Applications of Bayesian approach in modelling risk of malaria-related hospital mortality.

BACKGROUND: Malaria is a major public health problem in Malawi, however, quantifying its burden in a population is a challenge. Routine hospital data provide a proxy for measuring the incidence of severe malaria and for crudely estimating morbidity rates. Using such data, this paper proposes a method to describe trends, patterns and factors associated with in-hospital mortality attributed to the disease. METHODS: We develop semiparametric regression models which allow joint analysis of nonlinear effects of calendar time and continuous covariates, spatially structured variation, unstructured heterogeneity, and other fixed covariates. Modelling and inference use the fully Bayesian approach via Markov Chain Monte Carlo (MCMC) simulation techniques. The methodology is applied to analyse data arising from paediatric wards in Zomba district, Malawi, between 2002 and 2003. RESULTS AND CONCLUSION: We observe that the risk of dying in hospital is lower in the dry season, and for children who travel a distance of less than 5 kms to the hospital, but increases for those who are referred to the hospital. The results also indicate significant differences in both structured and unstructured spatial effects, and the health facility effects reveal considerable differences by type of facility or practice. More importantly, our approach shows non-linearities in the effect of metrical covariates on the probability of dying in hospital. The study emphasizes that the methodological framework used provides a useful tool for analysing the data at hand and of similar structure.

A Bayesian multinomial model to analyse spatial patterns of childhood co-morbidity in Malawi.

Children in less developed countries die from relatively small number of infectious disease, some of which epidemiologically overlap. Using self-reported illness data from the 2000 Malawi Demographic and Health Survey, we applied a random effects multinomial model to assess risk factors of childhood co-morbidity of fever, diarrhoea and pneumonia, and quantify area-specific spatial effects. The spatial structure was modelled using the conditional autoregressive prior. Various models were fitted and compared using deviance information criterion. Inference was Bayesian and was based on Markov Chain Monte Carlo simulation techniques. We found spatial variation in childhood co-morbidity and determinants of each outcome category differed. Specifically, risk factors associated with child co-morbidity included age of the child, place of residence, undernutrition, bednet use and Vitamin A. Higher residual risk levels were identified in the central and southern-eastern regions, particularly for fever, diarrhoea and pneumonia; fever and pneumonia; and fever and diarrhoea combinations. This linkage between childhood health and geographical location warrants further research to assess local causes of these clusters. More generally, although each disease has its own mechanism, overlapping risk factors suggest that integrated disease control approach may be cost-effective and should be employed.