Can hydroxychloroquine prevent Covid-19?

While scientists are working hard on vaccines that could prevent coronavirus infection and treatments for Covid-19 patients, there is currently no vaccine or proven treatment available. A common drug for treating malaria has shown conflicting results in recent studies, but a group of researchers wanted to see if this drug could possibly be used to stop people from developing Covid-19 after exposure to an infected person.

In Short: People were exposed to someone with Covid-19 were given either high doses of a malaria drug or a placebo for 5 days to see if this could prevent them from developing Covid-19 themselves. The malaria drug did not prevent subjects from developing Covid-19.

Can taking a high dose of an anti-malaria drug, hydroxychloroquine, prevent patient from developing Covid-19?

In a recent post, I wrote about a few early studies where a common malaria drug was found to possibly offer some benefit to Covid-19 patients. More thorough studies have since been done that showed little benefit from that same drug, hydroxychloroquine (HCQ).

Those earlier studies had a number of issues, including:

only a small number of people took part
the studies were not properly set up to avoid biases from doctors and researchers

This is important because biases can distort how a study is interpreted, resulting in incorrect conclusions that may delay scientific and medical advances. These biases can be amplified in smaller studies, which make the results even less reliable.

To show how a small sample size can bias a study, imagine a group of 15 yellow butterflies and 10 red butterflies, for a total of 25 butterflies (see below). Out of 25 butterflies, 60% are yellow and 40% are red. If you take a sample of 5 butterflies from the group, your ideal sample would be 60% yellow and 40% red. But you could also end up with 100% yellow butterflies, 100% red butterflies, or something in between. Neither Sample 1 or Sample 2 represent All Butterflies. By using a larger sample size, this sampling bias is less likely to happen, and more likely to resemble the real population of butterflies.

When designing a study, it is important to avoid biases in sampling that may distort an observation.

In an ideal study, the researchers would have a large number of participants, that well-represent the population as a whole, and perform a randomized controlled trial, where:

participants would randomly be assigned to take either a placebo or a drug,

The only difference between the two groups should be whether or not they are taking the drug (which may have an effect) or the placebo (which should have no effect). This is called a controlled study, where the group taking the placebo is the control group, and the experimental group is taking the drug.
the participants don’t know if they are taking the drug or the placebo,

This is called “blinded” because the participants don’t know which option they were given until the end of the study.
and the researchers don’t know who is taking the drug or the placebo

This is called “double-blinded” because the researchers also don’t know which participants were given the drug or placebo until the study is done.

By randomly assigning participants to a group, this reduces selection bias, and performing a double-blinded trial reduces the risk of bias from the participants and researchers. Altogether, this makes it easier to interpret the results from a study and draw conclusions.

Studies from earlier this year showed that a drug used to treat malaria, HCQ, may protect cells from coronavirus or shorten how long patients are sick. Because it can take up to 14 days to develop Covid-19 symptoms after people are exposed, a group of scientists wanted to test whether HCQ could help prevent Covid-19 infection after exposure.

Dr. Boulware and their colleagues first identified more than 800 people that had a higher risk of being exposed to the coronavirus, including the families of people that had tested positive for coronavirus. They selected participants who had been physically close (within 6 feet/2 metres) to someone who had tested positive for coronavirus (or showed symptoms) for at least 10 minutes and were not wearing protective equipment. If they were exposed less than 4 days ago, they could join the study.

Next, they were divided into 2 groups, one which would take high doses of HCQ for 5 days and the other which would be given a placebo for 5 days. The amount of HCQ given to each person was determined from a previous study. Vitamin B9 (also known as folate) was used as the placebo because it is relatively safe, and if you have too much Vitamin B9, your body can excrete it in your urine.

After 14 days, the researchers determined the number of people in each group that developed Covid-19. In the control group, about 14% of people tested positive for COVID-19, and almost 12% tested positive in the HCQ group. Because the difference between the two groups was so small, the researchers couldn’t confidently say that the difference was because of HCQ instead of random chance. The scientists interpreted this to mean that HCQ is not helpful in preventing people from developing COVID-19 after high-risk exposures.

HCQ and other drugs have shown promise in initial studies done in cells grown in a lab and in smaller human studies. As larger studies are performed, we gain a better understanding of which drugs might be helpful in reliably treating or preventing COVID-19. While we haven’t been able to identify a drug or treatment yet, there is more research being done every single day, there are currently more than 135 vaccines in development, and there is a lot of promise on the horizon.

The full study was published in the New England Journal of Medicine and can be found here (for free).