A novel methodological framework for multimodality, trajectory model-based prognostication
A novel methodological framework for multimodality, trajectory model-based prognostication
复制标题
DOI：
10.1016/j.resuscitation.2019.02.030
复制
发表时间：
2019-04-01
期刊：
影响因子：
6.5
通讯作者：
Nagin, Daniel
中科院分区：
文献类型：
--
作者：
Elmer, Jonathan;Jones, Bobby L.;Nagin, Daniel
预后工具通常结合几个时不变的临床预测指标，使用回归模型产生单一的时不变的结果预测。这将导致大量的信息丢失，因为重复或连续采样的数据被汇总到单个汇总度量中。我们描述了一种实时多变量结果预测方法，该方法既适应纵向数据，又适应时不变的临床特征。方法：我们纳入了心脏骤停复苏后接受>= 6 h脑电图监测的昏迷患者。我们使用Persyst v13 （Persyst Development Corp, Prescott AZ）生成定量脑电图（qEEG）特征，并计算全脑抑制比和振幅积分脑电图的每小时汇总。我们随机选择一半的受试者作为训练样本，另一半作为测试样本。对训练样本采用基于组的轨迹模型（GBTM），根据qEEG演化对患者进行分组，然后利用logistic回归估计组隶属度和临床协变量与昏迷苏醒、存活到出院之间的关系。我们利用这些参数计算检验样本的群体隶属性后验概率（PPGMs），并建立三种预后模型：调整后的逻辑回归（无GBTM）、未调整的GBTM（无临床协变量）和调整后的GBTM（所有数据）。我们比较了这些模型的性能特征。结果：纳入723例患者。7组GBTM的组特异性结局估计值从0到75%不等。与未调整的GBTM相比，调整后的GBTM校准在6和12小时以及到达结果估计的时间显著改善
Introduction: Prognostic tools typically combine several time-invariant clinical predictors using regression models that yield a single, time-invariant outcome prediction. This results in considerable information loss as repeatedly or continuously sampled data are aggregated into single summary measures. We describe a method for real-time multivariate outcome prediction that accommodates both longitudinal data and time-invariant clinical characteristics.Methods: We included comatose patients treated after resuscitation from cardiac arrest who underwent >= 6 h of electroencephalographic (EEG) monitoring. We used Persyst v13 (Persyst Development Corp, Prescott AZ) to generate quantitative EEG (qEEG) features and calculated hourly summaries of whole brain suppression ratio and amplitude-integrated EEG. We randomly selected half of subjects as a training sample and used the other half as a test sample. We applied group-based trajectory modeling (GBTM) to the training sample to group patients based on qEEG evolution, then estimated the relationship of group membership and clinical covariates with awakening from coma and surviving to hospital discharge using logistic regression. We used these parameters to calculate posterior probabilities of group membership (PPGMs) in the test sample, and built three prognostic models: adjusted logistic regression (no GBTM), unadjusted GBTM (no clinical covariates) and adjusted GBTM (all data). We compared these models performance characteristics.Results: We included 723 patients. Group-specific outcome estimates from a 7-group GBTM ranged from 0 to 75%. Compared to unadjusted GBTM, adjusted GBTM calibration was significantly improved at 6 and 12 h, and time to an outcome estimate