Association Between Dexamethasone Treatment After Hospital Discharge for Patients With COVID-19 Infection and Rates of Hospital Readmission and Mortality

Abstract

Importance: Current guidelines recommend use of dexamethasone, 6 mg/d, up to 10 days or until discharge for patients hospitalized with COVID-19. Whether patients who received less than 10 days of corticosteroids during hospitalization for COVID-19 benefit from continuing treatment at discharge has not been determined.

Objective: To assess whether continuing dexamethasone treatment at discharge is associated with reduced all-cause readmissions or mortality postdischarge.

Design, setting, and participants: A retrospective cohort study was conducted at 15 medical centers within Kaiser Permanente Southern California. The population included adults who received less than 10 days of dexamethasone, 6 mg/d, until discharge during hospitalization for COVID-19 and were discharged alive between May 1 and September 30, 2020.

Exposures: Continued dexamethasone treatment at discharge.

Main outcomes and measures: All-cause readmissions or mortality within 14 days from discharge.

Results: A total of 1164 patients with a median age of 55 (IQR, 44-66) years were identified. Most patients were of Hispanic ethnicity (822 [70.6%]) and male (674 [57.9%]) and required oxygen support during hospitalization (1048 [90.0%]). Of the 1164 patients, 692 (59.5%) continued dexamethasone, 6 mg/d, at discharge. A balanced cohort was created using propensity score and inverse probability of treatment weighting. The adjusted odds ratio (OR) for readmissions or mortality within 14 days was 0.87 (95% CI, 0.58-1.30) for patients who continued dexamethasone therapy at discharge compared with those who did not. Similar results were produced by a sensitivity analysis that restricted the treatment group to those who received exactly 10 days of dexamethasone (OR, 0.89; 95% CI, 0.55-1.43) and by subgroup analyses stratified by the duration of dexamethasone treatment as an inpatient (1-3 days: OR, 0.71; 95% CI, 0.43-1.16; 4-9 days: OR, 1.01; 95% CI, 0.48-2.12), oxygen requirement at discharge (room air: OR, 0.91; 95% CI, 0.53-1.59; supplemental oxygen use: OR, 0.76; 95% CI, 0.42-1.37), and disease duration at discharge (≤10 days: OR, 0.81; 95% CI, 0.49-1.33; >10 days: OR, 0.94; 95% CI, 0.48-1.86).

Conclusions and relevance: In this cohort study of patients with COVID-19, continuing treatment with dexamethasone, 6 mg/d, at discharge was not associated with a reduction in 14-day all-cause readmission or mortality. This finding suggests that dexamethasone should not be routinely prescribed beyond discharge for individuals with COVID-19.

Figure 1.. Population Flowchart

Data were missing on race and ethnicity (n = 7), body mass index (n = 2), and oxygen requirement at discharge (n = 4).

Figure 2.. Adjusted Odds Ratios of 14-Day All-Cause Readmission or Mortality by Subgroup After Inverse Probability of Treatment Weighting

An odds ratio less than 1 favors continuing dexamethasone at discharge. ^aThe variable for subgroup stratification was removed from the adjustment or propensity score. Otherwise, odds ratios were calculated with additional adjustment of symptom duration at discharge (<10 days) after inverse probability treatment weighting of propensity score including age, sex, race and ethnicity, body mass index, number of Elixhauser comorbidities (category), duration of inpatient dexamethasone treatment (days), remdesivir treatment, convalescent plasma, biologics therapy (including anakinra and tocilizumab), therapeutic anticoagulation, supplemental oxygen at discharge, and hypoxia (O₂ saturation <94%) at discharge.