Value Of Data From HIV Testing/counseling Centers Questioned

Use of data from voluntary HIV counselling and testing clinics for HIV surveillance in Africa is not appropriate, according to a Viewpoint in this week's issue of The Lancet.

The rapid expansion of antiretroviral treatment for HIV-infected adults, which has been helped by various international initiatives* has demanded a scale-up of voluntary HIV counselling and testing services. Increases in clinic coverage, the obligation to fulfil donor-driven requirements of programme monitoring and assessment, and the desire to measure the effect of HIV-related interventions on the evolution of the epidemic, have led to suggestions to use data from these clinics for HIV and behavioural surveillance.

Gabriel Mwaluko (Tanzania Netherlands Project to Support HIV/AIDS control, Mwanza, Tanzania) and colleagues believe that this use of these data is not appropriate and argue that the data are inherently biased, and their collection could compromise the function of the clinics. How data for HIV prevalence from these clinics should be interpreted is not clear as they are influenced by the self-selection of patients who come forward for testing as well as by the underlying prevalence of HIV in the population - both are likely to change with time, and thus estimations of prevalence and trends are likely to be biased.

Furthermore, the authors state that the collection of data from different types of counselling and testing services to obtain national estimates presents problems - patients referred from medical wards to a testing centre within a hospital are different from those in a walk-in centre located in the community, who are different again from women accessing counselling via services for the prevention of mother-to-child transmission of HIV.

According to the authors, obtaining detailed data on client sexual behaviour could help assessment of the biases in HIV-prevalence, but the need for detailed structured questions about sexual behaviour does not conform with the ethos of the voluntary counselling and testing interview, and data collected when the client is emotionally stressed are notoriously unreliable. The authors state that this data would be better obtained by a community based inquiry - such as Demographic and Health Surveys that have been undertaken in many African countries - which inquire about testing history and sexual behaviour, and would not overburden voluntary counselling and testing services.

The authors conclude: "In view of the problems inherent in both obtaining and interpreting surveillance data from voluntary HIV counselling and testing centres, prudence should be exercised before extensions to routine data collections by providers of HIV counselling and testing services are recommended…what is needed in many countries is the implementation of well-designed but simple data-collection devices to ensure that useful and meaningful data are obtained and that duplication is eliminated."

###

Miss Alison Wringe, London School of Hygiene and Tropical Medicine, UK.

*Such as WHO's 3 by 5 strategy, the Global Fund to fight AIDS, Tuberculosis and Malaria, and the President's Emergency Plan for AIDs Relief.

Contact: Joe Santangelo
Lancet