Evidence-based priority-setting for new systematic reviews: a case study for primary open-angle glaucoma

Article type
Authors
Li T, Dickersin K, Ssemanda E, Scherer R, Ervin A
Abstract
Background: It is essential to prioritize systematic reviews to ensure that important clinical questions are addressed. Practice guidelines reflect the community’s clinical questions and may provide a starting point for prioritization. Objectives: As part of the Cochrane prioritization project, we are testing a framework to prioritize clinical questions related to management of primary open-angle glaucoma (POAG) which can then be addressed by Cochrane reviews. Methods: Using an iterative process and the input of multiple individuals, we restated each recommendation in the 2005 American Academy of Ophthalmology guideline on the management of POAG as an answerable clinical question. To identify the available evidence for each question, we searched The Cochrane Library, PubMed and EMBASE for systematic reviews and the Cochrane Central Register of Controlled Trials for RCTs. We are asking international experts in the field to rank the importance of the clinical questions using an online Delphi survey conducted between April and August 2008 without informing them of the available evidence for each question. The survey is undergoing the ethics review and approval process. Results: We derived 45 clinical questions from the guideline. We identified 35 systematic reviews and over 400 potentially eligible RCTs pertaining to one or more clinical questions specified (see Table 1). We have encountered challenges in determining how best to: derive questions from clinical recommendations; decide the relevance of an RCT to a question without doing a systematic review; ask clinicians about the importance of a topic in a way that is meaningful to them; and incorporate information about the available evidence when we ask clinicians about prioritization of reviews. We will present the ranking of questions based on the results the Delphi survey. Conclusions: In testing a possible process for prioritizing systematic reviews using guidelines, we have encountered some challenges and are investigating solutions.