Sensitivity analysis of incomplete longitudinal data departing from the missing at random assumption: Methodology and application in a clinical trial with drop-outs
Sensitivity analysis of incomplete longitudinal data departing from the missing at random assumption: Methodology and application in a clinical trial with drop-outs
复制标题
DOI：
10.1177/0962280213490014
复制
发表时间：
2016-08-01
影响因子：
2.3
通讯作者：
Chavance, M.
中科院分区：
文献类型：
--
作者：
Moreno-Betancur, M.;Chavance, M.
基于直接可能性的退出纵向数据的统计分析，并使用所有可用数据，在有关退出机制的一些假设下提供无偏且完全有效的估计。不幸的是，这些假设永远无法从数据中得到检验。因此，应定期进行敏感性分析，以评估偏离这些假设的推论的稳健性。然而，在建立这种分析时，每个特定的科学背景都需要不同的考虑，不存在标准方法，这仍然是一个活跃的研究领域。我们提出了一种灵活的程序来在处理连续结果时执行敏感性分析，这些结果在初始似然分析中由线性混合模型描述。该方法依赖于完整数据可能性的模式混合模型分解，并在模拟研究中得到验证。该方法是由一项维持睡眠失眠治疗的随机临床试验推动的。该案例研究说明了我们方法的实用价值，并强调了在分析带有丢失的数据时进行敏感性分析的必要性：初步分析的一些结论被证明是可靠的，而其他结论则被发现是脆弱的并且强烈依赖于建模假设。提供了用于实现的 R 代码。
Statistical analyses of longitudinal data with drop-outs based on direct likelihood, and using all the available data, provide unbiased and fully efficient estimates under some assumptions about the drop-out mechanism. Unfortunately, these assumptions can never be tested from the data. Thus, sensitivity analyses should be routinely performed to assess the robustness of inferences to departures from these assumptions. However, each specific scientific context requires different considerations when setting up such an analysis, no standard method exists and this is still an active area of research. We propose a flexible procedure to perform sensitivity analyses when dealing with continuous outcomes, which are described by a linear mixed model in an initial likelihood analysis. The methodology relies on the pattern-mixture model factorisation of the full data likelihood and was validated in a simulation study. The approach was prompted by a randomised clinical trial for sleep-maintenance insomnia treatment. This case study illustrated the practical value of our approach and underlined the need for sensitivity analyses when analysing data with drop-outs: some of the conclusions from the initial analysis were shown to be reliable, while others were found to be fragile and strongly dependent on modelling assumptions. R code for implementation is provided.