Early epidemiological studies of U.S. patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have suggested a possible association between ABO/Rh blood group phenotype and both susceptibility and severity of COVID-19 infection. Given the remarkable heterogeneity of the host inflammatory response in this viral syndrome and the widespread expression of ABO/Rh antigens on vascular and alveolar endothelial cells, it is biologically plausible that certain blood group phenotypes, with their unique repertoire of anti-A and/or anti-B antibodies, may differentially augment the host-pathogen response. We conducted a retrospective review of patients hospitalized for COVID-19 within our regional healthcare network in San Diego County to identify an association between ABO/Rh blood group type and the severity of infection.


All patients hospitalized at one of five Scripps Health hospitals in San Diego County from March 1, 2020 to July 30, 2020 with a PCR confirmed diagnosis of COVID-19 and blood type on record were included in the initial analysis (n = 316). Demographic, laboratory and clinical data were extracted from the electronic medical record and included age, ethnicity, BMI, sex, medications, co-morbidities and admission white blood cell and lymphocyte count, hemoglobin, platelets, ESR, CRP and D-dimer. Outcomes of interest included length of hospitalization, intensive care unit (ICU) admission, mechanical ventilation need, and mortality. Significant associations between each parameter of interest and blood group type were determined using either linear or logistic regression analysis. To address potential confounding variables, an adjusted multivariate model accounting for potential significant (p< 0.1) predictors of each outcome on univariate analysis, in addition to blood type groups, was conducted to further refine any associations. The study was approved by the Scripps Health Institutional Review Board.


316 patients met inclusion criteria for analysis. Hospitalized COVID patients were predominantly male, obese (BMI 30.6) and were an average age of 63 years. Almost 70% of patients hospitalized were Hispanic. 57.0% of patients were blood type O, 30.4% were type A, 3.8% were type B and 8.9% were type AB. 7% were Rh negative. Median length of hospital stay was 16.5±14.7 days, 59% were admitted to the ICU, 37% were intubated, and 27% died. Further relevant laboratory values on admission, co-morbidities, and medications administered during hospitalization are summarized in Table 1. Blood type, with or without adjusting for other clinical variables, was not predictive of length of hospital stay, ICU admission, or intubation during the hospitalization. Type B blood alone was associated with decreased odds of death (OR: 0.27, 95% CI: 0.06-0.85, p<0.05), though this effect was not seen after adjusting for significant confounding variables (OR: 0.39, 95% CI: 0.08-1.43, p>0.18).


In this large, multi-hospital, retrospective analysis of patients hospitalized for COVID-19 in San Diego County, there was a low relative percentage of Rh negative blood type and type B blood compared to historical population averages. Blood type was not determined to be independently associated with hospital length of stay, mechanical ventilation, ICU admission or death. ABO/Rh blood typing appears to have a limited prognostic role in COVID-19 severity of hospitalized patients, though further analysis of the protective effects of type B and/or Rh negative blood type may be warranted in a larger sample.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.