Existing evidence suggests a link between ABO blood type and severe outcomes in coronavirus disease 2019 (COVID-19). We aimed to assess the relationship between blood type and severe outcomes across variant strains throughout the pandemic. This was a multicenter retrospective observational cohort analysis from a large health system in southeastern Michigan using electronic medical records to evaluate emergency encounters, hospitalization, and severe outcomes in COVID-19 based on ABO blood type. Consecutive adult patients presenting to the emergency department with a primary diagnosis of COVID-19 (U07.1) from March 1, 2020 through December 31, 2022 were assessed. Patients who presented during three distinct time intervals that coincided with Alpha, Delta, and Omicron variant predominance were included in the analysis. Exclusions included no record of ABO blood type, positive PCR COVID-19 test within the preceding 28 days, and if transferred from out of the health system. Severe outcomes were inclusive of intensive care unit admission, mechanical ventilation, or death, which, as a composite, represented our primary outcome. Secondary outcomes were hospital admission and length of stay. A logistic regression model was employed to test the association between ABO blood type and severe outcome, adjusting for age, sex, race, vaccination status, Elixhauser comorbidity indices, and the dominant variant time period in which the encounter occurred. Of the 33,796 COVID-19 encounters, 9416 met inclusion criteria; 4071 (43.2%) were type O, 3417 (36.3%) were type A, 459 (4.9%) were type AB, and 1469 (15.6%) were type B blood. Note that 66.4% of the cohort was female (p=0.18). The proportion of composite severe disease among the four blood types was similar and ranged between 8.6% and 8.9% (p=0.98). Note that 53.0% of type A blood patients required hospital admission, compared to 51.9%, 50.4%, and 48.1% of type AB, B, and O blood, respectively (p<0.001). Compared to patients with O blood type (43.2%), non-O blood type (58.8%; composite of A, AB, and B) exhibited no statistically significant difference in the proportion of composite severe disease (8.8%vs. 8.7%; p=0.81) Multivariable regression analyses exhibited no significant difference regarding the presence of severe outcomes among the four blood types or O versus non-O blood types during T1, T2, and T3. ABO blood type was not associated with COVID-19 severe outcomes across the Delta, Alpha, and Omicron dominant COVID waves across a large health system in southeastern Michigan. Further research is needed to better understand if ABO blood type is a risk factor for severe disease among evolving COVID-19 variants and other viral upper respiratory infections.