Antiretroviral adherence in resource-poor settings: Challenging the voices of doom

This article is more than 21 years old.

Dallas, Texas: The annual release of updated global HIV/AIDS prevalence data from UNAIDS is one of the gloomier events of what locals call the Holiday Season, and the rest of us call December. The ratcheting-up of numbers infected by additional millions, the 30%-plus prevalence rates in several African nations, and the emergence of rapidly-spreading epidemics in Eastern Europe and Asia paint a grim picture indeed. World-wide, HIV continues to work much more effectively than we do in our attempts both to contain it’s spread, and provide care to those affected.

Whilst UNAIDS’ AIDS Epidemic Update provides a vital wake-up call, approaching the epidemic from this macro level inevitably hides the scores of micro level successes which need to be reported in equal measure. Just one of these, from an antiretroviral access programme in South Africa, was reported at a meeting on treatment adherence here in Dallas this weekend.

As Dr Catherine Orrell, of the Diana, Princess of Wales HIV Research Unit at Somerset Hospital, Cape Town, noted, the prejudices of some Western scientists continue to add to the challenges which people living in resource-poor settings have been dealing with for many years. Adherence to antiretroviral therapies is a good example here. From brutish remarks about Africans’ ability to tell the time, to papers in well-respected medical journals, the assumption that people in resource-poor settings will struggle with adherence, and lose, has been trotted out again and again, and undoubtedly acts to prevent access to medicines in these communities.


relative risk

Comparing one group with another, expresses differences in the risk of something happening. For example, in comparison with group A, people in group B have a relative risk of 3 of being ill (they are three times as likely to get ill). A relative risk above 1 means the risk is higher in the group of interest; a relative risk below 1 means the risk is lower. 

multivariate analysis

An extension of multivariable analysis that is used to model two or more outcomes at the same time.


The Joint United Nations Programme on HIV/AIDS (UNAIDS) brings together the resources of ten United Nations organisations in response to HIV and AIDS.


A variant characterised by a specific genotype.


drug resistance

A drug-resistant HIV strain is one which is less susceptible to the effects of one or more anti-HIV drugs because of an accumulation of HIV mutations in its genotype. Resistance can be the result of a poor adherence to treatment or of transmission of an already resistant virus.

In a letter to AIDS in March of last year, Popp and Fisher, of the Center for HIV Intervention and Prevention at the University of Connecticut, argue that: "Without any behavioural science-based interventions aimed at ensuring adherence, the seemingly humanitarian efforts of drug companies, governments, and the UN could have explosive unintended negative consequences. Individual patients may not benefit, may become treatment resistant, and developing countries could become a veritable ‘petri dish’ for new, treatment-resistant HIV strains."

Undoubtedly these concerns are evidence-based. But the evidence base has largely been generated in high-income nations, where the days of using poor adherence as an argument to withhold antiretroviral treatment from vulnerable individuals are, one hopes, some way behind us.

Treatment success in Senegal

An early report from an evidently humanitarian effort; a government-supported antiretroviral treatment programme in Senegal, which was sponsored by the French National Agency for Research on AIDS and the European Union, demonstrated that Popp and Fisher were wrong on all counts. Published in the July 5th issue of AIDS, eighteen month follow-up of 58 Senegalese beginning HAART in Dakar in 1998, reported that 87.9% of participants adhered successfully. Viral load was undetectable in 71.2%, 51.4% and 59.3% at months 6, 12 and 18 respectively. Largely an advanced patient population, the immunological benefits of treatment were also clear: median CD4 count increase at eighteen months was 180 cells. Drug resistance emerged in just two people, and the evidence suggests that interrupted access to treatment (participants were required to contribute US$34 to the cost of their therapy) rather than poor adherence was the cause. When one considers that the regimen used in this programme was over-whelmingly d4T/ddI/indinavir, amongst the most demanding HAART combinations available, it’s tempting to suggest that this study presents a model of good practice which many in the West may benefit from adopting themselves.

Treatment success in Khayelitsha

Turning to South Africa, an antiretroviral programme in the Khayelitsha township of Cape Town provides further evidence that HIV medicines can be supplied successfully in resource-poor settings. A Medecins Sans Frontieres-supported project supplied NNRTI-based HAART to 177 individuals. From a median baseline CD4 count of 48 cells, the increase after eight months of treatment was 115 cells. Ninety-one per cent had suppressed viral load at this point, and importantly, participants benefited from an average 8.8kg increase in body weight.

New report from Cape Town

The findings of Catherine Orrell and colleagues are currently in press with AIDS. Previewed here, her cohort includes 289 people enrolled in antiretroviral therapy trials. Eighty per cent did not speak English as a first language, and 42% lived in informal dwellings or shacks. After a year of treatment, adherence as measured by pill count averaged 87.2%; the median value being 93.5%. At 48 weeks, 70.9% had viral load below 400 copies.

The strongest predictor of poor adherence by multivariate analysis was taking a three times daily regimen compared to twice daily (relative risk 3.07). Other significant factors were speaking English, which was protective with a relative risk of 0.41, and older age, also protective; relative risk 0.97. Gender, HIV stage and socio-economic status did not influence adherence, though medication was provided at no cost in this study.

Overall, these studies present treatment response rates which are at least comparable with those routinely observed in high-income nations. Though they support the notion that adherence is not an inherent barrier to antiretroviral access in resource-poor settings, Dr Orrell acknowledges that problems relating to finance, infrastructure and staff expertise remain significant. A recent surveillance report from a medical school in the north of South Africa found that 25% of medical students were HIV-positive. If these essential workers are to fulfil their role in future, these obstacles will need to be addressed. As Dr Orrell had demonstrated, this scenario is achievable.

Practice guidelines from the British HIV Association and Medical Society for Study of Venereal Diseases on supporting antiretroviral adherence are open for consultation here on


AIDS Epidemic Update. UNAIDS December 2002.

Popp D and Fisher JD. First do no harm: A call for emphasizing adherence and HIV prevention interventions in active antiretroviral therapy programs in the developing world. AIDS 2002;16:669-80.

Laurent C et al. The Senegalese government’s highly active antiretroviral therapy initiative: an 18-month follow-up study. AIDS 2002;16:1363-70.

Kasper T et al. Antiretroviral therapy in primary health care centers in a South African Township. 14th International AIDS Conference, Barcelona, July 7-12, 2002, abstract MoOrB1095.

Orrell C. The challenge of adherence in resource poor settings. International Conference on Adherence to Antiretroviral Therapy, Dallas, December 5-8, 2002.