Guidelines for Designing and Conducting Delphi Consensus Studies: An Expert Consensus Delphi Study.

Objective: The purpose of this study was to conduct a Delphi project to develop guidelines for the design and execution of Delphi studies within medical and surgical specialties.

Methods: Open-ended questions in round 1 and open-ended and semi-open questions in round 2 were answered. The results of the first two rounds were used to develop a Likert style questionnaire for round 3. The level of agreement and consensus was defined as 80%. Consensus was further categorized into specific percentage ranges for clarity: 100% unanimous consensus, 90-99% very strong consensus, 80-89% consensus.

Results: Consensus was achieved for 35 out of 63 items (56%). Unanimous agreement was reached for 4 items (6.3%), while very strong consensus was established for 12 items (19%). Consensus was reached for an additional 19 items (30.1%), and the panel remained undecided on 7 items (11.1%).

Conclusions: Unanimous agreement was reached for iteration, the ability to establish treatment guidelines, a proven track record of panel members, and the requirement for at least one steering committee member to be a Delphi expert. Very strong consensus was reached on several key requirements: a clear definition of consensus, controlled feedback between rounds, precise definitions of expert and expertise, and the need for panel members to demonstrate experience through publications and clinical practice. Criteria for panel selection should ensure diversity and specialization, with steering committee members being content experts and a minimum of 20-30 panel members for broader topics. Regional experts should provide consensus on specific topics only. The steering committee should develop questions, with open-ended questions in Round 1 and both types in Round 2. Limiting the process to three rounds is advisable, aiming for at least 80% consensus in the final round.