Haem Elham, Doostfatemeh Marziyeh, Firouzabadi Negar, Ghazanfari Nima, Karlsson Mats O
Department of Biostatistics, School of Medicine, Shiraz University of Medical Sciences, Shiraz, Iran.
Department of Pharmacology & Toxicology, School of Pharmacy, Shiraz University of Medical Sciences, Shiraz, Iran.
J Pharmacokinet Pharmacodyn. 2020 Jun;47(3):241-253. doi: 10.1007/s10928-020-09686-0. Epub 2020 Apr 13.
This manuscript aims to present the first item response theory (IRT) model within a pharmacometric framework to characterize the longitudinal changes of Aberrant Behavior Checklist (ABC) data in children with autism. Data were obtained from 120 patients, which included 20,880 observations of the 58 items for up to three months. Observed scores for each ABC item were modeled as a function of the subject's disability. Longitudinal IRT models with five latent disability variables based on ABC subscales were used to describe the irritability, lethargy, stereotypic behavior, hyperactivity, and inappropriate speech over time. The IRT pharmacometric models could accurately describe the longitudinal changes of the patient's disability while estimating different time-course of disability for the subscales. For all subscales, model-estimated disability was reduced following initiation of therapy, most markedly for hyperactivity. The developed framework provides a description of ABC longitudinal data that can be a suitable alternative to traditional ABC data collected in autism clinical trials. IRT is a powerful tool with the ability to capture the heterogeneous nature of ABC, which results in more accurate analysis in comparison to traditional approaches.
本手稿旨在提出药动学框架内的首个项目反应理论（IRT）模型，以描述自闭症儿童异常行为检查表（ABC）数据的纵向变化。数据来自120名患者，包括对58个项目长达三个月的20880次观察。每个ABC项目的观察分数被建模为受试者残疾情况的函数。基于ABC分量表的具有五个潜在残疾变量的纵向IRT模型用于描述易怒、嗜睡、刻板行为、多动和不当言语随时间的变化。IRT药动学模型在估计分量表不同残疾时间进程的同时，能够准确描述患者残疾情况的纵向变化。对于所有分量表，治疗开始后模型估计的残疾情况有所减轻，多动最为明显。所开发的框架提供了ABC纵向数据的描述，可作为自闭症临床试验中收集的传统ABC数据的合适替代方案。IRT是一种强大的工具，能够捕捉ABC的异质性，与传统方法相比，能进行更准确的分析。
