Cost-effectiveness of antiviral therapies for hospitalized COVID-19 patients at Dr. Hasan Sadikin Hospital, Indonesia: a retrospective real-world analysis

Background The COVID-19 pandemic placed substantial clinical and economic pressure on healthcare systems, particularly in low- and middle-income countries. Evidence on the cost-effectiveness of antiviral therapies using real-world hospital data in Indonesia remains limited. Objective To compare costs and CT-defined virological response among favipiravir, remdesivir, and oseltamivir in hospitalized COVID-19 patients at Dr. Hasan Sadikin Hospital. Oseltamivir was retained as a historical comparator because it was used during the early pandemic under evolving Indonesian treatment guidance. Methods A retrospective individual-level cost-effectiveness analysis was conducted using data from adults hospitalized with COVID-19 during 2020–2021 from healthcare-provider and limited societal perspectives. Effectiveness was defined as the change in RT-PCR cycle-threshold (CT) value from baseline to the last available measurement. This semi-quantitative virological surrogate was used because paired RT-PCR CT values were consistently available, whereas patient-centered outcomes were not available in the source dataset. Mean costs and CT changes were compared, and incremental cost-effectiveness ratios (ICERs) were calculated against oseltamivir. Uncertainty was assessed using 10,000 stratified non-parametric bootstrap replications and one-way sensitivity analyses. Exploratory regression models adjusted for age, sex, admission year, and baseline CT. Results Among 871 patients, 15 received oseltamivir, 554 favipiravir, and 302 remdesivir. Mean healthcare costs were IDR 43.95 million, 29.01 million, and 50.90 million, respectively, with mean CT changes of 10.97, 12.30, and 12.08 units. Compared with oseltamivir, favipiravir had an IDR 14.94 million lower mean cost and a 1.33-unit greater mean CT change, indicating point-estimate dominance. Remdesivir had an IDR 6.95 million higher mean cost and a 1.11-unit greater mean CT change (ICER: IDR 6.27 million per additional CT unit). Bootstrap intervals for incremental costs and effects crossed zero. Adjusted analyses showed no statistically significant CT differences between antivirals and oseltamivir, and both adjusted contrasts were negative in direction. Conclusion Point estimates favored favipiravir over the historical oseltamivir comparator, while remdesivir had higher costs with modest incremental CT improvement. Because no accepted willingness-to-pay threshold exists per CT unit, the remdesivir ICER is descriptive only. Given the surrogate nature of CT, the small oseltamivir group, and potential residual confounding, these findings should be interpreted as exploratory economic evidence rather than definitive guidance for treatment or reimbursement decisions.