When hematologic malignancies meet COVID-19 in the United States: Infections, death and disparities

Abstract

Scientific data is limited on the risks, adverse outcomes and racial disparities for COVID-19 illness in individuals with hematologic malignancies in the United States. To fill this void, we screened and analyzed a nation-wide database of patient electronic health records (EHRs) of 73 million patients in the US (up to September 1st) for COVID-19 and eight major types of hematologic malignancies. Patients with hematologic malignancies had increased odds of COVID-19 infection compared with patients without hematologic malignancies for both all-time diagnosis (malignancy diagnosed in the past year or prior) (adjusted Odds ratio or AOR: 2.27 [2.17-2.36], p < 0.001) and recent diagnosis (malignancy diagnosed in the past year) (AOR:11.91 [11.31-12.53], p < 0.001), with strongest effect for recently diagnosed acute lymphoid leukemia (AOR: 31.03 [25.87-37.27], p < 0.001), essential thrombocythemia (AOR: 20.65 [19.10-22.32], p < 0.001), acute myeloid leukemia (AOR: 18.94 [15.79-22.73], p < 0.001), multiple myeloma (AOR: 14.21 [12.72-15.89], p < 0.001). Among patients with hematologic malignancies, African Americans had higher odds of COVID-19 infection than Caucasians with largest racial disparity for multiple myeloma (AOR: 4.23 [3.21-5.56], p < 0.001). Patients with recently diagnosed hematologic malignancies had worse outcomes (hospitalization: 51.9%, death: 14.8%) than COVID-19 patients without hematologic malignancies (hospitalization: 23.5%, death: 5.1%) (p < 0.001) and hematologic malignancy patients without COVID-19 (hospitalization: 15.0%, death: 4.1%) (p < 0.001).

Keywords: COVID-19; Disparity; Health outcomes; Hematologic malignancies; Risk.

Fig. 1

Odds of COVID-19 infection in patients with recent versus all-time diagnosis of hematologic malignancies, adjusted for age, gender, race and potential COVID-19 risk factors (cardiovascular diseases, type 2 diabetes, obesity, chronic kidney diseases, chronic obstructive pulmonary disease (COPD), asthma, substance use disorders, cancer therapy (chemotherapy, radiotherapy, immunotherapy), transplant procedure (bone marrow transplant, solid organ transplant) and nursing home stay status.

Fig. 2

Effects of demographics on odds of COVID-19 infection among patients with recently diagnosed hematologic malignancies, adjusted for other demographics and potential COVID-19 risk factors (cardiovascular diseases, type 2 diabetes, obesity, chronic kidney diseases, chronic obstructive pulmonary disease (COPD), asthma, substance use disorders, cancer therapy (chemotherapy, radiotherapy, immunotherapy), transplant procedure (bone marrow transplant, solid organ transplant) and nursing home stay status.

Fig. 3

Hospitalization and death rates among three populations: patients with both recent diagnosis of hematologic malignancies and COVID-19, COVID-19 patients without hematologic malignancies, hematologic malignancy patients without COVID-19. ***: p < 0.0001; **: p < 0.001; ns: not statistically significant. The SNOMED-CT concepts “Hospital admission (procedure)” (ID 32485007) was used to obtain hospitalization status from patient EHRs. Explorys regularly imports from the Social Security Death index for the “deceased” status.