|
Can antiretoviral therapy contain a previously escalating TB epidemic in a high HIV prevalence community?
K. Middelkoop1, R. Wood1, L. Myer2,3, E. Sebastian4, L.-G. Bekker1
1University of Cape Town; Desmond Tutu HIV Centre, Institute of Infectious Disease and Molecular Medicine; Department of Medicine, Cape Town, South Africa, 2University of Cape Town, Infectious Diseases Epidemiology Unit, School of Public Health & Family Medicine, Cape Town, South Africa, 3Columbia University, Department of Epidemiology, Mailman School of Public Health, New York, United States, 4University of Cape Town; Desmond Tutu HIV Centre, Institute of Infectious Disease and Molecular Medicine, Cape Town, South Africa
Background: The World Health Organisation (WHO)-recommended directly observed therapy short course (DOTS) strategies are failing to contain the tuberculosis (TB) epidemic in high HIV prevalence countries. However there are few population level insights into how the availability of ART may impact on TB rates in areas where HIV is prevalent.
Methods: We monitored TB notifications in a South African community with 23% HIV prevalence and a well-functioning TB programme based on DOTS. Notification data came from the local TB clinic from 1998-2004 (prior to ARV availability) and from 2004-2008 (following ARVs roll-out; by end 2008, an estimated 31% of the HIV-infected population was on ARVs). National TB program guidelines for diagnosis and management of TB patients have not changed significantly over the study period. Rates were calculated using population denominators from community census data.
Results: Between 1998 and 2004 population annual TB notification rates (cases/100,000) increased significantly from 899 to 1,554 (p< 0.001). Increased notifications were due to a significant rate increase among HIV+ individuals from 4,268 in 1998 to 6,566 in 2004 (p< 0.001) while rates in HIV-ve patients remained stable around 550. Significant ARV roll-out commenced in 2005 when community TB notifications peaked at 2,132 after which rates declined to 1,804 in 2008. During this 4 year period HIV-ve TB rates remained stable around 570, and rates among HIV+ve patients not receiving ART decreased moderately from 8,266 to 7,785 (p=0.41) while rates of those receiving ART declined markedly from 6,806 to 1,650 (p< 0.001).
Conclusions: Wide-scale availability of ARVs, coupled with a well functioning TB program, appears to be associated with decreases in TB notifications in this community, and these decreases are most prominent among HIV infected individual on ART. The 2005 peak in TB rates may be due to increased screening for TB in patients accessing ARVs.
Download the e-Poster
|