Home About us Contact
|
Home About us Contact | ||
|
|
||
Within-subject Correlation (within-subject + correlation)
Selected AbstractsPenalized spline models for functional principal component analysisJOURNAL OF THE ROYAL STATISTICAL SOCIETY: SERIES B (STATISTICAL METHODOLOGY), Issue 1 2006Fang Yao Summary., We propose an iterative estimation procedure for performing functional principal component analysis. The procedure aims at functional or longitudinal data where the repeated measurements from the same subject are correlated. An increasingly popular smoothing approach, penalized spline regression, is used to represent the mean function. This allows straightforward incorporation of covariates and simple implementation of approximate inference procedures for coefficients. For the handling of the within-subject correlation, we develop an iterative procedure which reduces the dependence between the repeated measurements that are made for the same subject. The resulting data after iteration are theoretically shown to be asymptotically equivalent (in probability) to a set of independent data. This suggests that the general theory of penalized spline regression that has been developed for independent data can also be applied to functional data. The effectiveness of the proposed procedure is demonstrated via a simulation study and an application to yeast cell cycle gene expression data. [source] Knee pain reduces joint space width in conventional standing anteroposterior radiographs of osteoarthritic kneesARTHRITIS & RHEUMATISM, Issue 5 2002Steven A. Mazzuca Objective A suspected, but heretofore undemonstrated, limitation of the conventional weight-bearing anteroposterior (AP) knee radiograph, in which the joint is imaged in extension, for studies of progression of osteoarthritis (OA) is that changes in knee pain may affect extension, thereby altering the apparent thickness of the articular cartilage. The present study was undertaken to examine the effect of changes in knee pain of varying magnitudes on radiographic joint space width (JSW) in the weight-bearing extended and the semiflexed AP views, in which radioanatomic positioning of the knee was carefully standardized by fluoroscopy. Methods Fifteen patients with knee OA underwent a washout of their analgesic/nonsteroidal antiinflammatory drug (NSAID) agents (duration 5 half-lives), after which standing AP and semiflexed AP knee radiographs of both knees were obtained. Examinations were repeated 1,12 weeks later (median 4.5 weeks, mean 6.0 weeks), after resumption of analgesic/NSAID therapy. Knee pain was measured with the pain subscale of the Western Ontario and McMaster Universities Osteoarthritis (WOMAC) Index (Likert scale). JSW was measured with a pair of calipers and a magnifying lens. Mixed model analyses of variance were used to test the significance of changes in pain and JSW within and between 2 groups of knees with mild-to-moderate radiographic severity of OA: (a) "flaring knees," in which the patient rated standing knee pain as severe or extreme after the washout and in which pain decreased to any degree after resumption of analgesics and/or NSAIDs (n = 12) and (b) "nonflaring knees," in which standing knee pain was absent, mild, or moderate after the washout or did not decrease after resumption of treatment (n = 15). Results After reinstitution of treatment, WOMAC pain scores decreased significantly in both flaring and nonflaring knees (,44%; P < 0.0001 and ,18%; P < 0.01, respectively). After adjustment for the within-subject correlation between knees, mean JSW (±SEM) in the extended view of the flaring OA knee increased significantly from the first to second examination (0.20 ± 0.06 mm; P = 0.005). In contrast, the change in adjusted mean JSW in the extended view of the nonflaring OA knee was negligible (,0.04 ± 0.04 mm) and significantly smaller than that observed in flaring knees (P < 0.01). Mean JSW in the semiflexed AP view was unaffected by the severity or responsiveness of standing knee pain in flaring and nonflaring OA knees. Conclusion JSW in weight-bearing extended-view radiographs of highly symptomatic OA knees can be altered significantly by changes in joint pain. In clinical trials and in epidemiologic studies of OA progression that use this radiographic technique, longitudinal variations in pain may confound changes in the apparent thickness of the articular cartilage. [source] Bayesian Inference in Semiparametric Mixed Models for Longitudinal DataBIOMETRICS, Issue 1 2010Yisheng Li Summary We consider Bayesian inference in semiparametric mixed models (SPMMs) for longitudinal data. SPMMs are a class of models that use a nonparametric function to model a time effect, a parametric function to model other covariate effects, and parametric or nonparametric random effects to account for the within-subject correlation. We model the nonparametric function using a Bayesian formulation of a cubic smoothing spline, and the random effect distribution using a normal distribution and alternatively a nonparametric Dirichlet process (DP) prior. When the random effect distribution is assumed to be normal, we propose a uniform shrinkage prior (USP) for the variance components and the smoothing parameter. When the random effect distribution is modeled nonparametrically, we use a DP prior with a normal base measure and propose a USP for the hyperparameters of the DP base measure. We argue that the commonly assumed DP prior implies a nonzero mean of the random effect distribution, even when a base measure with mean zero is specified. This implies weak identifiability for the fixed effects, and can therefore lead to biased estimators and poor inference for the regression coefficients and the spline estimator of the nonparametric function. We propose an adjustment using a postprocessing technique. We show that under mild conditions the posterior is proper under the proposed USP, a flat prior for the fixed effect parameters, and an improper prior for the residual variance. We illustrate the proposed approach using a longitudinal hormone dataset, and carry out extensive simulation studies to compare its finite sample performance with existing methods. [source] Bayesian Quantile Regression for Longitudinal Studies with Nonignorable Missing DataBIOMETRICS, Issue 1 2010Ying Yuan Summary We study quantile regression (QR) for longitudinal measurements with nonignorable intermittent missing data and dropout. Compared to conventional mean regression, quantile regression can characterize the entire conditional distribution of the outcome variable, and is more robust to outliers and misspecification of the error distribution. We account for the within-subject correlation by introducing a,,2,penalty in the usual QR check function to shrink the subject-specific intercepts and slopes toward the common population values. The informative missing data are assumed to be related to the longitudinal outcome process through the shared latent random effects. We assess the performance of the proposed method using simulation studies, and illustrate it with data from a pediatric AIDS clinical trial. [source] Self, in-group, and out-group evaluation: bond or breach?EUROPEAN JOURNAL OF SOCIAL PSYCHOLOGY, Issue 5 2003Reinout E. De Vries A number of studies have looked at causes of in-group bias, but few studies have actually investigated whether the two components of in-group bias, i.e. in-group and out-group evaluation, are related to each other and whether they have similar or different predictors. In the Fiji Islands, self-, in-group, and out-group evaluations were obtained using within-subject correlations from a sample of 336 indigenous and Indian Fijians. Self-evaluation was positively related to in-group evaluation, and both were positively related to out-group evaluation, supporting a spillover model. After controlling for background variables and the other evaluation variables, regression analyses showed that in-group identification was positively related to in-group evaluation, and social distance and political ethnocentrism were negatively related to out-group evaluation. Additionally, ethnicity interacted with collective self-esteem in determining both in-group favouritism and out-group derogation. Copyright © 2003 John Wiley & Sons, Ltd. [source] Whole-body bone scintigraphy provides a measure of the total-body burden of osteoarthritis for the purpose of systemic biomarker validationARTHRITIS & RHEUMATISM, Issue 11 2009Shelby Addison Objective To evaluate the association of serum and synovial fluid cartilage oligomeric matrix protein (COMP) with systemic and local measures of osteoarthritis (OA) activity by bone scintigraphy. Methods Samples of serum and knee joint synovial fluid (275 knees) were obtained from 159 patients with symptomatic OA of at least 1 knee. Bone scintigraphy using 99mTc-labeled methylene diphosphonate was performed, and early-phase knee scans and late-phase whole-body bone scans of 15 additional joint sites were scored semiquantitatively. To control for within-subject correlations of knee data, generalized linear modeling was used in the correlation of the bone scan scores with the COMP levels. Principal components analysis was used to explore the contribution of each joint site to the variance in serum COMP levels. Results The correlation between synovial fluid and serum COMP levels was significant (r = 0.206, P = 0.006). Synovial fluid COMP levels correlated most strongly with the early-phase knee bone scan scores (P = 0.0003), even after adjustment for OA severity according to the late-phase bone scan scores (P = 0.015), as well as synovial fluid volumes (P < 0.0001). Serum COMP levels correlated with the total-body bone scan scores (r = 0.188, P = 0.018) and with a factor composed of the bone scan scores in the shoulders, spine, lateral knees, and sacroiliac joints (P = 0.0004). Conclusion Synovial fluid COMP levels correlated strongly with 2 indicators of knee joint inflammation: early-phase bone scintigraphic findings and synovial fluid volume. Serum COMP levels correlated with total-body joint disease severity as determined by late-phase bone scintigraphy, supporting the hypothesis that whole-body bone scintigraphy is a means of quantifying the total-body burden of OA for systemic biomarker validation. [source] | ||||||||||