The process was conducted in a quasi-anonymous manner. Registered participants' emails were known only to the research coordinator to allow for sending reminders. Respondents' judgements and opinions remained strictly anonymous to other members of the expert group. For each round, participants were emailed a link to the questionnaire in the language of their choice (English or French), and were allotted two weeks to complete it. Email reminders were sent 48 hours before the deadline for each round.
The international group of experts was then asked to evaluate whether the items would be relevant for a generic tool that could be easily adapted to any CPD activity, using a 5-point Likert scale (1 = item completely irrelevant, 5 = item completely relevant). A clinical vignette illustrating how the proposed items could be used in a CPD activity whose learning objective was to perform a knee evaluation was given as an example, but participants were asked to rate their response to each item formulated in general terms (e.g. I intend to adopt the behavior described in the training activity objectives in my practice). In the first round, participants were asked simply to rate their responses to each item. In the second round, distributions of respondents' answers to each item in the previous round were presented in percentage form. In both rounds, participants were encouraged to comment both on the relevance of particular items and on the relevance of the questionnaire as a whole to evaluating the impact of the CPD activity on adoption of a clinical behavior. As there are no definite criteria for determining consensus in a Delphi study [12] , content validity was set a priori when at least 75% of participants had reached agreement on the relevance of an item. A partial consensus was reached when more than 60% but less than 75% of participants agreed on an item's relevance. Absence of consensus was determined to be when less than 60% of participants agreed on the relevance of an item. Once the experts had completed this task, the partnership committee reviewed the final list of selected items. The committee analyzed the experts' comments on each item and reformulated the original items when judged necessary. Items that did not reach a consensus rate of at least 60% were excluded.
The international group of experts was then asked to evaluate whether the items would be relevant for a generic tool that could be easily adapted to any CPD activity, using a 5-point Likert scale (1 = item completely irrelevant, 5 = item completely relevant). A clinical vignette illustrating how the proposed items could be used in a CPD activity whose learning objective was to perform a knee evaluation was given as an example, but participants were asked to rate their response to each item formulated in general terms (e.g. I intend to adopt the behavior described in the training activity objectives in my practice). In the first round, participants were asked simply to rate their responses to each item. In the second round, distributions of respondents' answers to each item in the previous round were presented in percentage form. In both rounds, participants were encouraged to comment both on the relevance of particular items and on the relevance of the questionnaire as a whole to evaluating the impact of the CPD activity on adoption of a clinical behavior. As there are no definite criteria for determining consensus in a Delphi study [12] , content validity was set a priori when at least 75% of participants had reached agreement on the relevance of an item. A partial consensus was reached when more than 60% but less than 75% of participants agreed on an item's relevance. Absence of consensus was determined to be when less than 60% of participants agreed on the relevance of an item. Once the experts had completed this task, the partnership committee reviewed the final list of selected items. The committee analyzed the experts' comments on each item and reformulated the original items when judged necessary. Items that did not reach a consensus rate of at least 60% were excluded.
Free full text: Click here
