ABSTRACT Objectives This study investigated the outcomes ofunderweight patients with COVID-19 and the effectiveness of antiviral agents inthis population. Methods A retrospective cohort study using theTriNetX research network wasconducted. Propensity score matching (PSM) was employed to balance the firstcohort involving COVID-19 patients with underweight and normal-weight. In thesecond cohort, underweight patients receiving antiviral agents and untreatedindividuals were matched using PSM. The primary outcome was a composite ofall-cause hospitalization and death during the 7–30-day follow-up period. Results After PSM, the first cohort including eachgroup of 13,502 patients with balanced baseline characteristics were identifiedfor comparing the outcome of patients with underweight and normal weight. The underweightgroup had a higher risk of the composite primary outcome than those with normalweight (hazard ratio [HR], 1.251; 95% confidence interval [CI], 1.132–1.382). The second cohort included each 884 underweight patients withand without receiving antivirals.Compared with untreated patients,those receiving antiviral treatment had a lower risk of composite primaryoutcomes (HR, 0.426; 95% CI, 0.278–0.653). Conclusion Underweight status may be associated with ahigher risk of all-cause hospitalization and death in patients with COVID-19.Among underweight patients, antiviral agents demonstrated clinically beneficial effects.