Process mapping of ePRO-enabled remote symptom monitoring across academic and community practices.

Abstract

510 Background: Electronic patient-reported outcome (ePRO)-enabled remote symptom monitoring (RSM) improves outcomes during cancer treatment, but sustained implementation requires coordination across technologies, workflows, and clinical roles. As RSM expands nationally, oncology practices must translate a common intervention into locally feasible processes. Existing implementation tools provide general guidance for patient-reported data systems, but detailed approaches for comparing the clinical operation of RSM across diverse real-world practices are lacking. Methods: We conducted a multi-site qualitative and process mapping study across 12 sites participating in OncoPRO, a national consortium of practices implementing RSM facilitated by ASCO. Data included ethnographic field notes, site-level workflow documents, process maps, implementation tracking data, and semi-structured interviews with 78 healthcare professionals across 12 practices. Process maps anchored interviews to elicit descriptions of who does what, when, how, and why across four subprocesses: screening, enrollment, survey monitoring, and intervention. Applying a cultural models lens, we used process map-anchored coding and constant comparison to identify shared workflow components and cross-site variation. Results: Five components structured RSM workflows across practices: ownership, mechanism, timing, auditing, and challenges. These components were present across all four subprocesses but varied substantially by practice. Screening ranged from manual clinic-list review and staff judgment to automated treatment-plan triggers. Enrollment ranged from staff-led education and manual invitation to automated portal-based activation Variation existed in how practices supported patients who did not enroll. Survey monitoring differed by timing logic, delivery modality, reminder functionality, and whether nonresponse was audited or acted upon. Intervention workflows varied by alert threshold, routing to individual clinicians versus pools or workboards, documentation location, and whether alert closure was formally tracked. Across subprocesses, practices balanced automation with human reinforcement, centralized versus distributed ownership, and standardized build functionality versus local adaptation. Conclusions: This study provides a practical framework for comparing RSM implementation across diverse oncology settings. Identifying these components clarifies where local workflows diverge, where coordination breaks down, and where implementation support or technology optimization may be needed. This approach can guide role-specific training, support local adaptation, inform process measurement, and generate cross-site recommendations for implementing and sustaining ePRO-enabled RSM in routine cancer care.