Time to immunologic recovery and determinant factors among adults who initiated ART in Felege Hiwot Referral Hospital, northwest Ethiopia

Abstract

Background: CD4 cells are the major targets for human immunodeficiency virus (HIV) and treatment with Antiretroviral Therapy (ART) influences the CD4 cell count of HIV patients. In addition to ART, the time required to reach normal range of CD4 counts (500 cells/mm3) can be affected by clinical, socio-demographic, and behavioral factors. This retrospective cohort study was conducted to determine the incidence of having the normal range of CD4 cell counts and factors that affect the time required to reach this normal range among adult HIV patients who initiated into ART. Methods: Data of 4 years were retrospectively retrieved from routinely registered characteristics of 937 ART users enrolled in 2010. Survival time until immunologic recovery and its determinant factors were examined using the frailty model with different parametric distributions alternatively. Results: Most (80.8%) of the ART attendants had CD4 cell count of 200 cells/mm3 or less at initiation. The overall incidence rate of immunologic recovery was 12.67 persons per 1000 person-months (95% CI 11.30, 14.20). The dependency of frailties of immunologic recovery by residence was statistically significant (Theta = 0.05, p value = 0.006). Baseline age (Adjusted Hazard Ratio (AHR) = 0.98, 95% CI 0.97, 0.99), baseline CD4 count (AHR = 1.006, 95% CI 1.005, 1.008), and female sex (AHR = 1.34, 95% CI 1.03, 1.73) were significantly associated with shorter survival time for immunologic recovery. Conclusion: Higher baseline CD4 count, lower baseline age, and female sex were positively associated with the time to immunologic recovery, which also dependent on proximity/residence of ART users. Therefore, further scale up of ART services with due emphasis to patients with low CD4 count at baseline particularly for male and older ART users are recommended to reach the normal range of CD4 count in a shorter time of treatment.