Abstract 5945: Cancer care in rural America: factors associated with acute care utilization following chemotherapy at a community cancer center

Abstract Background Acute care use (ACU) events, including emergency department visits and hospitalizations following chemotherapy, are costly and often preventable with outpatient interventions. The Centers for Medicare & Medicaid Services track ACU within 30 days of outpatient chemotherapy through the OP-35 measure. While predictive models exist to identify at-risk patients, few have been validated on diverse populations. We aimed to quantify the differences between those who experience ACU and those who do not. Patients and Methods In a retrospective study of 5, 575 patients receiving chemotherapy between October 2021 and May 2024, we assessed ACU within 180 days of any chemotherapy visit using institutional records and claims data. Of these, 3, 230 patients (57.9%) experienced an ACU event, with a median of 75 days between chemotherapy and ACU. Among those with an ACU event, 2, 491 patients (77.1%) experienced an event within 30 days of chemotherapy, 509 (15.8%) within 90 days, and 230 (7.1%) by 180 days. Results Demographics and comorbidities differed significantly between groups. Patients with ACU events were older (66 vs. 64; p < .0001), more likely to be former smokers (38% vs. 33%; p < .0001) or alcohol users (14% vs. 9%; p < .0001), less likely to live rurally (28% vs. 31%; p = .003), more likely to use Medicare (59% vs 53%; p <.0001), and more likely to be mobility impaired (38% vs 17%; p <.0001). These patients were also more likely to have multiple primary tumors (40% vs. 23%; p < .0001) and had higher weighted Charlson Comorbidity Index scores (4.52 vs. 3.19; p < .0001). There were notable differences between groups in terms of prior hospital utilization. Patients with ACU had a greater number of previous inpatient admissions (7 vs. 4; p < .0001) and spent more days hospitalized per admission (4.7 vs. 3.8 days; p < .0001). There was also variation in laboratory markers. Patients with ACU events exhibited lower pre-chemotherapy albumin levels (3.83 vs. 4.03 g/dL; p < .0001) and higher neutrophil-to-lymphocyte ratios (3.16 vs. 2.79; p < .0001), suggestive of greater inflammation. A LASSO logistic regression model was applied to predict ACU within 30-day and 180-day windows, achieving prediction accuracies of 69% and 72%, respectively (AUC: 0.76 and 0.79). Conclusions These findings show various factors are associated with increased risk of ACU, which is essential for determining which patients may benefit most from proactive outpatient support to reduce ACU rates, drive patient satisfaction, and lower costs. This work will be applied prospectively to triage patients at highest risk. Additionally, future work will focus on distinguishing preventable from non-preventable ACU events. Citation Format: McKenna Deaton, Crystal Hattum, Sam Gaster, Samantha Olson, Sheena Lewis, Katherine Rehfeld, Christina Early, Lacey Roberts, Maren Ernst, Cynthia Kannenberg, Mary Hughes, Julia Althoff, Jamie Arens, Tobias Meissner. Cancer care in rural America: factors associated with acute care utilization following chemotherapy at a community cancer center [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5945.