Applying GRADE for diagnosis revealed methodological challenges: an illustrative example for guideline developers

Abstract

ObjectiveTo identify challenges in the application of GRADE for diagnosis when assessing the certainty of evidence in the test-treatment strategy (diagnostic accuracy, test burden, management effectiveness, natural course, linked evidence) in an illustrative example and to propose solutions to these challenges. Study Design and SettingA case study in applying GRADE for diagnosis that looked at the added value of IgE for diagnosing allergic rhinitis. ResultsEvaluation of the full test-treatment strategy showed a lack of (high-quality) evidence for all elements. In our example, we found a lack of evidence for test burden, natural course, and link between the test result and clinical management. Overall, systematically reviewing the evidence for all elements of a test-treatment strategy is more time-consuming than only considering test accuracy results and management effectiveness. For increasing efficiency, the guideline panel could determine critical elements of the test-treatment strategy that need a systematic review of the evidence. For less critical elements, a guideline panel can rely on gray literature and professional expertise. ConclusionA lack of high-quality evidence and time investment if the full test-treatment strategy is assessed, creating challenges in applying GRADE for diagnosis. Discussion within guideline panels about critical elements that need to be reviewed might help.