3653 - Emergency Department Visits and Hospital Admissions among Patients Undergoing Radiation Therapy: A Large-Scale Informatics Analysis by Disease Site

Purpose/Objective(s): Unplanned emergency department (ED) visits and hospital admissions during radiation therapy can significantly impact patient outcomes and healthcare resource utilization. This study utilizes a large-scale informatics-driven analysis to quantify ED visit and hospitalization rates across patients receiving radiation to various disease sites with an aim to develop future predictive models to stratify patient risk.

Materials/Methods: A retrospective analysis was conducted using data from 26,647 patients who underwent CT simulation for radiation therapy at a single institution between 2013 and 2024. Data on treatment sites, ED visits, and hospital admissions were extracted and analyzed. ED and hospital admissions were looked at within 5 days of the final documented treatment to include admissions only during or shortly after radiation treatments.

Results: CT simulations increased over time, with a notable rise from 1,440 in 2013 to 3,275 in 2024. The most common treatment sites included Brain (3,928), Genitourinary (3,383), and Breast (3,151), with lower representation in Gynecologic (661), Gastrointestinal (931), and Extremity (944). ED visits were highest in patients receiving radiation to the Spine (38.7%), Extremity (30.3%), Brain (26.3%), Thoracic (25.0%), Head & Neck (22.8%), GI (21.2%), and Pelvis (20.6%). Conversely, Genitourinary (4.3%), Breast (5.0%), and Gynecological (9.5%) sites had the lowest ED visit rates. Similarly, hospital admissions were most common in patients treated for Spine (36.4%), Brain (29.1%), Extremity (29.0%), Thoracic (24.1%), Head & Neck (21.7%), Gastrointestinal (20.5%), and Pelvis (20.2%) conditions, whereas the lowest hospitalization rates were observed in Genitourinary (3.1%), Breast (3.7%), and Gynecological (8.8%) patients. See Table 1.

Conclusion: Patients undergoing radiation therapy for Spine, Extremity, and Head & Neck cancers demonstrate the highest rates of unplanned ED visits and hospitalizations, suggesting a need for targeted supportive care interventions. Predictive modeling can be used to further refine risk stratification and guide clinical decision-making to mitigate these risks. Future research should focus on integrating these predictive tools into routine clinical practice to enhance patient outcomes and optimize healthcare utilization.