MODELLING THE CD4+ COUNT PROGRESSION OF HIV+ PATIENTS USING GENERALIZED ESTIMATING EQUATIONS: A CASE OF DEBRE BERHAN REFERRAL HOSPITAL, ETHIOPIA

CD4+ count is used to assess the clinical status of HIV-infected patients and making informed decisions about the initiation of Antiretroviral Therapy. To determine the progression of patient characteristics, the CD4+ count should be measured repeatedly per individual who used to evaluate the evolution of CD4+ count over time from baseline. Thus, the main objective of the study was modeling the progression of CD4+ counts of HIV+ patients under ART in Debre Berhan Referral Hospital, Ethiopia. A retrospective cohort study is used; and out of a population of HIV-patients who were taking ART in the hospital during September, 2013 to February, 2019, data on 322 patients was taken to the study from which 225 (69.88%) were females and 97(30.12%) were males. Generalized Estimating Equations model is used to analyze the non-independent data measured for each patients for specified period of time. The result of this study showed that CD4+ count patients improved from average baseline, 335.7 to 408.6 after all follow-up times. The average CD4+ count of female patients was changed 0.5793 squared times than average CD4+ count of male patients (β = 0.5793 , p-value = 0.02618). HIV+ patients’ age had negatively significant effect on the average progression of CD4+ count(β = -0.0321, p-value=0.0139). Its negative estimate indicated that the average CD4+ count older patients reduced by 0.0321 squared times than younger individuals. The predictors of patients’ baseline CD4+count, sex, age, functional status, marital status, treatment regimen class, WHO clinical stage and follow up time found to be significant association with the progression of CD4+ count. HIV+ patients who had lowest initial CD4+ count need special attention as patients with higher baseline CD4 count tend to have higher rate of change over time than those with lower baseline CD4+ count