Assessing the impact of distance from primary hemophilia treatment center on annualized bleeding rates in people with inheritable bleeding disorders

Abstract Introduction: Access to federally-supported hemophilia treatment centers (HTC) is critically important in people with inheritable bleeding disorders (PwBD). Distance to an HTC has been reported as a barrier to access to acute care for bleeding related complications. The impact of distance on long-term outcomes in PwBD is unknown. We aimed to assess difference in the annualized bleeding rates (ABR) in PwBD living within 50 miles, 50-100 miles and greater than 100 miles from their primary HTC. Methods: This retrospective cohort study was conducted using the American Thrombosis and Hemostasis dataset (ATHNdataset), a HIPAA-compliant, de-identified health dataset containing data from individuals with bleeding and clotting disorders receiving care in the United States at ATHN-affiliated centers. To avoid distance inaccuracy from three-digit zip codes centroids, we geocoded addresses to calculate driving distances. To avoid identification by geocoding, latitude-longitude coordinates of residential addresses and primary HTC location were used and managed by ATHN data scientists (not available to the investigators). We included people of all genders with Hemophilia A (HA), Hemophilia B (HB) and von Willebrand disease (VWD) with at least one recorded visit or bleeding event from January 1, 2021, to December 31, 2024. Baseline characteristics included age, sex at birth, race, primary diagnosis, and driving distance from an individual's residential address to their primary HTC. Bleeding events within the specified time were used to calculate annualized bleeding rate (ABR). Results: The ATHNdataset included a total of 43,649 individuals out of which 21,943 met our inclusion criteria and had at least one recorded visit or bleeding event; 10,206 (46.5%) with HA, 3294 (15%) with HB, and 8443 (38.5%) with VWD, of which 6855 (31.3%) had Type 1, 1121 (5.1%) had Type 2, 221 (1%) had Type 3 VWD, and 246 (1.1%) had an unknown subvariant. Females comprised 7,161 (32.6%) of the cohort. Of the included population, 14,343 (65.3%) lived 0-50 miles to their primary HTC. The remaining 4,145 (19%) lived within 50-100 miles and 3,455 (15.7%) lived >100 miles from their primary HTC. For PwBD living within 50 miles of their primary HTC, the mean (SD) ABR was 2.3 (27.6), for those living within 50-100 miles and >100 miles mean ABR was 1.9 (24.3) and 5.9 (50.9) respectively (p<0.001). We performed pairwise comparisons of mean ABR between individuals living 0-50 miles, 50-100 miles, and >100 miles from their primary HTC. The mean difference in ABR was 3.6 (95% CI 2.2, 5.0) and 3.9 (95% CI 2.2, 5.6) respectively (p <0.001). We further stratified the results based on diagnosis and severity of hemophilia. For males with severe HA, the mean ABR was 6.0 (43.7) for those living 0-50 miles, 0.5 (1.0) for those living 50-100 miles, and 14.3 (71.5) for those living >100 miles from their primary HTC. For males with severe HB, mean ABR was 0.9 (1.6) for those living 0-50 miles, 0.3 (0.5) for those living 50-100 miles, and 0.2 (0.4) for those living >100 miles. For people with VWD, mean ABR was 0.9 (15.1) for those living 0-50 miles, 1.3 (24.1) between 50-100 miles, and 3.3 (32.9) if living >100 miles. We stratified VWD by more severe bleeding phenotype (types 3, 2N, and 2B) and calculated ABR. In this cohort, the ABR for those living within 50 miles was 4.0 (35.4), 0.5 (1.3) for those living 50-100 miles and 7.6 (51.1) if living >100 miles from their HTC (p=0.09). Conclusion: Among people with inheritable bleeding disorders, our study is the first to demonstrate a statistically significant and clinically meaningful increase in annualized bleeding rate associated with greater driving distance from an individual's primary HTC, underscoring the impact of geographic access on bleeding outcomes. Future work includes the investigation whether driving distance to an individual's primary HTC impacts other long term morbidities such as development of target joints and assessing markers of bleeding phenotype such as age of diagnosis, use of bleeding prophylaxis, and comparing care provided to women, girls, and people with the potential to menstruate to men with the same diagnosis and clinical severity.Acknowledgments: This project is supported by the DREAM Award from the ATHN/HTRS awarded to the primary author.