PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 16389918

On the multivariate probit model for exchangeable binary data with covariates.

Catalina Stefanescu·Bruce W Turnbull
View original source ↗Export metadata ↓

Abstract

This paper considers the use of a multivariate binomial probit model for the analysis of correlated exchangeable binary data. The model can naturally accommodate both cluster and individual level covariates, while keeping a fairly flexible intracluster association structure. We discuss Bayesian estimation when a sample of independent clusters of varying sizes are available, and show how Gibbs sampling may be used to derive the posterior densities of parameters. The methodology is illustrated with two examples: the first involves epidemiological data from a study of familial disease aggregation; the second uses teratological data from a developmental toxicity application.

Explore related subjects

Abnormalities, Drug-InducedAnimalsBayes TheoremBiometryCarcinoma, HepatocellularData Interpretation, StatisticalFemaleHumansLiver NeoplasmsModels, StatisticalMultivariate AnalysisPregnancyToxicologyJournal ArticleResearch Support, N.I.H., ExtramuralResearch Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Catalina Stefanescu, Bruce W Turnbull. 2005. On the multivariate probit model for exchangeable binary data with covariates.. https://doi.org/10.1002/bimj.200410101

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2007

Assessment of birth defects according to maternal therapy among infants in the Women and Infants Transmission Study.

BACKGROUND: To evaluate rate and types of birth defects according to timing of antiretroviral exposure among babies born to HIV-infected women. METHODS: Anomalies identified during the prenatal, neonatal, or follow-up period were classified using criteria of the Antiretroviral Pregnancy Registry. Antiretroviral use was classified as none, second or third trimester only, or first trimester. RESULTS: From January 1, 1990 through June 30, 2004, 2527 live births (LBs) occurred to 2353 women. Defects were identified in 90 babies for a rate of 3.56 defects per 100 LBs. The rate of defects was 3.19 per 100 LBs (24 of 752 LBs) with first-trimester antiretroviral exposure, 3.54 per 100 LBs (41 of 1158 LBs) with exposure later in pregnancy, and 4.05 of 100 LBs (25 of 617 LBs) with no antiretroviral use. Only genital abnormalities, specifically hypospadias, were significantly increased among babies born to women with first-trimester exposure to antiretrovirals (7 of 382 male LBs) compared with the 2 other groups (2 of 892 male LBs; P = 0.007). On logistic regression, use of zidovudine in the first trimester was associated with hypospadias (adjusted odds ratio = 10.68, 95% confidence interval: 2.11 to 54.13; P = 0.004). CONCLUSIONS: In general, data were reassuring, although the frequency of exposure to newer agents was limited. The increased risk of hypospadias after first-trimester exposure must be explored, because this association has not been detected previously.

Abnormalities, Drug-Induced↗
PubMed2007

Prenatal testosterone exposure permanently masculinizes anogenital distance, nipple development, and reproductive tract morphology in female Sprague-Dawley rats.

In mammals, abnormal increases in fetal androgens disrupt normal development of the female phenotype. Due to the recent concern regarding environmental androgen-active chemicals, there is a need to identify sources of fetal androgen variation and sensitive developmental markers for androgenic activity in female rats. Anogenital distances (AGD), nipple retention, reproductive tract, and external genitalia are morphological parameters organized by prenatal androgens and are predictive of altered masculinized/defeminized phenotype in adult female mice and rats. The objectives of this study were to (1) characterize the natural prenatal androgen environment of rats including the magnitude of the intrauterine position (IUP) effect, (2) characterize the permanent effects of prenatal androgen exposure on female rats, and (3) determine the ability of AGD and areolas to predict these permanent androgenic alterations in female rats. Untreated male fetal rats had higher tissue testosterone (T) concentrations than females in the amniotic fluid, reproductive tract, gonad, and fetal body. The intrauterine position (IUP) of male and female fetuses did not affect T concentrations or AGD in male or female rats at gestational day (GD) 22. Female offspring exposed to 0, 1.5, and 2.5 mg/kg/day testosterone propionate (TP) on GDs 14-18 displayed increased AGD at postnatal day (PND) 2 and decreased nipples at PND 13 and as adults. TP-induced changes in neonatal AGD and infant areola number were reliable indicators of permanently altered adult phenotype in female rats. Further, females in the two high-dose groups displayed increased incidences of external genital malformations and the presence of prostatic tissue, not normally found in female rats.

Abnormalities, Drug-Induced↗
PubMed2007

Observations on juvenile myoclonic epilepsy amongst ethnic Bengalees in West Bengal--an Eastern Indian State.

BACKGROUND: Juvenile myoclonic epilepsy (JME) is not too uncommonly encountered in Indian neurological practice. A number of reports from different parts of India have documented the clinical phenomenology and EEG characteristics of this genetically determined epileptic syndrome. However, no study has yet been reported from the Eastern part of India and none done so far in patients in a specific ethnic group. Furthermore therapy response and follow up data are not available in detail in the Indian studies. OBJECTIVE: To study disease expression, EEG characteristics and therapy response of JME patients in ethnic Bengalees in West Bengal, an Eastern Indian State, in a clinic based study. MATERIAL AND METHODS: 200 patients with JME attending the Neurology Department of the Institute have been followed up for 5 years and different parameters of disease expression as outlined above have been studied. RESULTS: Overall clinical disease expression has been found to be similar in this clinic based study in ethnic Bengalees as compared to other reports from India and elsewhere. About 16% of patients showed a relative resistance to Valproate therapy. Hundred percent of patients in whom therapy withdrawal was attempted, relapsed within<1-2 years. Amongst female patients (132), 16 developed features of polycystic ovarian syndrome while on Valproate therapy. In over half of them, the symptoms regressed after successful switch over from Valproate to Clobazam. 12/132 female patients became pregnant during follow up and while on Valproate; teratogenic effect was evident in only one such patient. CONCLUSIONS: Phenotypic variations in disease expression including therapy response have been noted within a single ethnic group of patients attending the clinic and might account for genetic heterogeneity noted in molecular genetic studies. JME cannot really be called a very 'benign' epileptic syndrome; recurrence after therapy withdrawal almost invariably occurs.

Abnormalities, Drug-Induced↗