A Joint Model for Longitudinal CD4 Count and Body Weight of Adult HIV patients after ART initiation, Asella Hospital

Abstract

The clinical symptoms of HIV infection were evolved on the depletion of CD4 cells and the replication of HIV RNA. Therefore, the HIV cocktail therapy is a combination reagent which focuses on inhibits the replication of HIV RNA at different stages of HIV life-cycle. Wasting syndrome is included in the current case definition of acquired immunodeficiency syndrome (AIDS), as specified in 1993 by the Centers for Disease Control and Prevention (CDC). The CDC defines wasting syndrome as unexplained weight loss of more than 10 percent, accompany by fever or diarrhea. The development of wasting syndrome is considered an indicator condition for AIDS. Therefore this study focuses on the association of the two outcomes since both are an indicator of the progression of the disease. OBJECTIVE: The main objective of this study is to investigate the joint evolution of CD4 cell counts and body weight of HIV patients after ART initiation and identify factors affecting the two end points. METHODS: In this study secondary data were used from Asella referral hospital in HIV adult patients. The study population consists of 300 HIV patients, measured repeatedly at least three times on each patient who is 18 years old or older those treated with antiretroviral therapy from September 2000 to August 2005. Joint and separate model are considered to study the joint evolution and identify the potential risk factors affecting the two end points. RESULT: First the two outcomes are analyzed separately for purpose of identifying associated risk factors for the progress of CD4 count and weight separately. Include on the joint analysis those factors for to investigate the joint evolution and association of CD4 count & weight, and associated risk factors for the progress of the two end points by considering a joint linear mixed effects model. To capture the association of the two outcomes we use a common parameter which is the correlation of the evolution of the two outcomes 0.5054 implied the two outcomes have positive (direct) correlation. Analyzing the two outcomes separately and jointly the covariates are the same that is covariates associated with CD4 count separate analysis was also significant in the joint analysis the same for the weight outcome. The covariates sex, time, WHO 2 | P a g e stage, functional status, TB status and quadratic time were have significant association with CD4 count. Sex, time, functional status, TB status and sex by time interaction term were significantly associated with weight. Among all covariates, sex and time were negatively associated with weight that means when the time increases after sometime the weight decreases. CONCLUSION: The results of the separate and joint model analysis are consistent. When the joint model is compared with the separate model, the joint model fitted the data better than the separate model. The result from the joint model suggested a significant statistical association between the evolutions CD4 count and weight.