A Bayesian hierarchical model with spatial variable selection: the effect of weather on insurance claims.

Climate change will affect the insurance industry. We develop a Bayesian hierarchical statistical approach to explain and predict insurance losses due to weather events at a local geographic scale. The number of weather-related insurance claims is modelled by combining generalized linear models with spatially smoothed variable selection. Using Gibbs sampling and reversible jump Markov chain Monte Carlo methods, this model is fitted on daily weather and insurance data from each of the 319 municipalities which constitute southern and central Norway for the period 1997-2006. Precise out-of-sample predictions validate the model. Our results show interesting regional patterns in the effect of different weather covariates. In addition to being useful for insurance pricing, our model can be used for short-term predictions based on weather forecasts and for long-term predictions based on downscaled climate models.

A controlled intervention study to improve antibiotic use in a Russian paediatric hospital.

A controlled intervention study was performed in a paediatric hospital in Russia to improve antibiotic use and to see whether improvements persisted. During October-December 2002, clinical and microbiological data, antibiotic use, costs and outcome were recorded at two wards for gastrointestinal infections (GIIs) and two wards for respiratory tract infections (RTIs). Guidelines for diagnosis and treatment of infections were developed and implemented at one ward for GIIs and one ward for RTIs in 2003. The other two wards served as controls. The same data were recorded during the same 3-month periods in 2003 and 2004. At the intervention ward, the percentage of patients with GII who received antibiotics decreased from 94% in 2002 to 41% in 2003, but increased to 73% in 2004. In RTI patients these percentages were 90% in 2002, 53% in 2003 and 83% in 2004. The proportions of patients who received antibiotics in 2004 were still lower than in 2002: risk difference (RD)=0.217 (P

Analysis of the humoral immune response to immunoselected phage-displayed peptides by a microarray-based method.

We describe a novel approach for high-throughput analysis of the immune response in cancer patients using phage-based microarray technology. The recombinant phages used for fabricating phage arrays were initially selected via the use of random peptide phage libraries and breast cancer patient serum antibodies. The peptides displayed by the phages retained their ability to be recognized by serum antibodies after immobilization. The recombinant phage microarrays were screened against either breast cancer or healthy donor serum antibodies. A model-based statistical method is proposed to estimate significant differences in serum antibody reactivity between patients and normals. A significant tumor effect was found with most of the selected phage-displayed peptides, suggesting that recombinant phage microarrays can serve as a tool in monitoring humoral responses towards phage-displayed peptides.