Abstract
OBJECTIVE: In the United States, the incidence of hospital-associated venous thromboembolism (VTE) has been found to be higher among Blacks compared with Whites. However, the reason for this difference is not well-understood. Here, we utilize a large, nationwide cohort to evaluate the clinical factors that may contribute to the difference in rates of VTE in Black vs White patients. METHODS: We analyzed consecutive first hospital admissions of Black and White patients to all Veterans Affairs facilities from January 2016 to December 2021. We first compared VTE rates between White and Black patients within 90 days of hospital admission, identified 11 potential confounders for the relationship between race and VTE, and used these to develop a series of adjusted logistic regression models predicting the development of VTE. Odds ratios (ORs) obtained from these models were compared to examine the relationship between race, potential confounding variables, and VTE. RESULTS: The study cohort consisted of 142,456 patients (mean age, 54 years; 84.7% male), of whom 28.5% (n = 40,632) were Black and 71.5% (n = 101,824) were White. Of the cohort, 2.4% developed a VTE within 90 days (n = 3455), and the rate of VTE was higher in Black patients (2.7%) compared with White patients (2.3%; P < .001). A total of 12 models were developed to examine individual confounders of the relationship between race and VTE. In all models, Black race was associated with increased odds of VTE (OR, 1.06-1.17). The strength of this association remained unchanged in the range of OR 1.13 to 1.16 after adjusting for major covariates including social deprivation index, blood type, and surgery type. When adjusting for hemoglobin, the OR for the association between Black race and VTE was reduced. An inverse correlation was observed between Black race and hemoglobin, with Pearson and Spearman correlation coefficients of -0.17 and -0.180, respectively. CONCLUSIONS: In this cohort of hospitalized patients, Black race is a major risk factor for VTE, independent of social deprivation, blood type, and surgery. This effect may be modulated by hemoglobin, suggesting a potential clinical etiology underlying the trends observed between VTE and race.