In a short rejoinder, I thank Macklin and Benatar for their very helpful and thought-provoking comments and at the same time, I want to provide some additional information, which might be of interest to Macklin and our readers.
Residency Requirement
Macklin rightly comments that “the AIDS pandemic is a fast-moving target, necessitating appropriate changes in policies and practices where they are needed.” One such policy, she suggests, could be the residency criterion, which previously required that all those accepted for therapy in the Khayelitsha program be permanent residents of that township. ARV treatment is indeed more widely available in South Africa, as we suggested in our paper. However, this does not include all rural areas. For instance, most of Khayelitsha's population consists of persons of Xhosa origins from the Transkei region. One of their reasons for swapping a peaceful rural area for a filthy shack in an extremely violent partly urban environment is the lack of healthcare services in their homeland.
The other important point to consider is that the relationship of trust and confidence established between healthcare provider and patient is an essential element to ensure adherence in ARV treatment. Once this relationship is successfully established, it is very difficult to replicate it elsewhere. For example, we have several patients who prefer to travel more than 1,500 km every 2 months to be cared for in our clinics, rather than changing their treatment center. The same applies to the treatment of pregnant women after they have given birth.
Quantity of Persons Treated versus Quality of Treatment
It is very interesting that Macklin points out that the quality versus quantity dilemma stems from two different interpretations of the utilitarian principle. As is often the case with utilitarian principles, they require a fair amount of foresight and calculation, which might not always be feasible. To weigh the costs and benefits of treating as many people as possible versus providing the best possible treatment and care to those who are enrolled is not an exact science. To achieve an acceptable balance between quantity and quality in Khayelitsha, we defined some key outcome indicators (e.g., death rate, follow-up rate, viral suppression rate, CD4 increase rate), which are constantly monitored. This will, we hope, allow us to make the necessary predictions and calculations as accurately as possible in the future.
“First Come, First Served” versus “Jumping the Queue”
The dilemma, which Macklin also calls “best outcomes, versus fair chances,” is one of the most difficult ones we face at Khayelitsha. Shall we treat the most urgent cases in preference over those on the waiting list? Macklin's suggestion is to collect as much accurate empirical data as possible to help with those judgments. The problem is that one cannot maximize on the basis of just one particular outcome—for example, death or survival rates—as there are too many factors. Simplified, the empirical situation could be characterized as follows.
For fair chances: It is never too late to start ARV treatment. Patients who started with counts of <10 CD4 still have a survival rate of 75% at 36 months as compared to 86% survival rate overall. At the same time, HIV is only mildly symptomatic before it reaches the AIDS stage. It would be very difficult for family, friends, and the community to understand that HIV patients are treated in preference to serious AIDS cases. And it also has to be borne in mind that one of the main aims of MSF is to save lives, which might predispose staff to help those in most urgent need.
For best outcomes: Those in urgent need of ARV treatment show lower rates of immune restoration and more side effects, including severe immune reconstitution syndrome. This means that their treatment and care is more time consuming. Given the shortage of staff, this is a serious problem. Also, the presence of infection can very rapidly endanger the lives of those waiting faithfully in the queue for their treatment. This distinguishes them from, for instance, patients awaiting heart transplants, to whom Macklin alludes.
Even with accurate empirical evidence, this is not a straightforward decision to make. As a policy, we recently decided to include at least one patient with a CD4 above 100 for everyone selected with a count below 100.
Disclosure of HIV Status
Strict confidentiality about a potentially stigmatizing disease is a very laudable principle. However, in Khayelitsha we try to find a balance between strict confidentiality and what might be called semi-confidentiality for two main reasons: (1) feasibility and (2) desirability.
Nobody is forced to disclose their HIV status outside the clinic environment, with the exception of one treatment “buddy,” who could be a family member, a neighbor, or another ARV client met in the clinic. To maintain strict confidentiality and anonymity is impossible in clinics like those in Khayelitsha, which provide care for up to 1,000 patients suffering from HIV/AIDS at any one time. The pressure on clinics to manage the patient flow to guarantee strict privacy would be extremely high, thereby drawing resources, which are much needed for care and treatment. Evidence has also been gathered that openness about HIV/AIDS in high prevalence areas contributes to destigmatization of the disease and thereby encourages willingness to confront it. In addition, experience at Khayelitsha has shown that those who are unwilling to disclose their status to at least one person have not accepted the reality of their disease and may not achieve good adherence to long-term treatment.
