Global Medicines Program

January 23, 2014

Risk of Anemia Associated with Zidovudine (AZT)-based HAART in Namibia

Assegid Tassew Mengistu1, Ishmael Katjitae1, Catherine Corbell3, Johannes Gaeseb1, Jennie Lates1, Evans Sagwa2, David Mabirizi2, Jude Nwokike2, Francis Kalemeera1, Andy Stergachis3
1Ministry of Health and Social Services; 2Management Sciences for Health/ Strengthening Pharmaceutical Systems; 3University of Washington

Abstract #922 HIV/AIDS and TB. Keywords: Anemia, zidovudine, pharmacovigilance, Namibia, antiretroviral therapy

Problem Statement: Anemia among AZT-based antiretroviral therapy users was the most commonly reported adverse event to Namibia‘s Therapeutics Information and Pharmacovigilance Center (TIPC), accounting for 41% (106/256) of total adverse events reported in 2009. This generated a safety signal, which led to conduct of a pharmacoepidemiologic study to establish the incidence rate and risk factors of anemia among AZT users.

Objectives: To determine the incidence of and risk factors for anemia in adults on AZT-based HAART and to demonstrate the feasibility of using linked automated databases as a sustainable platform for assessing the safety and use of HAART to support evidence-based decision making in Namibia.

Design: A cohort of HIV adult patients newly initiated on AZT and d4T-based HAART from January 2007 to June 2008 were followed retrospectively using linked data from three automated data sources: (1) the ART electronic dispensing tool (EDT), a pharmacy-based database; (2) the electronic Patient Management System (ePMS), a clinical database; and (3) MEDITECH, a laboratory database from the Namibia Institute of Pathology. The paper-based clinical records stored at health facilities was used to validate the linked electronic data and to obtain additional information on risk factors of anemia for the nested case-control study.

Study Population: A total of 12,365 persons aged 19 to 65 years started on HAART between January 2007 and June 2008 whose unique person-records were identified from EDT and matched to ePMS and MEDITECH database.

Outcome Measure: Anemia, hemoglovin (Hb) value < 7.0 g/dl, diagnosed at least 30 days after starting HAART

Results: The adjusted relative risk (ARR) of severe anemia 29.76 (95% CI: 2.87,308.86) was highest during the first three months of AZT use compared to non-use. The risk dropped after the first three months of AZT use (ARR: 1.09 ) (95% CI: 0.17, 7.05) compared to non-use. In persons with baseline Hb values available, the incidence rate of developing severe anemia was 2.28 per 100 PY in the AZT cohort (95% CI: 1.81, 2.87). There was no significant difference in the incidence rate of severe anemia between the AZT and d4T cohorts during the entire period of follow-up among this subset of persons with baseline Hb values available. The incidence rate of severe anemia was similar to the incidence rate reported in South Africa and in Haiti but lower than the incidence rate reported in the Ivory Coast.The median number of Hb measurements during the first year was 2 for both AZT and d4T cohorts. Furthermore, the median time to a first Hb measurement was 37 days in the AZT cohort and 146 days in the d4T cohort.

Conclusions: Risk of severe anemia associated with AZT-based HAART was highest during first three months of AZT use and diminished thereafter. This study successfully demonstrated the benefit of records linkage in the examination of incidence rate and risk factors of adverse events and compliance with treatment guidelines.

Funding Source: USAID-funded Strengthening Pharmaceutical Systems