Wording Approaches to Convey the Strength of Recommendations in Health Guidelines

This study is a randomized controlled trial with a blinded non-clinical intervention. The three arms of this study are include different wording approaches to phrase health recommendations for the purpose of conveying the strength of a recommendation. Based on the Grading of Recommendations Assessment Development and Evaluation (GRADE) approach, the strength of a recommendation is determined to be either "strong" or "conditional" with a direction that either favours or does not favour an intervention. The wording of health recommendations to reflect this strength is not consistent, and there is no optimal approach that has been identified. Participants will be stratified based on self-identified recommendation user group (e.g., clinician, policymaker, member of public) and view one formatted recommendation intended for that group. Outcome data will be collected via an anonymous survey using multiple choice and 7-point likert scales to capture participant understanding and preference. Comparisons between the results of all participants within each intervention arm will be made, with the goal of clarifying whether there is a better wording approach for conveying the strength of recommendations to all audiences.

• GRADE approach ("recommends" and "suggest") (Other): "recommend" (strong) and "suggest" (conditional) Example: The guideline panel recommend(s)/suggest(s) \[against\] \[the intervention\], rather than \[the comparator\], for \[indication\].
• New approach ("strongly recommends" and "conditionally recommends") (Other): "strongly recommend" (strong) and "conditionally recommend" (conditional) Example: The guideline panel strongly recommend(s)/conditionally recommend(s) \[against\] \[the intervention\], rather than \[the comparator\], for \[indication\].
• Alternate approach ("should" and "may") (Other): "should" (strong) and "may" (conditional) Example: \[The intervention\] should/may \[not\] be \[verb\], rather than \[the comparator\], for \[indication\]

Primary outcomes

• Understanding of the Strength and Direction of Recommendation [During a single survey session (approximately 10 minutes)]: The primary outcome is understanding of the correct interpretation of recommendation strength and direction. This will be assessed with three multiple-choice questions where answer options represent interpretations for strong-for, conditional-for, conditional-against, and strong-against recommendation statements, as indicated by GRADE. Each question has 1 correct answer. Scores will be assessed per-question and aggregate (maximum score of 3/3 suggesting correct understanding versus minimum 0/3 suggesting poor understanding).

Secondary outcomes

• Intended Behaviour [During a single survey session (approximately 10 minutes)]: Defined as participants' intention in adopting and following the recommendation they read. Participants are asked how likely it is that they would follow the recommendation if the recommendation hypothetically applied to themselves. The question will be measured using a 7-point Likert-type scale with 1 being 'very unlikely', 4 'neutral', and 7 'very likely'.
• Preference [During a single survey session (approximately 10 minutes)]: This outcome aims to assess whether the choice of wording would lead to different decisions by participants. Defined as a greater liking of one wording approach over the others. After the blinded phase for prior questions, participants will be 'unblinded' to the purpose of the wording approaches to convey the strength of recommendation. Subsequently, they will be asked to choose between wording approaches or indicate no preference in a multiple-choice question format.

Sex: All, ages: 16 Years to any, accepts healthy volunteers

Inclusion Criteria:

* Individuals of 16 years or older of age
* Ability to read and understand English
* Previous use of, or intention to use, health guidelines or recommendations (e.g., intention to access clinical or public health guidelines, recommendations, or policy advice)

Exclusion Criteria:

* Previous participation in this study