Clinical versus Statistical Significance in Studies of Thoracic Malignancies.

作者：S. Dahlberg, E. Korn, J. Le-Rademacher, S. Mandrekar · 发表于：Journal of Thoracic Oncology · 年份：2020 · DOI：10.1016/j.jtho.2020.06.007 · 被引用次数：18 · 研究领域：Medicine

https://doi.org/10.1016/j.jtho.2020.06.007 The scientific synergy between statistics and medical research often leads to controversy when considering statistical versus clinical significance of findings from studies that are neither overwhelmingly practicechanging nor clearly negative. Studies can reach statistical significance but provide evidence that is not clinically meaningful, or results could not be statistically significant but very clinically relevant. To fully appreciate the debate about studies that fall in this gray area, one must understand how to interpret several features of statistical design and the interpretation of results. The conduct of clinical trials dictates that a rigorous statistical design a priori targets an effect size, which can be thought of as the true difference in outcome that a study hopes to detect when comparing two regimens with one another or when comparing a single regimen to a reliable historical control. Statistical inference, based on the distributional assumptions of the design, is used to formally test the prestated hypothesis about the effect size resulting in p values, which assist in the determination of a study’s success but do not provide any information regarding the magnitude of the observed effect size in the trial. The p values indicate statistical but not clinical significance; they are the metric for the determination of study success or failure after they are compared with the type I error rate, the probability of obser...