Analysis of longitudinal data using a finite mixture model

Bayesian nonparametric hierarchical modeling

In biomedical research, hierarchical models are very widely used to accommodate dependence in multivariate and longitudinal data and for borrowing of information across data from different sources. A primary concern in hierarchical modeling is sensitivity to parametric assumptions, such as linearity and normality of the random effects. Parametric assumptions on latent variable distributions can be challenging to check and are typically unwarranted, given available prior knowledge. This article reviews some recent developments in Bayesian nonparametric methods motivated by complex, multivariate and functional data collected in biomedical studies. The author provides a brief review of flexible parametric approaches relying on finite mixtures and latent class modeling. Dirichlet process mixture models are motivated by the need to generalize these approaches to avoid assuming a fixed finite number of classes. Focusing on an epidemiology application, the author illustrates the practical utility and potential of nonparametric Bayes methods.

Space-time mixture modelling of public health data

This paper aims to enlarge the usual scope of disease mapping by means of dynamic mixtures (DMDM) in case a time component is involved in the data. A special mixture model is suggested which looks for space-time components (clusters) simultaneously. The idea is illustrated using data on female lung cancer from the East German cancer registry for 1960-1989. The conventional mixed Poisson regression model is used as a third model for comparison. The models are discussed in terms of their benefits, difficulties and ease in interpretation, as well as their statistical meaning. Some ideas on evaluation of these models are also included.

Survival of vancomycin-resistant and vancomycin-susceptible enterococci on dry surfaces

We compared the abilities of Enterococcus faecium strains (three vancomycin-resistant enterococci [VRE] and five vancomycin-susceptible enterococci [VSE]) and Enterococcus faecalis strains (one VRE and 10 VSE) to survive under dry conditions. Bacterial suspensions of the strains were inoculated onto polyvinyl chloride and stored under defined conditions for up to 16 weeks. All strains survived for at least 1 week, and two strains survived for 4 months. A statistical model was used to distribute the 19 resulting survival curves between two types of survival curves. The type of survival curve was not associated with the species (E. faecalis versus E. faecium), the source of isolation (patient versus environment), or the susceptibility to vancomycin (VRE versus VSE). Resistance to dry conditions may promote the transmissibility of a strain, but VRE have no advantages over VSE with respect to their ability to survive under dry conditions.