A Decline in Community Viral Load in Cape Town and Johannesburg, South Africa between 2004 to 2011

Abstract

Background: The total number of people living with HIV in South Africa in 2011 is estimated at approximately 5.4 million, with a prevalence of 16.6% in the adult antenatal population: ranging from 6.2% in the Western Cape to 16.6% in Gauteng provinces, respectively. Antiretroviral therapy (ART) became accessible in 2004, with an estimated 1.4 million on treatment by December 2011. Prior work has suggested that a community viral load (CVL) can be used as a tool for monitoringthe success of ARV therapy. This pilot study measured the population based CVL in the cities of Cape Town (CT) and Johannesburg (JHB) and in a large urban clinic from 2004 and 2011. Methods: Retrospective analyses of HIV viral load (VL) tests performed over the 8-year period on patients from CT, JHB and an urban clinic were undertaken on data extracted from the central data warehouse of the National Health Laboratory Service and the clinic database. For each calendar year the CVL was represented as a mean (log) VL, and as a proportion of patients with suppressed VL. Each patient contributed only one VL per calendar year. Lower than detectable limit (LDL) results were assigned a theoretical minimum detectable level of the assay and values >3.0 million cp/mL were truncated at thatvalue. Results: We analyzed 848,325 VL tests matched to 550,527 patients. Patients were linked with a median of 2 tests in CT, 1 test in JHB, and 4 tests at the urban clinic. Total follow-up time after the first VL test ranged from no follow-up in the city communities to 24 months in the urban clinic. The proportion of LDL VL’s increased in the community and clinic sites between 2004 to 2009 and declined again in 2010. The proportion of subjects with VL <1000 cp/ml rose to approximately 84% in CT, 79% in JHB, and 88% in the urban clinic. CT demonstrates a higher proportion of subjects with suppressed viral loads than JHB, but the urban clinic shows the highest levels of suppression. Conclusions: The dramatic decline in CVL from 2004 to 2011 in two major cities in South Africa suggestssuccesses in the ARV treatment program. The proportion suppressed at the urban clinic generally exceed what was being observed at the community level. Our next step is apply this monitoring tool to district level and to evaluate outcomes of HIV control programmes for corrective action and appropriate direction of resources.

Conference: CROI conference 2013, Atlanta, USA