缺失二分类结局数据在频率派网络荟萃分析中排除、插补和建模方法的比较。

BACKGROUND: Missing participant outcome data (MOD) are ubiquitous in systematic reviews with network meta-analysis (NMA) as they invade from the inclusion of clinical trials with reported participant losses. There are available strategies to address aggregate MOD, and in particular binary MOD, while considering the missing at random (MAR) assumption as a starting point. Little is known about their performance though regarding the meta-analytic parameters of a random-effects model for aggregate binary outcome data as obtained from trial-reports (i.e. the number of events and number of MOD out of the total randomised per arm). METHODS: We used four strategies to handle binary MOD under MAR and we classified these strategies to those modelling versus excluding/imputing MOD and to those accounting for versus ignoring uncertainty about MAR. We investigated the performance of these strategies in terms of core NMA estimates by performing both an empirical and simulation study using random-effects NMA based on electrical network theory. We used Bland-Altman plots to illustrate the agreement between the compared strategies, and we considered the mean bias, coverage probability and width of the confidence interval to be the frequentist measures of performance. RESULTS: Modelling MOD under MAR agreed with exclusion and imputation under MAR in terms of estimated log odds ratios and inconsistency factor, whereas accountability or not of the uncertainty regarding MOD affected intervention hierarchy and precision around the NMA estimates: strategies that ignore uncertainty about MOD led to more precise NMA estimates, and increased between-trial variance. All strategies showed good performance for low MOD (<5%), consistent evidence and low between-trial variance, whereas performance was compromised for large informative MOD (> 20%), inconsistent evidence and substantial between-trial variance, especially for strategies that ignore uncertainty due to MOD. CONCLUSIONS: The analysts should avoid applying strategies that manipulate MOD before analysis (i.e. exclusion and imputation) as they implicate the inferences negatively. Modelling MOD, on the other hand, via a pattern-mixture model to propagate the uncertainty about MAR assumption constitutes both conceptually and statistically proper strategy to address MOD in a systematic review.

背景：在包含有报告参与者损失的临床试验的系统综述与网络荟萃分析（NMA）中，缺失的参与者结局数据（MOD）普遍存在。目前有多种策略可以处理聚合 MOD，特别是二项 MOD，同时考虑到缺失随机（MAR）假设作为起点。尽管如此，对于从试验报告中获得的随机效应模型汇总二项结局数据的荟萃分析参数（即每臂随机分配的总事件数和 MOD 数），关于这些策略的性能还知之甚少。

方法：我们使用四种策略在 MAR 下处理二项 MOD，并将这些策略分为对 MOD 进行建模与排除/插补以及对 MAR 不确定性进行处理与忽略的策略。我们通过使用基于电网络理论的随机效应 NMA 进行实证和模拟研究，来评估这些策略在核心 NMA 估计方面的性能。我们使用 Bland-Altman 图来说明比较策略之间的一致性，并将平均偏差、覆盖率概率和置信区间的宽度作为性能的频率测量。

结果：在 MAR 下对 MOD 进行建模与在 MAR 下排除和插补 MOD 的结果在估计的对数优势比和不一致性因子方面一致，而是否考虑 MOD 不确定性则会影响干预层级和 NMA 估计的精度：忽略 MOD 不确定性的策略会导致更精确的 NMA 估计，并增加试验间方差。所有策略在低 MOD（<5%）时表现良好，具有一致的证据和低试验间方差，而在高 MOD（>20%）时表现不佳，具有不一致的证据和较大的试验间方差，特别是对于忽略 MOD 不确定性的策略。

结论：分析人员应避免在分析前应用处理 MOD 的策略（即排除和插补），因为这些策略会对推论产生负面影响。另一方面，通过模式混合模型对 MAR 假设的不确定性进行建模，构成了在系统综述中处理 MOD 的概念上和统计学上正确的策略。