评估临床试验中治疗效应的收缩估计量
该研究基于Cochrane系统评价数据库(CDSR)的Meta分析集合,评估信噪比(SNR)较低时传统无偏估计量的表现,发现达到统计显著性的试验存在过度乐观和置信区间覆盖不足。作者提出的收缩估计量在均方根误差、覆盖率和偏倚幅度上均优于传统估计量,无论是否条件于统计显著性。
要点
- CDSR中许多试验的信噪比(SNR)较低,对真实效应的检验效能也低。
- 达到统计显著性的试验中,传统无偏估计量存在 considerable 过度乐观和置信区间覆盖不足。
- 收缩估计量在均方根误差、覆盖率和偏倚幅度上均优于传统无偏估计量。
核心结论
- 收缩估计量在均方根误差、覆盖率和偏倚幅度上均优于传统无偏估计量,无论是否条件于统计显著性。
研究设计与样本
- 利用Cochrane系统评价数据库(CDSR)作为Meta分析集合进行定量评估,比较收缩估计量与传统无偏估计量的性能。
主要发现
- CDSR中许多试验的信噪比(SNR)较低,对真实效应的检验效能也低。
- 达到统计显著性的试验中,传统无偏估计量存在 considerable 过度乐观和置信区间覆盖不足。
- 收缩估计量在均方根误差、覆盖率和偏倚幅度上均优于传统无偏估计量,条件和不条件于统计显著性均如此。
局限性
- 原文未报告局限性
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
The main objective of most clinical trials is to estimate the effect of some treatment compared to a control condition. We define the signal-to-noise ratio (SNR) as the ratio of the true treatment effect to the SE of its estimate. In a previous publication in this journal, we estimated the distribution of the SNR among the clinical trials in the Cochrane Database of Systematic Reviews (CDSR). We found that the SNR is often low, which implies that the power against the true effect is also low in many trials. Here we use the fact that the CDSR is a collection of meta-analyses to quantitatively assess the consequences. Among trials that have reached statistical significance we find considerable overoptimism of the usual unbiased estimator and under-coverage of the associated confidence interval. Previously, we have proposed a novel shrinkage estimator to address this "winner's curse." We compare the performance of our shrinkage estimator to the usual unbiased estimator in terms of the root mean squared error, the coverage and the bias of the magnitude. We find superior performance of the shrinkage estimator both conditionally and unconditionally on statistical significance.
引用
- van Zwet EW, Tian L, Tibshirani R, et al. Evaluating a shrinkage estimator for the treatment effect in clinical trials. Stat Med. 2023. PMID: 38111969. doi:10.1002/sim.9992
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。