Prediction of future HIV new Diagnosis Rates using Spatial Bayesian Method

Human immunodeficiency virus (HIV) infections are life-changing events that, if left untreated, can lead to acquired immunodeficiency disease (AIDS). The Centers for Disease Control and Prevention has reported dramatic outbreaks in regions that were not traditionally affected, besides the geographic regions that consistently exhibit high HIV rates. The National HIV/AIDS Strategy has identified a key goal of intensifying efforts in the communities with the greatest concentration of HIV cases. Regional prediction of disease is central to orchestrating appropriate public health responses. The presence of infection in a region is certainly partly influenced by the social and economic demographics and the prevalence of other sexually transmitted diseases in its population. However, these covariates alone are insufficient to explain the entire variability in the HIV data. After the effects due to the covariates are removed, the spread of infection across the U.S. still exhibits strong spatially and temporally varying patterns as values among neighboring regions and time periods tend to be similar. This presents both a challenge for data modeling as well as an opportunity to tackle the data sparsity issue due to the rarity of the HIV disease.

During her CAS appointment, Professor Li will develop prediction methods that take advantage of the spatial and temporal dependency structures so that the statistical inference at one location can borrow strength from neighboring regions in both space and time. She will generate algorithm and code for future new HIV diagnosis prediction at county level for the entire US, produce maps of predictions together with their uncertainties, and provide a detailed report of the methodology and results to the health department attempting to improve the health prevention system.