When should clinical practice guidelines consider recommending the development of patient decision aids: A modified Delphi study

Background: Clinical practice guidelines (CPGs) play a crucial role in evidence-based decision-making. Linking patient decision aids (PDAs) to CPGs can help to ensure that patient preferences are (better) incorporated in health decision-making. However, the need for a PDA may vary depending on the decision. Given the extensive number of recommendations in many CPGs and the resource-constrained environment in which guideline development groups often work, developing a PDA for each recommendation may not be necessary or practical. Objective: Our aim was to determine and rank criteria for prioritizing those recommendations within CPGs for which the development of a PDA is particularly relevant. Methods: The ethics committee of the Brandenburg Medical School granted a waiver for this Delphi Study. From July to December 2024, a modified Delphi process was conducted in three rounds with international experts in shared decision-making (SDM) and CPGs according to the RAND/UCLA Appropriateness Method. Potential criteria were initially generated through a scoping review, expert interviews, and a preliminary study, in which 19 (German) SDM experts were asked to evaluate two CPG scenarios to identify characteristics that could be used to determine CPG recommendations with SDM relevance. These criteria were discussed, prioritized, and reformulated in the three rounds to yield a final list. Each round consisted of a survey, which was pilot-tested before the start of the first round. Survey responses were evaluated descriptively. Following a survey round, experts were invited to discuss the survey results in a focus group, each of which was recorded, transcribed, and analyzed using qualitative content analysis. The results were provided to the experts as controlled feedback to facilitate an iterative consensus-building process. Results: 25 experts from eight countries participated. Of the initially identified 42 criteria, 30 were either excluded or merged in a workshop with the core team of the EDELL project prior to the initiation of the Delphi process. Of the remaining 12, two were merged, and two were removed due to insufficient consensus in the first round (<75% agreement). The remaining eight criteria were then ranked in the second and third round according to their importance for prioritizing CPG recommendations for which the development of a PDA is particularly relevant. Consensus was reached: multiple options with different benefit-harm profiles were ranked first, considered the most important, with an average ranking value of 1.61. Decision with impact and potential discrepancy in preferences were ranked second and third, with average ranking values of 2.78 and 3.96 respectively. These were followed by treatment burden (mean: 4.09), uncertainty of evidence (mean: 5.26), life values (mean: 5.83), adherence (mean: 6.17), and financial aspects (mean: 6.30). Implications for research and practice: This study provides a structured approach for linking CPG recommendations to priorities for PDA development. The identified criteria may help guideline developers allocate resources effectively, ensuring that PDAs are created for decisions where they are most beneficial. Future research will refine these criteria into a structured tool for CPG developers, supporting more patient-centered guideline implementation.