Senior GRADE methodologists encounter challenges as part of WHO guideline development panels: an inductive content analysis

BACKGROUND: The World Health Organization (WHO) classifies a substantial proportion of their recommendations as strong despite low or very low confidence (certainty) in estimates of effect. Such discordant recommendations are often inconsistent with Grading of Recommendations Assessment, Development and Evaluation (GRADE) guidance. OBJECTIVE: To gain the perspective of senior WHO methodology chairs regarding panels' use of GRADE, particularly regarding discordant recommendations. DATA SOURCES: Senior active GRADE methodologists who had served on at least two WHO panels and were an author on at least one peer-reviewed published article on GRADE methodology. METHODS: Five eligible methodologists participated in detailed semistructured interviews. Respondents answered questions regarding how they were viewed by other panelists and WHO leadership, and how they handled situations when panelists made discordant recommendations they felt were inappropriate. They also provided information on how the process can be improved. Interviews were recorded and transcribed, and inductive content analysis was used to derive codes, categories, and emergent themes. RESULTS: Three themes emerged from the interviews of five methodologists: (1) The perceived role of methodologists in the process, (2) Contributors to discordant recommendations, and (3) Strategies for improvement. Salient findings included (1) a perceived tension between methodologists and WHO panels as a result of panel members' resistance to adhering to GRADE guidance; (2) both financial and nonfinancial conflicts of interest among panel members as an explanation for discordant recommendations; and (3) the need for greater clarity of, and support for, the role of methodologists as co-chairs of panels. CONCLUSIONS: These findings suggest that the role of the GRADE methodologist as a co-chair needs to be clarified by the WHO leadership. They further suggest the need for additional training for panelists, quality monitoring, and feedback to ensure optimal use of GRADE in guideline development at WHO.