如何正確解讀和應用臨床試驗結果的P值

How to Correctly Interpret and Apply the P Values of Clinical Trial Results

一篇臨床試驗的結果常簡單地被分成兩個面向：「正面結果」或「負面結果」。如果一項試驗被宣稱具有「正面結果」，則主要療效結果的顯著性水平最少應達5%，但這是不夠的。要切記的是，達統計顯著差異不代表就具臨床重要性，具臨床重要性也不一定會有統計上的意義。實際上，對臨床試驗結果要能更細微的解讀需要徹底檢視所有的證據，包括次要終點、安全性問題、研究樣本數以及對主要和重要次要結局的治療效果的一致性等。本篇綜論透過數篇已發表的臨床試驗結果來闡述這些論點，此外本文還討論了提前終止試驗、療效-風險平衡、試驗設計瑕疵，以及試驗結果在本地環境中的應用等問題。健康照護者有責精準地解釋臨床試驗結果，以便對病人做出最佳的治療決定。

The results of a clinical trial are often simply divided into two dimensions: ''positive results'' or ''negative results.'' A significance level of 5% for the primary efficacy outcome is the minimum requirement if a trial is to be declared to have“positive outcome,”but it is not sufficient. It is important to remember that statistically significant difference does not imply clinical importance and clinical importance does not imply statistical significance. In reality, a more subtle interpretation for clinical trial outcome requires a thorough examination of the totality of the evidence, including secondary endpoints, safety issues, study sample size, the consistency of the treatment effect on the primary and key secondary outcomes, and so on. This review article elaborates on those arguments to interpret the results of several published clinical trials. In addition, issues related to stopping a trial early, the efficacy-risk balance, the flaws in trial design, and the application of trial results in a local setting are also discussed in this article. Health care professionals bear the responsibility for accurately interpreting clinical trial results in order to make the best treatment decisions for each patient.