COVID-19–Related In-Hospital Deaths

COVID-19–associated hospitalization outcomes provide valuable information about trends in COVID-19 severity and mortality risk. In this section, we examine in-hospital deaths based on reporting from CDC’s sentinel hospital surveillance system, COVID-NET and from the Premier Healthcare Database Special COVID-19 Release. In addition to examining temporal trends in the risk of in-hospital death, including whether risk varied by different predominant variant periods, we identify patient risk factors associated with in-hospital death such as age and patient vaccination status. Lastly, we look at indicators of severe COVID-19 among hospitalized patients by predominant variant period.

Risk of COVID-19–Related In-Hospital Death Declined for All Adult Age Groups

The risk of in-hospital death among patients hospitalized with COVID-19 during March 2022–August 2022 was lower than during the previous nine months for all adult age groups. During June 2021–February 2022, the monthly proportion of COVID-19–associated hospitalizations among adults aged ≥65 years reported to COVID-NET that resulted in in-hospital death ranged from 10–20% (Figure 13). During March–August 2022, monthly proportions decreased to 4–7%. Among adults aged 18–49 years and 50–64 years, monthly proportions of COVID-19–associated hospitalizations that resulted in in-hospital death during March–August 2022 ranged from 0–6% compared to 1.5–14.5% during the previous nine months.

The decline in the risk of in-hospital death for adult hospitalized patients reported to COVID-NET was also observed among hospitalized patients with COVID-19 in the Premier Healthcare Database Special COVID-19 Release. Hospitalized COVID-19 patients in this dataset had lower crude mortality risk during April–June 2022 compared to all other months of the pandemic. During April–June 2022, the monthly crude mortality risk for total COVID-19 hospitalizations was 3.2–4.8%; for the subset of hospitalizations identified as being primarily for COVID-19, the monthly crude mortality rate was 4.0–6.2% (Figure 14). Although the crude mortality risk for patients with COVID-19 declined during April–June 2022, risk of in-hospital death remained higher among patients hospitalized with COVID-19 compared to hospitalized patients without COVID-19 (1.7–2.1%). However, the gap in the risk of in-hospital death for patients hospitalized with COVID-19 compared to patients hospitalized without COVID-19 during April–June 2022 narrowed to the lowest level since April 2020.

Vaccination Status of Hospitalized COVID-19 Patients with In-Hospital Death Varied by Age

During May–August 2022, the average weekly COVID-19–associated hospitalization rate per 100,000 population among adults aged 18–49 years was 4.3, compared to 7.9 among adults aged 50–64 and 28.7 among adults aged ≥65 years. Although COVID-19–associated hospitalizations that result in in-hospital death are infrequent among adults aged 18–49 years (1% of COVID-19–associated hospitalizations in this age group), these deaths continue to occur. During May–August 2022, among adults aged 18–49 years who died in-hospital due to COVID-19, 79% were unvaccinated against SARS-CoV-2 (Figure 15). Among hospitalized patients aged 18–49 years who did not experience in-hospital death, 43% were unvaccinated against SARS-CoV-2 (data not shown).

Among adults aged ≥65 years, 78% of COVID-19–associated in-hospital deaths were among patients who had completed at least a primary series of COVID-19 vaccination. Interpretation of the proportion of hospitalized people who are vaccinated is complicated, especially among older adults and persons with multiple or advanced comorbidities who have both greater risk for severe COVID-19 outcomes and high vaccination coverage rates.¹⁸ Primary series and booster vaccination rates in the United States are highest for adults aged ≥65 years. While ~5% of U.S. adults aged ≥65 years remained unvaccinated during May–August 2022, unvaccinated adults in this age group comprised 22% of all COVID-19–associated hospitalizations and 22% of in-hospital deaths (Figure 16). In contrast, at the midpoint of the same May–August period, 34% of U.S. adults aged ≥65 years had received at least two booster or additional doses of COVID-19 vaccine but comprised only 14% of COVID-19–associated hospitalizations and 6% of in-hospital deaths.

Similarly, vaccinated adults hospitalized with COVID-19 tend to have more underlying comorbidities than hospitalized unvaccinated adults. An analysis of COVID-NET data found that 96% of adults who received a booster and were hospitalized through April 2022 had at least one underlying comorbidity and 79% of adults who received a booster and were hospitalized had ≥3 underlying comorbidities. Among hospitalized unvaccinated adults, 89% had at least 1 underlying comorbidity and 52% had ≥3 underlying comorbidities. An increasing number of studies have shown that COVID-19 booster vaccination, especially for older adults, reduces risk of severe COVID-19 outcomes.¹⁵ Furthermore, as shown earlier in this report, COVID-19 vaccination continues to reduce risk of death, especially among persons who stay up to date with receiving additional recommended booster doses. Some persons might remain at high risk for severe outcomes due to COVID-19 even after vaccination for multiple reasons, including suboptimal immune response or waning immunity due to time since the receipt of most recent vaccination.¹⁹

Risk of COVID-19 In-Hospital Death Varied by the Predominant Variant Period

This section examines whether risk of in-hospital death varied in periods with different predominant circulating SARS-CoV-2 strains. Hospitalization data from the Premier Healthcare Database Special COVID-19 Release are examined for three variant periods: Delta, early Omicron, and later Omicron. Differences in the risk for in-hospital death across these periods must be interpreted with caution as additional factors, such as increased population immunity due to prior infection, widespread COVID-19 vaccination, and increased availability of COVID-19 treatment, coincided with changes in the predominant circulating strains. Note that the crude mortality risk rates presented do not adjust for patient SARS-CoV-2 immunity due to COVID-19 vaccination status or previous infection.

Of the three periods examined, the Delta period had the highest overall crude mortality risk, with 15% of patients hospitalized primarily for COVID-19 experiencing in-hospital death. Mortality risk during the Delta period was particularly high for patients with ≥5 underlying medical conditions, with approximately 27% of patients dying while hospitalized primarily for COVID-19 (Figure 17). Among patients with one underlying medical condition, 8% died, and among those with no underlying medical conditions the approximated death rate fell to 3% (data not shown).

During the early Omicron period, the crude mortality risk for patients hospitalized primarily for COVID-19 was 13%; it declined to 5% during the later Omicron period. During the later Omicron period, the crude mortality risk also declined for patient groups at higher risk of severe outcomes from COVID-19, ranging from 6–8% for adults aged ≥80 years, persons with disabilities, and persons with multiple underlying medical conditions. Although mortality risk declined substantially for these patient groups during the later Omicron period, these patients continued to have higher risk of in-hospital death compared to the overall population of patients hospitalized primarily for COVID-19.

The proportion of in-hospital COVID-19–related deaths accounted for by different age groups has varied over the course of the pandemic. Adults aged ≥65 years consistently accounted for the majority (54%–82%) of in-hospital deaths among patients hospitalized primarily for COVID-19 in all three periods examined (Figure 18). However, during the Delta period, adults aged <65 years accounted for 46% of in-hospital deaths; during the later Omicron period, this proportion decreased to 18%. During the later Omicron period, even as the crude mortality risk for adults aged ≥80 years declined (Figure 17), this age group accounted for an increasing proportion of COVID-19–associated in-hospital deaths. During the Delta period, adults aged ≥80 years accounted for 17% of in-hospital deaths; during the later Omicron period, they accounted for 45% of in-hospital deaths (Figure 18).

Hospitalized patients with ≥3 underlying medical conditions have accounted for a higher proportion of COVID-19–related in-hospital deaths compared to patients with ≤2 underlying medical conditions during all three variant periods. During the later Omicron period, this patient population accounted for an even larger proportion of deaths (73%) than during the Delta period (62%) (Figure 19). This increase is driven primarily by patients with ≥5 underlying medical conditions, who accounted for a larger proportion of in-hospital deaths during the later Omicron period (30% of deaths) than during the Delta period (19% of deaths). Although patients with underlying medical conditions accounted for a higher proportion of in-hospital deaths in the later Omicron period, the risk of dying while hospitalized was lower overall compared to both the Delta and early Omicron periods (Figure 17). It is also important to note that older age and multiple medical conditions are highly correlated. For example, among Medicare Fee-for-Service beneficiaries aged ≥65 years, 40% had ≥5 chronic conditions in 2018. ²⁰

Indicators of Disease Severity Among Hospitalized Patients with COVID-19 Declined

Among patients hospitalized primarily for COVID-19, the proportion requiring intensive medical intervention such as non-invasive mechanical ventilation or invasive medical intervention declined between the Delta period and the later Omicron period (Table 3). For example, during the Delta period, 17% of hospitalizations required invasive mechanical ventilation compared to 6% during later Omicron. Less severe COVID-19 disease among hospitalized patients likely contributed to the lower rate of in-hospital deaths observed during the later Omicron period and the potential divergence of the trends in hospitalization rates from COVID-19–related mortality rates. Additional information about this topic can be found in the section: Risk of COVID-19-Related Mortality.

Use of intensive medical interventions among patients who died while hospitalized primarily for COVID-19 also declined from the Delta period to the later Omicron Period.  For example, during the Delta period, 72% of patients hospitalized primarily for COVID-19 who experienced in-hospital death received invasive mechanical ventilation compared to 44% of patients during the later Omicron period. Although the reasons for this observed difference require further investigation, it might reflect the increased occurrence of in-hospital deaths among older persons with multiple comorbidities during later Omicron who might not have been able to tolerate or benefit from such interventions or, who did not agree to intensive medical intervention.