Clinical Characteristics and Outcomes Among Adults Hospitalized with Laboratory-Confirmed SARS-CoV-2 Infection During Periods of B.1.617.2 (Delta) and B.1.1.529 (Omicron) Variant Predominance — One Hospital, California, July 15–September 23, 2021, and December 21, 2021–January 27, 2022

In mid-December 2021, the B.1.1.529 (Omicron) variant of SARS-CoV-2, the virus that causes COVID-19, surpassed the B.1.617.2 (Delta) variant as the predominant strain in California.§ Initial reports suggest that the Omicron variant is more transmissible and resistant to vaccine neutralization but causes less severe illness compared with previous variants (1-3). To describe characteristics of patients hospitalized with SARS-CoV-2 infection during periods of Delta and Omicron predominance, clinical characteristics and outcomes were retrospectively abstracted from the electronic health records (EHRs) of adults aged ≥18 years with positive reverse transcription-polymerase chain reaction (RT-PCR) SARS-CoV-2 test results admitted to one academic hospital in Los Angeles, California, during July 15-September 23, 2021 (Delta predominant period, 339 patients) and December 21, 2021-January 27, 2022 (Omicron predominant period, 737 patients). Compared with patients during the period of Delta predominance, a higher proportion of adults admitted during Omicron predominance had received the final dose in a primary COVID-19 vaccination series (were fully vaccinated) (39.6% versus 25.1%), and fewer received COVID-19-directed therapies. Although fewer required intensive care unit (ICU) admission and invasive mechanical ventilation (IMV), and fewer died while hospitalized during Omicron predominance, there were no significant differences in ICU admission or IMV when stratified by vaccination status. Fewer fully vaccinated Omicron-period patients died while hospitalized (3.4%), compared with Delta-period patients (10.6%). Among Omicron-period patients, vaccination was associated with lower likelihood of ICU admission, and among adults aged ≥65 years, lower likelihood of death while hospitalized. Likelihood of ICU admission and death were lowest among adults who had received a booster dose. Among the first 131 Omicron-period hospitalizations, 19.8% of patients were clinically assessed as admitted for non-COVID-19 conditions. Compared with adults considered likely to have been admitted because of COVID-19, these patients were younger (median age = 38 versus 67 years) and more likely to have received at least one dose of a COVID-19 vaccine (84.6% versus 61.0%). Although 20% of SARS-CoV-2-associated hospitalizations during the period of Omicron predominance might be driven by non-COVID-19 conditions, large numbers of hospitalizations place a strain on health systems. Vaccination, including a booster dose for those who are fully vaccinated, remains critical to minimizing risk for severe health outcomes among adults with SARS-CoV-2 infection.

Compared with adults considered likely to have been admitted because of COVID-19, these patients were younger (median age = 38 versus 67 years) and more likely to have received at least one dose of a COVID-19 vaccine (84.6% versus 61.0%). Although 20% of SARS-CoV-2-associated hospitalizations during the period of Omicron predominance might be driven by non-COVID-19 conditions, large numbers of hospitalizations place a strain on health systems. Vaccination, including a booster dose for those who are fully vaccinated, remains critical to minimizing risk for severe health outcomes among adults with SARS-CoV-2 infection.
Periods of Delta and Omicron predominance (July 15-September 23, 2021, and December 21, 2021-January 27, 2022, respectively) were defined to correspond to peaks in SARS-CoV-2 hospitalizations during which each variant accounted for ≥50% of sequenced SARS-CoV-2 isolates in California (Supplementary Figure, https://stacks.cdc.gov/view/ cdc/113987). RT-PCR-positive test results were determined via the hospital's internal flagging system for SARS-CoV-2 admissions, which incorporated laboratory results and provider documentation. ¶ Vaccination status was ascertained through electronic linkage from the EHR to the California Immunization Registry (CAIR).** Patient demographic and clinical characteristics were abstracted from the EHR. For early Omicron-period hospitalizations (December 21-January 2), detailed chart review was performed by one of four clinicians to determine whether the reason for admission was likely or not likely due to COVID-19, following prespecified criteria. † † Patient demographic and clinical characteristics were compared between Delta-and Omicron-period hospitalizations, ¶ This flagging can be triggered in one of two ways: either by a laboratory report indicating a positive SARS-CoV-2 RT-PCR test result from any time during hospitalization or during a 14-day lookback window preceding admission, which included tests performed in ambulatory and inpatient settings, or by admitting physician confirmation of RT-PCR positivity from an outside facility via patient interview, during which time a patient was queried about positive RT-PCR test results and any related COVID-19 symptoms over the preceding 14 days. Hospitalizations were included in the study if they occurred among adults without another hospitalization associated with a positive SARS-CoV-2 RT-PCR result during the preceding 90 days, or if there was a hospitalization associated with a positive RT-PCR result during the preceding 90 days, but the patient's symptoms had resolved before readmission as determined by the admitting provider.  Table 1). The median age increased both overall and among unvaccinated persons (Omicron = 64 years; Delta = 54 years; p<0.01), but not among fully vaccinated persons. The proportion of fully vaccinated adults who were Hispanic was higher during Omicron predominance (21.9%) than during Delta predominance (10.6%) (p = 0.02). Conversely, non-Hispanic White persons accounted for fewer admissions among fully vaccinated adults during Omicron predominance than during Delta predominance (46.6% versus 62.4%; p = 0.01). Fewer patients admitted during Omicron predominance than during Delta predominance received COVID-19-directed therapies, both among unvaccinated (57.9% and 81.7%; respectively) (p<0.01) and fully vaccinated adults (52.4% and 76.5%, respectively) (p<0.01). Compared with Delta-period patients, fewer Omicron-period patients required ICU admission (16.8% versus 23.3%; p = 0.01) or IMV (9.2% versus 13.6%; p = 0.03), and fewer died while hospitalized (4.0% versus 8.3%; p = 0.01). When stratified by vaccination status, however, differences in ICU admission and IMV between the two periods were not significant, despite lower percentages during Omicron predominance. Fewer fully vaccinated adults hospitalized during Omicron predominance died while hospitalized (3.4%) compared with those hospitalized during Delta predominance (10.6%) (p = 0.02). Among adults hospitalized during Omicron predominance, increasing vaccination was associated with lower likelihood of ICU admission (p = 0.02) and, among adults aged ≥65 years, lower likelihood of death while hospitalized (p = 0.04) ( Figure). Fully vaccinated patients who had received a booster dose had the lowest likelihood of these outcomes.
Of 131 early Omicron-period hospitalizations (December 21-January 2), 105 (80.2%) patients were assessed to have been likely admitted for COVID-19, and 26 (19.8%) were admitted primarily for non-COVID-19 conditions ( Table 2). Compared with adults hospitalized for COVID-19, those hospitalized for other conditions were younger (median age 38 versus 67 years; p<0.01), more likely to have received at least one dose of a COVID-19 vaccine (84.6% versus 61.0%; p = 0.02), less likely to experience symptoms and signs of a COVID-like illness, and less likely to receive COVID-19-directed therapies. * Vaccination status was ascertained from the California Immunization Registry. Fully vaccinated adults were those who were not immunocompromised and had received the second dose of a 2-dose COVID-19 vaccine series or a single dose of a 1-dose product ≥14 days before receiving a positive SARS-CoV-2 test result associated with their hospitalization. Immunocompromised adults were considered fully vaccinated if they had received a third dose of a 3-dose primary series or a single dose of a 1-dose product ≥14 days before receiving a positive SARS-CoV-2 test result associated with their hospitalization. Fully vaccinated adults were considered to have received a booster dose if they had received an additional dose (third or fourth) of an mRNA COVID-19 vaccine ≥14 days before receiving a positive SARS-CoV-2 test result associated with their hospitalization. Adults whose positive SARS-CoV-2 test date was ≥14 days after the first dose of a 2-dose series (or second dose of a 3-dose series) but <14 days after receipt of the second dose (or third dose) were considered partially vaccinated, as were those who had received only a single dose of a 2-dose product (or 1 or 2 doses of a 3-dose series). Adults with no documented receipt of any COVID-19 vaccine dose before the test date were considered unvaccinated. † Partially vaccinated adults were not included in analyses stratified by vaccination status because of small sample size. However, they were included in overall proportions and comparisons not stratified by vaccination status; thus, the total number of patients exceeds the sum of fully vaccinated and unvaccinated patients. § Vaccination status was ascertained from the California Immunization Registry. Booster status was unavailable for hospitalizations before December 1, 2021. ¶ Includes Native Hawaiian, other Pacific Islander, American Indian, and Alaska Native persons, and persons of unknown race or ethnicity. ** Includes baricitinib, casirivimab-imdevimab, convalescent plasma, sotrovimab, and tocilizumab. † † Denominator excludes 70 fully vaccinated patients who also received a booster dose. § § Denominator excludes 188 patients who remained hospitalized as of January 27, 2022. ¶ ¶ Denominator excludes 129 patients who remained hospitalized as of January 27, 2022 *** Denominator excludes 70 fully vaccinated patients who also received a booster dose and 43 patients who remained hospitalized as of January 27, 2022.

Discussion
Among adults hospitalized with SARS-CoV-2 infection at a single hospital in California during the Omicron-predominant period (December 21, 2021-January 27, 2022), COVID-19 vaccination, particularly receipt of a booster dose, was associated with lower likelihood of ICU admission, and, among adults aged ≥65 years, lower likelihood of death while hospitalized. Compared with the period of Delta predominance, a higher proportion of adults hospitalized during Omicron predominance were fully vaccinated. Consistent with earlier findings (3), Omicron-period hospitalizations were associated with a lower likelihood of ICU admission, IMV, and death while hospitalized, compared with Delta-period hospitalizations. However, the proportion requiring ICU admission and IMV did not differ significantly when stratified by vaccination status, suggesting that much of the lower disease severity observed during Omicron predominance might be driven by increased population-level vaccine-conferred immunity. These findings support the continued importance of COVID-19 vaccination, including booster doses, in mitigating the risk of severe illness associated with SARS-CoV-2 infection. From mid-July through mid-December 2021, the proportion of fully vaccinated adults in Los Angeles County increased nearly 20%, from approximately 65% to 77%,*** but the proportion of SARS-CoV-2 hospitalizations occurring in fully vaccinated adults increased almost 60%, from approximately 25% to 40%. The increase in the percentage of fully vaccinated Hispanic adults and the decrease in the percentage of non-Hispanic White adults hospitalized between the two periods likely reflect increased vaccination coverage among Hispanic persons during fall 2021. Increases in infections among vaccinated persons during the period of Omicron predominance were likely driven both by waning vaccine-derived immunity over time and by relative resistance to vaccine neutralization in the Omicron variant compared with the Delta variant (2,4). This is consistent with the observed decline in effectiveness *** https://covid19.ca.gov/vaccination-progress-data/ of 2-dose vaccination against COVID-19 hospitalization during the Omicron period (5). A previous study also found that, compared with the period of Delta predominance, the period of Omicron predominance in Los Angeles County was associated with a decrease in the degree of protection against COVID-19 and hospitalization (6). Despite this, COVID-19 vaccination, including a booster dose, was associated with lower likelihood of ICU admission during the Omicron period, and lower likelihood of death among adults aged ≥65 years, who are at higher risk for severe outcomes when hospitalized with COVID-19 (7,8).

FIGURE. Intensive care unit admission, use of invasive mechanical ventilation, and death while hospitalized among 737 adults hospitalized with SARS-CoV-2 infection during Omicron variant predominance, by age group and vaccination status* , † -one hospital, California
Early reports suggest that the Omicron variant has lower replication competence in lung parenchyma, † † †, § § § possibly contributing to a decreased severity of illness compared with earlier variants (3). However, among patients hospitalized for COVID-19 during the early Omicron predominant period, most had lower respiratory symptoms and abnormal chest imaging, approximately one third had hypoxemia, and 10% required IMV. These findings demonstrate that, despite observed changes compared with Delta, Omicron variant infection still causes severe lower respiratory illness. Similar data on † † † https://www.researchsquare.com/article/rs-1189219/v1 § § § https://www.researchsquare.com/article/rs-1211792/v1 Approximately 20% of SARS-CoV-2 admissions during early Omicron predominance were likely for reasons other than COVID-19, a proportion even higher among young and vaccinated adults. Given high rates of SARS-CoV-2 community transmission, this is not unexpected. This estimate stands in contrast to an estimated 63% of patients admitted with incidental SARS-CoV-2 infection reported from South Africa (9). While this difference might be driven, in part, by differences in demographics and population immunity, the present study's classification methodology might have overestimated the number of persons whose admission was driven by COVID-19. One third of patients classified as having been admitted for COVID-19 received no COVID-19-directed therapies. Alternatively, high population-level immunity from vaccination, previous SARS-CoV-2 infection, or both might have modulated the clinical presentation of patients with COVID-19 during Omicron predominance and atypical presentations might have been underrecognized (e.g., exacerbations of chronic medical conditions), or lesser illness severity might have resulted in fewer therapies. However, the pandemic health care burden is not limited to hospitalizations for symptomatic COVID-19. Even patients with positive SARS-CoV-2 test results admitted for non-COVID-19 conditions require isolation rooms and use of personal protective equipment and might transmit infection to health care workers, exacerbating staff shortages.
The findings in this report are subject to at least six limitations. First, sequencing data were not available to identify the SARS-CoV-2 variant. However, based on California genomic surveillance data, which is based on sequencing of ≥10% of all positive RT-PCR tests in the state, ¶ ¶ ¶ and on recent genomic surveillance for Los Angeles County (7), the Delta and Omicron variants accounted for the majority of sequenced isolates throughout their respective predominance periods. Second, the proportion of Omicron-period hospitalizations attributed to COVID-19 could not be compared with earlier periods, so it is unclear whether the proportion represented a change from an earlier period. Third, the study might have been underpowered to detect Omicron-specific reductions in illness severity after stratifying by vaccination status. Fourth, the analysis could not account for the interval since the last ¶ ¶ ¶ https://www.cdph.ca.gov/Programs/CID/DCDC/Pages/COVID-19/ COVID-Variants.aspx dose of COVID-19 vaccine, which might have been longer among Omicron-period patients. Fifth, there might have been incomplete ascertainment of deaths in the recent weeks of Omicron predominance; severely ill patients might remain hospitalized and might be at high risk of death. A longer period of observation might have reduced differences in death between the two periods. Finally, these findings are from a single hospital in Los Angeles and cannot be generalized to the United States. However, the hospital has a large catchment area in a racially and ethnically diverse region.
In this single-hospital study, adults hospitalized with SARS-CoV-2 infection during Omicron predominance had less severe illness compared with adults hospitalized during Delta predominance. Much of this effect appears to be driven by increased proportion of patients who were fully vaccinated. Approximately 20% of Omicron-period hospitalizations among adults with a positive SARS-CoV-2 RT-PCR result were driven by non-COVID-19 conditions, which might be attributed to high SARS-CoV-2 community transmission and high population vaccination coverage. COVID-19 vaccination was associated with lower likelihood of ICU admission during Omicron predominance. COVID-19 vaccination, including a booster dose for those who are fully vaccinated, is critical to minimizing the risk for severe health outcomes among adults with COVID-19.

Summary
What is already known about this topic?
The SARS-CoV-2 Omicron variant became predominant in the United States in mid-December 2021, coinciding with a rise in SARS-CoV-2-associated hospitalizations.
What is added by this report?
Among adults hospitalized with SARS-CoV-2 infection during Omicron predominance, COVID-19 vaccination, including with a booster dose, was associated with lower likelihood of intensive care unit admission. Compared with patients during the period of Delta predominance, Omicron-period patients had less severe illness, largely driven by an increased proportion who were fully vaccinated. Approximately 20% of early Omicronperiod hospitalizations were for non-COVID-19 conditions, particularly among young and vaccinated adults.
What are the implications for public health practice? COVID-19 vaccination, particularly a booster dose, continues to be critical in mitigating the health care burden of the Omicron variant.