Prognostic factors in the selection of antivirals for COVID-19 in high-risk outpatients: A retrospective cohort study.

Abstract

Background: Early oral antiviral treatment reduced disease progression and hospitalisation in high-risk COVID-19 outpatients. In practice, prescribing decisions were influenced by clinical risk, drug interaction concerns, and physician judgment, even in settings where guidelines prioritised nirmatrelvir/ritonavir. This study aimed to identify the patient characteristics that independently predicted oral antiviral selection.

Objectives: To identify (1) independent predictors of nirmatrelvir/ritonavir versus molnupiravir prescribing, and (2) predictors of receiving any potent antiviral versus no potent antiviral, among high-risk COVID-19 outpatients.

Design: Retrospective cohort study.

Methods: Using data from Bhumibol Adulyadej Hospital in Bangkok, Thailand, this retrospective cohort study enrolled 450 COVID-19 outpatients between April 2022 and June 2023. Participants were randomly selected into three treatment groups of 150 each: nirmatrelvir/ritonavir, molnupiravir, and no antiviral. Multivariate logistic regression was applied to identify independent predictors of drug selection across demographics, comorbidities, and concomitant medications.

Results: Treated patients were older than those who received no antiviral, with mean ages of 61.68 and 62.91 years in the nirmatrelvir/ritonavir and molnupiravir groups versus 43.89 years in the no-treatment cohort (p < 0.001). Comorbid disease was more prevalent in both antiviral groups. When nirmatrelvir/ritonavir and molnupiravir recipients were compared directly, concurrent statin use was the only variable that independently predicted a lower probability of nirmatrelvir/ritonavir being chosen (adjusted OR 0.46; 95% CI 0.27 to 0.79; p = 0.005).

Conclusion: Older patients and those carrying a greater comorbidity burden were prioritised for antiviral treatment in this Thai outpatient cohort. Drug-drug interaction concerns, particularly between ritonavir and statins, drove physicians toward molnupiravir even when nirmatrelvir/ritonavir would otherwise have been the preferred agent. Systematic incorporation of pharmacist-conducted medication assessment before antiviral prescribing represents a practicable strategy for reducing avoidable drug substitution in high-risk patients.