Background Impact Statement: This first nationally representative analysis focused exclusively on newly diagnosed AML demonstrates modifiable racial and ethnic disparities in critical complications and in-hospital outcomes during initial hospitalization, with variation by age and hospital context.

Racial and ethnic disparities in AML survival are well documented, but inequities in the early hospitalization period—when complications can strongly influence treatment trajectories—remain poorly understood. Identifying where disparities are most pronounced can inform targeted, evidence-based supportive care and quality improvement initiatives.

Methods We conducted a retrospective cohort study using the National Inpatient Sample (2018–2020). Adults with newly diagnosed AML were identified by primary ICD-10-CM codes C92.00, C92.50, C92.A0, and C92.60. We excluded remission, relapse, acute promyelocytic leukemia, and interhospital transfers. Analyses included all newly diagnosed AML hospitalizations regardless of treatment status; chemotherapy regimen intensity could not be reliably determined from NIS data. Outcomes included in-hospital mortality, sepsis, acute kidney injury (AKI), mechanical ventilation (MV), tumor lysis syndrome (TLS), blood product transfusion, length of stay (LOS), and total hospital costs. Multivariable logistic and linear regression adjusted for demographics, comorbidities, and hospital characteristics. Analyses were stratified by age (<60 vs. ≥60 years) and assessed for hospital-level effect modification. No formal multiple-comparisons adjustment was applied given the exploratory nature; statistical power for interaction analyses was estimated at 0.89.

Results We identified 74,595 weighted hospitalizations for newly diagnosed AML: White 71.5%, Black 10.0%, Hispanic 9.1%, Asian 4.3%, Other race 4.1%.

Compared with White patients:

  • Black: Higher AKI (OR 1.55; 95% CI 1.35–1.77; p<0.001), TLS (OR 1.35; 95% CI 1.08–1.70; p=0.010), transfusion (OR 1.24; 95% CI 1.08–1.41; p=0.002), longer LOS (+1.36 days; p=0.0011), and $17,675 higher hospital costs (p=0.020).

  • Hispanic: Lower AKI (OR 0.82; 95% CI 0.69–0.97; p=0.019), TLS (OR 0.73; 95% CI 0.54–0.98; p=0.035), higher transfusion (OR 1.21; 95% CI 1.04–1.40; p=0.015), longer LOS (+1.10 days; p=0.012), and $28,707 higher costs (p<0.001).

  • Asian: Higher sepsis (OR 1.36; 95% CI 1.06–1.74; p=0.016), transfusion (OR 1.29; 95% CI 1.06–1.56; p=0.011), and $22,368 higher costs (p=0.044).

  • Other race: Higher mortality (OR 1.34; 95% CI 1.02–1.76; p=0.035), sepsis (OR 1.45; 95% CI 1.15–1.83; p=0.002), MV (OR 1.76; 95% CI 1.21–2.57; p=0.003), longer LOS (+2.72 days; p<0.001), and $54,753 higher costs (p<0.001).

Hospital-level effect modification:

  • Disparities were amplified in certain settings. In urban non-teaching hospitals, Hispanic patients had higher mortality (OR 2.38; p=0.005) and MV use (OR 3.68; p=0.020); Black patients had higher MV use (OR 4.17; p=0.002).

  • In large-bed hospitals, Hispanic patients had lower sepsis risk (OR 0.55; p=0.025).

Age-stratified:

  • <60 years: Black patients had higher AKI (OR 1.97; p<0.001) and MV (OR 1.74; p=0.010).

  • ≥60 years: Other race had higher mortality (OR 1.40; p=0.030) and MV (OR 2.19; p<0.001).

Conclusions In newly diagnosed AML, substantial racial and ethnic disparities affect survival-critical complications such as AKI, sepsis, MV, and TLS. Patterns vary by age and hospital setting: Black patients had consistently higher AKI risk, while Other race older adults experienced excess mortality and MV use. Disparities were most pronounced in specific hospital contexts, revealing clear targets for intervention.

Key opportunities include early AKI prevention for Black patients, targeted quality improvement in urban non-teaching hospitals, and supportive care strategies tailored to sepsis and MV use patterns. Focusing on newly diagnosed AML uncovers disparities that may be masked in pooled AML analyses, guiding actionable quality improvement in the most modifiable phase of care.

This content is only available as a PDF.
2025
Sign in via your Institution