Modeling repeated self-reported outcome data: a continuous-time longitudinal Item Response Theory model

Item Response Theory (IRT) models have received growing interest in health science for analyzing latent constructs such as depression, anxiety, quality of life or cognitive functioning from the information provided by each individual's items responses. However, in the presence of repeated item measures, IRT methods usually assume that the measurement occasions are made at the exact same time for all patients. In this paper, we show how the IRT methodology can be combined with the mixed model theory to provide a longitudinal IRT model which exploits the information of a measurement scale provided at the item level while simultaneously handling observation times that may vary across individuals and items. The latent construct is a latent process defined in continuous time that is linked to the observed item responses through a measurement model at each individual- and occasion-specific observation time; we focus here on a Graded Response Model for binary and ordinal items. The Maximum Likelihood Estimation procedure of the model is available in the R package lcmm. The proposed approach is contextualized in a clinical example in end-stage renal disease, the PREDIALA study. The objective is to study the trajectories of depressive symptomatology (as measured by 7 items of the Hospital Anxiety and Depression scale) according to the time from registration on the renal transplant waiting list and the renal replacement therapy. We also illustrate how the method can be used to assess Differential Item Functioning and lack of measurement invariance over time.

Bayesian Nonparametric Longitudinal Data Analysis

Practical Bayesian nonparametric methods have been developed across a wide variety of contexts. Here, we develop a novel statistical model that generalizes standard mixed models for longitudinal data that include flexible mean functions as well as combined compound symmetry (CS) and autoregressive (AR) covariance structures. AR structure is often specified through the use of a Gaussian process (GP) with covariance functions that allow longitudinal data to be more correlated if they are observed closer in time than if they are observed farther apart. We allow for AR structure by considering a broader class of models that incorporates a Dirichlet Process Mixture (DPM) over the covariance parameters of the GP. We are able to take advantage of modern Bayesian statistical methods in making full predictive inferences and about characteristics of longitudinal profiles and their differences across covariate combinations. We also take advantage of the generality of our model, which provides for estimation of a variety of covariance structures. We observe that models that fail to incorporate CS or AR structure can result in very poor estimation of a covariance or correlation matrix. In our illustration using hormone data observed on women through the menopausal transition, biology dictates the use of a generalized family of sigmoid functions as a model for time trends across subpopulation categories.

Effects of Dexamphetamine and Music on Reversal Learning

Background:

Reversal learning has proven to be a valuable task in assessing the inhibitory process that is central to executive control. Psycho-stimulants and music are prevalent factors that influence cognition.

Objectives:

The present study aimed at investigating the influences of dexamphetamine and music on inhibitory control.

Materials and Methods:

This experimental study was conducted between May and June 2014 in the laboratory animal center of Shiraz University of Medical Sciences, Shiraz, Iran. Thirty mice were divided to five groups including a control group, a witness group, and three experimental groups. Food availability was restricted in order to maintain the subjects at 85% of their free-feeding body weight for behavioral testing. After discrimination learning, animals received four injections of 2 mg/kg dexamphetamine at two-hour intervals. The music group was exposed to music half an hour before reversal learning.

Results:

According to the results of the repeated measure analysis of variance (ANOVA), music increased errors (mean difference: -2.40, 95% CI: -3.59 to -1.22), yet dexamphetamine had no significant effect on reversal learning. Due to various advantages, we transited to the mixed model that showed increasing (Beta: 2.2 95% CI: 0.26 to 4.13) and borderline (Beta: 1.8 95% CI: -0.13 to 3.73) effects on the number of errors for dexamphetamine and music group, respectively.

Conclusions:

Drug-treated subjects were impaired in their ability to modulate behavior, based upon changing information about stimulus-reward associations, possibly due to the inability to inhibit their response. These effects may have relevance to some mental disorders such as drug-abuse, schizophrenia, and obsessive-compulsive disorder.

Genomic heritability of bovine growth using a mixed model

This study investigated heritability for bovine growth estimated with genomewide single nucleotide polymorphism (SNP) information obtained from a DNA microarray chip. Three hundred sixty seven Korean cattle were genotyped with the Illumina BovineSNP50 BeadChip, and 39,112 SNPs of 364 animals filtered by quality assurance were analyzed to estimate heritability of body weights at 6, 9, 12, 15, 18, 21, and 24 months of age. Restricted maximum likelihood estimate of heritability was obtained using covariance structure of genomic relationships among animals in a mixed model framework. Heritability estimates ranged from 0.58 to 0.76 for body weights at different ages. The heritability estimates using genomic information in this study were larger than those which had been estimated previously using pedigree information. The results revealed a trend that the heritability for body weight increased at a younger age (6 months). This suggests an early genetic evaluation for bovine growth using genomic information to increase genetic merits of animals.

Genetic relationships of carcass traits with retail cut productivity of hanwoo cattle

This study aimed to estimate genetic correlation between carcass grading and retail productivity traits and to estimate the correlated response on retail productivity traits through selection for carcass grading traits in order to assess the efficacy of indirect selection. Genetic parameters were estimated with the data from 4240 Hanwoo steers using mixed models, and phenotypes included carcass weight (CWT), back fat thickness (BFT), eye muscle area (EMA), marbling (MAR), and estimated lean yield percentage (ELP) as the carcass grading traits, and weight and portion of retail cuts (RCW and RCP), trimmed fats (TFW and TFP) and trimmed bones (TBW and TBP) as the lean productivity traits. The CWT had positive genetic correlations with RCW (0.95) and TFW (0.73), but its genetic correlation with RCP was negligible (0.02). The BFT was negatively correlated with RCP (−0.63), but positively correlated with TFW and TFP (0.77 and 0.70). Genetic correlations of MAR with TFW and TFP were low. Among the carcass grading traits, only EMA was positively correlated with both RCW (0.60) and RCP (0.72). The EMA had a relatively strong negative genetic correlation with TFW (−0.64). The genetic correlation coefficients of ELP with RCP, TFW, and TFP were 0.76, −0.90, and −0.82, respectively. These correlation coefficients suggested that the ELP and EMA might be favorable traits in regulating lean productivity of carcass.