Predicting Future CD4 Cell Counts of HIV/AIDS Patients by Non Stationary Markov Chain: A Case Study of Anambra State

Journal of Data Science

article

U.A. Osisiogu¹  Nwosu C.A.² 

[1] Department of Industrial Mathematics and Applied Statistics Ebonyi State University;Department of Statistics Anambra State University

关键词: Non Stationary Markov chain model;    Transition probability;    Transition probability matrix;   

DOI  :  10.6339/JDS.201501_13(1).0006

学科分类：土木及结构工程学

来源: JDS

【 摘 要 】

A total of 1094 HIV patients were involved in a cohort study (from January-December 2010) with follow-up in their CD4 cell transition counts and grouped according to their immunological states into five(5) states developed by Guiseppe Di Biase et al (2007). The five states (5) considered were: State one (CD4 > 500 cells/mm3 ), State two (350 < CD4 500 cells /mm3 ) State three(200 < CD4 350 cells/mm3 ), State four(CD4 200 cells/mm3 ), State five(Death). These states de ne the seriousness of the sickness based on the epidemiological states of the patients CD4 cell counts. We use the non-stationary Markov chain model for the prediction. The estimation of the non-stationary probabilities were done using the exponential smoothing technique. The result of the prediction showed a gradual decrease of the CD4 cells as we move from Jan-Dec. Furthermore, the result showed that the patients in the study cannot survive death from the month Dec. 2011, if they are not subjected to therapy, using highly active antiretrovirals (HAART). The results also showed that the model can be used for the testing of the drug e efficacy administered to patients within a given period.

【 授权许可】

CC BY