Attrition From Human Immunodeficiency Virus Treatment Programs in Africa: A Longitudinal Ecological Analysis Using Data From 307 144 Patients Initiating Antiretroviral Therapy Between 2005 and 2010

Abstract

Background: As access to antiretroviral therapy (ART) in Africa has increased dramatically, concerns have been raised regarding patient attrition, an important measure of program quality.

Methods: We examined aggregate data from 307144 patients initiating ART in 5638 successive cohorts at 638 facilities in 9 African countries from 2005 to 2010, a period characterized by massive treatment expansion. Poisson regression assessed trends in 6- and 12-month cohort attrition (ie, the proportion of patients in each cohort no longer receiving ART at their initiating facility) over calendar time and as ART services matured, and identified factors associated with attrition.

Results: Across all 9 countries, 6- and 12-month cohort attrition was 21% and 29%, respectively, with no decrease over calendar time (6-month P = .8735; 12-month P = .5717) or as ART services matured (6-month P = .3005; 12-month P = .2277). Additionally, attrition remained stable or decreased across both measures in nearly all countries. Initiating ART in facilities with more documented transfers and fewer women on ART, and in cohorts with poor CD4 count documentation and lower median CD4 count at ART initiation was associated with increased 6-month attrition. Increased 12-month attrition was observed in semiurban facilities and those with more documented transfers, and in cohorts with poor CD4 count documentation, whereas higher patient load was associated with decreased attrition.

Conclusions: Stable or decreasing trends in attrition for ART patients were observed in most countries, suggesting programs can be expanded without compromising quality. However, further reductions in attrition are needed to maximize individual and population benefits of ART.

Keywords: ART; Africa.; HIV/AIDS; attrition; retention.