Search

Children are an important group harmed by others' alcohol consumption. This study 1) compared the risk of occurrence of child maltreatment among children exposed vs not exposed to parents with an alcohol-attributable hospitalisation or service use for mental health/addiction and 2) conducted sensitivity analyses to estimate the cases of child maltreatment that could be attributable to alcohol under two different conditions in New Zealand.This cohort study was conducted among 58,359 children 0-17 years and their parents (years 2000-2017) using the Statistics New Zealand Integrated Data Infrastructure. The prevalence of hazardous drinking among parents was obtained from the New Zealand Health Survey 2017 (n = 13,869).Survival analysis based on a Bayesian piecewise exponential model was used to estimate the risk of time-to-first substantiated child maltreatment event (identified from social service, hospital, mortality and police data) related to exposure to parents with an alcohol-attributable hospitalisation or who used a mental health/addiction service (vs no exposure). Potential confounders were included for parents and children. The sensitivity analyses i) estimated an alcohol-attributable admissions/service use fraction for maltreatment in 2017 and ii) calculated a population-attributable fraction using the relative risk from the cohort and prevalence of hazardous drinking (AUDIT 8+) among parents in 2017.There was a 65.1% (1.65; 95% confidence interval [CI]: 1.46 - 1.86) increased risk of child maltreatment if a child was exposed to parents who had an alcohol-attributable hospitalised or mental health/addictions service use. The sensitivity analyses estimated that in 2017 14.6% (CI: 14.0% - 15.3%) and 11.4% (95% CI: 8.4% - 14.3%) of the documented cases of child maltreatment in New Zealand could be attributable to parents with severe or hazardous consumption.In New Zealand, exposure to parents with an alcohol-attributable hospitalisation or service use is a risk factor for substantiated child maltreatment.

Introduction: New Zealand’s Alcohol Licencing Trusts (ALTs) are community-owned entities that sell alcohol in their districts and return profits back to the local community.Aim: To investigate supp...

Background There are increasing reports of social media being used to buy and sell illegal drugs internationally. Studies of social media drug markets to date have largely involved general explorations of social media drug transactions in select countries. Social media drug markets may operate differently for different drug types and in different country contexts. Aims To identify predictors of the use of social media to purchase cannabis, methamphetamine, LSD and ecstasy/MDMA in New Zealand using a large-scale online survey sample. Methods The annual New Zealand Drug Trends Survey (NZDTS), an online convenience survey promoted via targeted Facebook™ campaign, was completed by 23,500 respondents aged 16+. Logistic regression models were fitted to identify demographic, drug use and drug market predictors of reporting the use of social media to purchase cannabis, methamphetamine, LSD and ecstasy/MDMA. Results Fifty-one percent of the sample had purchased cannabis, methamphetamine, ecstasy/MDMA or LSD in the past six months, of which 22% had done so via social media (n = 2,650). Cannabis was the drug type most purchased from social media among drug purchasers (24%), followed by ecstasy/MDMA (13%). Sixty percent of social media purchasers had used Facebook™, 48% Snapchat™, and 20% Instagram™. The leading advantages of social media purchasing were “high convenience” (74%), and “fast transaction speed” (43%). Younger age was a significant predictor of social media purchasing for all drug types (particularly among 16–17 year olds). Purchasing from someone identified as a “drug dealer” was a predictor of social media purchasing among respondents for all drug types. Conclusion Social media drug markets may have significant implications for drug purchasing by youth through providing greater access to supply and breaking down age-related barriers between social supply and commercial drug markets. Our findings highlight the importance of incorporating social media platforms into youth drug prevention and harm minimisation strategies.

Introduction This paper aimed to assess purchasing and drinking behaviour during the first COVID‐19 pandemic restrictions in New Zealand. Method A convenience sample was collected via Facebook from 2173 New Zealanders 18+ years during pandemic restrictions April/May 2020. Measures included: the quantity typically consumed during a drinking occasion and heavier drinking (6+ drinks on a typical occasion) in the past week; place of purchase including online alcohol delivery. Descriptive statistics were generated, logistic and linear regression models predicted heavier drinking and typical occasion quantity, respectively. Weighting was not applied. Results During pandemic restrictions, around 75% of respondents purchased from supermarkets, 40% used online alcohol delivery services (18% for the first time during COVID‐19). Purchasing online alcohol delivery during pandemic restrictions was associated with heavier drinking (75% higher odds) in the past week, while purchasing from supermarkets was not. About 58% of online purchasers under 25 reported no age checks. Sixteen percent of those purchasing online repeat ordered online to keep drinking after running out. Of respondents who had tried to buy alcohol and food online, 56% reported that alcohol was easier to get delivered than fresh food. Advertising for online alcohol delivery was seen by around 75% of the sample. Half of the sample reported drinking more alcohol during the restrictions. Discussion and Conclusions Online alcohol delivery during the COVID‐19 pandemic restrictions was associated with heavier drinking in the past week. The rapid expansion of online alcohol delivery coupled with a lack of regulatory control requires public health policy attention.

Aims To estimate the effect of national restrictions on late-night availability of alcohol on alcohol-related assault at a population level as indicated by (1) change in hospitalizations for weekend assaults and (2) change in the proportion of assaults documented by police that occur at night. Design Evaluation of a natural experiment, involving: (1) pre-post comparisons of age-specific incidence rates, adjusted for seasonality and background trend using Poisson regression; and (2) interrupted time-series analyses, using seasonal autoregressive integrated moving average (SARIMA) models of national data with no control site. Setting New Zealand. Participants (1) Inpatients discharged from NZ hospitals following assault during the weekend (Friday-Sunday) from 2004 to 2016 (n = 14 996) and (2) cases of assault recorded by NZ Police from 2012 to 2018. Intervention introduction of national maximum trading hours for all on-licence (8 a.m.-4 a.m.) and off-licence premises (7 a.m.-11 p.m.), abolishing existing 24-hour licences, on 18 December 2013. Measurements (1) Age-specific incidence of hospitalization for assault on Friday, Saturday or Sunday from the national hospital discharge data set, excluding short-stay emergency department admissions and (2) proportion of weekly police-documented assaults occurring between 9 p.m. and 5.59 a.m., from NZ Police Demand and Activity data set. Findings Following the restrictions, weekend hospitalized assaults declined by 11% [incidence rate ratio (IRR) = 0.89; 95% confidence interval (CI) = 0.84, 0.94], with the greatest reduction among 15-29-year-olds (IRR = 0.82; 95% CI = 0.76, 0.89). There was an absolute reduction (step change) of 1.8% (95% CI = 0.2, 3.5%) in the proportion of police-documented assaults occurring at night, equivalent to 9.70 (95% CI = 0.10, 19.30) fewer night-time assaults per week, out of 207.4. Conclusions The 2013 implementation of national maximum trading hours for alcohol in NZ was followed by reductions in two complementary indicators of alcohol-related assault, consistent with beneficial effects of modest nation-wide restrictions on the late-night availability of alcohol.

Background Small towns in New Zealand have reported high availability of methamphetamine, and conversely a shortage of cannabis. Stakeholders have suggested drug dealers are purposely promoting methamphetamine rather than cannabis. Aims (1) To compare the availability of methamphetamine and cannabis in different size communities; (2) Identify determinants of the high availability of methamphetamine, including low availability of cannabis. Method An online drug survey was promoted via a broadly targeted Facebook™ campaign. Participants were asked if they lived in a “city”, “small town” or “rural area”, their drug use patterns, and local drug market characteristics, including current availability. A total of 6311 people completed the survey. Logistic regression models were constructed to identify independent predictors of reporting high availability of methamphetamine, cannabis, ecstasy and LSD respectively, with low availability of cannabis included as a predictor in the non-cannabis markets. Results Methamphetamine was reported to be more available than cannabis in all regions. Methamphetamine was more available in towns/rural areas than in cities. Significant predictors of high availability of methamphetamine were living in a town/rural area (OR = 1.38), purchasing from a gang member (OR = 1.88), daily methamphetamine use (OR = 2.41), Maori ethnicity (OR = 1.36) and reporting low availability of cannabis (OR = 1.89). Low availability of cannabis was not a predictor of high availability of ecstasy or LSD. Living in a town/rural area was not a predictor of high availability of cannabis, LSD or ecstasy. Purchasing from a gang member was a predictor of high availability of cannabis (OR = 1.80) and LSD (OR = 4.61). Conclusions Further research is required to identify what causal relationships, if any, there are between the statistical associations of high methamphetamine availability, living in a small town, purchasing from a gang, and low cannabis availability. It may be the case that small towns offer an environment where a gang can control the local drugs market.

This work presents a Bayesian approach for estimating the limiting availability of an one-unit repairable system. A Bayesian analysis is developed considering an informative prior and a less informative prior distribution, respectively. Simulations are presented to study the performance of the Bayesian solutions. The maximum likelihood method is also revisited. Finally, a case study is considered, the Bayesian methodology is applied to estimate the limiting availability of a palletizer, which is used in the packaging of glass bottles. Extensions to a coherent system are also discussed. Resumen En este trabajo se presenta un enfoque bayesiano para estimar la disponibilidad límite de un sistema reparable uni-componente. Un análisis bayesiano es desarrollado considerando distribuciones a priori informativa y poco informativa, respectivamente. Simulaciones son presentadas para estudiar el desempeño de las soluciones bayesianas. El método de máxima verosimilitud también es reconsiderado. Finalmente, un caso de estudio es considerado, la metodología bayesiana es aplicada para estimar la disponibilidad límite de un paletizador. el cual es usado en el embalaje de botellas de vidrio. Extensiones a un sistema coherente, también son discutidas.

INTRODUCTION AND AIMS This study aimed to assess the early impacts of a 2013 law change to restrict the social supply of alcohol to under 18s in New Zealand by assessing changes in social suppliers' behaviour. DESIGN AND METHODS National surveys of drinkers aged 16-65 years were collected before (2013) and after (2015) the law change. Suppliers were asked: to whom they supplied alcohol, how often and how much; if the supplier thought the alcohol they provided would be shared; whether the supplier had permission to supply from parent/guardian, and if they supervised their supply. General estimating equation models were used to conduct analyses. RESULTS Sons and daughters were commonly supplied to, as were friends, but friends were supplied greater quantities on average [13 drinks compared to four drinks for sons/daughters (at baseline)]. Following the law change, friends were less commonly supplied to (8% decrease), were supplied with fewer drinks (down from 13 to 11 drinks) and there was greater supervision of social supply to friends (16% increase) (and to other relatives). However, the number of drinks supplied by parents increased from four to six drinks. DISCUSSION AND CONCLUSIONS There was evidence of some early reductions in social supply in relation to the law change, in particular where the legislation aimed to have effect; specifically, less supply to friends under 18 years. We found no effect of the new law on parental supply. As quantities supplied are still very high, further policy restriction and public health interventions would be appropriate.

The family of power series cure rate models provides a flexible modeling framework for survival data of populations with a cure fraction. In this work, we present a simplified estimation procedure for the maximum likelihood (ML) approach. ML estimates are obtained via the expectation-maximization (EM) algorithm where the expectation step involves computation of the expected number of concurrent causes for each individual. It has the big advantage that the maximization step can be decomposed into separate maximizations of two lower-dimensional functions of the regression and survival distribution parameters, respectively. Two simulation studies are performed: the first to investigate the accuracy of the estimation procedure for different numbers of covariates and the second to compare our proposal with the direct maximization of the observed log-likelihood function. Finally, we illustrate the technique for parameter estimation on a dataset of survival times for patients with malignant melanoma.

Introduction New Zealand has a number of unusual geographical features that may influence the price of illegal drugs including international isolation, numerous rural communities, and two separate islands (North and South Islands). Aims (1) Collect data on the prices of different drug types from all regions of New Zealand; (2) Identify independent predictors of the reported prices. Method An online drug survey was promoted via a targeted Facebook™ campaign. A total of 6331 respondents completed the survey. Normal regression models were constructed to identify predictors of the reported price of an ounce of cannabis, gram of methamphetamine, pill of ecstasy and tab of LSD using demographics, drug use frequency, drug market indicators, region, community size, type of seller, and location of purchase as predictor variables. Results Higher availability was a predicator of lower prices for cannabis (-3% per higher availability category), methamphetamine (-4%) and ecstasy (-5%). Those living in the North Island (-26%) and buying from gangs (-5%) paid lower prices for methamphetamine. Those living in rural communities reported higher prices for methamphetamine (+5%) but lower prices for cannabis (-3%). Daily cannabis users (+4%) and those on social welfare (+2%) paid higher prices for cannabis. Conclusion Lower prices for methamphetamine in the North Island may reflect the concentration of methamphetamine manufacture there. The fact that gangs offer lower prices for methamphetamine suggests they do not have monopoly control of this market. Gangs may be able to offer lower prices for methamphetamine due to scale of production/trafficking and lower risk of victimisation. Higher prices for methamphetamine in rural areas may reflect less competitive markets. Lower prices for cannabis in rural communities may be due to proximity to growing locations. Daily cannabis users and those on social welfare may be less able to delay their cannabis purchases to low price harvest months.

The analysis of recurrent event data is of particular importance in medical statistics where patients suffering from chronic diseases often present with multiple recurring relapses or cancer patients experience several tumor recurrences. Whereas individual subjects can be assumed to be independent, the times between events of one subject are neither independent nor identically distributed. Apart from the marginal approach by Wei et al. (1989), the shared frailty model, see for example Duchateau and Janssen (2008), has been used extensively to analyze recurrent event data, where the correlation between sequential times is implicitly taken into account via a random effect. Oakes (1989) and Romeo et al. (2006) showed and exemplified the equivalence of frailty models for bivariate survival data to Archimedean copulas. Despite the fact that copula-based models have been used to model parallel survival data, their application to recurrent failure time data has only recently been suggested by Lawless and Yilmaz (2011) for the bivariate case. Here, we extend this to more than two recurrent events and model the joint distribution of recurrent events explicitly using parametric copulas within a Bayesian framework. This framework allows for parametric as well as a nonparametric modeling of the marginal baseline hazards and models the influence of covariates on the marginals via a proportional hazards assumption. Furthermore, the parameters of the copula may also depend on the covariates. We illustrate the flexibility of this approach using data from an asthma prevention trial in young children.

This work presents a Bayesian hierarchical model with the dual objective to analyze stratified survival data and to automatically classify each stratum into a finite number of groups. This is achieved by specifying parametric as well as piecewise stratum-specific baseline hazards and a finite mixture distribution for the stratum-specific shape parameters. A proportional hazards or accelerated failure time regression component allows to identify the influence of covariates on the survival distribution. We illustrate the model using a dataset of Atlantic salmon, stratified by families, that have been challenged with infectious pancreatic necrosis virus (IPNV). The main objectives are to model the survival time in terms of certain covariates as well as to classify the salmon families into either an IPNV susceptible or resistant group with the ultimate goal of improving resistance to IPNV through a selective breeding program. We compare the fit of different models that include stratum-specific baselines and covariate effects. The classifications show a certain degree of robustness with respect to model choice.

This work presents a Bayesian approach for estimating the limiting availability of an one-unit repairable system when the data are subject to right censorship. It is assumed that the failure and repair times of the components are independent exponential random variables. A conjugate Bayesian analysis is performed considering an informative and non informative prior distributions. Simulations are presented to study the performance of the Bayesian solutions. Some observations are made in relation to the maximum likelihood method. The Weibull case is also discussed.

This work is concerned with asymptotic properties of the bivariate survival function estimator using the functional relationship between marginal survival functions and a class of copulas for the dependence structure. Specifically, we study consistency and weak convergence of the bivariate survival function estimator obtained considering a two-step procedure of estimation. The obtained results are found from a key decomposition of the bivariate survival function in quantities that can be studied separately. In particular, we use relating results to almost sure and weak convergence of estimators, almost sure convergence of uniformly equicontinuous functions, and the delta method for functionals.

Buffers are used to deal with the detrimental impacts of uncertainty on projects. However, methods for the allocation of buffers often provide single unique solutions, which are inefficient in the multiobjective decision-making environment of construction. This paper discusses a probabilistic-based buffer allocation method (PBAL), which enables the final decision on buffer size to be made by the project planners based on their preferences about project completion time. It investigates the construction projects where each activity starts as early as possible. Accordingly, the decision involves determining the size of time buffer at the end of the project network. The accuracy of the results is subjected to approximation and numerical errors in the mathematical models among others. Most buffer allocation heuristics for projects have approximation errors and simulation-based techniques introduce numerical errors by their iterative sampling calculation approach. PBAL can minimize these errors by suppo...

Copula models have become increasingly popular for modelling the dependence structure in multivariate survival data. The two-parameter Archimedean family of Power Variance Function (PVF) copulas includes the Clayton, Positive Stable (Gumbel) and Inverse Gaussian copulas as special or limiting cases, thus offers a unified approach to fitting these important copulas. Two-stage frequentist procedures for estimating the marginal distributions and the PVF copula have been suggested by Andersen (Lifetime Data Anal 11:333–350, 2005), Massonnet et al. (J Stat Plann Inference 139(11):3865–3877, 2009) and Prenen et al. (J R Stat Soc Ser B 79(2):483–505, 2017) which first estimate the marginal distributions and conditional on these in a second step to estimate the PVF copula parameters. Here we explore an one-stage Bayesian approach that simultaneously estimates the marginal and the PVF copula parameters. For the marginal distributions, we consider both parametric as well as semiparametric models. We propose a new method to simulate uniform pairs with PVF dependence structure based on conditional sampling for copulas and on numerical approximation to solve a target equation. In a simulation study, small sample properties of the Bayesian estimators are explored. We illustrate the usefulness of the methodology using data on times to appendectomy for adult twins in the Australian NH&MRC Twin registry. Parameters of the marginal distributions and the PVF copula are simultaneously estimated in a parametric as well as a semiparametric approach where the marginal distributions are modelled using Weibull and piecewise exponential distributions, respectively.

BACKGROUND CONTEXT: Acute low back pain (LBP) is a frequent cause of physician visits and sick leave. Patients with longer sick leave account for most costs associated with LBP. Most research on risk factors for prolonged sick leave because of LBP has been done in Anglo-Saxon or Northern European populations with occupational LBP. Few studies have been conducted in less affluent Latin countries. PURPOSE: To investigate the prevalence of acute LBP as a cause of sick leave and the variables associated with longer work absence (WA) because of acute LBP in Chile. STUDY DESIGN: A retrospective study of nonoccupational sick leave certificates issued in a 1-year period by the Chilean Public Health System. PATIENT SAMPLE: Ten thousand cases with nonoccupational sick leave certificates issued with the diagnosis of acute LBP. OUTCOME MEASURES: Prevalence of acute LBP as a cause of sick leave, days of WA, and subsidy payment. METHODS: A review of sick leave certificates from patients with different diagnoses was done to determine the prevalence of acute LBP as a cause of sick leave. We investigated whether age, gender, history of at least one episode of sick leave in the previous year because of acute LBP or because of other diagnoses, stipendiary subsidy paid (converted to US $), work activity, occupation, and the specialty of the attending physician influenced the length of sick leave. RESULTS: Acute LBP represented 5.4% of cases causing sick leave. Patients with acute LBP who had significantly longer sick leave than the rest of the population were patients with an episode of WA because of LBP in the previous year (14% longer sick leave than patients without that history), manual workers (35% longer than nonmanual workers), and patients seen by orthopedic surgeons (43% longer than patients seen by other medical specialists). Nonpregnant women and men had similar durations of sick leave because of acute LBP. CONCLUSIONS: Physicians who treat patients with LBP should be aware of these variables associated with longer sick leave because of acute LBP, which represents a significant burden to any society and is an important epidemiologic problem. 2009 Elsevier Inc. All rights reserved.

This study aims to identify at the expression level the immune-related genes associated with IPN-susceptible and resistant phenotypes in Atlantic salmon full-sibling families. We have analyzed thirty full-sibling families infected by immersion with IPNV and then classified as resistant or susceptible using a multivariate survival analysis based on a gamma-Cox frailty model and the Kaplan–Meier mortality curves. In four families within each group head kidneys were pooled for real-time PCR and one-color salmon-specific oligonucleotide microarray (21K) analysis at day 1 and 5 post-infection. Transcripts involved in innate response (IL-6, IFN-α), antigen presentation (HSP-70, HSP-90, MHC-I), TH1 response (IL-12, IFN-γ, CRFB6), immunosuppression (IL-10, TGF-β1) and leukocyte activation and migration (CCL-19, CD18) showed a differential expression pattern between both phenotypes, except in IL-6. In susceptible families, except for IFN-γ, the expressions dropped to basal values at day 5 post-infection. In resistant families, unlike susceptible families, levels remained high or increased (except for IL-6) at day 5. Transcriptomic analysis showed that both families have a clear differential expression pattern, resulting in a marked down-regulation in immune related genes involved in innate response, complement system, antigen recognition and activation of immune response in IPN-resistant. Down-regulation of genes, mainly related to tissue differentiation and protein degradation metabolism, was also observed in resistant families. We have identified an immune-related gene patterns associated with susceptibility and resistance to IPNV infection of Atlantic salmon. This suggests that a limited immune response is associated with resistant fish phenotype to IPNV challenge while a highly inflammatory but short response is associated with susceptibility.

The analysis of recurrent event data is of particular importance in medical statistics where patients suffering from chronic diseases often present with multiple recurring relapses or cancer patients experience several tumor recurrences. Whereas individual subjects can be assumed to be independent, the times between events of one subject are neither independent nor identically distributed. Apart from the marginal approach by Wei et al. (1989), the shared frailty model, see for example Duchateau and Janssen (2008), has been used extensively to analyze recurrent event data, where the correlation between sequential times is implicitly taken into account via a random effect. Oakes (1989) and Romeo et al. (2006) showed and exemplified the equivalence of frailty models for bivariate survival data to Archimedean copulas. Despite the fact that copula-based models have been used to model parallel survival data, their application to recurrent failure time data has only recently been suggested by Lawless and Yilmaz (2011) for the bivariate case. Here, we extend this to more than two recurrent events and model the joint distribution of recurrent events explicitly using parametric copulas within a Bayesian framework. This framework allows for parametric as well as a nonparametric modeling of the marginal baseline hazards and models the influence of covariates on the marginals via a proportional hazards assumption. Furthermore, the parameters of the copula may also depend on the covariates. We illustrate the flexibility of this approach using data from an asthma prevention trial in young children.

Since many authors have emphasized the need of asymmetric link functions to fit binary regression models, we propose in this work two new skew-probit link functions for the binary response variables. These link functions will be named power probit and reciprocal power probit due to the relation between them including the probit link as a special case. Also, the probit regressions are special cases of the models considered in this work. A Bayesian inference approach using MCMC is developed for real data suggesting that the link functions proposed here are more appropriate than other link functions used in the literature. In addition, simulation study show that the use of probit model will lead to biased estimate of the regression coefficient.

Molecular mechanisms of plant-pathogen interaction have been studied thoroughly because of its importance for crop production and food supply. This knowledge is a starting point in order to identify new and specific resistance genes by detecting similar expression patterns. Here we evaluate the usefulness of clustering and data-mining methods to group together known plant resistance genes based on expression profiles. We conduct clustering separately on P.infestans inoculated and not-inoculated tomatoes and conclude that conducting the analysis separately is important for each condition, because grouping is different reflecting a characteristic behavior of resistance genes in presence of the pathogen.

Publicly available genomic data are a great source of biological knowledge that can be extracted when appropriate data analysis is used. Predicting the biological function of genes is of interest to understand molecular mechanisms of virulence and resistance in pathogens and hosts and is important for drug discovery and disease control. This is commonly done by searching for similar gene expression behavior. Here, we used publicly available Streptococcus pyogenes microarray data obtained during primate infection to identify genes that have a potential influence on virulence and Phytophtora infestance inoculated tomato microarray data to identify genes potentially implicated in resistance processes. This approach goes beyond co-expression analysis. We employed a quasi-likelihood model separated by primate gender/inoculation condition to model median gene expression of known virulence/resistance factors. Based on this model, an influence analysis considering time course measurement was performed to detect genes with atypical expression. This procedure allowed for the detection of genes potentially implicated in the infection process. Finally, we discuss the biological meaning of these results, showing that influence analysis is an efficient and useful alternative for functional gene prediction.

A nonparametric Bayesian approach is applied to estimate a survival curve by means of a functional of the subsurvival functions associated with censored and non-censored events. In order to actually compute the Bayesian estimator, a numerical algorithm based on the Runge-Kutta fourth-order method is introduced. It provides good accuracy and is simple to program. Using a simulated data set, the performance of the Bayesian estimator is compared to the Product-Limit. A descriptive analysis of the results from the simulations is presented. The conclusions are given in terms of the proportion of the censored data and sample size. Also, the predictive value of the estimations is investigated using a cross-validation measure, namely the mean square predictive value. The numerical methodology is illustrated considering the original Kaplan-Meier data. Finally, the Bayesian analysis is applied to a real case of cervix uterine cancer, where the elicitation of the prior distribution considers the high proportion of censoring in the sample.

Copula models have become increasingly popular for modeling multivariate survival data. In this paper we review some of the recent work that has been appeared for copula model for bivariate survival data and propose a Bayesian modeling. Our approach is very flexible with respect to the choice of marginal distributions and, depending on the copula model employed, it is possible to have a class of variation for the dependence parameter. We compare some of the copula models using a descriptive diagnostic method and three popular Bayesian model selection criteria. Our methodology is illustrated with the Diabetic Retinopathy Study (1976).

Background: Molecular mechanisms of plant–pathogen interactions have been studied thoroughly but much about them is still unknown. A better understanding of these mechanisms and the detection of new resistance genes can improve crop production and food supply. Extracting this knowledge from available genomic data is a challenging task. Results: Here, we evaluate the usefulness of clustering, data-mining and regression to identify potential new resistance genes. Three types of analyses were conducted separately over two conditions, tomatoes inoculated with Phytophthora infestans and not inoculated tomatoes. Predictions for 10 new resistance genes obtained by all applied methods were selected as being the most reliable and are therefore reported as potential resistance genes. Conclusion: Application of different statistical analyses to detect potential resistance genes reliably has shown to conduct interesting results that improve knowledge on molecular mechanisms of plant resistance to pathogens.