Spreading routine serious illness conversations at a single cancer center using a multidisciplinary and patient-family advisory approach.

367 Background: ASCO Guidelines recommend oncologists conduct serious illness conservations (SIC) for all patients with advanced cancer. We describe the spreading of a multidisciplinary and patient-family advisor (PFA) quality improvement (QI) initiative to conduct routine SIC at a NCI-designated comprehensive cancer center. Methods: This single center study describes the second phase of a learning health system initiative to conduct routine SIC for all seriously ill patients with cancer. Prior work completed included defining patient eligibility (e.g. two-year surprise question), and deploying a SIC template in the electronic health record (EHR) to capture SIC in a centralized location. Phase II can be categorized into three steps: a) increasing communication coach and PFA capacity; b) refining EHR tools to automate tracking and reporting of outcomes; and c) adoption of the Model for Improvement as the QI methodology to guide testing and implementation. For the first three months, each team engaged in preparatory work including: process mapping, key driver diagram development, and SIC Guide training. In the last six months, each team met weekly to set their own team goals, conducted iterative PDSAs, and reviewed run charts of their performance. Patient-level data on SIC documentation was collected through automated EHR tools and provided to the teams on a weekly basis. Results: Over nine months (3/1/2021 to 12/31/21), four teams screened 510 patients with cancer, identified 272 (53%) patients as eligible for SIC, and 178 (65%) of those eligible had a documented SIC from a baseline of 0%. The breakdown of this combined SIC patient count by clinician author is as follows: team A 40 (22%); team B 45 (25%); team C 9 (5%); team D 14 (8%) and specialty palliative care 70 (39%). Each team set modest, initial SIC documentation goals (range 5-15%), and attainment of first SIC documentation goal varied (range 2 to 4 months). We retained all clinician team members during the study period, but 2 out of 3 PFAs left by study period end. Noted challenges with PFA recruitment and retention included: lengthy recruitment, integration into clinical teams, and resistance to change by teams. Conclusions: The multidisciplinary approach, inclusive of specialty palliative care, increased SIC documentation. PFA involvement, as implemented, met with challenges and yielded mixed results. Additional follow-up will be required to assess if gains can be sustained and/or increased.