分组Cox模型及其在癌症预后因素研究中的应用

Ｃｏｘ比例危险回归模型是医学随访研究、临床试验研究中分析生存资料最常用的多因素分析方法，但它不适合于处理分组生存资料或重叠严重的大样本生存数据。笔者对分组比例危险回归模型及其在大样本寿命表生存资料分析中的应用进行了讨论。最后结合实例借助于ＧＬＩＭ软件探讨它在肺癌随访资料预后因素分析中的应用。     

多水平模型和潜变量增长曲线模型在纵向数据分析中的应用及比较

比较多水平模型和潜变量增长曲线模型在纵向数据分析中的应用。文中以结直肠癌患者术后的生命质量情况为实例，比较两种方法的异同。结果表明两方法的参数估计值结果非常接近，多水平模型在模型构建时较为容易，而潜变量增长曲线模型在模型评价等方面具有优势。两方法均可很好地分析纵向观测的数据，且各有优点，研究者应根据需要选择合适的方法分析数据。     

神经网络模型在纵向随访数据分析中的研究进展

流行病学随访研究中会产生大量的纵向数据,但该类数据的分析一直是统计学的难点。目前大多采用传统的线性混合模型来处理。该模型对数据的分布要求严格,且假定疾病是线性进展的,在应用过程中会受到一定限制。近年来,有研究者提出采用神经网络模型来处理纵向随访数据。本文就神经网络模型在纵向随访数据分析中的研究现状进行探讨,为纵向随访数据的分析提供一个新思路。     

纵向数据潜变量增长曲线模型及其在Mplus中的实现

探讨纵向数据潜变量增长曲线模型及其在Mplus中的实现方法。通过实例采用Mplus软件处理某高校大学生心理健康状况纵向数据。结果表明潜变量增长曲线模型可以处理含有潜变量的纵向数据，能够比较总体发展趋势和个体发展的差异，纳入协变量可以提高模型拟合效果；采用Mplus软件实现潜变量增长曲线模型，程序简单，操作方便。纵向数据潜变量增长曲线模型及其在Mplus中的实现程序，可为实际应用尤其是流行病学队列研究提供统计方法学方面的指导和参考。     

潜在类别混合模型及其在纵向数据轨迹分析中的应用北大核心CSCD

目的介绍潜在类别混合模型及其在纵向数据轨迹分析中的应用。方法以一项限制能量摄入的随机对照临床试验为例,应用潜在类别混合模型进行轨迹分析,结合贝叶斯信息准则、平均后验概率及高后验概率个体所占比例判断最佳轨迹数目及形状。结果四组三次模型最优,人群分为四类减重模式:高体重快速减重组、低体重快速减重组、高体重缓慢减重组及对照组。结论潜在类别混合模型既能识别群体中的异质性,又能考虑到类别内个体发展轨迹,有望广泛应用于纵向数据的处理分析。     

基于机器学习与生存模型建立脂肪肝Joint联合预测模型

目的 利用机器学习算法与生存模型建立脂肪肝Joint联合预测模型，为有关单位进行脂肪肝健康管理提供理论依据。方法 选取2006—2016年某人群体检数据为研究对象，据模拟实验结果选择机器学习方法建立纵向亚模型，利用时依Cox模型建立生存亚模型，再联合建模。结果 XGBoost算法F-measure值最大，均方误差最小，建立纵向亚模型。XGBoost-Joint联合模型稳定性和拟合效果优于其他组合的Joint模型。结论 Joint联合模型将纵向亚模型与生存亚模型相联系，关联变量在模型中对脂肪肝患病风险的影响明显上升，说明纵向过程对生存结果的影响很大。     

样本选择模型的似然估计与两步估计

[目的]介绍样本选择模型及其估计方法。[方法]应用样本选择模型对模拟数据进行分析,并与传统线性回归模型进行比较。[结果]对于特定条件下的缺失数据,样本选择模型比传统的线性回归模型更合理,并可以提高模型参数估计的效率。[结论]样本选择模型在医学研究中拥有广泛的应用前景。     

似乎不相关回归模型及其在老年认知功能研究中的应用

目的 探讨似乎不相关回归模型在医学研究中的应用。方法 利用似乎不相关回归模型对老年人认知功能减退的影响因素进行分析。结果 特定的非独立结构情形下似乎不相关回归估计方法比传统最小二乘估计方法更合理，并可以提高参数估计的效率。结论 似乎不相关回归模型在医学科研领域具有良好的应用前景。     

零膨胀负二项回归模型在医学研究中的应用

在医学研究中存在许多零膨胀数据,传统模型对其进行拟合时容易发生模型不收敛、参数估计与实际情况偏差较大等问题。零膨胀负二项回归模型是近些年提出的处理零膨胀数据的一种方法,在医学研究中获得了极大的关注和应用。本文阐述了零膨胀负二项回归模型的原理及使用方法,梳理了该模型在慢性病研究、传染病研究、临床实验研究、卫生服务调查研究等医学研究中的应用,并对其在应用中的不足之处以及研究趋势进行探讨,以便医学科研工作者能更好地利用零膨胀负二项回归模型对零膨胀计数资料进行统计分析。     

广义估计方程与多水平logistic回归模型在临床纵向数据分析中的应用及比较

目的探讨广义估计方程和多水平模型的应用与临床纵向研究以解决个体重复观测数据内部的相关性问题。方法根据临床纵向实例数据的特点,拟合因变量为二分类的广义估计方程和多水平模型,并与一般logistic模型比较。结果广义估计方程和多水平模型的分析结果与一般logistic模型不同。由于未能考虑个体内重复观测数据的相关性,一般logistic模型错误显示临床分期与近期疗效相关,而广义估计方程和多水平模型分析结果则显示相关无统计学意义。经分层分析也未发现临床分期与近期疗效的关联。结论广义估计方程和多水平模型都能有效地考虑重复观测数据内部相关性并能处理有缺失值的资料。与多水平模型相比,广义估计方程的参数估计较为稳定,可有效的估计各解释变量的效应。     

Assessment of Studies Evaluating Incremental Costs,Effectiveness, or Cost-Effectiveness of Systemic Therapies in Breast Cancer Based on Claims Data: A Systematic Review

ObjectivesLarge secondary databases, such as those containing insurance claims data, are increasingly being used to compare the effects and costs of treatments in routine clinical practice. Despite their appeal, however, caution must be exercised when using these data. In this study, we aimed to identify and assess the methodological quality of studies that used claims data to compare the effectiveness, costs, or cost-effectiveness of systemic therapies for breast cancer.MethodsWe searched Embase, the Cochrane Library, Medline, Web of Science, and Google Scholar for English-language publications and assessed the methodological quality using the Good Research for Comparative Effectiveness principles. This study was registered with the International Prospective Register of Systematic Reviews (PROSPERO) under number CRD42018103992.ResultsWe identified 1251 articles, of which 106 met the inclusion criteria. Most studies were conducted in the United States (74%) and Taiwan (9%) and were based on claims data sets (35%) or claims data linked to cancer registries (58%). Furthermore, most included large samples (mean 17 130 patients) and elderly patients, and they covered various outcomes (eg, survival, adverse events, resource use, and costs). Key methodological shortcomings were the lack of information on relevant confounders, the risk of immortal time bias, and the lack of information on the validity of outcomes. Only a few studies performed sensitivity analyses.ConclusionsMany comparative studies of cost, effectiveness, and cost-effectiveness have been published in recent decades based on claims data, and the number of publications has increased over time. Despite the availability of guidelines to improve quality, methodological issues persist and are often inappropriately addressed or reported.     

医学科研数据仓库的建设

应用面向主题的数据仓库技术,将我院现有的电子化业务数据中有关临床医学的海量数据进行整合,以集成的电子化数据形成医学科研的专用数据仓库。利用该独立的医学科研数据仓库,使得科研人员可对历史的全部海量数据进行在线的、实时的、深入的分析,挖掘和共享临床科研数据,帮助提高各级医学技术人员的科研能力和效率。     

Picking up the pieces—Applying the DISEASE FILTER to health data

This contribution presents systematic biases in the process of generating health data by using a step-by-step explanation of the DISEASE FILTER, a heuristic instrument that we designed in order to better understand and evaluate health data. The systematic bias in health data generally varies by data type (register versus survey data) and the operationalization of health outcomes. Self-reported subjective health and disease assessments, for instance, underlie a different selectivity than do data based on medical examinations or health care statistics. Although this is obvious, systematic approaches used to better understand the process of generating health data have been missing until now. We begin with the definitions and classifications of diseases that change (e.g. over time), describe the selective nature of access to and use of medical health care (e.g. depending on health insurance and gender), present biases in diagnoses (e.g. by gender and professional status), report these biases in relation to the decision for or against various treatment (e.g. by age and income), and finally outline the determinants of the treatments (ambulant versus stationary, e.g. via mobility and age). We then show how to apply the DISEASE FILTER to health data and discuss the benefits and shortcomings of our heuristic model. Finally, we give some suggestions on how to deal with biases in health data and how to avoid them.     

Visualizing the quality of partially accruing data for use in decision making

Secondary use of clinical health data for near real-time public health surveillance presents challenges surrounding its utility due to data quality issues. Data used for real-time surveillance must be timely, accurate and complete if it is to be useful; if incomplete data are used for surveillance, understanding the structure of the incompleteness is necessary. Such data are commonly aggregated due to privacy concerns. The Distribute project was a near real-time influenza-like-illness (ILI) surveillance system that relied on aggregated secondary clinical health data. The goal of this work is to disseminate the data quality tools developed to gain insight into the data quality problems associated with these data. These tools apply in general to any system where aggregate data are accrued over time and were created through the end-user-as-developer paradigm. Each tool was developed during the exploratory analysis to gain insight into structural aspects of data quality. Our key finding is that data quality of partially accruing data must be studied in the context of accrual lag—the difference between the time an event occurs and the time data for that event are received, i.e. the time at which data become available to the surveillance system. Our visualization methods therefore revolve around visualizing dimensions of data quality affected by accrual lag, in particular the tradeoff between timeliness and completion, and the effects of accrual lag on accuracy. Accounting for accrual lag in partially accruing data is necessary to avoid misleading or biased conclusions about trends in indicator values and data quality.     

A Virtual Data Repository Stimulates Data Sharing in a Consortium

Research data may have substantial impact beyond the original study objectives. The Collaborating Consortium of Cohorts Producing NIDA Opportunities (C3PNO) facilitates the combination of data and access to specimens from nine NIDA-funded cohorts in a virtual data repository (VDR).Unique challenges were addressed to create the VDR. An initial set of common data elements was agreed upon, selected based on their importance for a wide range of research proposals. Data were mapped to a common set of values. Bioethics consultations resulted in the development of various controls and procedures to protect against inadvertent disclosure of personally identifiable information. Standard operating procedures govern the evaluation of proposed concepts, and specimen and data use agreements ensure proper data handling and storage.Data from eight cohorts have been loaded into a relational database with tables capturing substance use, available specimens, and other participant data. A total of 6,177 participants were seen at a study visit within the past six months and are considered under active follow-up for C3PNO cohort participation as of the third data transfer, which occurred in January 2020. A total of 70,391 biospecimens of various types are available for these participants to test approved scientific hypotheses. Sociodemographic and clinical data accompany these samples.The VDR is a web-based interactive, searchable database available in the public domain, accessed at www.c3pno.org. The VDR are available to inform both consortium and external investigators interested in submitting concept sheets to address novel scientific questions to address high priority research on HIV/AIDS in the context of substance use.     

面向HIS的数据综合仓库体系结构

目的：研究设计面向HIS的数据综合仓库体系。方法：通过研制专用的数据综合仓库、数据综合集市及脚本化扩充类XML（SQL）接口等构建面向HIS的数据综合仓库体系。结果：采用此体系结构可以达到较理想的应用效果。结论：以数据综合集市、数据综合仓库等专用智能积木箱为主的体系结构的构建是可行的，可以适用于面向HIS的医学数据挖掘和处理。     

医院数据系统的安全性备份方案实施

本文分析了医院数据系统可能存在的安全风险,运用ORACLE数据库提供的数据备份和恢复工具,提出了一些针对性的数据备份和数据恢复的方法,建立一套简单实用的医院数据系统高速、有效的备份方案,使医院数据系统运行安全稳定。     

维护我院HIS数据安全的措施

本文分析了影响HIS数据安全的硬件和软件因素,并指出相应的维护措施。