Determinants of hospitalization in Chinese patients with type 2 diabetes receiving a peer support intervention and JADE integrated care: the PEARL randomised controlled trial.

BACKGROUND: In a randomized controlled trial of 628 Chinese patients with type 2 diabetes receiving multidisciplinary care in the Joint Asia Diabetes Evaluation (JADE) Progam, 372 were randomized to receive additional telephone-based peer support (Peer Empowerment And Remote communication Linked by information technology, PEARL) intervention. After 12 months, all-cause hospitalization was reduced by half in the PEARL group especially in those with high Depression Anxiety and Stress Scale (DASS) scores. METHODS: We used stratified analyses, negative binomial regression, and structural equation modelling (SEM) to examine the inter-relationships between emotions, self-management, cardiometabolic risk factors, and hospitalization. RESULTS: Hospitalized patients were older, more likely to have heart or kidney disease, and negative emotions than those without hospitalization. Patients with high DASS score who did not receive peer support had the highest hospitalization rates. After adjustment for confounders, peer support reduced the frequency of hospitalizations by 48% with a relative risk of 0.52 (95% CI 0·35-0·79;p = 0·0018). Using SEM, improvement of negative emotions reduced treatment nonadherence (Est = 0.240, p = 0.034) and hospitalizations (Est=-0.218, p = 0.001). The latter was also reduced by an interactive term of peer support and chronic kidney disease (Est = 0.833, p = < 0.001) and that of peer support and heart disease (Est = 0.455, p = 0.001). CONCLUSIONS: In type 2 diabetes, improvement of negative emotions and peer support reduced hospitalizations, especially in those with comorbidities, in part mediated through improving treatment nonadherence. Integrating peer support is feasible and adds value to multidisciplinary care, augmented by information technology, especially in patients with comorbidities. TRIAL REGISTRATION: NCT00950716 Registered July 31, 2009.

Bayesian analysis of transformation latent variable models with multivariate censored data.

Transformation latent variable models are proposed in this study to analyze multivariate censored data. The proposed models generalize conventional linear transformation models to semiparametric transformation models that accommodate latent variables. The characteristics of the latent variables were assessed based on several correlated observed indicators through measurement equations. A Bayesian approach was developed with Bayesian P-splines technique and the Markov chain Monte Carlo algorithm to estimate the unknown parameters and transformation functions. Simulation shows that the performance of the proposed methodology is satisfactory. The proposed method was applied to analyze a cardiovascular disease data set.

The genetic basis of population fecundity prediction across multiple field populations of Nilaparvata lugens.

Identifying the molecular markers for complex quantitative traits in natural populations promises to provide novel insight into genetic mechanisms of adaptation and to aid in forecasting population dynamics. In this study, we investigated single nucleotide polymorphisms (SNPs) using candidate gene approach from high- and low-fecundity populations of the brown planthopper (BPH) Nilaparvata lugens Stål (Hemiptera: Delphacidae) divergently selected for fecundity. We also tested whether the population fecundity can be predicted by a few SNPs. Seven genes (ACE, fizzy, HMGCR, LpR, Sxl, Vg and VgR) were inspected for SNPs in N. lugens, which is a serious insect pest of rice. By direct sequencing of the complementary DNA and promoter sequences of these candidate genes, 1033 SNPs were discovered within high- and low-fecundity BPH populations. A panel of 121 candidate SNPs were selected and genotyped in 215 individuals from 2 laboratory populations (HFP and LFP) and 3 field populations (GZP, SGP and ZSP). Prior to association tests, population structure and linkage disequilibrium (LD) among the 3 field populations were analysed. The association results showed that 7 SNPs were significantly associated with population fecundity in BPH. These significant SNPs were used for constructing general liner models with stepwise regression. The best predictive model was composed of 2 SNPs (ACE-862 and VgR-816 ) with very good fitting degree. We found that 29% of the phenotypic variation in fecundity could be accounted for by only two markers. Using two laboratory populations and a complete independent field population, the predictive accuracy was 84.35-92.39%. The predictive model provides an efficient molecular method to predict BPH fecundity of field populations and provides novel insights for insect population management.

A Bayesian modeling approach for generalized semiparametric structural equation models.

In behavioral, biomedical, and psychological studies, structural equation models (SEMs) have been widely used for assessing relationships between latent variables. Regression-type structural models based on parametric functions are often used for such purposes. In many applications, however, parametric SEMs are not adequate to capture subtle patterns in the functions over the entire range of the predictor variable. A different but equally important limitation of traditional parametric SEMs is that they are not designed to handle mixed data types-continuous, count, ordered, and unordered categorical. This paper develops a generalized semiparametric SEM that is able to handle mixed data types and to simultaneously model different functional relationships among latent variables. A structural equation of the proposed SEM is formulated using a series of unspecified smooth functions. The Bayesian P-splines approach and Markov chain Monte Carlo methods are developed to estimate the smooth functions and the unknown parameters. Moreover, we examine the relative benefits of semiparametric modeling over parametric modeling using a Bayesian model-comparison statistic, called the complete deviance information criterion (DIC). The performance of the developed methodology is evaluated using a simulation study. To illustrate the method, we used a data set derived from the National Longitudinal Survey of Youth.

Prediction of ocular irritancy of 26 chemicals and 26 cosmetic products with isolated rabbit eye (IRE) test.

OBJECTIVE: This study aims to establish and evaluate the methodology of isolated rabbit eye (IRE) test. METHODS: IRE test was performed according to modifications of the in vitro toxicology (INVITTOX) Protocol No.85: Rabbit enucleated eye test by European Centre for the Validation of Alternative Methods (ECVAM), and then 26 chemicals and 26 cosmetic products were tested in both in vitro IRE and in vivo Draize tests. A statistical analysis was conducted to determine the relevance of the IRE test to the data generated in the Draize test. RESULTS: IRE test was established successfully in our laboratory. It was shown that ranking correlation and class concordance were fairly well between the IRE test and the Draize test for 26 reference chemicals (Fisher's Exact Test χ(2)=51.314, P<0.001; McNemar P=0.261; Gamma=0.960, P<0.001; Kappa=0.843, P<0.001) and 26 cosmetic products (Fisher's Exact Test χ(2)=15.522, P<0.001; McNemar P=0.311; Gamma=0.967, P<0.001; Kappa=0.611, P<0.001). CONCLUSION: IRE test was established successfully for in vitro testing of eye irritation as an alternative to Draize test.

A Bayesian analysis of mixture structural equation models with non-ignorable missing responses and covariates.

In behavioral, biomedical, and social-psychological sciences, it is common to encounter latent variables and heterogeneous data. Mixture structural equation models (SEMs) are very useful methods to analyze these kinds of data. Moreover, the presence of missing data, including both missing responses and missing covariates, is an important issue in practical research. However, limited work has been done on the analysis of mixture SEMs with non-ignorable missing responses and covariates. The main objective of this paper is to develop a Bayesian approach for analyzing mixture SEMs with an unknown number of components, in which a multinomial logit model is introduced to assess the influence of some covariates on the component probability. Results of our simulation study show that the Bayesian estimates obtained by the proposed method are accurate, and the model selection procedure via a modified DIC is useful in identifying the correct number of components and in selecting an appropriate missing mechanism in the proposed mixture SEMs. A real data set related to a longitudinal study of polydrug use is employed to illustrate the methodology.

Bayesian analysis of mixtures in structural equation models with non-ignorable missing data.

Structural equation models (SEMs) have become widely used to determine the interrelationships between latent and observed variables in social, psychological, and behavioural sciences. As heterogeneous data are very common in practical research in these fields, the analysis of mixture models has received a lot of attention in the literature. An important issue in the analysis of mixture SEMs is the presence of missing data, in particular of data missing with a non-ignorable mechanism. However, only a limited amount of work has been done in analysing mixture SEMs with non-ignorable missing data. The main objective of this paper is to develop a Bayesian approach for analysing mixture SEMs with an unknown number of components and non-ignorable missing data. A simulation study shows that Bayesian estimates obtained by the proposed Markov chain Monte Carlo methods are accurate and the Bayes factor computed via a path sampling procedure is useful for identifying the correct number of components, selecting an appropriate missingness mechanism, and investigating various effects of latent variables in the mixture SEMs. A real data set on a study of job satisfaction is used to demonstrate the methodology.

Non-linear structural equation models with correlated continuous and discrete data.

Structural equation models (SEMs) have been widely applied to examine interrelationships among latent and observed variables in social and psychological research. Motivated by the fact that correlated discrete variables are frequently encountered in practical applications, a non-linear SEM that accommodates covariates, and mixed continuous, ordered, and unordered categorical variables is proposed. Maximum likelihood methods for estimation and model comparison are discussed. One real-life data set about cardiovascular disease is used to illustrate the methodologies.