耕莘护理专科学校.ppt

Clinical Trial 報告者：郭梅珍、江倫志、王美華 指導教授：周桂如教授 報告日期：98.12.08 大綱 一、Definition ? 二、Design Principles 1.Sample size and power estimates 2.Randomization 3.Treatment masking 4.Design factors ?三、Stages Definition Is a planned experiment designed to assess the efficacy of a treatment by comparing the outcomes in a group of patients treated with the test treatment with those observed in a comparable group of patients receiving a control treatment,where patient in both groups are enrolled,treated,and followed over the same time period. Sample size and power estimate The number of subjects in a clinical trial should be large enough to provide a reliable answer to the questions addressed,but should also be the minimum necessary to achieve this aim Statistic considerations Practice considerations Literature review Power Effect size variable scale precision Variability Sample size compliance Sample size calculation 最常使用-Single outcome, dichotomous response 需要四項資訊 P1 –對照組成功的比例 P2 –治療組成功的比例 α–type I error，通常設為0.05 1-β–能正確地辨明兩組之間差異的統計檢力(power) 各組需要的個案數可由下列公式計算 舉例來說，如果假設p1=90%、p2=95%、α= 0.05、β= 0.1 quantitative measurements，如血壓值等? 需要四項資訊? μ1 –對照組的平均值、及σ(SD)? δ(μ2 -μ1) –預期治療組比上對照組的差異 α 1-β ?各組需要的個案數可由下列公式計算 Sample Size DeterminationInformation Required Study objectives Test for equivalence Test for non-inferiority Test for superiority (clinically/statistically) Study design Parallel or crossover Group sequential design Other designs Primary study endpoint(s) Continuous or discrete Multiple study endpoints Sample Size DeterminationInformation Required Clinically meaningful difference Clinically important difference Non-inferiority/superiority margin Equivalence/similarity limit Significance level 1% or 5% Desired power 80% or 90% Other information, e.g., Stratification? 1:1 ratio or 2:1 ratio? Log-transformation? Major principles in clinical trials Minimizing bias Maximizing precision Identify possible sources of bias. Are the trial results robust relative to these bias? Rando