Working‐correlation‐structure identification in generalized estimating equations

Selecting an appropriate working correlation structure is pertinent to clustered data analysis using generalized estimating equations (GEE) because an inappropriate choice will lead to inefficient parameter estimation. We investigate the well‐known criterion of QIC for selecting a working correlation structure, and have found that performance of the QIC is deteriorated by a term that is theoretically independent of the correlation structures but has to be estimated with an error. This leads us to propose a correlation information criterion (CIC) that substantially improves the QIC performance. Extensive simulation studies indicate that the CIC has remarkable improvement in selecting the correct correlation structures. We also illustrate our findings using a data set from the Madras Longitudinal Schizophrenia Study. Copyright © 2008 John Wiley & Sons, Ltd.     

增龄过程中心脏径限的变化及其影响因素

为探讨年龄增长及伴随的一些生理性变化对心脏径限的影响,对300余例20—81岁(平均46±16岁)的正常成人进行了超声心动图研究。多元逐步回归及协方差分析表明,左室壁厚度随年龄俱增,而左室舒张未期内径变化不大;室间隔/左室后壁厚度比值亦渐有增长,反映了室间隔的随龄增长稍快于左室后壁,故在增龄过程中,左室流出道稍有缩小;左房、主动脉根部和右室内径均有随年龄而增长的变化。二尖瓣舒张期 EF 斜率随年龄增长渐有下降,可能反映了左室顺应性的降低及二尖瓣弹性的减退。此外,一些心脏测值有性别差异,不少径限尚与体格大小或血压水平有相关关系。对于不同年龄组间心脏径限差异的原因,以及心脏径限彼此间的关系进行了分析和讨论,认为在临床上评价心脏径限正常与否和异常程度时,应考虑年龄、性别、体格大小、血压等因素的影响。     

Simultaneous genetic analysis of longitudinal means and covariance structure in the simplex model using twin data

A longitudinal model based on the simplex model is presented to analyze simultaneously means and covariance structure using univariate longitudinal twin data. The objective of the model is to decompose the mean trend into components which can be attributed to those genetic and environmental factors which give rise to phenotypic individual differences and a component of unknown constitution which does not involve individual differences. Illustrations are given using simulated data and repeatedly measured weight obtained in a sample of 82 female twin pairs on six occasions.     

Mixed General Linear Model Analysis of Quantitative Electroencephalographic (qEEG) Data

ABSTRACT

This paper describes a mixed general linear analysis of the quantitative electroencephalogram (qEEG). The modeling is similar to regression, which builds a regression or ‘best-fit’ model for the data structure but, in addition, provides for correlations between observations. A mixed linear model states that data consists of two parts: fixed effects and random effects. Fixed effects determine the expected values of the observations, while random effects account for the stochastic deviations from these expected values both between and within individuals. Since errors are independent between subjects, the deviations from the expected values may also be modeled using a repeated measures approach. The term ‘repeated measures’ in this model refers to data with multiple observations from one specific source. It is reasonable to assume that these observations from the same source are correlated, even if only slightly, in some measurable way. Consequently, statistical analysis of repeated measures data gives a more accurate prediction capability when the issue of covariation between these measures is addressed.

With mixed model methodology now available (e.g., the mixed procedure [Mixed PROC] of the SAS® system), the covariance structure can be incorporated into the statistical model. Disregarding potential random effects not specific to single individuals and absorbing potential within-subject random effects into the covariance matrix allows one to work with a simplified model.

The use of a mixed procedure and its method of modeling the data structure appear to provide an accurate and objective method of analysis resulting in quantifiable equations for testing predictions. Essentially, this method allows the physiological pattern of each individual in the study, not related to any other variable, to be represented and accounted for in the model. Several comparative examples will be used to highlight the information that can be hidden in data structures depending on the type of statistical analysis used.     

Surgical myectomy versus alcohol septal ablation for obstructive hypertrophic cardiomyopathy: A propensity score–matched cohort

Objectives

In patients with hypertrophic cardiomyopathy, obstruction of the left ventricular outflow tract can be relieved by surgical septal myectomy or alcohol septal ablation, but uncertainty remains regarding long-term results and comparative effectiveness of alcohol septal ablation. This study aims to compare short- and long-term outcomes of the 2 procedures.

Decreased centrality of cortical volume covariance networks in autism spectrum disorders

Autism spectrum disorders (ASD) are a group of neurodevelopmental conditions characterized by atypical structural and functional brain connectivity. Complex network analysis has been mainly used to describe altered network-level organization for functional systems and white matter tracts in ASD. However, atypical functional and structural connectivity are likely to be also linked to abnormal development of the correlated structure of cortical gray matter. Such covariations of gray matter are particularly well suited to the investigation of the complex cortical pathology of ASD, which is not confined to isolated brain regions but instead acts at the systems level. In this study, we examined network centrality properties of gray matter networks in adults with ASD (n = 84) and neurotypical controls (n = 84) using graph theoretical analysis. We derived a structural covariance network for each group using interregional correlation matrices of cortical volumes extracted from a surface-based parcellation scheme containing 68 cortical regions. Differences between groups in closeness network centrality measures were evaluated using permutation testing. We identified several brain regions in the medial frontal, parietal and temporo-occipital cortices with reductions in closeness centrality in ASD compared to controls. We also found an association between an increased number of autistic traits and reduced centrality of visual nodes in neurotypicals. Our study shows that ASD are accompanied by atypical organization of structural covariance networks by means of a decreased centrality of regions relevant for social and sensorimotor processing. These findings provide further evidence for the altered network-level connectivity model of ASD.     

Differences in the Impact and Management of Heart Disease Between Older Women and Men

Findings are presented from an exploratory study of differences in the impact and mana ernent of heart disease between men and wornen aged 60 and over (N = 323). The sample consisted of ambulatory, non-institutionalized, cardiac patients who had agreed to participate in a self-management intervention study. Differences between men and women were found regarding sym tom experience, physical functioning, exercise compliance, use oPprescribed drugs, and physicianlclinic visits. Differences in emotional well-being and psychosocial functioning were not found; however, women reported more stress and limitat~on in the area of household management. Implications for future research and practice are discussed.     

Pravastatin restored the infarct size-limiting effect of ischemic preconditioning blunted by hypercholesterolemia in the rabbit model of myocardial infarction

OBJECTIVES

We tested to find out whether pravastatin restores the infarct size (IS)-limiting effect of ischemic preconditioning (IP) and if it has any effect on the IP-induced activation of adenosine producing enzyme ecto-5′-nucleotidase which plays a key role in the IP-induced cardioprotection.

BACKGROUND

The IS-limiting effect of IP is blunted by hypercholesterolemia. Recently, HMG-CoA reductase inhibitors are shown to have direct cytoprotective effects.

METHODS

Rabbits were fed with a normal or cholesterol (1%) added diet with or without pravastatin (5 mg/kg/day) treatment. Infarct size was measured after 30 min occlusion and 3 h reperfusion of circumflex coronary artery with or without the IP procedure (5 min occlusion and 10 min reperfusion). Additionally, ecto-5′-nucleotidase activities of ischemic and nonischemic myocardium were measured immediately after IP procedure.

RESULTS

This dose of pravastatin did not normalize the increased level of serum cholesterol. The IS-limiting effect of preceding IP (IS reduced from 36.7% to 9.6%, p < 0.001) was abolished by hypercholesterolemia (from 46.1% to 31.3%, p = NS) and restored by pravastatin treatment (from 35.2% to 9.4%, p < 0.001). Pravastatin treatment did not affect IS or the effect of IP under normocholesterolemia. The activation of ecto-5′-nucleotidase presented as the activity ratio of ischemic to nonischemic myocardium (3.1-fold in normocholesterolemia) was blunted by hypercholesterolemia (1.8-fold, p < 0.05) and restored by pravastatin treatment (2.9-fold).

CONCLUSIONS

Pravastatin, at the dose serum cholesterol was not normalized, restored the IS-limiting effect of IP and IP-induced ecto-5′-nucleotidase activation, which were both blunted by hypercholesterolemia. The activation of ecto-5′-nucleotidase may be worth further investigation as a possible mechanism for the hypercholesterolemia-induced retardation and pravastatin-mediated restoration of the cardioprotective effect of IP.