Effectiveness of nirmatrelvir/ritonavir (Paxlovid®) in preventing hospitalisation and death among COVID-19 patients: a prospective cohort study

INTRODUCTION: Previous trials and real-world studies have shown that nirmatrelvir/ritonavir (Paxlovid®) reduces hospitalisation and deaths in symptomatic, high-risk, nonsevere COVID-19 patients. However, there was a scarcity of data on its effectiveness in the local setting. This study aimed to determine the effectiveness of Paxlovid® in reducing hospitalisation and mortality among COVID-19 patients and to identify the types of adverse events that occur after taking Paxlovid®.

MATERIALS AND METHODS: A two-arm prospective cohort study was conducted among adult patients with COVID-19 categories 2 and 3 treated with Paxlovid® and a matched control group. A standard risk-stratified scoring system was used to establish Paxlovid® eligibility. All patients who were prescribed Paxlovid® and took at least one dose of Paxlovid® were included in the study. The control patients were selected from a centralised COVID-19 patient registry and matched based on age, gender and COVID-19 stage severity.

RESULTS: A total of 552 subjects were included in the study and evenly allocated to the treatment and control groups. There was no statistically significant difference in 28-day hospitalisation after diagnosis [Paxlovid®: 26 (9.4%), Control: 34 (12.3%), OR: 0.74; 95%CI, 0.43-1.27; p=0.274] or all-cause death [Paxlovid®: 2 (0.7%), Control: 3 (1.1%), OR 1.51; 95%CI, 0.25-9.09; p=0.999]. There was no significant reduction in hospitalisation duration, intensive care unit admission events or supplementary oxygen requirement in the treatment arm. Ethnicity, COVID-19 severity at diagnosis, comorbidities and vaccination status were predictors of hospitalisation events.

CONCLUSION: In this two-arm study, Paxlovid® did not significantly lower the incidence of hospitalisation, all-cause death and the need for supplemental oxygen. Adverse effects were frequent but not severe. Paxlovid® efficacy varied across settings and populations, warranting further real-world investigations.