Outcomes by race for cancer patients hospitalized with SARS-CoV-2 infection

Purpose: Disparities in COVID-19 outcomes have been widely reported, with disproportionate negative impacts onthe African American (AA) population The purpose of this study was to evaluate the impact of race on COVID-19outcomes for cancer patients hospitalized in a large Michigan health care system Methods: A cohort of hospitalized, laboratory-confirmed SARS-CoV-2 positive patients was identified through theHenry Ford Health System Institutional COVID prospective patient registry between March 1st-May 2020 Those with a diagnosis of cancer were identified using our institutional tumor registry and electronic health record (EHR) Patient self-reported race/ethnicity data were extracted from the system's centralized EHR, as were otherdemographic and clinical covariates Racial differences in cumulative incidence of mortality and hospital discharge were tested To further evaluate the effect of race on the mortality, Fine-Gray competing-risks model was performed with discharge alive as a competing event A P<0 05 was considered statistically significant Results: Out of the 204 COVID+ cancer patients hospitalized in our health care system, 69 6% were AA (N=142) AA patients were slightly younger than non-AA patients (70 35 v 74 58, p=0 023) No difference in mean BMI wasdetected (30 33 AA v 29 87 non-AA, p = 0 68) A smaller proportion of AA patients had active cancer (36 6% v 40 3%, p = 0 73) Outcomes were generally inferior in the AA cohort, although these differences were notstatistically significant The rate of ICU admission was 41 5% in AA and 37 1% in non-AA (p=0 659) 34 5% of AApatients required intubation compared to 25 8% of non-AA patients (p=0 288) In our model, older age was the onlyvariable that significantly increased the risk of death (standard hazard ratio SHR 1 05, p = 0 002) The risk of death was higher for AA patients (SHR 1 92, p=0 068) and males (SHR 1 62, p = 0 078) but did not meet statisticalsignificance