Delphi method consensus on radiographic characteristics influencing management decisions for proximal humerus...

An individualized, and evidence-based approach is needed for effective management of proximal humerus fractures (PHF). As a first step toward this goal, expert agreement is needed on the fracture features that are relevant to management decisions. In this study, we utilize an e-Delphi method to obtain consensus on a comprehensive list of fracture features important to PHF management.

A literature review was conducted to identify all PHF features of importance. A panel of trauma and shoulder surgeons with expertise in treating PHF were recruited. The e-Delphi method consisted of two rounds of 4-point Likert scale questions. In the first-round participating surgeons were asked to assess each feature's level of importance on management decisions, and had the option of proposing modifications to current feature definitions and recommending new features for inclusion in Round 2. Final consensus was reached for a fracture feature if at least 78% of surgeons indicated it was "Very Important" or "Important" in Round 2. The 78% consensus threshold was chosen based on prior item content validity threshold recommendations.

The e-Delphi participants consisted of 29 physicians which completed two rounds of surveys. Round 1 began with 10 features and two additional features, bone quality and humeral head subluxation, were recommended by participants. Of a total of 12 features, 8 reached consensus in Round 2 which included: dislocation (100%), head split fracture (100%), topographical parts (97%), head shaft translation (93%), displacement of parts (90%), bone quality (83%), head impaction (83%), and head shaft angulation (83%).

Consistent with prior research, we identified dislocation, head split fracture, topographical parts, and displacement of parts as important features guiding management decisions for PHF. Our results also distinguish head shaft angulation, head shaft translation, head impaction, and bone quality as key PHF features that are relevant when making treatment decisions. We anticipate that these results can provide a foundation for future research efforts to identify patient subgroups and evaluate comparative treatment outcomes for PHF.