Rates of emergency department visits and acute-care hospital admissions after starting chemotherapy or having surgery across six community-based cancer centers

6621 Background: Cancer patients who start chemotherapy or have surgery are at risk for serious adverse events, which can include emergency department visits (EDV) and acute-care hospital admissions (ADM). These events are relevant to patients, clinicians, health systems, and policy makers. Few studies have described the frequency and predictors of EDVs and ADMs in the routine care setting. We sought to address this knowledge gap, and to describe the impact of the COVID pandemic on these events, using data from the SIMPRO consortium of six US-based health systems. Methods: In 2019, SIMPRO started a phased deployment of an electronic health record-integrated symptom management program (eSyM) as part of the routine care provided to patients with a suspected or confirmed gastrointestinal, thoracic, and gynecologic malignancy. eSyM prompts patients to complete symptom questionnaires and provides symptom management resources for 180 days after starting chemotherapy and 60 days following surgery. This analysis includes all patients, whether they used eSyM resources or not. Outcomes, which were measured using EHR-based encounter data, included the rates of EDVs and ADMs at 30 and 90-days after starting chemotherapy or following discharge from surgery. To account for potential confounding, we used multivariable logistic regression for the 30-day outcomes and Cox regression for the 90-day outcomes. Results: From September 2019-December 2022, 8664 patients started chemotherapy and 11,578 had surgery, with 3075 and 2587 patients, respectively, experiencing at least one outcome event (Table). 90-day outcomes varied significantly across health systems for chemotherapy (EDV 2-21%, ADM 5-28%) and surgery (EDV 7-14%, ADM 9-18%) patients. In multivariable analyses, patients who reported an employment status of disabled were more likely to report an EDV (OR 1.73-1.96) and an ADM (ORs 1.52-1.96). Widowed status was associated with greater odds of having an ADM (ORs 1.29-1.30), but not an EDV. Cox regression found chemotherapy recipients experienced lower hazards of an EDV, but not an ADM, during 2020 (coinciding with the start of the COVID pandemic). For surgery patients, there was no significant association between calendar year and EDVs or ADMs. [p<0.001 for all reported values]. Conclusions: A meaningful proportion of patients who started chemotherapy or had surgery for a suspected or confirmed malignancy in the routine-care setting had an EDV or ADM within 90 days. Most events occurred within 30 days. Further work is required to ascertain how many of these encounters are avoidable and what structural or procedural interventions could improve outcomes. Clinical trial information: NCT03850912 . [Table: see text]