Socioeconomic disparities in the receipt of allogeneic stem cell transplantation among patients with relapsed Acute Myeloid Leukemia: A national inpatient sample analysis, 2016–2020

Abstract Introduction: Allogeneic stem cell transplantation (alloSCT) remains a cornerstone of curative-intent therapy for patients with relapsed acute myeloid leukemia (AML), offering the potential for durable remission and improved survival. However, access to alloSCT is not equitably distributed across patient populations. While clinical eligibility is primarily determined by disease biology and performance status, sociodemographic factors—such as race, insurance coverage, and income level—may also influence access to this life-saving treatment. In this study, we used data from a nationally representative inpatient sample to examine the impact of sociodemographic and hospital-level factors on alloSCT receipt among hospitalized patients with relapsed AML. Methods: We conducted a retrospective cross-sectional study of hospitalized patients with relapsed AML using the National Inpatient Sample (NIS) from 2016 to 2020. Cases were identified using ICD-10 diagnosis codes. Assessed variables included age, sex, race, insurance type, and socioeconomic class (based on ZIP code–linked median household income: lowest [<$50,000], low [$50,000–$64,999], high [$65,000–$85,999], and highest [≥$86,000]). Hospital-level variables included U.S. census region and comorbidity burden. Multivariable logistic regression was used to identify independent predictors of alloSCT receipt. Results: There were 35230 hospitalizations of patients with relapsed AML from 2016 to 2020. Of these, 9015 patients had received alloSCT (25.6%). The mean age of AML patients who received alloASCT was 58 ± 16 years. Slightly more were males (n=4,795, 53.2%), most were non-Hispanic Whites (n=6,110, 67.8%), on Medicare (n=3,990, 44.3%), and a slightly larger population belonged to the high (n=2260, 25.1%) and highest SES groups (n=2320, 25.7%). Geographically, most alloSCT recipients were hospitalized in the South (n=3520, 39.0%) and had a mild comorbidity burden (n=4275, 47.4%). In multivariable logistic regression analysis, various sociodemographic and hospital-level factors were independently linked to receiving alloSCT. Females had slightly higher odds of receiving transplantation compared to males (OR: 1.06, p=0.045). Being non-White was associated with lower odds of alloSCT (OR: 0.94, p < 0.001). Insurance type was a key predictor, with patients holding non-private insurance less likely to undergo alloSCT than those with private insurance (OR: 0.92, p < 0.001). Likewise, patients from lower income quartiles were less likely to have alloSCT (OR: 0.96, p=0.003), indicating income-based disparities. Age was not significantly linked to transplantation (OR: 1.001, p=0.488). Interestingly, higher comorbidity burden was modestly associated with increased odds of transplantation (OR: 1.06, p=0.008). Compared to the Northeast, patients treated in the Midwest (OR: 1.46), West (OR: 1.26), and South (OR: 1.13) were more likely to receive alloSCT (all p<0.005). Conclusion: Significant disparities exist in the receipt of alloSCT among patients with relapsed AML. Non-White race, non-private insurance, lower income, and hospitalization in the Northeast were associated with a lower likelihood of transplantation. These findings underscore the need for targeted policy reforms and institutional strategies to improve equitable access to curative therapies for socioeconomically disadvantaged populations.