051 Can We Automatically Produce Generic Decision Aids for the Clinical Encounter Directly from GRADE Guideline Recommendations? Experience from the Share-It Project. (15th August 2013)

• Record Type: Journal Article
• Authors: Agoritsas, T; Brandt, L; Heen, A Anja; Kristiansen, A; Alonso-Coello, P; Akl, E; Neumann, I; Tikkinen, K; Montor, V; Guyatt, G; Vandvik, P
• Is Part Of: BMJ quality & safety. Volume 22(2013)Supplement 1
• Page Start: A28
• Page End: A28
• Publication Date: 2013-08-15
• Subjects: Medical care -- Quality control -- Periodicals; Health facilities -- Risk management -- Periodicals; Medical errors -- Prevention -- Periodicals; 362.106805
• DOI: 10.1136/bmjqs-2013-002293.82
• Languages: English
• ISSNs: 2044-5415

Abstract

Abstract : Background: Although decision aids (DA) can help to communicate evidence to patients, their production is time consuming, often not based on the best available evidence or rapidly outdated. Linking trustworthy guidelines and DA for shared-decision making could both overcome these limitations and enhance guideline dissemination. Objectives: To test the feasibility of automatically translating any recommendations from GRADE guidelines into generic and interactive DA accessible on tablet computers for clinicians and their patients in the clinical encounter. Methods: As part of the DECIDE project, we developed a framework for translating components of GRADE into DA, following the International Patient Decision Aid Standards. Using a recently published guideline, we implemented that framework in our MAGIC (Making Grade the Irresistible Choice) application – a prototype electronic guideline tool and publication platform that can automatically display recommendations in multilayered presentation formats. Results: Our prototype was able to automatically translate a large number of GRADE recommendations and their supporting evidence into electronic and interactive DA. Preliminary results of user-testing in real patient-clinician interactions suggest that these DA can be used at the point of care to discuss estimates of treatment effects for patient relevant outcomes, confidence in estimates, burden of treatment, and cost issues. Discussion: This study provides a proof-of-concept that components of GRADE recommendations can be interactively displayed in generic tools for interactive shared-decision making in a wide range of treatment alternatives. Implications for Guideline Developers/Users: Our electronic DA offer promising opportunities to disseminate guidelines at the point of care.