A Core Components Framework for Evaluating Implementation of Competency-Based Medical Education Programs

Abstract

PURPOSE: The rapid adoption of competency-based medical education (CBME) provides an unprecedented opportunity to study implementation. Examining "fidelity of implementation"-that is, whether CBME is being implemented as intended-is hampered, however, by the lack of a common framework. This article details the development of such a framework. METHOD: A two-step method was used. First, a perspective indicating how CBME is intended to bring about change was described. Accordingly, core components were identified. Drawing from the literature, the core components were organized into a draft framework. Using a modified Delphi approach, the second step examined consensus amongst an international group of experts in CBME. RESULTS: Two different viewpoints describing how a CBME program can bring about change were found: production and reform. Because the reform model was most consistent with the characterization of CBME as a transformative innovation, this perspective was used to create a draft framework. Following the Delphi process, five core components of CBME curricula were identified: outcome competencies, sequenced progression, tailored learning experiences, competency-focused instruction, and programmatic assessment. With some modification in wording, consensus emerged amongst the panel of international experts. CONCLUSIONS: Typically, implementation evaluation relies on the creation of a specific checklist of practices. Given the ongoing evolution and complexity of CBME, this work, however, focused on identifying core components. Consistent with recent developments in program evaluation, where implementation is described as a developmental trajectory toward fidelity, identifying core components is presented as a fundamental first step toward gaining a more sophisticated understanding of implementation.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 494 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.