Insurance-related disparities in emergency department presentation among hospitalized colorectal cancer patients: A national analysis.

Abstract

e15682 Background: Insurance status may influence how and when patients with colorectal cancer (CRC) access care, leading to emergency department (ED)–based hospitalizations rather than planned admissions. We evaluated insurance-related disparities in ED-based admission and in-hospital mortality among CRC patients. Methods: Using the National Inpatient Sample (2018–2020), we identified adults hospitalized with CRC (ICD-10-CM codes C18–C20), excluding inter-hospital transfers and history-only codes. Insurance was categorized as private (reference), Medicare, Medicaid, and other (self-pay/no charge/other). ED-based admission was defined as hospital admission originating in the ED rather than direct or elective admission. Survey-weighted multivariable logistic regression examined associations between insurance type and (1) ED-based admission and (2) in-hospital mortality, adjusting for demographics, income quartile, comorbidities, and hospital characteristics. We assessed whether accounting for ED admission patterns reduced insurance-related mortality differences. Results: Among 421,350 CRC hospitalizations, ED admission rates varied significantly by insurance: private 25%, Medicare 33%, Medicaid 35%, and other 40% (p < 0.0001). After adjustment, odds of ED-based admission were substantially elevated for the other (self-pay/no charge/other) group (aOR 1.96, 95% CI 1.81–2.12), Medicaid (aOR 1.70, 1.60–1.81), and Medicare (aOR 1.37, 1.30–1.44) versus private insurance. In-hospital mortality was higher for the other insurance group (aOR 1.97, 95% CI 1.59–2.44), while Medicaid mortality was similar to private and Medicare showed slightly lower adjusted mortality. Accounting for ED admission patterns modestly reduced excess mortality for the other group (from aOR 1.97 to 1.81, ~17% attenuation). Notably, ED-based admission itself was associated with lower in-hospital mortality (aOR 0.51, 95% CI 0.46–0.57), likely reflecting case-mix differences in direct admissions. Conclusions: Substantial insurance-related disparities exist in how CRC patients enter the hospital, with ED-based admission markedly more common among publicly insured and uninsured patients. However, admission patterns explain only a small portion of mortality disparities, and ED admissions were paradoxically associated with lower mortality. These findings suggest that barriers to timely pre-diagnostic care contribute to emergency presentations, while survival disparities likely reflect upstream factors (stage at presentation) and downstream factors (treatment intensity and care coordination). Expanding insurance coverage, strengthening navigation for abnormal screening or diagnostic findings, and improving urgent outpatient evaluation pathways may reduce preventable emergency presentations and advance equity in CRC outcomes.