Community Oncologists have a Favorable View of Algorithm-Based Default Palliative Care Referrals with Opt-Out

Outcomes 1. Understand and describe the role of qualitative analysis for evaluating a pragmatic randomized controlled trial implementing a risk-algorithm based default palliative care referrals with opt-out nudge for advanced lung and non-colorectal GI cancer patients.2. Understand and describe the results of our qualitative study exploring community oncology clinician perspectives about this novel intervention and evaluate how it might apply to their own practice setting specifically surrounding the goal of increasing palliative care access to advanced cancer patients. Key Message In a qualitative study of community-based oncologists, participants found risk algorithm-based default palliative care referrals within the electronic health record to be a helpful reminder and non-intrusive to their workflow. Contrary to existing findings on oncologist perspectives, this study shows a favorable view of the process of an opt-out nudge and goal of increasing earlier palliative care access. Introduction/Context This qualitative study is part of a pragmatic randomized control trial evaluating risk algorithm-based default referral nudges to palliative care (PC) in the oncologists’ electronic medical record (EMR) task flow across a large community oncology practice spanning multiple sites. This qualitative study is the first of its kind in the community oncology setting. Methods We conducted semi-structured interviews with oncology providers, including with physicians and nurse practitioners, in a large community oncology setting in the South. Purposive sampling was used to select clinicians involved in our intervention group. We asked providers for their perspectives on the risk algorithm, the opt out nudge for PC referrals, impact on their clinical workflow, facilitators and barriers to scaling this intervention, and overall perspectives on PC. Interviews lasted approximately 30 minutes and were conducted over Zoom, transcribed, and analyzed using Dedoose software. Results Of the twelve providers interviewed, the majority were physicians (11). Risk algorithm criteria were considered appropriate, with some confusion around age guidelines and suggestions to add more symptoms, screening for social support, and comorbidities. The nudge was seen as seamless and beneficial. The EMR nudge alert was viewed as a helpful reminder to busy clinicians, with the option to opt out on individual patients. Other facilitators included access to an internal outpatient PC program seen as a welcomed service, and a nurse coordinator as a resource to introduce PC to patients. Barriers included perceived staffing limitations; concerns about patient inconvenience around cost, transportation, and time; and individual patient factors, such as emotional fragility and appropriateness for PC due to low symptom burden or stable disease status. Conclusion This intervention marks a novel way to improve access to PC. Contrary to existing findings on oncologist perspectives, this study shows a general acceptance of this process, which has minimal impact on workflow. Keywords Scientific Research / Models of Palliative Care Delivery