Authors: Ayumu Matsuoka, Tomonori Mizutani, Yuki Kaji, Akiko Yaguchi-Saito, Miyuki Odawara, Aki Otsuki, Junko Saito, Maiko Fujimori, Yosuke Uchitomi, Taichi Shimazu
Published: 2021-06-02
DOI: 10.1200/jco.2021.39.15_suppl.12012
Source: Full article
12012 Background: Geriatric assessment (GA) is recommended in various guidelines for older adults with cancer, but is not widely used in daily practice. This study aims to identify multi-level barriers and facilitators of GA implementation in daily oncology practice, based on a theoretical implementation framework. Methods: We conducted 20 semi-structured interviews with healthcare providers and managers in 14 hospitals treating older adults with cancer in Japan. The Consolidated Framework for Implementation Research (CFIR) was used to guide the collection and analysis of interview data using a deductive approach. CFIR consists of 5 major domains (I. intervention characteristics, II. outer setting, inner setting,. individual characteristics, and V. process), including 39 constructs. Differences in the constructs influencing GA implementation between hospitals where GA is routinely performed (high implementation, HI) and hospitals where GA is not performed or performed only in clinical trials (low implementation, LI) were explored. Results: Among constructs identified as barriers or facilitators of GA implementation, 15 multi-level constructs greatly differed between 5HI and 5LI, including 4 constructs from intervention characteristics, 6 from inner setting, 1 from individual characteristics, and 4 from process. In HI, GA was self-administered (I. adaptability), or administered on a mobile app with interpretation (I. design quality and packaging). In HI, healthcare providers strongly perceived the need to change the practice for older adults (III. tension for change), and recognized GA as fitting in with existing workflow as part of their jobs (III. compatibility). In LI, they did not realize the need to change practice, and rejected GA as an extra burden on their heavy workload. In HI, the usefulness of GA was widely recognized by healthcare providers (IV. knowledge and beliefs about the intervention), GA was given high priority (III. relative priority), had strong support from hospital directors and nursing chiefs (V. leadership engagement), and multiple stakeholders were successfully engaged, such as healthcare providers, especially nurses (V. key stakeholders), directors and nursing chiefs (V. opinion leaders), and those who dedicated themselves to implementing GA (champions). Conclusions: This is the first study to reveal the multi-level barriers and facilitators of GA implementation in daily oncology practice. The findings highlight the need to focus not only on individual or intervention characteristics, but also on the inner setting and the process of implementing GA. Our findings suggest future strategies, such as devising the administration of GA using technology, conducting local needs assessment and consensus discussions about the usefulness and priority of GA, and engaging multiple stakeholders.