Publication Type(s): Journal: Conference Abstract
Author(s): Johnston G.; Exton L.; Mohd Mustapa M.F.; Coulson I.; English J.; Bourke J.
Abstract:The British Association of Dermatologists (BAD) is one of only a few specialist societies that fund independently produced guidelines with a qualified information technology team. In the speciality of contact dermatitis it is recognized that evidence may be lacking or of poor quality, and so employing a robust methodology is important in developing guidance even in areas where the evidence base is weak.
The Grading of Recommendations Assessment, Development and Evaluation (GRADE) is the gold-standard methodology for developing systematic reviews and clinical guidelines. It offers a transparent process from evidence to recommendations, with set criteria to guide the guideline development group (GDG) in formulating a recommendation and determining its strength, with clear, pragmatic interpretation of strong vs. weak recommendations for clinicians and patients. It also acknowledges patient values and preferences. The new BAD guidelines adopted this system for contact dermatitis. The GDG specified key clinical questions to form the basis of the guideline using the PICO method. This technique frames and answers a clinical question and considers the patient population, an intervention, a comparator and the outcome. Literature searches covering the PICO questions were run across multiple databases.
The GDG then identified all possible outcomes of interest and of importance to patients, and ranked them by those that were (i) critical for decision making, (ii) important but not critical for decision making and (iii) not important for decision making. Those outcomes agreed as critical for decision making determined the overall quality of the evidence. Unlike previous systems where each study is assessed and graded individually, GRADE assesses the quality of evidence per comparison (intervention vs. comparator/control), per outcome and then across all the studies (i.e. for the body of evidence) that report that comparison and that outcome. Evidence from randomized controlled trials would be graded initially as 'high' quality, and observational studies would start with 'low' quality ratings due to inherent selection bias, and then be upgraded or downgraded depending on the quality of the study. An overall quality rating of the evidence was then obtained for each outcome.
A GRADE evidence summary table, which displays the quality assessments, study features and summary of findings and a template for linking the evidence to the recommendations, was used to formulate evidence. Statements/summaries, recommendations and their grading were produced 'live' during meetings following discussion of the synthesized evidence. The authors suggest that all GDGs adopt a systematic approach to investigation, management and treatment of contact dermatitis.