Methods for measuring implementation fidelity to the transitional care model.

backgroundFidelity is defined as the extent to which an intervention is delivered as intended. Without a clear understanding of fidelity to an intervention, variability in implementation may be confounded with observed outcomes, making valid conclusions about the intervention's efficacy challenging. The MIRROR-TCM (Multisite Replication of a Randomized Controlled Trial - Transitional Care Model) implementation evaluation, discussed here, sought to examine the delivery of the transitional care model (TCM) intervention and factors that influenced fidelity to the implementation of the intervention protocol over time. The TCM includes 10 core components, which aim to support older adults transitioning from the hospital back to the community. The objective of this paper is to describe the operationalization of a fidelity measure for the TCM intervention. METHODOLOGICAL DEVELOPMENT OF THE FIDELITY SCORE: Fidelity was operationalized at the element, component, and overall intervention levels. The TCM fidelity score examines 38 elements, which are broken into 8 components. Rules to evaluate an element-level, binary (0/1) fidelity scores were established by the study team, which consisted of clinicians and statisticians, and updated through feedback from the advanced practice registered nurses delivering the intervention. Component-level fidelity was assessed as the proportion of total elements completed from each component. Finally, overall fidelity is calculated as the proportion of required elements completed multiplied by 38 or the total possible elements to complete.

conclusionProper evaluation of fidelity to the delivery of an intervention is key in evaluating its implementation. This paper provides new methods to calculate fidelity for a complex intervention, which required the incorporation of 38 elements and nuances that did not require all patients to receive all elements, requiring the creation of clear rules for the exclusion of elements for a participant. The fidelity measure was employed throughout implementation and used to give feedback to the study team and improve fidelity. This paper outlines a novel methodology for flexible fidelity scoring in multicomponent interventions, incorporating rule-based exemptions to accommodate variability in real-world implementation. Future work needs to include validation of the metric through evaluation of fidelity at the participant level to examine its association with clinical outcomes in the parent MIRROR-TCM study, overall and by site.