Systematic reviews are described as the gold standard in the evidence-based healthcare hierarchy. They are supposed to be transparent, reproducible, and follow a set structure. So how can systematic reviews – and within them meta-analyses – that look at the exact same question find radically different answers? The reality, of course, is that meta-analyses are not immune to biases.
This talk will use reviews of electronic cigarettes for smoking cessation as a case study, looking across meta-analyses in this area. Variation in outcomes and conclusions from these reviews will be interrogated, to improve understanding of how different methods shape outcomes, and how different biases shape conclusions.
This free talk is given in conjunction with the Meta-analysis course, part of the Evidence-Based Healthcare programme.