Almost everyone has heard of this anti-COVID drug called Remdesivir. After all, it was used to treat President Trump after he tested COVID positive. This is a medication that was developed by Gilead Sciences in order to suppress Ebola, but was not effective against Ebola in clinical trials. You might also have heard different opinions about whether it actually works in COVID. In fact, there are two recent scientific reports that say almost the exact opposite. An American study from the National Institutes of Health, or NIH, claims that Remdesivir helps patients with COVID, and a World Health Organization, or WHO study claims that it does not. So, if you are wondering who is right, stick with me for a bit. In the next few minutes, I am going to take you right past all the opinions to go straight to the facts, so we can draw our own conclusions about this controversial drug. Welcome to my world of evidence-based medicine. Let’s start with that American study…
First lets focus on the time it took patients to recover. After all, this was the main point of the study. When I magnify this, we can see that the control group recovered in an average of fifteen days, while remdesivir-treated patients recovered on average in ten days. Those are the two key numbers. This is a noticeable improvement in recovery time and it is statistically meaningful. Therefore, the investigators declared that this drug does help patients. What the researchers did was to divide patients into groups based on how sick they were. Mild, Moderate, and Severe. Patients at Level 4 were hospitalized but did not require oxygen treatment. Patients at Level 5 required some oxygen but not a lot. This is apparently the level that President Trump was at. Patients at Level 6 required high levels of oxygen. There was also a Level 7, patients who required a ventilator but I am going to ignore them because their results were pretty much identical to those at Level 6. Let’s start with Level 4, the mild group, by looking at their mortality rate, which I will magnify for our benefit.
Note that both groups had very similar mortality rates, as a little more than four percent of subjects in each group died. The bottom line is that everyone at this mild level did pretty well, with very few patients dying in either group and no real advantage from taking the medication. So is it fair to single out this one group in this very large comprehensive study? Isn’t that an example of cherry picking, slicing and dicing the results until we come up with something we like? One must be careful but I would argue that it is okay to do that in this case for two reasons. First, the difference in outcomes in the level 5 group was quite striking and statistically strong. Even more important, it makes biological sense, that a drug has its biggest impact on the mama bear group. All through my world of medicine, it is often true that patients with mild disease do not benefit from a treatment because they will do fine whether they get it or not. And patients with severe disease may not derive benefit because its too late to help them.
It is often the folks in the middle for whom the treatment makes all the difference. That’s our sweet spot. So I am willing to hang my hat on this finding from the NIH trial, which was well designed and free from bias. Remdesivir is not a cure-all, not a magic bullet for COVID. But hospitalized patients on modest amounts of oxygen will likely recover faster and be less likely to die if they are treated with this drug. Now, what about that WHO study that was reported last week, shortly after the final report of the NIH trial? In the WHO trial, which was also large and free from drug company bias, there was no apparent benefit of remdesivir on recovery time or mortality. However, the WHO trial lumped all the bears together; it didn’t sort out the mama bears, and it had a serious flaw in design. It was not a blinded trial. Every treating physician knew whether his or her patient was in the treatment group or the control group, which created an unacceptable opportunity for bias to creep into the results.
If I know my patient is on remdesivir, it might influence all the other decisions I make about that patient, including what other drugs to use or not use, and whether or when to put my patient on a ventilator. So in the lineup of study designs, the American study is a gold standard, and the WHO study is a silver. For that reason alone, I am going to take as my final word, the positive results of the NIH study, and continue to recommend remdesivir for mama bears like President Trump. As a final disclaimer, I do not have any financial relationship with either Gilead Sciences, the NIH, the WHO, or the US government, except that I pay federal taxes, of course, and write political essays that are mostly critical of our national leadership.