生存资料回归模型分析--Cox比例风险假设的图形检验法

本文目的是介绍目前使用图形检验比例风险的常用方法。经典的Cox比例风险回归模型要求生存资料满足比例风险假设,而在临床资料中,这个假设往往并不成立。鉴于此,本文首先阐述了比例风险假设的概念;然后介绍了一些检验比例风险假设是否成立的常用图示方法,主要包括Kaplan-Meier生存曲线图、ln[-ln(St)]生存时间关系图、缩放Schoenfeld残差与时间的关系图、SAS软件PHREG过程中ACCESS语句的PH和RESAMPLE选项产生的模拟路径图;最后,基于SAS软件并通过实例演示上述方法的实现。     

生存资料回归模型分析——生存资料Cox比例风险回归模型分析

本文目的是介绍生存资料Cox比例风险回归模型分析的概念、作用及使用SAS软件实现计算的方法。首先介绍相关基本概念,包括"Cox比例风险回归模型简介""模型假定及其检验""参数解释"和"参数估计与假设检验";然后通过一个实例并基于SAS软件演示如何实施生存资料Cox比例风险回归模型分析,内容包括"产生SAS数据集""绘制生存曲线图""判断PH假定是否成立"和"算出参数估计值与假设检验结果"。结果表明:当生存资料满足PH假定时,Cox比例风险回归模型可用于生存资料影响因素分析、校正混杂因素后的组间比较以及对每个个体进行预后指数和生存率的预测。     

生存资料回归模型分析——生存资料参数回归模型分析SAS实现

本文目的是介绍生存资料参数回归模型的SAS实现,包括创建SAS数据集、依据图示法选择模型、拟合参数模型和似然比检验。利用SAS中的LIFEREG过程绘制生存函数关于生存时间的关系图,拟合对应的参数分布回归模型,通过拟合优度检验选择最优的参数回归模型,最后对相关结果进行解释。     

生存资料回归模型分析——生存资料参数回归模型分析基础

本文目的是介绍生存资料参数回归模型有关的基础知识。首先,介绍了构建三个常见的生存资料参数回归模型的基本原理,包括指数分布回归模型、Weibull分布回归模型和Log-logistic分布回归模型;其次,介绍了基于图示法判断生存时间服从何种概率分布的方法;最后,介绍了基于最大似然估计法求解参数回归模型中的参数和两个参数回归模型拟合优度的比较。得到如下结论:①当资料中的生存时间服从特定概率分布时,应选用相应的参数回归模型;②图示法可用于粗略判断生存时间服从何种概率分布;③似然比检验可用于包含不同参数数目的两个参数回归模型之间拟合优度的比较。     

如何正确运用χ~2检验——高维表资料相对危险度分析与SAS实现

本文目的是介绍g×2×2表资料的相对危险度分析方法和基于SAS软件实现计算的方法。内容包括以下几个方面:其一,g×2×2表资料的齐性检验;其二,当资料满足齐性要求时,基于校正的方法对共同相对危险度进行点估计和置信区间估计;其三,当资料不满足齐性要求时,基于DerSimonian-Laird法(简称DL法)估计共同相对危险度RRDL及其95%置信区间;其四,资料满足与不满足齐性要求时,检验"共同相对危险度是否等于1"。本文结合两个实例并基于SAS软件,分别实现了针对相对危险度的齐性检验、资料满足齐性要求与不满足齐性要求条件下,共同相对危险度的点估计及其置信区间估计,并对SAS输出结果进行了解释,做出统计结论和专业结论。     

如何正确运用χ~2检验——高维表资料齐性检验与SAS实现

本文目的是介绍一种特殊高维表(即g×2×2表)资料优势比的齐性检验方法及SAS实现。在SAS/STAT的FREQ过程中,详细介绍了5种优势比齐性检验方法,分别是"Breslow-Day检验""Breslow-Day-Tarone检验""Q检验""I~2度量统计量及其不确定性限值"和"Zelen?s精确检验"。另外,还介绍了相对危险度与危险率差的齐性检验方法。文章结合两个实例,介绍了基于SAS软件实现优势比齐性检验的具体内容,对输出结果进行了解释,并做出了统计结论和专业结论。     

如何正确运用χ2检验——横断面设计四格表资料的χ2检验

本文目的是介绍横断面设计四格表资料的χ²检验及SAS和R软件实现。具体介绍了3种方法,即“Pearson?sχ²检验”“校正的Pearson?sχ²检验”和“似然比χ²检验”。内容涉及前述提及的3种统计分析方法的检验假设、计算原理与公式、前提条件、SAS与R软件的实现、结果解释以及结论陈述。     

如何正确运用χ~2检验——高维表资料独立性检验与SAS实现

本文目的是介绍一种特殊高维表(即g×2×2表)资料的独立性检验方法及SAS实现。在SAS软件和统计学教科书中,有三种方法可用于进行高维表资料的独立性检验,分别为广义CMHχ~2检验(简称为"方法1")、公式中含有权重系数的加权χ~2检验(简称为"方法2")和公式中没有权重系数的加权χ~2检验(简称为"方法 3")。本文通过公式推导和变形,揭示出"方法2"与"方法 3"在本质上是完全相同的加权χ~2检验,但具有不同的表现形式;还揭示出加权χ~2检验统计量的估计值与"方法1"中的CMHχ~2检验统计量的估计值近似相等。本文结合一个实例,介绍基于SAS软件实现g×2×2表资料独立性检验的具体方法,对输出结果进行解释,并做出统计结论和专业结论。     

如何正确运用Z检验——总体方差已知时两算术均值比较一般差异性Z检验及SAS实现

本文的目的是介绍总体方差已知时两算术均值比较一般差异性Z检验及SAS实现。首先，介绍了选用Z检验的前提条件；其次，介绍了单组设计和成组设计条件下进行均值比较的Z检验统计量，并介绍了上单侧检验、下单侧检验和双侧检验时所对应的拒绝域；最后，结合两个实例，基于SAS软件实现3种Z检验，对输出结果做出了解释。并针对如何合理选择单侧检验还是双侧检验给予说明。     

如何正确运用t检验——两算术均值比较非劣效性t检验及SAS实现

本文主要介绍临床试验中非劣效性检验的概念、原理和作用以及成组设计一元定量资料非劣效性检验的SAS实现。通过实例展示SAS在非劣效性检验中的应用,分别基于原始的定量数据或者基于给定样本含量、均数、标准差两种数据结构下进行操作,对结果进行解释并作出结论。     

Proportional Hazards Modeling of Saccadic Response Times During Reading

In this article we use proportional hazards models to examine how low‐level processes affect the probability of making a saccade over time, through the period of fixation, during reading. We apply the Cox proportional hazards model to investigate how launch distance (relative to word beginning), fixation location (relative to word center), and word frequency affect the hazard of a saccadic response. This model requires that covariates have a constant impact on the hazard over time, the assumption of proportional hazards. We show that this assumption is not supported. The impact of the covariates changes with the time passed since fixation onset. To account for the non‐proportional hazards we fit step functions of time, resulting in a model with time‐varying effects on the hazard. We evaluate the ability to predict the timing of saccades on held‐out fixation data. The model with time‐varying effects performs better in predicting the timing of saccades for fixations as short as 100 ms and as long as 500 ms, when compared both to a baseline model without covariates and a model which assumes constant covariate effects. This result suggests that the time‐varying effects model better recovers the time course of low‐level processes that influence the decision to move the eyes.     

Intérêt de la proprioception du tronc et de membre supérieur au moyen de « l’épreuve nez et doigt »

The capacities to keep upright or to move the trunk involve significant functional consequences for cerebral palsy patients. The quality of information related to the perception of the body position and how they are then treated contribute to the execution of the movement. Yet, they have not been much explored until now. The finger-and-nose test was created by M. Le Metayer as an assessment method. In this paper, we present the first results obtained with cerebral palsy patients using these testing methods.     

生存资料回归模型分析——生存资料及其统计分析方法概述

本文目的是全面介绍生存资料的特点及其常用统计分析方法。生存资料具有以下四个特点:①同时具有生存结局和生存时间;②生存时间可能含有删失数据或截尾数据;③生存时间的分布通常不服从正态分布,常呈指数分布、Weibull分布、对数正态分布;④影响生存时间的因素较复杂且不易控制。生存资料统计分析方法涉及统计描述、差异性分析和回归分析三大类,其中,统计描述主要有Kaplan-Meier(卡普兰-迈耶)估计法和Life table(寿命表)估计法;差异性分析主要有对数秩检验(log-rank test)和威尔考克森检验(Wilcoxon test);而回归分析主要有Cox比例和非比例风险回归模型、参数回归模型。在对生存资料进行统计分析时,需要合理选择统计分析方法,方可全面而又深入地揭示生存资料的内在变化规律。     

Testing for a deficit in single-case studies: effects of departures from normality

In neuropsychological single-case research inferences concerning a patient's cognitive status are often based on referring the patient's test score to those obtained from a modestly sized control sample. Two methods of testing for a deficit (z and a method proposed by Crawford and Howell [Crawford, J. R. & Howell, D. C. (1998). Comparing an individual's test score against norms derived from small samples. The Clinical Neuropsychologist, 12, 482-486]) both assume the control distribution is normal but this assumption will often be violated in practice. Monte Carlo simulation was employed to study the effects of leptokurtosis and the combination of skew and leptokurtosis on the Type I error rates for these two methods. For Crawford and Howell's method, leptokurtosis produced only a modest inflation of the Type I error rate when the control sample N was small-to-modest in size and error rates were lower than the specified rates at larger N. In contrast, the combination of leptokurtosis and skew produced marked inflation of error rates for small Ns. With a specified error rate of 5%, actual error rates as high as 14.31% and 9.96% were observed for z and Crawford and Howell's method respectively. Potential solutions to the problem of non-normal data are evaluated.     

Methods for estimating release rates during high frequency quantal secretion and for testing such methods

The rate of spontaneous quantal release must be estimated in some investigations of synaptic transmission, even when frequencies are so high that individual quanta cannot be distinguished. An obvious method is to measure the time integrals of the summed MEPPs and then dividing this value by the integral of an average MEPP. The method was tested by recording miniature end-plate potentials (MEPPs) at frog neuromuscular junctions, counting the number of MEPPS, and then adding together records from the same junction to simulate high frequencies. The estimates from the integral method agreed well with the actual counts. The method can readily be used with a microcomputer and does not require stationary data. Methods based on fluctuation analysis were also used to estimate quantal frequencies, but they did not always give good estimates. This was not a thorough test of the fluctuation method, but an example of testing with MEPP data. The integral method might be reasonably reliable, but there are further potential complications, like changes in MEPP size and short-circuiting of the end-plate membrane, which may make it difficult to obtain reliable measurements of high frequency quantal secretion without voltage clamping and protocols that permit measurements of individual MEPCs during the course of the frequency measurements.     

Adaptive cognitive testing in dementia

The objective of this study was to investigate whether the Cambridge Cognitive Examination (CAMCOG), a widely used screening test for dementia, can be tailored to the individual patient with Computerized Adaptive Testing (CAT). CAT accomplishes this by only using items that are appropriate for the level of ability of the patient under investigation. Potential advantages of CAT for clinical practice and research are efficient cognitive testing and a reduction of the test burden in elderly patients and consequently less measurement error during testing. In a two step method with previously collected CAMCOG data (n = 797) (1) patient abilities and CAMCOG item difficulties were estimated with the One Parameter Logistic Model (OPLM), a Rasch type of model. CAT was then used (2) to re‐estimate the patient abilities. Despite an average test reduction of 60%, CAT estimates were in excellent agreement (intra‐class correlation >0.98) with the results based on the entire CAMCOG and they also had similar accuracy for the diagnosis of dementia (area under the curve 0.91) as the original CAMCOG. These results were replicated in an independent sample (n = 170). We conclude that tailored testing with CAT enables much more efficient screening for dementia than testing with an extensive instrument. Copyright © 2009 John Wiley & Sons, Ltd.     

Schizophrenia: gender and familial risk.

The morbid risks for schizophrenia and any nonaffective psychosis in the first degree relatives of male and female schizophrenic probands were compared utilizing Cox proportional hazards models. The schizophrenic probands (275 male; 106 female) were drawn from a larger sample of hospitalized patients obtained by systematically screening all psychiatric admissions to 15 facilities over a six-year period. Proband diagnoses (DSM-III) were based on a direct assessment of the patient and a review of medical records. The family history method was used to obtain information about the first degree relatives of the probands. Cox proportional hazards models were adjusted for duration of illness of the proband and gender of the relatives. First degree relatives of female probands had significantly higher morbid risks for schizophrenia and nonaffective psychosis than relatives of male probands. The differential risk for schizophrenia in the relatives of male and female probands demonstrated in this study, as well as others, suggests that males and females may be at different risk for subtypes of the disorder.