1008 — Necessary but Not Sufficient: A Multimethod Study of the Role of Champions in Healthcare-Related Implementation

Lead/Presenter: Edward Miech,  COIN - Indianapolis
All Authors: MIECH EJ (PRIS-M QUERI (Indianapolis)), Rattray NA (PRIS-M QUERI), Damush TM (PRIS-M QUERI)

Objectives:
One factor often cited as crucial to effective implementation in healthcare is the presence of champions. Few studies, though, have attempted to describe and explain how champions influence implementation outcomes. This analysis sought to identify the specific causal paths by which champions influenced implementation success in a large, prospective and national study of VA acute stroke programs.

Methods:
This multimethod analysis of champions integrated three discrete research methods: construct scoring, Configurational Comparative Methods and qualitative analysis. This research was based on the VA RE-INSPIRE study, which examined the implementation of acute stroke programs at VA medical centers across the United States between 2012-2015. The Consolidated Framework for Implementation Research (CFIR) served as the study's conceptual framework. Context-related data was collected through semi-structured interviews conducted at three annual site visits at each of the participating 11 VA medical centers. Interview data was transcribed, de-identified and then qualitatively coded by study team members. Next, the RE-INSPIRE project team systematically applied 20 CFIR constructs (including the "champion" construct) directly to the qualitative data for each of the 33 site visits, scoring each CFIR construct for valence (i.e., positive, neutral, negative) and magnitude (i.e., weak or strong) that reflected the influence of that construct on the implementation process; scores varied from +2 (strong positive influence) to -2 (strong negative influence). This dataset of over 600 different CFIR scores was analyzed with Configurational Comparative Methods (CCMs), a mathematical approach based on Boolean algebra and set theory, to determine the specific role of champions vis-a-vis implementation outcomes as well as other particular CFIR constructs (e.g., Reflecting and Evaluating). These cross-case solutions were then explored in detail using the rich qualitative data, using within-case analyses to explain how the presence of champions related to implementation outcomes.

Results:
Over 150 VA staff from 11 VA medical centers participated in the study, resulting in a total of over 300 interviews. The presence of effective champions by itself was not consistently linked to implementation success. However, effective champions were necessary for effective reflecting and evaluating, which in turn proved sufficient for implementation success: effective champions ? effective reflecting and evaluating ? implementation success. The presence of champions, though, did not always translate into effective reflecting and evaluating; in these cases, implementation was not successful. Qualitative analyses specified the particular causal mechanisms linking champions with reflecting and evaluating.

Implications:
The presence of champions was found to be a "necessary but not sufficient" condition for healthcare-related implementation success.

Impacts:
Multimethod research offers new ways to parse the necessity/sufficiency of specific conditions like the presence of champions for healthcare-related implementation success, as well as determine the causal pathways by which certain conditions link to other particular conditions and together yield an outcome of interest.