The Kaiser Permanente Breast Cancer Survivorship National Clinical Algorithm: Regional variation in implementation processes/strategies within an integrated health care system.

104 Background: Kaiser Permanente (KP) is an integrated healthcare system providing comprehensive services to over 10 million members. The KP Interregional Breast Care Leaders, a multidisciplinary clinician group, developed a comprehensive evidence/consensus-based algorithm for breast cancer survivors, the KP Breast Cancer Survivorship National Clinical Algorithm (BCSNCA). The BCSNCA is intended to reduce variation in medical and psychosocial surveillance and improve outcomes by providing guidance to locally implemented survivorship programs, ie, recommendations for surveillance, late effects management, and risk reduction. As a quality improvement project, we evaluated regional/ facility level implementation.

METHODS: Qualitative data on BCSNCA implementation was collected from key informant interviews with oncology providers for 20 sites in 6 regions: Georgia, Colorado, Hawaii, Southern (KPSC) and Northern California (KPNC), and Mid-Atlantic, and by attending BCSNCA meetings. Implementation activities were recorded, categorized, and compared to BCSNCA.

RESULTS: Facilities in 3 regions implemented discrete BCSNCA components: Northwest and Georgia facilities implemented dedicated survivorship clinics; a KPSC facility piloted a nurse navigator standardized psychosocial assessment; KPNC implemented local guidelines, similar to the BCSNCA. One region has not implemented; Georgia implemented all BCSNCA components. There is variation within each region. Implementation drivers include available resources, competing QI priorities/leadership preferences, and adaptability of extant programs.

CONCLUSIONS: We found variation between and within regions. The BCSNCA content accommodates variation in implementation, guided by a complex set of factors, including resource availability, leadership preferences, and local organizational goals. Even in integrated systems, the need for locally driven guideline adaptation is critical. Next step: assessment of BCSNCA components on patient-level outcomes.