Detecting influential subjects in intensive longitudinal data using mixed-effects location scale models.

Detecting influential subjects in intensive longitudinal data using mixed-effects location scale models.
复制标题

DOI:
10.1186/s12874-023-02046-9
复制
发表时间:
2023-10-18
影响因子:
4
通讯作者:
--
中科院分区:
医学3区
文献类型:
--
作者:

文献摘要

参考文献

相似文献

密集的纵向健康结果的收集允许其平均值(位置)和可变性(规模)的联合建模。关注结果的位置,使用标准混合效应回归模型(MRM)在纵向数据中检测有影响力的主题的措施已被广泛讨论。然而,没有现有的方法能够检测严重影响结果规模的受试者。我们建议应用混合效应位置尺度(MELS)建模结合常用的影响措施,如库克距离和DFBETAS来填补这一空白。在本文中,我们提供了一个框架,研究人员遵循时,试图检测有影响力的主题的规模和位置的结果。该框架允许详细检查每个主题对模型拟合的影响,以及MELS模型不同组成部分中系数的点估计和精度。我们模拟了纵向医疗研究中的两种常见场景,发现我们框架中的影响力指标在99%以上的时间内成功捕获了有影响力的受试者。我们还重新分析了健康行为研究的数据,发现了4个特别有影响力的主题,其中两个无法通过常规MRM进行影响分析。所提出的框架可以帮助研究人员检测有影响力的主题,否则会被使用常规MRM的有影响力的分析所忽略,并在一个模型中分析所有数据,尽管有影响力的主题。在线版本包含补充材料,可通过10.1186/s12874-023-02046-9获得。
Collection of intensive longitudinal health outcomes allows joint modeling of their mean (location) and variability (scale). Focusing on the location of the outcome, measures to detect influential subjects in longitudinal data using standard mixed-effects regression models (MRMs) have been widely discussed. However, no existing approach enables the detection of subjects that heavily influence the scale of the outcome. We propose applying mixed-effects location scale (MELS) modeling combined with commonly used influence measures such as Cook’s distance and DFBETAS to fill this gap. In this paper, we provide a framework for researchers to follow when trying to detect influential subjects for both the scale and location of the outcome. The framework allows detailed examination of each subject’s influence on model fit as well as point estimates and precision of coefficients in different components of a MELS model. We simulated two common scenarios in longitudinal healthcare studies and found that influence measures in our framework successfully capture influential subjects over 99% of the time. We also re-analyzed data from a health behavior study and found 4 particularly influential subjects, among which two cannot be detected by influence analyses via regular MRMs. The proposed framework can help researchers detect influential subject(s) that will be otherwise overlooked by influential analysis using regular MRMs and analyze all data in one model despite influential subjects. The online version contains supplementary material available at 10.1186/s12874-023-02046-9.
DOI: 10.1249/01.mss.0000139806.53824.2e
发表时间: 2004-09-01
期刊: MEDICINE AND SCIENCE IN SPORTS AND EXERCISE
影响因子: --
作者:
Tudor-Locke, C;Ham, SA;Kimsey, CD
通讯作者: Kimsey, CD
DOI: 10.1037/a0028085
发表时间: 2012-09-01
影响因子: 7
作者:
Meade, Adam W.;Craig, S. Bartholomew
通讯作者: Craig, S. Bartholomew
DOI: 10.1111/1467-985x.00094
发表时间: 1998-01-01
影响因子: 2
作者:
Langford, IH;Lewis, T
通讯作者: Lewis, T
DOI: 10.1111/j.1541-0420.2007.00924.x
发表时间: 2008-06-01
期刊: BIOMETRICS
影响因子: 1.9
作者:
Hedeker, Donald;Mermelstein, Robin J.;Demirtas, Hakan
通讯作者: Demirtas, Hakan
DOI: 10.1080/00273171.2012.658328
发表时间: 2012-01-01
影响因子: 3.8
作者:
Rast, Philippe;Hofer, Scott M.;Sparks, Catharine
通讯作者: Sparks, Catharine