Mixed-effects models for repeated measures (MMRMs) with an ‘unstructured’ (UN) covariance structure are frequently used in primary analyses for group comparisons of incomplete continuous longitudinal data from drug development trials. However, MMRM-UN analysis could lead to convergence problems in numerical optimisation, especially in trials with a small sample size or high dropout rate. Although the so-called sandwich covariance estimator is robust against the misspecification of the covariance structure, its performance deteriorates for small sample sizes. We review eight modified covariance estimators adjusted for small-sample bias and compare their performances in the framework of MMRM analysis through simulations. In terms of the type 1 error rate and coverage probability of confidence intervals for group comparisons, Mancl and DeRouen's covariance estimator (MD) shows the best performance among the modified covariance estimators, followed by Fay and Graubard's estimator. The performance of MD is nearly equivalent to that of the Kenward–Roger method with a UN structure. The Kenward–Roger method with first-order autoregressive structure results in substantial inflation of the type 1 error rate in the scenario where the variance of measurements increases across visits. In summary, we recommend the use of MD in MMRM analysis if the convergence problem involving a UN structure occurs in small clinical trials.

中文翻译：

---

缺失数据的小样本纵向研究中线性混合模型修正协方差估计量的实践回顾和比较

具有“非结构化”(UN) 协方差结构的重复测量 (MMRM) 混合效应模型经常用于药物开发试验中不完整的连续纵向数据的组比较的主要分析中。然而，MMRM-UN 分析可能会导致数值优化中的收敛问题，特别是在样本量小或辍学率高的试验中。尽管所谓的夹心协方差估计器对协方差结构的错误指定具有鲁棒性，但对于小样本量，它的性能会下降。我们回顾了针对小样本偏差调整的八个修正协方差估计量，并通过模拟比较了它们在 MMRM 分析框架中的表现。就组比较的置信区间的类型 1 错误率和覆盖概率而言，Mancl 和 DeRouen' s 协方差估计器 (MD) 在修改后的协方差估计器中显示出最好的性能，其次是 Fay 和 Graubard 的估计器。MD 的性能几乎等同于具有 UN 结构的 Kenward-Roger 方法。具有一阶自回归结构的 Kenward-Roger 方法在测量方差随访问增加的情况下导致类型 1 错误率的大幅膨胀。总之，如果涉及 UN 结构的收敛问题发生在小型临床试验中，我们建议在 MMRM 分析中使用 MD。具有一阶自回归结构的 Kenward-Roger 方法在测量方差随访问增加的情况下导致类型 1 错误率的大幅膨胀。总之，如果涉及 UN 结构的收敛问题发生在小型临床试验中，我们建议在 MMRM 分析中使用 MD。具有一阶自回归结构的 Kenward-Roger 方法在测量方差随访问增加的情况下导致类型 1 错误率的大幅膨胀。总之，如果涉及 UN 结构的收敛问题发生在小型临床试验中，我们建议在 MMRM 分析中使用 MD。