Background: Background: Essentia Health St Mary’s Medical Center, a tertiary referral center serving the rural areas of Northeastern Minnesota and Northwestern Wisconsin, identified disparities in hospice utilization of Medical Shared Savings Programs (MSSP) decedents in 2021 – 2022 (40% vs 53% Medicare national, 2nd quarter 2022). Health system metrics on hospital readmission and length of stay as well as a lack of documented advanced care planning were identified opportunities for improvement.

Purpose: The intervention chosen to address this concern was to use a patient-centered approach of improving communication around goals and values with the Serious Illness Conversation Guide (SICG). Our approach was to begin the system-wide effort with a focus on documented conversations in the hospitalist group. The hospitalist group was chosen as being most likely to affect improvement in the observed disparity in our population.

Description: The health system partnered with Ariadne labs to implement a SICG training program in early 2023. The hospitalist group of 74 clinicians agreed to initial implementation and adopted SICG training as their compensation linked group quality measure. Hospitalist training (including billing for conversations) was accomplished in March through June of 2023. Electronic medical record, health system and individual clinician outcome metrics were tracked. A 4-month post implementation clinician survey was conducted.Results: During the implementation phase 80 clinicians received SICG training with measurement of documented conversations in the electronic medical record. 357 SICG conversations were documented by 60% of trained clinicians during the 4-month implementation phase, including the entire hospitalist group. 80% of documented conversations originated from non-palliative care clinicians. The increase in documented conversations was associated with a 2-fold increase in hospice referrals (from 18 / month to 37/ month) and a 9.8% increase in outpatient hospice average daily census (from 255 to 280 patients). Implementation was also associated with a transient decrease in palliative care consultation referrals from 6.6% of unplanned inpatient admissions per month to 4.4 %, subsequently recovering to 6%. Health system reported observed/expected inpatient mortality ratio declined from 1.01 to 0.9 during this time period. Ongoing longitudinal metrics will be presented. A 6-month post implementation survey in the hospitalist group (43% participation) showed that 78% of survey respondents reported using the SICG. 95% felt adequately supported in their practice to have serious illness conversations with patients or surrogates in their practice and 67% reported increased satisfaction with their role in patient care because of these discussions. 84% expressed confidence in their ability to bill for a serious illness conversation.

Conclusions: Rapid implementation of the Serious Illness Conversation Guide documentation in a hospitalist program was associated with early improvements in health system outcome metrics and high participating clinician satisfaction. Results from hospitalist implementation are being used to facilitate system-wide adoption of the SICG.