Background: Previous research has established that treating COVID-19 hospitalized patients with remdesivir (RDV) was associated with significantly lower hospital readmission rates. We extend prior research with an additional year of more recent data by examining the association between RDV use and 30-day readmission rates due to any cause as well as due to COVID-19 among patients hospitalized for COVID-19 during December 2020 to April 2023.
Methods: The PINC AI Healthcare Database was utilized to identify adults admitted with a primary discharge diagnosis for COVID-19 (ICD-10: U07.1) flagged as “present-on-admission” from December 2020 to April 2023 and discharged alive. The multivariable logistic regression analysis was used to examine 30-day all-cause and COVID-19 related readmission to the same hospital comparing RDV vs no RDV treatment, adjusting for key parameters of the initial COVID-19 hospitalization including patient age, corticosteroids use, Charlson Comorbidity Index, intensive care unit admission, and maximum supplemental oxygenation requirements.
Results: Of the 590,493 patients included in this study, 55% of the patients received RDV treatment during the initial COVID-19 hospitalization. Non-RDV patients were similar in age to RDV patients (median [interquartile range]: 66 [54-78] vs 64 [52-75]), had a higher proportion with no supplementary oxygen charges (NSOc) (55% vs 34%), and had a lower proportion with supplemental oxygen, including low-flow oxygen (33% vs 43%), high-flow oxygen/non-invasive ventilation (10% vs 20%), and invasive mechanical ventilation/ECMO (2% vs 3%) (Table 1).Overall, RDV-treated patients had lower readmission rates within 30 days of discharge for all-cause readmissions (7.7% vs 10.1%, p<.0001) and COVID-19 related readmissions (3.6% vs 5.8%, p<.0001) compared to patients treated without RDV (Figure 1). After adjusting for age and initial COVID-19 hospitalization characteristics, RDV patients had significantly lower odds of 30-day all-cause readmissions compared to non-RDV patients overall (adjusted odds ratio [95% CI]: 0.76 [0.75-0.78], p<.0001), as well as among the subgroups of NSOc (0.75 [0.72-0.77], p<.0001) and any supplemental oxygen use (0.79 [0.77-0.81], p<.0001) (Figure 1). Similarly, 30-day COVID-19-related readmissions were significantly lower for RDV-treated patients than non-RDV patients overall (0.63 [0.61-0.65], p<.0001), as well as among the subgroups of NSOc (0.59 [0.57-0.62], p<.0001) and any supplemental oxygen use (0.67 [0.64-0.69], p<.0001) (Figure 1).
Conclusions: Among adults hospitalized with COVID-19 during the study period (December 2020 to April 2023), RDV treatment was associated with a 24% lower likelihood of all-cause and 37% lower likelihood of COVID-19-related readmission. Patients treated with RDV had a lower rate of readmission in spite of having higher supplemental oxygenation support in their initial COVID-19 hospitalization compared to patients not treated with RDV.
