Results
Clinical characteristics of the study population
The study population consisted of 3,029 patients who were hospitalized and received at least one dose of remdesivir for COVID-19 during the one-year study period. The median age was 67 years [IQR 55, 77]; 1,293 (42.7%) were female and 2,684 (88.6%) were white. Median BMI was 31 [IQR 27, 37] kg/m2. The median MASS was 3 [IQR 2, 6]. Most common medical comorbidities were hypertension (38.1%), cardiovascular disease (33.6%) and diabetes mellitus (31.7%).
Systemic steroids and other immunomodulators (baricitinib or tocilizumab) were used in 2,384 (78.7%) and 381 (12.6%) patients, respectively. Most patients (84.1%) required oxygen supplementation, as follows: 1,373 (45.3%) by nasal cannula, 829 (27.4%) by high-flow nasal cannula or non-invasive positive-pressor ventilator support, 152 (5.0%) by mechanical ventilator, and 5 (0.2%) by extracorporeal membrane oxygenation (ECMO). The details of the demographic and other clinical characteristics of patients are in Table 1 .
Overall outcomes of the study population
Two hundred and sixty-five (8.7%) patients died within 28 days of starting remdesivir; 219 (82.6%) were COVID-19 related. Older age was significantly associated with higher rates of all-cause and COVID-19 related deaths (Table 2 ). Higher all-cause and COVID-19 related death was also observed among patients who did not complete the 5-day course of remdesivir (Figure 1 ). Patients who improved clinically and were dismissed to complete the remdesivir course in the outpatient setting had a lower risk of all-cause and COVID-19 related death (Table 2 ).
Among patients who survived during the COVID-19 hospitalization, 141 (5.1%) patients had subsequent ED visits or re-admission within 30 days. There was a higher rate of ED visits and re-admissions within 30 days among patients who were dismissed prior to completion of 5 days of remdesivir (Table 2 ). Twenty-two (0.8%) patients had healthcare encounters due to long COVID within 180 days.
Characteristics and outcomes of patients who completed remdesivir inpatient versus those who were transitioned to the outpatient setting
Of 3,029 patients, the majority (2,711; 89.5%) completed the remdesivir for a total of 5 days. Among 2711 patients, the majority (n=2,169 [80%]) completed the 5 days of remdesivir during hospitalization while 542 patients (20.0%) were dismissed early and transitioned care to complete remdesivir in the outpatient setting. There was a total of 1,077 doses of remdesivir infused in hospital-based outpatient infusion centers.
Patients transitioned to complete the remdesivir course as an outpatient were significantly younger than those who remained hospitalized (median age, 61 [IQR 48, 72] versus 69 [IQR 57, 79]; p<0.001), and they had significantly lower number of comorbidities, as shown by MASS (median MASS 3 [IQR 1, 5] versus 4 [IQR 2, 6];p <0.001). The proportion of patients who required oxygen supplementation and systemic corticosteroids and immunomodulators were significantly higher among patients who remained in the inpatient setting (p <0.001). Abnormal liver tests were observed more often among patients who remained in the hospital (n=1,315 [60%]) compared to those transitioned to the outpatient setting (n=231 [42.6%]).
The 28-day mortality rate was 9.7% (n=211 patients) and 0.9% (n=5 patients) for the inpatient and outpatient groups, respectively. Adjusting for age, sex and MASS, the outpatient groups had lower odds of death within 28 days (adjusted Odd Ratio (aOR) 0.14, 95% confidence interval (CI) 0.06-0.32, p <0.001). However, the rate of subsequent ED visits or re-admission within 30 days was significantly higher in patients dismissed early to complete the remdesivir course in the outpatient setting (adjusted Hazard Ratio [aHR]: 1.88, 95% CI 1.27-2.79, p =.002) (Table 2 ). After excluding those who died in the hospital, there was no significant difference in the odds of death within 28 days between two groups (aOR: 0.69, 95% CI 0.28-1.68,p =0.41), but the rate of ED visits or re-admission within 30 days remained higher in the outpatient cohort (aHR: 1.75, 95% CI 1.15-2.64,p =0.009) (Supplemental Table 1 ).
Characteristics and outcomes of patients who completed versus did not complete remdesivir during hospitalization
Of 3,029 patients, the majority (n=2465 [81.4%]) received remdesivir only in the hospital setting. Among these 2465 patients, 296 (12.0%) did not complete the recommended 5-day course of remdesivir in the hospital (or in the outpatient setting). Twenty-two of 296 (7.4%) patients did not complete remdesivir because they died within five days after starting treatment.
There were no statistically significant differences in age, sex, and MASS between the inpatients who did and did not complete remdesivir in the hospital. However, the proportion of patients who required oxygen supplementation and received steroids or immunomodulators were significantly higher among patients who completed the remdesivir course for 5 days (p <0.001).
The adjusted odds of abnormal liver panel were lower among hospitalized patients who did not complete versus those who completed the 5-day course of remdesivir (aOR 0.56; 95% CI 0.43-0.72,p <0.001). There was no significant difference in subsequent ED visits or re-admission within 30 days, but the adjusted odds of 28-day mortality was two-fold higher among patients who did not complete remdesivir in the hospital (aOR 2.07, 95% CI 1.45-2.95,p <0.001) (Table 2 ). The results were similar, after excluding the patients who died in the hospital (Supplemental Table 1 ).
Characteristics and outcomes of dismissed patients who completed versus did not complete remdesivir in the outpatient setting
A total of 564 patients were dismissed from the hospital prior to completion of the 5-day course of remdesivir. Among them, 22 (0.7%) patients received at least one dose of outpatient remdesivir but did not complete the full 5-day course. There were no statistically significant differences in the clinical characteristics between patients who completed versus those who did not complete outpatient remdesivir.
The all-cause and COVID-19 mortality among 22 patients who did not complete the remdesivir course in the outpatient setting was similar to those who remained hospitalized for the 5-day course of remdesivir. In contrast, the all-cause and COVID-19 mortality among those who completed the 5-day course of remdesivir in the outpatient was lower than those who remained in the hospital (Table 2 ).