Impact of social vulnerability on end-of-life care in hematologic malignancies.

Abstract

12036 Background: Patients with hematologic malignancies frequently receive intensive inpatient care near theend of life. While disparities in end-of-life (EOL) care are well described in oncology, how cumulative social vulnerability shapes treatment intensity and palliative care involvementamong hospitalized patients with hematologic cancers remains unclear. We examined the association between social vulnerability burden and inpatient EOL care patterns using a nationally representative database. Methods: We performed a retrospective cohort study using the 2021–2022 National Inpatient Sample of adult hospitalizations with hematologic malignancies. A composite hematologic social vulnerability index (cSVI-Heme) was constructed using validated proxies available in NIS: minoritized race/ethnicity, Medicaid or uninsured status, and lowest ZIP-code income quartile. Hospitalizations were classified as low (0), moderate (1), or high (≥2) vulnerability.The primary outcome was high-intensity EOL care (invasive mechanical ventilation, cardiopulmonary resuscitation, or acute dialysis). Secondary outcomes included palliative care utilization and in-hospital mortality. Associations were assessed using multivariable logistic regression and inverse probability of treatment weighting, adjusting for demographics, comorbidity burden, acute organ failure, sepsis, malignancy subtype, and hospital characteristics. Results: The cohort included 202,802 hospitalizations, of which 54.2% were low, 31.0% moderate, and 14.8% high vulnerability. Overall, 5.9% involved high-intensity EOL care, increasing stepwise across vulnerability tiers (5.1% vs 6.5% vs 7.9%; p < 0.001). After adjustment, compared with low vulnerability hospitalizations, moderate vulnerability was associated with higher odds of high-intensity EOL care (aOR 1.25, 95% CI 1.19–1.30), and high vulnerability with substantially higher odds (aOR 1.50, 95% CI 1.42–1.58), demonstrating a clear dose–response relationship. Findings were confirmed in propensity-weighted analyses.This gradient was driven primarily by acute dialysis (3.1% vs 4.6% vs 6.2%; p < 0.001), while rates of mechanical ventilation and CPR were similar. High vulnerability was independently associated with higher in-hospital mortality (aOR 1.20, 95% CI 1.12–1.28), whereaspalliative care utilization did not increase after adjustment. Conclusions: Among hospitalized patients with hematologic malignancies, greater social vulnerability is associated with higher-intensity end-of-life care and increased mortality, with a clear dose–response relationship. These differences are not accompanied by greater palliative care involvement, highlighting a mismatch between treatment intensity and supportive care. Integrating social vulnerability into inpatient hematologic oncology care may support earlier goals-of-care discussions and more equitable EOL care.