A Markov chain based model to predict HIV/AIDS epidemiological trends

Abstract

The emerging and growing use of electronic medical records (EMRs) nowadays gives the possibility of exploiting the huge amount of collected clinical data for epidemiological research purpose, together with the opportunity to address and verify intervention policies and facility management. In this paper we present a Markov chain based model that makes use of real clinical data to simulate epidemiological scenarios for HIV epidemic at a district level. Ad hoc original software has been used, that can be adopted in every similar scenario. This research project is conducted within the Drug Resource Enhancement Against AIDS and Malnutrition (DREAM) Program to fight HIV/AIDS in sub-Saharan Africa. The results of this paper show in a clear and robust fashion how the proposed Markov chain based model can be helpful to predict epidemiological trends and hence to support decision making to face the diffusion of HIV/AIDS and other sexually transmitted diseases. © Springer-Verlag 2013.