Grant: Project Grant

Systematic reviews in healthcare are a longstanding cornerstone of evidence based medicine. These reviews are commonly used to inform many vital initiatives including the development of clinical practice guidelines, generation of reimbursement decisions, evaluation of clinical equipoise for clinical trials and other applications. The range of knowledge users is broad, and includes physicians, decision makers, researchers and patients alike. While systematic reviews are of tremendous importance in progressing patient healthcare, users of these studies often express interest in finding ways to complete them faster and at lower costs. Reviews may often take as long as a year to complete, and often cost $100,000 or more to cover all resources and staffing. Recently, experts in systematic reviews and computer science have developed artificial intelligence methods which offer a way in which to have computers assist with the systematic review process to enhance speed and lower staff needs. While appealing, there is great need to establish the dependability of these tools in this role to ensure the scientific quality of systematic reviews are not compromised. The current study proposes a plan to perform primary research studies to evaluate the performance of artificial intelligence tools in the context of systematic reviews. We will share our findings with key organizations in Canada and abroad whose objectives are to use systematic reviews to optimize healthcare for Canada and elsewhere.