Success and failure in HIV treatment: contrasting lessons from South Africa

This article is more than 21 years old.

Contrasting stories about using antiretrovirals where resources are limited emerge from two letters in the latest issue of the journal AIDS. One tells of the progress made by Médecins Sans Frontières (MSF) in showing that HIV treatment can be delivered as effectively as in industrialised countries. It is complemented by other MSF reports published by the Global Health Council and the World Health Organization. Another describes a private clinic in KwaZulu Natal, South Africa, where financial and other constraints led to many patients receiving suboptimal treatment.

Jean-Michel Tassie and colleagues from MSF outline, in their letter to AIDS, the experience up until May 2002 of the patients who started HAART in seven projects in different countries across Africa (Cameroon, Kenya, Malawi, South Africa), Asia (Cambodia, Thailand) and Latin America (Guatemala). Despite treating people in advanced stages of disease - the median CD4 cell count was just 48 cells per mm3 - treatment was clearly effective in most of those who could tolerate it, with a disproportionate number of deaths in the first 30 days on treatment. The probability of survival at 6 months was estimated at 89.5%, and among 118 patients whose viral load could be tested after 6 months on treatment, 106 (89.8%) were below 300 copies/ml.

Rachel Cohen and Kevin Phelan remind us that this is what MSF reported at the Barcelona International Conference on AIDS last year. Since then, things have moved on: "As of June 2003, MSF was providing antiretroviral (ARV) therapy for approximately 5,000 patients in 23 programs in 14 countries: Burkina Faso, Cambodia, Cameroon, Guatemala, Honduras, Indonesia, Kenya, Malawi, Mozambique, Myanmar, South Africa, Thailand, Uganda and the Ukraine. By the end of 2003, MSF expects to be treating around 10,000 patients with ARVs and, in 2004, anticipates running a total of nearly 50 ARV treatment programs in 30 countries. Additional countries will include Angola, Benin, Burundi, Chad, China, Democratic Republic of Congo, El Salvador, Ethiopia, Guinea, Laos, Nigeria, Peru, Rwanda, Sierra Leone, Sudan, Zambia and Zimbabwe."


pilot study

Small-scale, preliminary study, conducted to evaluate feasibility, time, cost, adverse events, and improve upon the design of a future full-scale research project.



Taking a drug on its own, rather than in combination with other drugs.

Lessons from one of those projects, in the Khayelitsha, 30 kilometres outside Cape Town, South Africa, are set out in an article by Toby Kasper and others in the latest issue of WHO's Essential Drugs Monitor.

This article stresses the value of involving the community in provision, which is aided by giving treatment at primary health centres, and of involving the patients themselves. The patients have been active politically, in speaking up through local media for the need to sustain the programme, at community level in working with the Treatment Action Campaign on local education efforts, and individually, by informing themselves about their own treatment and taking responsibility for their own adherence to it.

It is argued that the effect of ARV treatment in preventing opportunistic infections makes it much easier to manage patients on ARVs, especially after the first few months, than if ARVs had not been provided. In Khayelitsha, the development of standardised protocols to manage adverse events that might be linked to treatment has allowed an increasing proportion of the care of patients on ARVs to be taken on by nurses rather than doctors.

Part of the success of the programme in Khayelitsha has clearly been due to selection of patients most likely to adhere to treatment on criteria which include a history of attendance at the health centre and may take account of other criteria such as their level of openness about their HIV status in their household. Some priority is given to people who have another household member on treatment, to alleviate any pressure to share medication inappropriately.

One of the challenges and questions must obviously be, how to sustain the same level of adherence while extending access to a higher proportion of those who actually need it.

A contrasting picture of ARV treatment emerges from a study by two researchers affiliated to Canadian and Kenyan universities who reviewed the records of a private primary care clinic in rural KwaZulu-Natal, South Africa.

72 patients were prescribed ARVs between March 1999 and February 2002. Only 57 of these patients actually received ARVs - because some of them had run out of insurance cover or didn't meet an insurance scheme's criteria for treatment (e.g. through having a CD4 count above 350). Only 31, all of whom had medical aid - were treated with triple therapy, mostly efavirenz, AZT and 3TC. 17 received double therapy, seven received regimens containing hydroxyurea, and two received monotherapy. This means that all of the patients who paid out-of-pocket for their treatment, and some who had support from a medical aid scheme, received suboptimal treatment according to the current standard set by WHO.

Adherence was assessed on the basis of repeat prescription records which were available for 36 patients. Only eight patients were more than 90% adherent and 21 were less than 70% adherent. "The reasons for non-adherence were rarely known, but included patients thinking that a month of therapy was sufficient, sharing with a spouse and medical aid funds being exhausted."

Follow-up data were available for 37 patients, 21 of them on triple therapy, and found only 12 to have achieved a viral load below 400.

Despite relatively advanced health care infrastructure and monitoring, this study shows that when ARV treatment is provided with inadequate financial underpinning and without sufficient training and support for staff and patients, the results do not match those seen in properly managed pilot studies such as those reported by MSF.

The authors conclude, "Studies performed in centres with considerable support for providers and patients have shown that ART can be safe and effective in resource-limited settings. Our study shows that when ART is introduced without adequate support for providers and patients the results can be disappointing." ... "There is an immense need for operational research in antiretroviral settings that not only advocates the use of these medications but describes ways to integrate them into existing private and public sector activities."


Cohen R et al. How Do We Scale-Up Access to ARVs? Treatment Now Global Health Council, Washington DC: AIDSLink, Issue 81, 2003 [available online here].

Kasper T et al. Demystifying antiretroviral therapy in resource-poor settings. WHO, Geneva: Essential Drugs Monitor, Issue 32:20-21, 2003 [available in English online in pdf format here].

Livesley N et al. Antiretroviral therapy in a primary care clinic in rural south Africa (letter). AIDS 17:2005-2006, 2003.

Tassie J-M et al. Highly active antiretroviral therapy in resource-poor settings: the experience of Medecins Sans Frontieres (letter). AIDS 17:1995-1997, 2003.