Adaptive Mixture Modelling Metropolis Methods for Bayesian Analysis of Non-linear State-Space Models.

We describe a strategy for Markov chain Monte Carlo analysis of non-linear, non-Gaussian state-space models involving batch analysis for inference on dynamic, latent state variables and fixed model parameters. The key innovation is a Metropolis-Hastings method for the time series of state variables based on sequential approximation of filtering and smoothing densities using normal mixtures. These mixtures are propagated through the non-linearities using an accurate, local mixture approximation method, and we use a regenerating procedure to deal with potential degeneracy of mixture components. This provides accurate, direct approximations to sequential filtering and retrospective smoothing distributions, and hence a useful construction of global Metropolis proposal distributions for simulation of posteriors for the set of states. This analysis is embedded within a Gibbs sampler to include uncertain fixed parameters. We give an example motivated by an application in systems biology. Supplemental materials provide an example based on a stochastic volatility model as well as MATLAB code.

The Pratt Pouch Provides a Three-Fold Access Increase to Antiretroviral Medication for Births outside Health Facilities in Southern Zambia.

INTRODUCTION: Modern day antiretroviral therapy allows HIV+ pregnant women to lower the likelihood of viral transmission to their infants before, during, and after birth from 20-45% to less than 5%. In developing countries, where non-facility births may outnumber facility births, infant access to safe antiretroviral medication during the critical first three days after birth is often limited. A single-dose, polyethylene pouch ("Pratt Pouch") addresses this challenge by allowing the medication to be distributed to mothers during antenatal care. METHODS: The Pratt Pouch was introduced as part of a one year clinical feasibility study in two districts in Southern Province, Zambia. Participating nurses, community health workers, and pharmacists were trained before implementation. Success in achieving improved antiretroviral medication access was assessed via pre intervention and post intervention survey responses by HIV+ mothers. RESULTS: Access to medication for HIV-exposed infants born outside of a health facility increased from 35% (17/51) before the introduction of the pouch to 94% (15/16) after (p<0.05). A non-significant increase in homebirth rates from 33% (pre intervention cohort) to 50% (post intervention cohort) was observed (p>0.05). Results remained below the national average homebirth rate of 52%. Users reported minimal spillage and a high level of satisfaction with the Pratt Pouch. CONCLUSION: The Pratt Pouch enhances access to infant antiretroviral medication in a rural, non-facility birth setting. Wide scale implementation could have a substantial global impact on HIV transmission rates from mother to child.