South Africa exceeded national targets for new patients starting antiretroviral treatment (ART) by around 50% between 2007 and 2011 – achieving treatment coverage of close to 80% of eligible adults – according to new research carried out by Dr Leigh F Johnson, actuarial scientist at the University of Cape Town, published in the March issue of The Southern African Journal of Medicine.
From mid-2004 to mid-2011, the total numbers of people receiving ART increased from 47,500 (95% CI: 42,900 to 51,800) to 1.79 million people (95% CI: 1.65 to 1.93 million). The latter figure represents close to 80% of adult treatment coverage, according to eligibility criteria in use during this period (CD4 cell counts under 200 cells/mm3). Using current South African CD4 cell count eligibility criteria (under 350 cells/mm3), coverage achieved decreases to 52% (95% CI: 46-57%).
While the targets were still exceeded, children and men started ART at considerably lower ratios than women.
Women accounted for 61%, men 31% and children 8% of the total.
Effective HIV treatment significantly reduces illness and death resulting from HIV, as well as onward transmission of HIV. Evaluating the effectiveness of HIV treatment and prevention programmes requires monitoring access to ART.
Previous monitoring assessments have shown a dramatic increase in access to ART in South Africa. While these assessments have suggested South Africa was on track to meet the targets of its HIV & AIDS & STI National Strategic Plan 2007-2011 (the NSP), no formal assessment has been made, Dr Johnson adds.
Indeed, his study shows just how difficult it is to be certain about the exact level of ART coverage being achieved in South Africa. The study finds that the total number of people receiving treatment lies somewhere between 1.65 and 1.93 million – a variation of 300,000 people.
Dr Johnson cites the challenges associated with monitoring access to ART in South Africa:
Understanding public sector statistics is complicated by a change in reporting in late 2009 from cumulative numbers on ART to those starting ART.
There has been no routine collection, nor reporting, of statistics from disease management programmes or those run by non-governmental organisations (NGOs).
Information about age and gender is lacking; of particular concern since the rates of men starting ART may be lower than among women.
Uncertainties about how ‘treatment need’ is defined – the denominator used to estimate ART coverage. Mathematical models have been used to estimate numbers of HIV-positive individuals with CD4 counts below different thresholds. Yet the rates of CD4 cell count decline assumed in these models and how they may differ between populations are unclear.
Cross-sectional measures may not account for recent programme performance.
ART data were collected from public and private ART providers. Independent demographic project models were used to assess HIV incidence estimates. To estimate the numbers of adults needing ART, a model was developed to simulate population growth over time, HIV incidence and estimated rates of progression (through CD4 decline), and fitted to local CD4 and HIV prevalence data.
Extraordinary progress has been made in South Africa in the roll out of ART since 2004 and the start of the public sector ART programme. The NSP 2007-2011 targets have been far exceeded and include a 32% reduction in unmet need for ART.
Of the 1.79 million people accessing treatment in 2011, 85% received ART through the public sector, 11% through disease management programmes in the private sector, and 4% through community treatment programmes run by NGOs.
The two provinces of KwaZulu Natal and Gauteng accounted for 56% of all those receiving ART.
Dr Johnson suggests that the lower rate of men starting ART compared to women may be linked to gender differences in health-seeking behaviours or the perception that men who seek care are ‘weak’. Or, he suggests, higher rates of women starting ART may be linked to higher rates of HIV diagnosis through antenatal care. He suggests that lower access rates in children may reflect lower rates of testing but cautions against making a comparison with adult measures of access since the course of the illness differs considerably.
The author had limited success in validating the reported ART programme statistics. Crude estimates of numbers of public sector patients on treatment in each quarter were calculated according to lamivudine (3TC, Epivir) sales figures from Aspen Pharmacare, which supplied 80% of lamivudine to the public sector until recently. While they did not differ significantly from the model estimates until the end of 2008, and again from October 2009 until March 2010, they were significantly lower than model estimates from January to September 2009.
Assuming patients on ART went for viral load testing twice a year, the numbers of viral load tests undertaken by the National Laboratory Service for public sector clinics were used to get a theoretical estimate of the numbers receiving ART. The estimates were both higher and lower than the model estimates, depending on the time period looked at.
Similarly, the model estimate of the proportion of 18 to 45 year-olds on ART at mid-2008 was significantly lower than the corresponding proportion estimated in the 2008 Human Sciences Research Council Survey (HSRC).
High levels of testing and counselling and increased capacity to deliver ART are needed if the ambitious NSP 2012-2016 ART enrolment targets (80% of those considered to be in 'new need' in that year, and 80% of the unmet need from previous years) are to be met and successfully monitored, writes Dr Johnson.
Dr Johnson concludes with the hope that the proposed measures to strengthen monitoring and evaluation of South Africa’s ART programme, which include a single patient identifier in the health sector and a single registry at the primary level, “will lead to greater precision in estimating future ART coverage, as well as a deeper understanding of the factors determining access to and retention in care”.