Should I take Paxlovid for COVID?

Four and a half years into the pandemic, after being cautious for so long, I finally contracted COVID-19. While I might have continued to avoid the illness by maintaining strict precautions—avoiding groups, masking, and so forth—I was tired of living such a restricted life. COVID-19 is here to stay, and I felt it was time to resume living more normally. It’s interesting how differently one views medical treatment when you, as a physician, are the patient. The night I tested positive, I wasn’t severely ill, experiencing only mild to moderate symptoms. I agonized over whether to take Paxlovid. I consulted my wife, a retired physician with a Master’s degree in Public Health from Johns Hopkins, my trusted physician partners, and my own doctor. There was no consensus on what I should do. I’ve treated many patients with Paxlovid and am familiar with the side effects, drug interactions, and potential for rebound, which can lead to disease relapse and prolonged viral shedding. Given that I’m 69 and have at least one other risk factor besides age, I decided to take Paxlovid. Within 24 hours, I felt fine, and over the next few days, I continued to feel well. Whether this was the natural course of the illness or the effect of Paxlovid, I’ll never know, although current studies suggest that Paxlovid doesn’t significantly affect symptom duration. Although I am not seriously ill, on day eight, I began to feel sick again, with symptoms as bad as, and actually worse than, those on days one and two. My test, which had reverted to negative, was now strongly positive again. I am now at home, recovering from rebound COVID-19 after taking Paxlovid.

The decision to use Paxlovid is complex. The data is not as compelling as many current articles and public health officials suggest. The original Paxlovid studies were conducted more than two years ago with different strains of COVID-19 and included both vaccinated and unvaccinated groups. In those studies, Paxlovid was associated with a substantially reduced risk of hospitalization and death. We all remember that phase of the pandemic—people in hospitals on respirators, morgue trucks parked outside, and an atmosphere of fear. But that was a few years ago, and much has changed since then. The incidence of serious disease has decreased significantly with recent variants, and most of my patients are fully vaccinated, which is the most critical factor in reducing the severity of COVID-19 infection.

While the initial trials of Paxlovid were compelling, showing substantial reductions in hospitalization and death, more recent studies published this year in the New England Journal of Medicine are less impressive. The EPIC-SR trial of Paxlovid, published in April this year, did show a slight benefit with the medication, but the benefit seems to be limited to those at the highest risk for complications. It’s important to note that even this trial is based on data from two years ago and does not reflect the current, less severe strains of COVID-19.

The EPIC-SR study found that although fewer participants in the Paxlovid group were hospitalized for COVID-19 or died from any cause compared to the placebo group (0.8% vs. 1.6%, with the only death occurring in the placebo group), the difference was not statistically significant. This result suggests that the benefits observed among both vaccinated and unvaccinated high-risk individuals do not extend to those at lower risk for severe COVID-19, supporting guidelines that recommend Paxlovid only for those at high risk for disease progression.

Additionally, Paxlovid did not significantly shorten the duration of illness or symptoms in this study. The average duration of symptoms was 13 days in the placebo group and 12 days in the Paxlovid group. While this study reported a low risk of rebound, a recent study, reported in the Annals of Internal Medicine, from Mass General Hospital, found that one in five individuals taking Paxlovid experienced relapse of infection with a positive test result and shedding of live, potentially contagious virus after an initial recovery and negative test—a phenomenon known as virologic rebound. By contrast, rebound occurred in about 2% of those not taking Paxlovid.

This finding aligns with what many physicians have observed in practice: rebound is quite common. While many cases of rebound illness after Paxlovid are mild, mine was not. Although rebound illness does not seem to be overly severe, and people generally recover, I have not seen any reports fully addressing the implications of rebound COVID-19. The public health aspects of rebound illness, such as prolonged viral shedding and increased contagiousness, are significant. When people feel well and test negative, they will resume normal activities, stop wearing masks, attend events and stop testing thereby missing the fact that they are once again contagious. If they are experiencing mild rebound and shedding the virus, they can infect many others, who in turn spread the virus further. I have not seen any studies that adequately address the impact of contagiousness and the spread of infection in this context. From a public health standpoint, reducing disease transmission is likely the most effective way to reduce COVID-19 complications. I believe this is a critical omission in the current decision-making process regarding COVID-19 treatment.

The effects of Paxlovid on long COVID-19 are even less certain. A 2022 study suggested a significant reduction in long COVID, but this study was conducted two years ago with different viral strains in a less vaccinated, higher-risk population. It has not been shown that Paxlovid reduces long COVID with current strains of COVID-19 in either high-risk or lower-risk individuals. Furthermore, the incidence of long COVID has steadily diminished over the past four years, and it is not clear that Paxlovid plays any role in reducing this serious complication. Data from a New England Journal of Medicine letter reviewing data published in August 2024 indicate that the incidence of long COVID has decreased from a high of 10.42 cases per 100 persons at one year after infection during the pre-delta era to a low of 3.50 cases per 100 persons at one year among vaccinated individuals infected during the omicron era. The current incidence is likely even lower, given the ongoing evolution of variants during the Omicron era. The authors attribute 28% of this decrease to factors related to the pandemic era and 72% to vaccination. That being said, I have seen several articles and blogs from reputable sources stating that Paxlovid does not appear to increase the risk of long COVID, so there’s no harm in taking it. But that is a low bar for recommending treatment—”it probably won’t hurt!” Every medication has side effects, and Paxlovid is no different. Reconsider the above scenario of treating 1,000 patients and imagine scaling that up to 10 million patients. I believe we probably have better and more effective ways to spend our limited healthcare resources.

The bottom line is that I am uncertain about the best course of action for many of my patients with acute COVID-19. I would not recommend Paxlovid for LOW RISK people such as those under 65 unless they have significant medical conditions that place them at high risk. For those truly at high risk, I would recommend Paxlovid, but it’s important to recognize that not every condition listed by the CDC as increasing risk significantly raises the complication rate from COVID-19. Many people consider themselves “HIGH RISK” for COVID-19 complications even when they are not. Even for healthy individuals aged 65-74, it’s unclear whether this medication provides a major benefit, although I will continue to recommend it for those 65 and older, as per current guidelines, without insisting on its use. I am willing to provide Paxlovid to any of my patients who request it, as each person must make an individual decision about its benefits versus the side effects and risks. In clinical medicine, we generally treat patients, not society. What we truly need are realistic, science-based guidelines and adequate funding and communication from our public health experts. I encourage everyone to get updated vaccines, the most significant way to reduce COVID complications and long COVID, and to remain vigilant about avoiding high-risk situations during COVID-19 surges. If you do take Paxlovid, please be aware of the 20% incidence of rebound, test after completing treatment, and continue to isolate and take precautions if symptoms recur, especially if tests turn positive again—you do not want to infect others. Based on my experience and the current data, I do not plan to take Paxlovid if I contract COVID-19 again in the near future unless there is better data that I should. There is still much we do not know about this virus, and I hope that future studies will provide better guidance so we can improve our advice over time.

Mitchell B. Dunn, MD
DC Internists
August 2024