Incidence and predictors of Tuberculosis among patients enrolled in Anti-Retroviral Therapy after universal test and treat program, Addis Ababa, Ethiopia. A retrospective follow -up study

Abstract

Introduction Tuberculosis (TB) is the leading killer of people living with HIV (PLHIV) and almost one-third of deaths in the world are attributed to it and many of these deaths occur in developing countries. Despite these evidences, after the implementation of universal test and treat (UTT) strategy, information regarding the incidence and predictors of tuberculosis among PLHIV is limited in Ethiopia. Therefore, this study aimed to assess the incidence and predictors of tuberculosis among patients enrolled in Anti-Retroviral Therapy (ART) after universal test and treat program at St. Peter hospital and Zewditu Memorial Hospital, Addis Ababa, Ethiopia. Methods Institutional-based retrospective cohort study was conducted from November 1 to 30, 2020. Simple random sampling was used to select a total of 539 adults records which was enrolled on ART. Data was collected and entered into EPI DATA 3.1 and analyzed using STATA version 14.1. Time-to-event distributions were estimated using Kaplan–Meier estimates. Hazards across different categories were compared using log-rank tests. Predictors were identified using the Cox proportional hazards model. The hazard ratio (HR) and 95% confidence interval (CI) were computed. Variables having P-value < 0.05 from the multivariable analysis were considered as a statistically significant. Result Among 539 records reviewed, 529 (98%) were included in the final analysis. The total follow-up period was 1529 Person-Year (PY). The incidence rate in this cohort was found to be 4.84 per 100-person year (95%CI,3.83–6.11). CD4 count<200 (AHR: 3.14,95% CI:1.64–7.10), poor adherence (AHR:2.16, 95% CI:1.21–3.85), underweight (AHR:2.42, 95% CI: 1.30–4.51), not taking isoniazid prophylaxis therapy (AHR: 2.78,95% CI: 1.06–7.30), being bedridden 3.06; (AHR: 3.06, 95% CI: 1.50–6.24), and baseline WHO stage three or four (AHR:2.33, 95% CI:1.08–5.02) were independent predictors for the incidence of TB among HIV positive patients. Conclusion In this study, the incidence of tuberculosis is relatively low as compared to studies done before the initiation of test and treat program in Ethiopia. low CD4count, poor level of adherence, low BMI, not taking IPT prophylaxis, bedridden functional status, and being on baseline WHO stage III or IV were found to increase the hazard of tuberculosis. Hence, close follow up, reminders, surveillance, and tracing mechanisms targeting this higher risk group would decrease Tuberculosis among PLHIV.