The Forest and the Trees: Evidence-Based Medicine in the Age of Information

In 2005, John Hamilton² wrote in the Journal on how best to practice evidence-based medicine, namely develop an answerable question from a clinical situation and then work your way down a hierarchy of evidence to identify the best available evidence to inform your clinical decision. He advised formulating each clinical question requiring the use of evidence as population, intervention, comparison, outcome, and timing (PICOT). Actually, Hamilton referred to PECOT, where "E" stands for "exposure," but we prefer the acronym Haynes et al.³ used. For Alyssa's query, the question would be, "Among adolescents with depression (P), is venlafaxine (I) more effective than placebo (C) in decreasing depressive symptoms (O) after 12 weeks (T)?" In this Translations article, we set out to answer the question, "What approach is currently recommended for clinicians searching for high-quality evidence to guide optimal clinical decision making?" When Hamilton wrote his piece, he focused on individual randomized controlled trials as the primary source of good evidence. However, in 2018, given the amount of information available, it has become increasingly clear that it is more helpful and efficient to start with high-quality evidence syntheses of primary studies (ie, systematic reviews [SRs], meta-analyses [MAs], and clinical practice guidelines [CPGs]) rather than with the individual primary studies. This means that clinicians now need to become versed in how to distinguish a high-quality synthesis from low-quality ones.