As SARS-CoV-2 continues to evolve in the context of emergence of new SARS-CoV-2 variants, increasing vaccination and hybrid immunity, timely monitoring of COVID-19 antiviral effectiveness remains critical, particularly in high-risk populations. This study evaluated the feasibility of using routinely collected health data in Victoria, Australia, to estimate the real-world effectiveness of oral antivirals (molnupiravir and nirmatrelvir–ritonavir) against severe COVID-19 outcomes. We analysed record-linked, routinely collected health data for individuals aged ≥70 years with confirmed COVID-19 between July 2022 and March 2023. Three analytical methods (multivariable logistic regression, Cox regression, and target trial emulation) were applied to assess associations between antiviral treatment and risk of hospitalisation or death within 35 days of COVID-19 onset. Across all approaches, antiviral treatment was consistently associated with reduced rate of severe outcomes compared to no treatment. Estimated effects ranged from hazard ratios of 0.62 (95% CI: 0.57–0.68) from the adjusted Cox model and 0.45 (95% CI: 0.41-0.49) in the target trial emulation, to an odds ratio of 0.28 (95% CI: 0.25-0.30) in the adjusted logistic model. Overall, the findings support the use of oral antivirals in older adults. Variation in effect estimates reflect differing methodological assumptions and estimands, rather than direct differences in effectiveness, and highlight the value of triangulating multiple analytical approaches when using observational data. The study also underscores the need to improve the quality and standardisation of routinely collected health data to strengthen pandemic preparedness, and support timely, evidence-based public health responses in future outbreaks.
Fathima et al. (Fri,) studied this question.