Hospital Readmissions Among Veterans Within 90 Days of Discharge Following Initial Hospitalization for COVID-19

Abstract

Introduction: Some patients experience ongoing sequelae after discharge, including rehospitalization; therefore, outcomes following COVID-19 hospitalization are of continued interest. We examined readmissions within 90 days of hospital discharge for veterans hospitalized with COVID-19 during the first 10 months of the pandemic in the US.

Methods: Veterans hospitalized with COVID-19 at a Veterans Health Administration (VA) hospital from March 1, 2020, through December 31, 2020 were followed for 90 days after discharge to determine readmission rates.

Results: Of 20,414 veterans hospitalized with COVID-19 during this time period, 13% (n = 2,643) died in the hospital. Among survivors (n = 17,771), 16% (n = 2,764) were readmitted within 90 days of discharge, with a mean time to readmission of 21.6 days (SD = 21.1). Characteristics of the initial COVID-19 hospitalization associated with readmission included length of stay, mechanical ventilator use, higher comorbidity index score, current smoking, urban residence, discharged against medical advice, and hospitalized from September through December 2020 versus March through August 2020 (all P values <.02). Veterans readmitted from September through December 2020 were more often White, lived in a rural or highly rural area, and had shorter initial hospitalizations than veterans hospitalized earlier in the year.

Conclusion: Approximately 1 of 6 veterans discharged alive following a COVID-19 hospitalization from March 1 through December 31, 2020, were readmitted within 90 days. The longer the hospital stay, the greater the likelihood of readmission. Readmissions also were more likely when the initial admission required mechanical ventilation, or when the veteran had multiple comorbidities, smoked, or lived in an urban area. COVID-19 hospitalizations were shorter from September through December 2020, suggesting that hospital over-capacity may have resulted in earlier discharges and increased readmissions. Efforts to monitor and provide support for patients discharged in high bed-capacity situations may help avoid readmissions.

Figure 1

Number and percentage of US veterans (N = 17,771) hospitalized for a diagnosis of COVID-19 who were readmitted to hospital within 90 days of discharge following their initial hospitalization, by number of days between discharge and readmission, March 1–December 31, 2020.

Figure 2

Logistic regression model showing estimated odds of readmission to a Veterans Administration hospital among US veterans (N = 17,771) hospitalized for a diagnosis of COVID-19 who were readmitted to hospital within 90 days of discharge following their initial hospitalization, March 1– December 31, 2020. The Elixhauser score ranges from –0.4 to 15.8 by using the Moore et al method (19) scaled down by 10. Odds ratio and associated 95% CIs for estimated coefficients are represented by points and bars, respectively, and by patient characteristics. Abbreviations: AMA, against medical advice; LTC, long-term care.