With COVID-19 infections becoming more common, experts have recently urged doctors to prescribe the antiviral drug Paxlovid more than they have been to minimize patients’ symptoms and reduce the chance that they’ll develop severe disease. The drug is authorized for people at higher risk of developing severe COVID-19, including those who are older and people with underlying health conditions. But many patients who have taken Paxlovid have reported developing rebound infections shortly after: testing positive again for the virus after first supposedly clearing the infection and testing negative. Studies have documented that the repeat positive tests are due to the same virus that caused the original infection returning again, rather than a new infection. In 2022, the U.S. Centers for Disease Control and Prevention advised physicians about the potential for rebound infections; the agency continues to recommend the drug for those at high risk of severe COVID-19 disease, but warned doctors to be aware of rebound since people could be contagious when their infection returned.
So-called “Paxlovid rebound” has raised questions about how common rebound infections are, both with and without Paxlovid. Pfizer, Paxlovid’s manufacturer, found in its own study of the drug that rebound occurs in about 1.7% of Paxlovid patients, which is slightly lower than what they found in the untreated, placebo group. Larger studies have not yet established how often rebound occurs in people who are infected and not treated. But a new study published in the Annals of Internal Medicine sheds some light on the question, reporting on how likely rebound infections are without taking an antiviral.
“When I heard reports of people telling me they were getting better [on Paxlovid] and then got worse again, there was always a question in my mind about whether this happens during the natural COVID-19 infection recovery period,” says Dr. Jonathan Li, associate professor of medicine at Harvard Medical School and Brigham and Women’s Hospital and author of the study. “Only by understanding what happens in untreated infections can we interpret the data we are getting from patients receiving Paxlovid.”
In the trial—which was part of a larger network of trials testing various antiviral treatments for people with mild-to-moderate COVID-19—Li tracked the symptoms and viral levels, measured by weekly nasal swabs, of about 560 people who received a placebo over the course of their infection. Everyone swabbed at the start of the study and at two, three, and four weeks later. They also kept a daily log of symptoms, including fever, headaches, and coughs.
About 26% of these untreated people reported that their symptoms returned around 11 days after their onset, and 31% had higher virus levels after they had initially dropped. Overall, 3% of people reported both a return of symptoms and a higher viral load during the month-long study period. (All scenarios indicate infection rebound.)
“Those results tell us that symptom improvement is not a linear process but, in fact, waxes and wanes over time,” says Li. “It’s also very rarely associated with high-level viral rebound. Even without Paxlovid…patients will have symptom rebound, and potentially viral rebound as well. We have to be careful when saying Paxlovid will cause a significant side effect of rebound, when we still just don’t know.”
Read More: A One-Shot COVID-19 Treatment Shows Promise
Various studies and anecdotal data have found vastly different rates of rebound among both Paxlovid users and untreated people. Anecdotal reports, for instance, suggest a much higher rate of rebound among people who take Paxlovid than the Pfizer study did. But variations between all of this research, including the threshold of viral load that the researchers set to record virus levels, could account for the differences. One advantage of the current study, says Li, is that the patients were swabbed every week—but the small number of positive rebound samples also means it’s hard to draw any definitive conclusions about the incidence of rebound.
So how should the results be interpreted?
Li says it’s most important to remember the reason why people take Paxlovid. “The reason we recommend Paxlovid is not to prevent rebound [infections] but to prevent hospitalization and death,” he says. “When I counsel my patients, I tell them that the clinical trial [that the U.S. Food and Drug Administration reviewed to authorize Paxlovid] showed 90% protection from hospitalization and death despite any viral rebound after treatment. We need to keep our eyes on the prize.” Scientists are also looking into whether Paxlovid can help reduce the risk of Long COVID, though that research is still early and conclusions can’t yet be drawn.
Rebound infections are not unusual with viruses, and researchers are learning more about why this particular virus bounces back after waning and how extensive the repeat infections are. One possibility, related to way Paxlovid works, is that the recommended five days of pills may not be enough to properly suppress the virus, so it comes back when the medication stops. Another theory is that in response to the immune system, the virus may be moving to different parts of the body and finding new cells to infect, causing surges in virus levels and the return of symptoms. “We need more intensive data looking at both those taking Paxlovid and those who are not, to better understand what’s going on,” says Li.
Deciding who might benefit from Paxlovid should come after a detailed discussion between doctor and patient, Li says. “I counsel patients depending on their overall risk,” he says. “Risk factors like age do not lead to dichotomous ‘yes-no’ answers; it’s a continuous spectrum.”
Data from his study should help in those discussions, to better balance the risks and benefits of the treatment for individual patients. More studies are also needed to clarify the cycle of rebound, since people who test positive again after testing negative can still spread the virus to others.
More Must-Reads From TIME