trial if you can get it outside of the trial. So it is hurting the ability to recruit patients into the studies, and it hurts our ability to create generalizable knowledge that can actually save lives when we have a lot of off-label treatments available.

And I think the debate over hydroxychloroquine epitomizes that, because we’ve had a number of studies now, large, randomized controlled trials, that show that hydroxychloroquine does not help and may even hurt for those hospitalized patients to whom it is given. However, we don’t have great evidence for some of the earlier stage patients, and it’s partly because now we’ve got all these people saying, “well, I took it and I didn’t get COVID.” The off-label use of hydroxychloroquine really kind of hurt the cause to create generalizable knowledge. It dissuaded people from enrolling in certain types of studies, and then it also created this shadow, non-official group of case studies that are being promoted as evidence that it works, when they’re not actually evidence that it works. So it has prolonged the debate beyond what would have been necessary had it only been used in the context of clinical trials.

Are you confident saying that physicians are better equipped to fight the virus than they were back in April and May? Have outcomes improved in that way?

Yes, absolutely. I’m very confident that we have better evidence, and that we have more effective treatment protocols, and that because of that we are able to save more lives. And I expect that to only improve.

And that goes back to this ongoing debate, when we talk about, ‘Oh, just let people get infected, just create herd immunity.’ People forget that the longer you can wait to get COVID—hopefully we’ll get a vaccine quickly, so we’ll be able to protect us all from getting infected—but if you get infected today, your chance of living through that infection is better than if you got infected in February or March. And I expect that to continue to improve. So if there’s not a vaccine, the longer you can wait to get infected, the better, because we are going to continue to have better treatments—assuming people continue to enroll in trials.

But also because we are accumulating knowledge through the scientific method, the same way that we always do for medicine. It’s not a linear process. It’s two steps forward, one step back in terms of that knowledge generation. So there are going to continue to be things that change, there are going to continue to be things that we think help and maybe don’t. That’s normal in medicine. But overall, the trajectory is absolutely a positive and encouraging trajectory in terms of our ability to effectively treat and save lives of patients who are sick with COVID.

Photograph by Dirk Waem/Belga Mag/AFP/Getty Images.