Validating the Comprehensive Framework for Guideline Implementability (CFGI): a factorial randomized experiment on clinical practice guideline users’ implementation willingness

Poor adoption of clinical practice guidelines (CPGs) can compromise healthcare quality and patient safety. Optimizing intrinsic guideline factors (e.g., language, format) is a low-cost and effective strategy to improve uptake. We previously developed the Comprehensive Framework for Guideline Implementability (CFGI) to describe factors that influence guideline implementability. This study aimed to validate the relevance of CFGI domains and identify which domains best predict guideline implementability. We conducted a factorial randomized experiment among CPG users from pediatrics, obstetrics and gynaecology (O&G), and pharmacy in China. Six CFGI domains were examined: scope and purpose, clarity and consistency, development and evidence base, structure and contents, development team and transparency, and format and presentation. For each specialty, guidelines were adapted into 64 versions representing all combinations of presence or absence of the six domains. Participants were randomly assigned to evaluate one version and rated their implementation willingness on a 1-100 scale using a REDCap slider interface. Multiple linear regression with effect coding was used to test main and interaction effects, adjusting for guideline category and hospital level. Of 768 participants, 539 were included in the final analysis: 176 (32.7%) evaluated the pediatrics guideline, 154 (28.6%) the O&G guideline, and 209 (38.8%) the pharmacy guideline. Six domains were significantly associated with higher implementation willingness: “format and presentation” with the greatest impact ( β = 21.50, P < 0.001); “structure and contents” ( β = 21.11, P < 0.001) followed by “development and evidence base” ( β = 17.65, P = 0.042), “clarity and consistency of recommendations” ( β = 12.89, P < 0.001), “development team and transparency” ( β = 11.82, P < 0.001) and “scope and purpose” ( β = 8.97, P = 0.046). All six CFGI domains were experimentally supported as significant contributors to implementation willingness and are therefore retained in the CFGI framework pending further tool development. This study provides initial experimental support for six CFGI domains as predictors of guideline users’ implementation willingness, a key proximal indicator of guideline implementability. When a guideline incorporates these six domains, guideline users report higher implementation willingness; whether this translates into actual implementation behavior warrants further investigation. China Clinical Trails Registry (ChiCTR2400086931); registered July 15, 2024. https://www.chictr.org.cn/.