非配对设计多值有序资料一水平多重Logistic回归分析

本文目的是介绍非配对设计多值有序资料一水平多重logistic回归模型的构建与求解方法。本文详细介绍了构建累积logistic回归模型的原理和具体方法,并结合实例介绍如何使用SAS软件中的LOGISTIC过程来拟合此回归模型,并对逐步回归法的输出结果进行了解释;其次讨论了有关构建累积logistic回归模型的过程中自变量筛选、模型评价以及拟合模型时需注意的问题。     

非配对设计多值名义资料多水平多重Logistic回归模型

本文目的是介绍非配对设计多值名义资料多水平多重logistic回归模型的构建与求解方法。首先介绍了有关的基本概念,涉及“多值名义结果变量”“分层或多水平数据结构”和“扩展的多重logistic回归模型”;其次,呈现了一个具有二水平结构的横断面调查资料,该资料涉及多个影响因素和一个多值有序的结果变量（在本文中,将其视为多值名义结果变量）;最后,借助SAS中的两个过程（即GLIMMIX和NLMIXED）对给定的资料进行统计分析,即构建和求解“非配对设计多值名义资料多水平多重logistic回归模型”,并对相关结果进行比较和解释。     

复杂抽样调查设计多值名义资料一水平多重Logistic回归分析

本文目的是介绍复杂抽样调查设计多值名义资料一水平多重logistic回归模型构建，并探讨不同策略之间的差异。采用SAS中的LOGISTIC过程和SURVEYLOGISTIC过程，分别按照是否考虑抽样设计与是否考虑抽样权重共4种分析策略对数据构建广义logistic回归模型，并比较结果。不同分析策略所得结果显示，不仅参数估计值、回归系数标准误、OR值及其置信区间的估计值有所差别，而且对纳入模型的解释变量也有影响。因此，在对复杂抽样调查设计多值名义资料构建广义logistics回归模型时，既要考虑抽样设计，又要兼顾抽样权重，否则即使样本量足够大，也会导致错误的推断结论。     

复杂抽样调查设计多值有序资料一水平多重Logistic回归分析

本文目的是比较不同分析策略对复杂抽样调查设计多值有序资料一水平多重logistic回归分析结果的异同。通过实例分析,利用四种不同的分析策略(将复杂抽样视为单纯随机抽样,考虑抽样设计不考虑抽样权重,考虑抽样权重不考虑抽样设计,同时考虑抽样设计和抽样权重)对复杂抽样设计多值有序资料进行建模。在四种不同分析策略的累积logistic回归模型拟合的结果中,自变量的偏回归系数、标准误差及P值均有所不同。在对复杂抽样调查设计的多值有序资料回归建模时,将抽样设计和抽样权重纳入统计分析,会得到更准确、更稳健的分析结果。     

非配对设计多值名义资料一水平多重Logistic回归分析

本文目的是介绍非配对设计多值名义资料一水平多重logistic回归分析的基本原理、建模策略及注意事项。结合实例,应用SAS 9.4构建未经变量筛选和经变量筛选的多值名义资料多重logistic回归模型。通过回归分析的计算结果可知,同一变量的回归系数在不同logit函数中存在代数关系。多值名义多重logistic回归分析可以用来处理结果变量为多值名义变量的回归建模问题,并可以结合SAS实现对自变量的筛选,以获得简洁的回归模型。     

非配对设计二值资料多水平多重Logistic回归分析

本文目的是介绍非配对设计二值资料多水平多重logistic回归模型的构建与求解方法。首先介绍模型的有关概念及模型的构建原理,基于实例使用SAS软件对列联表资料进行分析,以proc glimmix和proc nlmixed过程构建和求解模型,并对相关结果进行解释和比较。     

合理进行多重Logistic回归分析——结合多水平模型分析

本文目的是介绍如何结合多水平模型分析,合理地进行多重Logistic回归分析的方法。第一,介绍了与多水平模型分析有关的4个基本概念。第二,介绍了构建多水平模型的3个步骤。第三,通过一个多中心药物临床试验的实例,介绍了如何用SAS软件进行分析的全过程,其内容如下：①检验各中心优势比之间是否具有齐性;②对试验中心产生哑变量后构建多重Logistic回归模型;③将试验中心视为分层变量构建多重Logistic回归模型;④构建随机截距多水平多重Logistic回归模型;⑤构建随机截距和随机斜率多水平多重Logistic回归模型。得到的结论是,当具有二值结果变量的各层级资料间存在差异时,最合适的做法是构建多水平多重Logistic回归模型。     

非配对设计二值资料一水平多重Logistic回归分析

本文的目的是介绍非配对设计二值资料一水平多重Logistic回归模型的构建与求解方法。基于SAS软件分别对以列联表和数据库形式呈现的定性资料进行全面分析,并得出了4个对提高模型拟合优度很有价值的结论:第一,若资料以列联表形式呈现,应拟合"加权"Logistic回归模型;第二,若资料中包含定量自变量,不适合将其定性化;第三,若资料中包含定量自变量,应依据定量自变量和二值自变量产生出派生自变量;第四,若资料中有定性自变量时,必须将多值名义或有序自变量进行哑变量变换,不需要依据二值自变量产生出派生自变量。     

提高回归模型拟合优度的策略(Ⅳ)——优化计分变换与其他变量变换

本文目的是介绍第四种提高回归模型拟合优度的策略,即优化计分变换与其他变量变换。具体方法包括以下几个方面:①第一,对多值名义自变量采取"优化计分变换";②对有序自变量分别采取"单调变换"与"优化计分变换";③对定量自变量分别采取"样条变换"和"单调样条变换";④对定量因变量分别采取"样条变换""单调样条变换"和"BOX-COX变换"。全部变量变换方法组合起来共12种,共创建了12个多重非线性回归模型。依据"拟合优度评价指标"的取值,从12个回归模型中挑选出一个,即本文中的"模型1",其"均方误差平方根=0.30935、R~2=0.9586、调整R~2=0.9527"。结合本期科研方法专题同类文章的结果和结论,得出提高回归模型拟合优度的策略主要在于以下四点:①应对"定量因变量""定量自变量"和"多值有序自变量"采取合适的变量变换方法;②在拟合回归模型的过程中,应尽可能多地引入派生变量;③应假定回归模型中不含截距项;④在构建回归模型的过程中,应尽可能多地使用筛选自变量的策略,如"前进法""后退法"和"逐步法"。     

零膨胀Poisson分布模型回归分析

本文目的是介绍零膨胀Poisson分布模型回归分析。首先,介绍零膨胀计数资料及其零膨胀Poisson分布回归模型构建原理,包括"零膨胀Poisson分布回归模型的形式"和"零膨胀Poisson分布回归模型的求解";其次,介绍"零膨胀Poisson分布回归模型的SAS实现",包括"创建SAS数据集""呈现因变量Y的频数分布""求出因变量Y的均值和方差"和"基于全部自变量对因变量Y构建多重零膨胀Poisson分布回归模型"。本文结果提示,当计数资料为非严重过离散的零膨胀计数资料时,拟合"多重零膨胀Poisson分布回归模型",可获得满意的拟合效果。     

A method for ordinal outcomes: The ordered stereotype model

Objective : The collection and use of ordinal variables are common in many psychological and psychiatric studies. Although the models for continuous variables have similarities to those for ordinal variables, there are advantages when a model developed for modeling ordinal data is used such as avoiding “floor” and “ceiling” effects and avoiding to assign scores, as it happens in continuous models, which can produce results sensitive to the score assigned. This paper introduces and focuses on the application of the ordered stereotype model, which was developed for modeling ordinal outcomes and is not so popular as other models such as linear regression and proportional odds models. This paper aims to compare the performance of the ordered stereotype model with other more commonly used models among researchers and practitioners. Methods : This article compares the performance of the stereotype model against the proportional odd and linear regression models, with three, four, and five levels of ordinal categories and sample sizes 100, 500, and 1000. This paper also discusses the problem of treating ordinal responses as continuous using a simulation study. The trend odds model is also presented in the application. Results : Three types of models were fitted in one real‐life example, including ordered stereotype, proportional odds, and trend odds models. They reached similar conclusions in terms of the significance of covariates. The simulation study evaluated the performance of the ordered stereotype model under four cases. The performance varies depending on the scenarios. Conclusions : The method presented can be applied to several areas of psychiatry dealing with ordinal outcomes. One of the main advantages of this model is that it breaks with the assumption of levels of the ordinal response are equally spaced, which might be not true.     

复杂抽样调查设计二值资料一水平多重Logistic回归分析

本文目的是介绍复杂抽样调查设计二值资料多重logistic回归分析方法。通过一个实例,利用八种不同的分析策略(不考虑抽样设计和抽样权重、考虑抽样设计不考虑抽样权重、不考虑抽样设计考虑抽样权重、同时考虑抽样设计和抽样权重以及分别不考虑与考虑派生变量)对数据进行建模。对所得结果进行比较得出如下结论:在对复杂抽样设计资料进行统计分析的过程中,同时考虑抽样设计和抽样权重可以得到符合数据内部变量间依赖关系真实情况的结论。此外,本研究还介绍了采用SAS软件中SURVEYLOGISTIC过程对复杂抽样调查数据进行多重Llogistic回归分析的详细步骤。     

Factors associated with utilization of healthcare resources among epilepsy patients

PURPOSE: To determine those variables associated with utilization of healthcare resources in epilepsy patients. METHODS: We interviewed 256 epilepsy patients. Target variables included the number of clinic visits, ER visits and in-patient admissions over the past year and AEDs currently being used. Predictor variables were age, race/ethnicity, marital status, education, income, insurance, seizure frequency and QOLIE-10 results. We used univariate analysis to determine those factors associated with the target variables and multivariate analysis to ascertain those independently significant. RESULTS: On univariate analysis, higher seizure frequency and poorer QOLIE-10 scores were associated with the number of clinic visits, ER visits and in-patient admissions. Increased seizure frequency and male gender were associated with higher use of AEDs. Using ordinal logistic regression, QOLIE-10 scores was the only variable associated with the number of clinic visits. Both seizure frequency and QOLIE-10 scores were independently associated with the number of in-patient admissions while seizure frequency and male gender remained independently associated with AED use. Using binary logistic regression, QOLIE-10 scores and seizure frequency were independently associated with the number of ER visits. CONCLUSION: Seizure frequency and quality of life are major factors associated with utilization of healthcare resources in epilepsy patients.     

Smoking and suicidality in patients with a psychotic disorder

Cigarette smoking has been associated with an increased risk of suicide. Patients with psychosis are more likely to smoke cigarettes and are also at an increased risk of suicide. The aim of this study was to compare risk for suicidal behavior among patients with psychosis who were current smokers, previous smokers and nonsmokers. We studied 1812 of the 1825 participants who took part in the Australian Survey of High Impact Psychosis (SHIP) for whom smoking data was available. We identified predictors for lifetime suicide attempts using univariate logistic regression analysis. These variables were retained for the multiple logistic regression models if they were a significant predictor of lifetime suicide attempts. A series of multiple logistic regressions were then conducted to predict lifetime suicide attempts using current smoking status and lifetime smoking status as independent variables, respectively, while controlling for the retained predictor variables. Current smoking and lifetime smoking were statistically significant predictors of lifetime suicide attempts. However adding the covariates to a logistic regression model reduced this association to non-significance. The strongest predictors were self-harm in the past 12 months, the presence of lifetime depressive symptoms and a diagnosis of psychotic depression. Identification of suicide risk factors is essential for successful suicide prevention. While previous research highlights the importance of cigarette smoking as an important risk factor for suicidal behaviors including in patients with psychosis, these results must be interpreted within the context of methodological issues.     

Artificial neural networks improve the prediction of mortality in intracerebral hemorrhage.

BACKGROUND: Artificial neural network (ANN) analysis methods have led to more sensitive diagnosis of myocardial infarction and improved prediction of mortality in breast cancer, prostate cancer, and trauma patients. Prognostic studies have identified early clinical and radiographic predictors of mortality after intracerebral hemorrhage (ICH). To date, published models have not achieved the accuracy necessary for use in making decisions to limit medical interventions. We recently reported a logistic regression model that correctly classified 79% of patients who died and 90% of patients who survived. In an attempt to improve prediction of mortality we computed an ANN model with the same data. OBJECTIVE: To determine whether an ANN analysis would provide a more accurate prediction of mortality after ICH when compared with multiple logistic regression models computed using the same data. METHODS: Analyses were conducted on data collected prospectively on 81 patients with supratentorial ICH. Multiple logistic regression was used to predict hospital mortality, then an ANN analysis was applied to the same data set. Input variables were age, gender, race, hydrocephalus, mean arterial pressure, pulse pressure, Glasgow Coma Scale score, intraventricular hemorrhage, hydrocephalus, hematoma size, hematoma location (ganglionic, thalamic, or lobar), cisternal effacement, pineal shift, history of hypertension, history of diabetes, and age. RESULTS: The ANN model correctly classified all patients (100%) as alive or dead compared with 85% correct classification for the logistic regression model. A second ANN verification model was equally accurate. The ANN was superior to the logistic regression model on all objective measures of fit. CONCLUSIONS: ANN analysis more effectively uses information for prediction of mortality in this sample of patients with ICH. A well-validated ANN may have a role in the clinical management of ICH.