While the COVID-19 pandemic appears to be winding down, its effects are still felt by the millions of people worldwide experiencing post–COVID-19 condition (PCC, or long COVID).1 The antiviral drug nirmatrelvir (marketed as Paxlovid [Pfizer], in combination with ritonavir) and molnupiravir (Lagevrio [Merck]) are recommended as first- and second-line treatments for acute illness in patients with specific risk factors (eg, diabetes).2 However, there are still no US Food and Drug Administration–approved drugs for the treatment or prevention of PCC. Recent studies among US veterans (mostly male) suggest that nirmatrelvir and molnupiravir reduce the risk of some sequelae of COVID-19.3,4 We performed a cohort study of the 2 drugs in PCC in older patients who were Medicare enrollees.
The cohort came from Medicare enrollees aged 65 years or older diagnosed with COVID-19 between January and September 2022. COVID-19 was identified with an outpatient International Statistical Classification of Diseases, Tenth Revision, Clinical Modification code of U07.1. In January 2022, free home COVID-19 tests became available and not all positive self-tests were captured in Medicare data. Therefore, we also considered the prescription of nirmatrelvir or molnupiravir to be indicative of COVID-19 because no other indications existed. Following previous work,5 we identified PCC based on the World Health Organization (WHO) consensus clinical definition.6 Any new occurrence (not present prior to COVID-19 diagnosis) of the 11 symptoms between 4 to 12 weeks after infection was considered as PCC. We used an extended Cox regression with propensity score adjustment to examine the 2 drugs and the incidence of PCC. We included age, sex, race, geographic region, dual eligibility, low-income subsidy, and 51 chronic comorbidities as covariates as included in the Medicare data (eMethods, eTable in Supplement 1). This study was declared not human participant research by the Office of Human Research Protection at the National Institutes of Health. Statistical analyses were conducted using SAS version 7.15 (SAS Institute Inc) and a 2-sided significance at P < .05. This study followed the STROBE reporting guideline.
Overall, among 3 975 690 outpatients with COVID-19, 57% remained in our study after exclusion. Among them, 19.5% received nirmatrelvir and 2.6% received molnupiravir. PCC incidence among patients receiving nirmatrelvir was 11.8%, 13.7% for molnupiravir, and 14.5% for neither, absolute risk reduction was 2.7% for nirmatrelvir, 0.8% for molnupiravir, with hazard ratios (HRs) of 0.87 (95% CI, 0.86-0.88; P < .001) for nirmatrelvir and 0.92 (95% CI, 0.90-0.94; P < .001) for molnupiravir, compared with no treatment (Table 1). Sensitivity analysis of only patients with the COVID-19 code showed a similar pattern but smaller effect sizes (nirmatrelvir: HR, 0.93 [95% CI, 0.92-0.94; P < .001], molnupiravir: HR, 0.96 [95% CI, 0.93-0.99; P = .001]). In an interaction analysis, we found significantly smaller effect sizes in females than males (HRs for nirmatrelvir: 0.89 vs 0.84; molnupiravir: 0.95 vs 0.88). Female sex; Asian, Black, and Hispanic races; and indicators of low income were associated with increased risk of PCC. The most common symptoms in PCC were fatigue (29.9%), dyspnea (22.4%), and cough (21%) (Table 2).
Consistent with the findings of Xie et al,3,4 we found that nirmatrelvir and molnupiravir were associated with a small reduction in incidence of PCC. Our effect sizes are smaller than those of Xie et al3,4 (absolute risk reduction, nirmatrelvir 4.5%; molnupiravir 3.0%) but our sample size is 8-fold larger. We also have a more balanced sex ratio (female 59% vs 14%), which is important because PCC is more common in females. The smaller effect sizes in females may explain our overall smaller effect sizes. We used the WHO consensus definition based on symptoms rather than disease diagnosis (eg, ischemic heart disease), which is more akin to how PCC is identified clinically. Limitations of our study include not incorporating vaccination status because of incomplete data, use of prescription of the drugs as evidence of COVID-19, and restriction to patients 65 years or older. The current approved use of the 2 drugs is for the prevention of severe acute COVID-19. Our findings suggest that they may also have a role in preventing PCC.
Accepted for Publication: July 22, 2023.
Published Online: October 23, 2023. doi:10.1001/jamainternmed.2023.5099
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2023 Fung KW et al. JAMA Internal Medicine.
Corresponding Author: Kin Wah Fung, MD, 8600 Rockville Pike, Bethesda, MD 20894 (kfung@mail.nih.gov).
Author Contributions: Drs Baye and Baik had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Fung, Baik, McDonald.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Fung, Baik, McDonald.
Critical review of the manuscript for important intellectual content: All authors.
Statistical analysis: Baye, Baik.
Administrative, technical, or material support: McDonald.
Supervision: Baik, McDonald.
Conflict of Interest Disclosures: None reported.
Funding/Support: This research was supported by the Intramural Research Program of the NIH, National Library of Medicine.
Role of the Funder/Sponsor: The Intramural Research Program of the NIH, National Library of Medicine had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Data Sharing Statement: See Supplement 2.