Hospital Readmissions After COVID-19, March–August 2020

This report covers the first months of the COVID-19 pandemic in the United States, in 2020, before vaccines were available.

COVID-19 is a complex illness whose complications may need continuing care, but in 2020 few studies had followed large groups of patients after their first COVID-19 hospital stay. CDC used electronic health records and administrative data from the Premier Healthcare Database, drawing on 865 nongovernmental community and teaching hospitals, to see how patients left the hospital, how often they came back, and who was most likely to be readmitted.

The first hospitalization with a COVID-19 diagnosis counted as the index stay. Any later admission to the same hospital within two months of discharge, through August 2020, for COVID-19 or anything else, counted as a readmission; the database records only returns to the same hospital.

The patients

From March to July 2020, 126,137 patients were hospitalized with COVID-19. Most (81%) came from outside health care settings and 18% were transferred from another hospital, clinic or skilled nursing facility. During the index stay, 15% were admitted to intensive care, 13% needed invasive mechanical ventilation and 4% noninvasive ventilation.

Sixty-two percent had a diagnosis for at least one of five chronic conditions linked to severe COVID-19: chronic obstructive pulmonary disease (21%), heart failure (16%), diabetes (27%), chronic kidney disease (21%) or obesity (27%). Eight percent (10,008) had been in the same hospital in the three months before. About 15% (19,594) died during the index stay.

Where survivors went

Of the 106,543 who were discharged:

Readmissions

• How many: 9,504 patients (9%) were readmitted within two months, and 1,667 (1.6%) more than once.
• How soon: the median time to first readmission was 8 days (interquartile range 3–20). Patients sent home for self-care who came back did so within a median of 7 days; about two thirds of them had at least one of the five chronic conditions.
• Deaths: fewer than 0.1% of patients died during a readmission.
• Why: the most common main diagnoses at readmission were infectious and parasitic diseases, mostly COVID-19 (45%), followed by circulatory (11%) and digestive (7%) diseases.

After accounting for other factors, the odds of readmission were higher for patients:

• aged 65 or older, rising with age, compared with those aged 18–39;
• with chronic obstructive pulmonary disease (odds ratio 1.4), heart failure (1.6), diabetes (1.2) or chronic kidney disease (1.6);
• hospitalized in the three months before the COVID-19 stay (2.6 times the odds);
• discharged to a skilled nursing facility (1.4) or with home health care (1.3), rather than home.

Non-Hispanic white patients were more likely to be readmitted than patients of other racial and ethnic groups. Non-Hispanic Black (23%) and Hispanic (21%) patients made up larger shares of the cohort than of the U.S. population (13% and 18%), yet their odds of readmission were lower. A slight link between readmission and length of stay needed more study.

What it means

The findings match other studies linking chronic conditions to readmission, which could reflect complications of those conditions combined with COVID-19, lasting effects of COVID-19, or indirect effects of the pandemic. Knowing how often and why patients return can guide clinical care and discharge decisions, especially about what level and place of follow-up care patients who seem stable at discharge will need. With cases rising nationwide in late 2020, the authors advised hospitals to plan for at least 9% of COVID-19 patients needing readmission, and the extra beds and resources that implies, and called for continued prevention efforts for older people and those with underlying conditions.

Limits

COVID-19 was identified from diagnosis codes, not lab results, so some cases may be misclassified. Chronic conditions were counted only if they were coded. The main discharge diagnosis only approximates the reason for a stay. Patients who went to a different hospital could not be followed, and the lasting effects of COVID-19 in people who were never readmitted were not captured.

Sources

• Lavery AM, Preston LE, Ko JY, et al. Characteristics of Hospitalized COVID-19 Patients Discharged and Experiencing Same-Hospital Readmission — United States, March–August 2020. MMWR Morb Mortal Wkly Rep 69(45); posted as an Early Release on November 9, 2020. https://www.cdc.gov/mmwr/volumes/69/wr/mm6945e2.htm
• The report's discussion lists the most common readmission diagnoses as circulatory, digestive and respiratory diseases, which differs from its results; this page uses the figures in the results.