Using Stakeholder Engagement to Overcome Barriers to Implementing Patient-reported Outcomes (PROs) in Cancer Care Delivery Approaches From 3 Prospective Studies

Introduction: Patient-reported outcome (PRO) measures used during cancer care delivery improve communication about symptoms between patients and clinicians and reduce service utilization for uncontrolled symptoms. However, uptake of PROs in routine cancer care has been slow. In this paper, we describe stakeholder engagement activities used to overcome barriers to implementing PROs. Implementation occurred in 2 study settings: PROs completed in the waiting room and reviewed during clinical visits to guide symptom management for multiple myeloma (visit-based PROs); and weekly PROs completed by cancer patients between chemotherapy visits to monitor symptoms at home (remote PROs). Methods: PRO implementation steps across studies included: (1) clinician and patient input on key symptoms, PRO measures, and identifying which PRO responses are clinically concerning to better target nursing actions; (2) developing PRO-based clinical decision support (CDS) for responding to concerning PROs; (3) training clinicians and clinical research assistants to interpret PROs and use software; and (4) describing implementation impact (frequency of concerning PRO responses and nursing actions). Discussion: Clinician and patient input was critical for identifying key symptoms, PRO measures, and clinically concerning response options. For the visit-based PRO observational study, all symptom scores appeared on a clinician dashboard, and those rated >= 1 by patients (on a 0-4 or 0-10 scale) had PRO-based CDS available for access. For the 2 remote PROs trials, stakeholders recommended that the 2 worst response options (eg, PRO responses of often/always or severe/very severe) would trigger an automated email alert to a nurse along with PRO-based CDS. In each study, PRO-based CDS was tailored based on clinician input. Across studies, the most common nursing response to concerning PROs was counseling patients on (or providing care plans for) self-management of symptoms. In the trials, the percentage of weekly remote PROs generating an alert to a nurse ranged from 13% at an academic center to 36% in community oncology practices.