Novel approach to the codesign of a tailored intervention bundle: applying theory with pragmatism

Theory-based design of tailored intervention bundles to address low-value care is an evolving science. Codesign approaches are increasingly being used to engage end-users in implementation projects. This manuscript outlines a pragmatic, theory-informed approach to codesigning an intervention bundle with physicians to address laboratory test overuse in hospitals.

We used a qualitative approach to collect data during virtual audit and group feedback sessions based on the Calgary Audit and Feedback Framework between January 2021 and June 2021 across four hospitals in Calgary, Canada. Sessions were audio-recorded and transcribed, and detailed field notes were taken. In introductory sessions, participants identified barriers, facilitators, and intervention ideas for laboratory test overuse in hospitals. In follow-up sessions, they contextualised ideas using a patient-journey map and prioritised them with a matrix. We used the Theoretical Domains Framework to analyse barriers and facilitators and applied behaviour change techniques (BCT) to map physician-generated interventions to theory-based implementation strategies.

We conducted twelve sessions with 83 physicians representing family medicine and internal medicine specialties. Our pragmatic codesign approach with physicians emphasised assessing clinical stability when considering ordering practices and generated intervention ideas focused on stable patients. Physician-generated intervention ideas were mapped to the BCT taxonomy to inform the development of a tailored behavioural change intervention bundle to reduce overutilisation of laboratory testing in hospitals, consisting of (1) policy changes, (2) order system modifications, (3) education, and (4) audit and feedback.

Our approach allowed the recognition of context-specific barriers and facilitators, engaged end-users to codevelop intervention ideas that addressed the identified barriers, and led to the development of a tailored multicomponent intervention bundle. This approach emulated effective application of BCTs to systematically identified barriers and facilitators. Further research is required to adapt this approach to other contexts to determine generalisability and to evaluate the effectiveness of this intervention bundle.