2342 - Site-Specific Patterns of Acute Care Utilization during Radiation Therapy: A Large-Scale Informatics Analysis Identifying Differential Prevention Opportunities
Presenter(s)
M. J. Lane, L. Pandey, K. Chang, A. J. Chang, and R. R. Savjani; Department of Radiation Oncology, University of California, Los Angeles, Los Angeles, CA
Purpose/Objective(s):
Unplanned emergency department (ED) visits and hospitalizations during radiation therapy disrupt treatment, compromise outcomes, and increase costs. While disease sites at highest risk for acute care utilization have been identified, the specific clinical drivers for these encounters remain poorly characterized. This study uses large-scale informatics to identify reasons for acute care encounters and characterize site-specific patterns, revealing differential opportunities for targeted intervention.
Materials/Methods:
A retrospective analysis was conducted using data from 24,178 patients who underwent 39,245 CT simulations for radiation therapy at a single institution between January 2013 and February 2026. Disease sites were defined by primary diagnosis associated with the simulation order. Acute care encounters (ED visits, hospitalizations) were captured within 90 days of simulation. Encounter-level diagnoses were categorized with fractional weighting for multifactorial visits.
Results:
A total of 8,183 acute care encounter diagnoses were identified. The most frequent encounter reasons were gastrointestinal (14.9%), respiratory (12.7%), neurologic (10.4%), pain (10.0%), and infection/fever (9.9%). Complaint categories revealed distinct acuity gradients: respiratory symptoms comprised 28.4% of ICU admissions versus 8.0% of ED-only visits while pain showed the inverse (3.2% ICU versus 11.9% ED-only). Neurologic complaints escalated with acuity (16.8% ICU versus 10.6% ED-only), whereas gastrointestinal symptoms predominated in lower-acuity encounters (7.4% ICU versus 15.2% ED-only).
Site-specific patterns emerged: head and neck, gynecologic, gastrointestinal, and bony malignancies predominantly presented with concerns amenable to initial outpatient management (gastrointestinal symptoms, pain), while CNS and thoracic malignancies presented with concerns likely less amenable to prevention (neurologic and respiratory symptoms).
Conclusion:
Acute care utilization varies by treatment site with acuity gradients suggesting differential preventability. Lower-acuity complaints (pain, gastrointestinal symptoms such as nausea or diarrhea) may represent missed prevention opportunities; higher-acuity complaints (respiratory, neurologic) may reflect disease progression. Disease sites such as head and neck, gynecologic, gastrointestinal, and bony malignancies may benefit from proactive outpatient symptom management targeting pain control, hydration, and nutrition. CNS and thoracic malignancies may require enhanced surveillance and rapid-access pathways rather than prevention-focused interventions. These findings provide a framework for site-specific quality improvement to reduce avoidable acute care utilization, minimize treatment interruptions, and optimize resource allocation in radiation oncology.