Is remdesivir treatment associated with reduced risk of inpatient mortality among US patients hospitalized with COVID-19?

In a recent study posted to JAMA Network Open, researchers assessed the effect of remdesivir on inpatient coronavirus disease 2019 (COVID-19)-associated mortality among individuals hospitalized with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections in an enormous dataset of United States (US) residents beyond clinical trial settings.

Study: Association of Remdesivir Treatment With Mortality Among Hospitalized Adults With COVID-19 in the United States. Image Credit: Sonis Photography/Shutterstock


The COVID-19 pandemic has caused considerable mortality and morbidity across the globe. Randomized controlled trial (RCT) findings are considered the standard criterion standard; however, mixed results have been obtained for remdesivir use in prior documented trials and remdesivir efficacy in pandemic mitigation.

Regular clinical practice data can add to RCT findings to yield required scientific evidence, particularly during a worldwide pandemic with swiftly transforming clinical management strategies outpacing the performance of new clinical trials. Moreover, clinical practice data can increase the study population for assessing outcomes among individual subgroups.

About the study

In the present retrospective cohort study, researchers investigated whether remdesivir medication could reduce inpatient death rates among hospitalized COVID-19 patients.

Deidentified HealthVerity system data, including pharmacy and medical claims data from Medicare, Medicaid, and commercial health plans, and chargemaster data from 400 hospitals, obtained between 1 December 2018 and 3 May 2021. Only adult hospitalized COVID-19 patients who were treated with remdesivir between 1 May 2020 and 3 May 2021, with activity in data sources beginning ≥1.0 year before the index date and ≥1 chargemaster record or medical claim in the year before the index date, were included for the analysis.

The study outcomes and measures were all-cause inpatient deaths within four weeks of commencing remdesivir drug treatment for remdesivir-treated individuals and the matched index dates for controls. The diagnosis of COVID-19 was based on ICD-10 (international statistical classification of diseases and related health problems, 10th revision) code U07.1.

The dates of initiating remdesivir treatment during the inpatient stay for remdesivir-treated individuals and the concordant hospitalization dates for controls were considered index dates. Individuals prescribed as part of clinical trials, history of SARS-CoV-2 infection-associated inpatient hospitalizations or previous remdesivir use were excluded from the analysis. The team followed a two-stage approach involving 1:1 RSS (risk-set sampling) and propensity score matching (PS) to attain a balance in baseline parameters between remdesivir-treated individuals and controls.

The severity of SARS-CoV-2 infections was evaluated on the basis of the index dates. Comorbid conditions were assessed the year before, exclusive of the index dates. Co-medications were recorded as those received between the hospitalization dates and index dates. Cox proportional hazards regression modeling was performed, and the hazard ratios (HR) were calculated.


A total of 1,143,770 individuals were hospitalized between 1 May 2020 and 3 May 2021, and 113,579 of them were eligible for further analysis. Of the eligible patients, 35% (n=39,775) were treated with remdesivir. The median value for the duration between hospitalization and the initiation of remdesivir treatment was one day. After PS and RSS matching, a total of 24,856 individuals exposed to remdesivir treatment, and an equal number of controls, were considered for the final analysis.

The majority of the study participants were male, with an average age of 67 years. The median values for the duration of follow-up in the remdesivir-treated group and control group were six days and five days, respectively. A total of 3,557 deaths (14%) and 3,775 deaths (15%) were reported in the remdesivir group and control group, respectively. 

The 28.0-day death rates among remdesivir-treated individuals and controls were 0.5 and 0.6 per individual month, respectively. Remdesivir use significantly reduced inpatient deaths among hospitalized COVID-19 patients by 17.0% in comparison to controls (HR 0.8). At baseline, 64%, 22%, 11%, and 3.0% of individuals in both groups breathed ambient air and required low-flow oxygenation, required high-flow oxygenation with non-invasive mechanical ventilation, required invasive ventilation or ECMO (extracorporeal membrane oxygenation), respectively.

Significant reductions in deaths were found in all oxygen supplementation subgroups, inclusive of individuals breathing room air (HR 0.9), those requiring low-flow oxygenation (HR 0.8), those requiring high-flow oxygenation or non-invasive mechanical ventilators (HR 0.8), and those with invasive mechanical ventilator or ECMO requirements (HR 0.8).

The most commonly observed comorbidities among the participants were hypertension (72%), respiratory (69%), and metabolic conditions (77%). Corticosteroids were administered to 70% of the remdesivir-treated and control individuals. The as-treated analysis and post hoc initial-treatment analysis yielded similar results.

Among controls, 25% (n=6,259) were eventually administered remdesivir medication and, in the as-treated-type analysis, were restricted during the initiation of treatment. The tipping point analysis showed that the E-value for HR 0.8 was 1.7, indicating that the HR could be completely explained by residual confounders in the case of a ≥1.7 relative risk associated with remdesivir drug exposure and with inpatient deaths for unmeasured confounders.


Overall, the study findings showed that remdesivir use significantly lowered inpatient deaths among hospitalized COVID-19 patients, complementing RCT findings. Further research must be conducted to evaluate the effects of remdesivir use by the SARS-CoV-2 variant in relation to the days elapsed from the onset of symptoms. 

Source link

Back to top button