Clinical trial design and evidence-based outcomes in the study of liver diseases

Current medical training often does not include the formal study of trial design, forcing clinicians to acquire this knowledge independently. This article reviews the foundational elements of clinical trial design. An overarching hierarchy of clinical evidence is introduced, and the relative strengths and limitations of the major types of study designs are discussed. A corollary to the hierarchy of evidence in trial designs is proposed for trial outcomes: the "pyramid of endpoints." This pyramid represents a spectrum of outcomes from tangible health events to intermediate markers with no direct physical impact on an individual. The potential advantages and difficulties inherent in the use of surrogate endpoints for final health outcomes are explored. Randomized controlled trials utilizing "hard" clinical endpoints are advocated as the most efficient and reliable way to directly assess the benefits and harms of a therapy; however, using a case study of treatments for hepatocellular carcinoma, we highlight the challenges that can complicate even the highest levels of evidence. All trials have a "signal-to-noise" ratio - this review emphasizes the need for careful and deliberate consideration of the potential limitations of every study, and provides basic tools to assist the practitioner in identifying common pitfalls of clinical trials. Published by Elsevier B.V. on behalf of the European Association for the Study of the Liver.