Grant: Project Grant

Being able to critically appraise the findings of studies is central to informed decision-making in patient care. Patients, healthcare providers and policy makers (i.e. knowledge users [KUs]) need the highest quality studies to make decisions about which treatments should be used in healthcare practice. The most reliable way to answer questions using research is a systematic review, which is the most common type of knowledge synthesis used to summarize all existing medical literature on a topic. These reviews are needed to support patient care, develop policies, and assess new drugs and devices. Currently there are many different types of systematic reviews, such as scoping reviews, reviews with meta-analysis, and reviews with network meta-analysis (NMA). Reviews with NMA can rank and order medicines used for the same medical condition by their safety and effectiveness. Many tools are used to assess the potential biases in these different reviews. For example, the AMSTAR checklist assesses whether a review with pairwise meta-analysis (MA) has been well conducted. The ROBIS tool assesses whether a MA is at risk of containing dangerous biases that may distort the findings. However, no tool currently exists specifically to assess the risk of bias in a review with NMA. Our project has 4 objectives. First we will identify biases that are found in NMAs through a methodological review (a type of knowledge synthesis where all the evidence on a given methods topic is identified and described). Second, we will select the most important biases to be included in the tool through an agreement-building process (Delphi) and finalise the tool. Third, we will use 3 examples to test the usability of the tool through an evaluative pilot study, and four, develop a KT strategy. Our outputs will be 4 journal articles, presentations, video, webinar and plain language products. Our newly developed tool will help KUs choose the highest quality NMA, which will improve evidenced-based care.