Introduction

In their book, Tragic Choices, Guido Calabresi and Phillip Bobbit write:

We cannot know why the world suffers. But we can know how the world decides that suffering shall come to some persons and not to others. In the distribution of scarce goods, society has to decide which methods of allotment to use…and when attention is riveted on such distributions they arouse emotions of compassion, outrage and terror. It is then that conflicts are laid bare between on the one hand, those values by which society determined the beneficiaries of the distributions and the perimeters of scarcity, and on the other hand, those humanistic moral values which prize life and well being. In such conflicts, at such junctures, societies confront the tragic choice.[1]

Without a doubt, HIV/AIDS is the greatest health crisis in the world today. The demand for treatment for people living with HIV and AIDS worldwide cannot be ignored. At the same time, there are countless challenges to meeting this demand, not the least of which are the cost of drugs, the complexity of treatment regimens, the inadequacy of health and delivery systems, the lack of knowledge about treatment, and the threat of drug resistance.[2] In order to meet some of these challenges, several developing countries are engaged in efforts to create national HIV/AIDS programs that steadily expand their public health sector capacity with the long-term goal of providing universal HIV/AIDS treatment.

As countries start or continue to scale-up their national antiretroviral (ARV) programs they will encounter difficult ethical decisions, including decisions about who should receive the limited available treatment. These choices will undoubtedly test their moral and ethical values.  In the end, because not everyone will have immediate access to life-saving drugs, tragic choices will have to be made. However, if the scale-up programs are successful they will offer a world of hope to millions of people who are in dire need of treatment and have no other means of receiving effective therapy. Because there is a population in immediate and dire need of treatment, decisions must be made with a degree of boldness.  But because the process is being scaled-up in an environment of great uncertainty, and because tragic choices will inevitably be made, these decisions must also be approached with a level of humility,

Progress in ARV Scale-up Programs

In the last two decades, over 30 million people have died of HIV/AIDS. Today, an estimated 40 million people live with HIV/AIDS - approximately 28.5 million of them in sub-Saharan Africa.[3] Outside of Africa, the Caribbean is the region hardest hit by HIV/AIDS,[4] but the HIV epidemic is also quickly growing in other areas such as India, Russia, and China.  Unfortunately, of the 6 million people in developing countries who currently need ARV therapy, fewer than 8% are receiving it, and without rapid access to properly managed treatment these millions of women, children, and men will die.[5]  Antiretroviral drugs (ARVs), especially when used in combinations of three or more, can dramatically improve the health of people living with HIV/AIDS around the world.[6] Unfortunately, for the vast majority of infected people in need, ARVs have been out of reach - until now.

In a span of about two years, mainly due to pressure from human rights and civil society organizations, we have seen the initiation of new programs such as The Global Fund to Fight AIDS,[7] the United States Presidential Emergency Plan for AIDS Relief,[8] and the World Health Organization’s Three by Five Program, which aims to treat three million people in five years.[9] Global funding for HIV/AIDS in resource constrained countries has, as a result, increased from just over $300 million in 1999 to an unprecedented $3 billion in 2002 and $4.7 billion in 2003, with additional funding promised by foreign governments and international donor agencies.[10] In less than a year, international discourse on HIV/AIDS in resource constrained countries has changed from “Is there a moral obligation on the part of industrialized nations to provide treatment?” (there clearly is), to “How can we best administer AIDS treatment programs?” Unfortunately, managing patients with HIV/AIDS is not just about providing drugs. As succinctly put by Dr. Gordon Perkin:

Even if we had free and unlimited supplies of ARVs and other essential HIV/AIDS commodities, they still would not be available to the majority of people who need them because of poor infrastructure.[11]

Building capacity and mobilizing resources on the scale needed to meet national treatment targets is daunting for many countries. Most health systems in sub-Saharan Africa and other resource constrained regions are over-stretched, under-developed and operate at sub-optimal levels.  There is a real danger that the benefits of drug availability could be undone by ineffective distribution and misuse, both of which could lead to the development of drug resistant viruses and the withdrawal of donor funding. In order for scale-up programs to be successful they must provide a package of services, including voluntary counseling and testing, monitoring of disease progression, prophylaxis, diagnosis and treatment of opportunistic infection, delay of viral replication with antiretrovirals, management of drug side effects, prevention of mother-to-child transmission, and provision of psychological and moral support to patients and their caregivers.[12]  Therefore, to maximize the benefits of ARV drug regimens, countries have to urgently make concrete plans to scale-up their treatment programs, establish or strengthen their national drug procurement and distribution policies, develop quality control mechanisms and engage in the relevant operational research.

On average, the treatment targets set by countries remain cautious, amounting to a combined total in 52 countries of approximately 500,000 people on antiretroviral treatment by 2005, less than 10% of those currently in need.[13] Some countries are moving beyond pilot treatment programs to set targets that more accurately match feasibility with need. For example, Thailand is currently providing treatment to 13,000 people with HIV/AIDS and aims to provide universal access by 2005.[14] The Economic Community of West African States (ECOWAS) is aiming to expand coverage to at least 400,000 people in 15 countries by 2005.[15] In addition to these country programs, non-government organizations such as Medicins Sans Frontiers (MSF) are providing ARV treatment through small pilot schemes and community programs in several developing countries.[16] The general consensus emerging at the international platform is that the ethical and political battle to reduce drug prices and provide funding for treatment has largely been won; the next battle is to develop and implement fair, effective, and equitable national treatment plans. 

Ethical Challenges Ahead

Ethical challenges are to be found at every level of the scale-up process, including in relationships between infected persons and their families/communities, between health care workers and patients, among institutions (both public and private), and among nations and/or international donor agencies.  As scale-up programs progress, new ethical questions continue to emerge. This paper will focus on five major ethical problems that are of current concern, namely:

        I.      The Allocation of Treatment

     II.      Administration and Delivery of Care

   III.      The Need for Operational Research

  IV.      Fairness across Disease Groups and Priority Setting

     V.      Sustainability of Treatment Programs

I.                   Allocation of Treatment

Among the most difficult questions scale-up programs face are: Who should receive the limited supply of therapy? What is the legitimate authority to decide who gets treatment? Justice and equity are often, and rightly, invoked as the underlying values in the distribution of limited resources. In the human rights framework, justice is defined as the quality of being fair. Equity is an ethical and value-based concept that is grounded in the idea that people should be treated as having equal worth. How can the concept of justice be used to promote equitable distribution in an environment with limited resources?

There are several theories of justice that can be applied to the allocation of scarce resources. For example, some people believe that a just allocation is one that provides the greatest good to the greatest number of people. They would therefore support the seeking of cures for more common (rather than rare) diseases, adopt programs that help many rather than few, and generally use funds where they will have the largest aggregate positive impact on the most people.[17] Others attempt to solve allocation dilemmas by holding that access to the same benefits, goods, and services should be provided to everyone on the same basis. For expensive and scarce resources, lotteries might be used so that all similarly situated individuals have an equal opportunity and are recognized as of equal worth (and need).[18]

Further, some believe that the primary responsibility for health care lies with private citizens and not with the state. They assert, therefore that market forces and personal choice should shape distribution of scarce resources.[19] Still others believe that distributions of social goods are fair when impartial people agree on the procedures that should be used for distribution. They contend that the way in which we form stable and just societies is through the process of building a consensus that merits endorsement by rational and informed people of good will.[20] Finally, some argue that allocation is not primarily an ethical problem, but an economic one. That is, that the allocation of resources on the basis of efficiency would obviate the need for ethical principles.

What has become increasingly clear is that there is no consensus at the policy level on which principles should apply to resource allocation decisions. In addition, even if everyone agreed on certain principles, there would still be disagreement on how those principles should be translated into health policy. Having said that, it is important for countries to engage in a discussion on distributive justice because the very concept of justice is not a luxury in many countries and there is a real danger that some allocation programs may be manifestly unjust and unacceptable. I believe that it is possible (and necessary) for countries to make informed judgments on allocation methods that are ethically robust and better suited to them based on their own socio-economic, political and cultural circumstances.

In the real world, countries and institutions are already setting up treatment eligibility criteria based on a variety of factors and ethical principles. For example:

Technical Criteria: The World Health Organization has recommended using medical criteria (disease progression) as the way to select candidates for its Three by Five program.[21] Under this program, patients with CD4 counts that are 200 or below or those clinically diagnosed with AIDS (where CD4 testing is not possible) are given priority. This approach, although technical on its face, makes certain value judgments about those individuals who do not meet the set criteria. People with CD4 counts greater than 200 might argue that they stand to benefit more than those whose disease has progressed significantly if they are given early access to treatment. That is, they might argue that this particular technical criterion does not maximize the good that could result from the program. 

Level of Health Care and Delivery Infrastructure: Some countries have decided to establish scale-up programs only where a certain level of basic health and delivery infrastructure already exists. In many developing countries, this limits the programs to urban areas. It can be argued that situating a treatment program in an existing health system that inherently favors the better-off (who generally live in urban areas) only serves to exacerbate existing inequities in health care systems that discriminate against the rural poor.

Social Worth and other Value-Based Criteria: In Uganda some doctors have argued that priority for treatment must go to those with the greatest risk of transmitting infections, such as HIV infected pregnant women.[22] Other communities feel that every person suffering from HIV/AIDS ought to have an equal chance of receiving treatment and that a random, lottery type selection process is therefore the fairest procedure for distributing treatment. Yet others are making decisions on whom to treat on the basis of indicators like age, level of education, and vocation. For example, arguments are being made that the treatment of infected health care personnel should be prioritized in order to maintain the crucial health delivery infrastructure. Some countries ravaged by war and political instability might make the argument that, before any national treatment program can be implemented, infected military and security personnel should be treated so that they can maintain the civil order necessary for scale-up programs. Countries like Kenya argue that there are certain groups that merit special consideration for treatment, such as victims of rape, people who became infected during vaccine trials or through administration of infected blood, HIV/AIDS orphans, and infected pregnant women. In South Africa, Medicins Sans Frontiers has an elaborate selection process which includes giving preference to patients on the basis of the number of dependents they have, their health status, level of income, and the voluntary disclosure of their HIV status.[23]

From the above, it is clear that different countries and communities will decide who receives treatment based on a variety of factors not limited to community values, local context, and/or the available health and delivery infrastructure. In the absence of established or structured decision making processes, there is a real danger that some countries will determine who gets treatment in ethically questionable ways, or through processes that are likely to be inequitable (biased in access) and open to abuse and corruption.[24]

II.                Administration and Delivery of Care

Once patients are selected, further ethical issues such as what standard of care is applicable arise at the level of administration of treatment.  For example: What minimum infrastructure is required to implement treatment plans and how does this infrastructure affect the quality of care provided to patients? What should the standard of care be for the administration of treatment in poor countries? Should there be uniform standards of treatment globally? Can the “Partners-in-Health” Haiti model,[25] which is often cited as the gold standard in the provision of ARV treatment in resource constrained settings, be successfully replicated in similarly situated developing countries? As stated earlier, one of the major obstacles faced by resource constrained countries is the poor state of their health and delivery systems. If these systems do not operate efficiently and effectively, the proper delivery and administration of drugs will be difficult or impossible. Problems with these systems arise from a lack of resources, including the inability to procure and sustain affordable drug supplies, the inability to manage complicated laboratory monitoring, a lack of trained personnel, inadequate community education strategies to encourage compliance, corruption at various levels, and the absence of sufficient political will in some countries. The tuberculosis example from Zimbabwe provides a practical and parallel illustration of this problem.

Tuberculosis drugs in Africa cost less than $10 dollars per course, for which the treatment duration is six months. In the 1990s, the Zimbabwean government initiated a national tuberculosis program to treat 50,000 tuberculosis patients annually. Unfortunately, this program has been fraught with problems such as a weak political commitment to tuberculosis control, funding gaps, low access to treatment due to poor infrastructure in new settlements, and limited involvement of communities in tuberculosis control. Consequently, the program has not been able to meet its target goals and Zimbabwean health officials are struggling to meet organizational requirements needed to continue administering the program. A conservative estimate of the number of people in need of antiretroviral therapy in Zimbabwe is 250,000 (Zimbabwe has about 1.5 million HIV infected people). The challenges that the government is encountering as it tries to implement the tuberculosis program will be greatly amplified in ARV scale-up programs unless the underlying infrastructural and political issues are adequately addressed.[26]

Tanzania also provides a vivid example of the importance of infrastructure, in this case health care personnel, in the scale-up process. There are fewer than 100 physician specialists in the public sector in Tanzania, serving a population of approximately 32 million. Regional and district hospitals have few or no physicians and are mostly staffed by clinical officers with only the most basic medical training. It is estimated that if physicians are required to prescribe ARV therapy, 400 more physicians would be needed immediately in the country.[27] This presents a difficult challenge for Tanzania, which may have to find other creative ways (such as recruiting and training traditional healers) to fill the health care personnel gap. This limited human resources pool poses further ethical challenges. If all or the majority of doctors and/or  health care workers are recruited to work in ARV scale-up programs, what does that mean for the entire health care system? Will other critical health care programs be deprived of the necessary personnel? Relatedly, if scale-up program responsibilities fall mainly on non-governmental organizations (NGOs), they will have to recruit health officers in the country and there is a great danger that there will be a ‘brain-drain’ from public programs to the better-paying private NGO sector. Safeguards have to be put in place to prevent such losses to public programs.

In summary therefore, the cost and logistical requirements for delivering, administering, monitoring, and evaluating treatment programs comprise a major obstacle to the implementation of ARV therapy in resource constrained settings. Because we are dealing with an emergency, there is a great temptation to expedite the dispensing of drugs to those in need and in so doing, bypass certain recommended processes. While the desire to expedite treatment is good, it creates an ethical imperative to explore, anticipate and make concrete provisions for any negative consequences that may arise due to these expedited actions. In the face of real threats of drug resistance, this issue is not merely academic. A careful balance has to be drawn between the ethical imperative to treat the sick immediately and the need to ensure that the expedited emergency treatment strategies employed will not fail dramatically for lack of sufficient infrastructure to the detriment of the affected population.

The Need for Operational Research

Operational research refers to the application of advanced analytical techniques to help make better decisions and to solve problems. In the context of ARV programs, it would include the carrying out of observational studies, outcome studies and cost-effectiveness studies.[28] As mentioned earlier, many scale-up programs are being conducted in environments where there more questions than answers. For example, what are the best ARV regimens in a given setting? How can barriers to adherence be assessed? Who derives the most benefit from ARVs (taking in to account behavioral factors, stage of disease and laboratory parameters)?[29] How does HIV interact with other endemic diseases and what is the effect on the patient’s response to treatment? How effective is the program in prolonging survival and ensuring quality of care? How can the frequency and predictors of ARV resistance be properly and easily detected?[30] What are the side effect profiles of ARVs in different populations?[31] Conducting operations research can therefore help in defining the optimal parameters of care in a variety of ways.

Because there are long and short-term uncertainties in ARV scale-up programs, there is an ethical obligation on the part of treatment teams to set up the necessary structures to learn from such programs, with a view to improving them. Operations research is thus a critical element to the iterative process of improving quality of care. In addition, monitoring and evaluation measures should not simply focus on process measure, but also the quality measures. [32] In order to do this, donors and countries have to invest in the creation of information systems to gather, analyze and manage the relevant data. Because of the sensitivity of the information being gathered including patient records, privacy safeguards will have to be built into the systems.

Without a systematic approach to learning by doing, there is a great danger that programs may end up not only wasting limited life-saving resources, but also perpetuating potentially harmful health care practices – an ethically unacceptable state of affairs.

IV.             Fairness across Disease Groups and Priority Setting

Consider this hypothetical:  In a rural family in Kenya, the father, a farm worker, has been diagnosed as HIV+ and because Kenya is one of the recipients of donor finances to fund treatment for HIV/AIDS, and because the father fits within the eligibility criteria, he is put on a free treatment regimen complete with monitoring and support from the community clinic. The mother suddenly falls ill with malaria, is bedridden and cannot work. The father’s wages are not enough to provide school fees and other necessities for the children. The father, at a complete loss for what to do, decides to sell his HIV treatment in the black market (the drugs will benefit an HIV patient who did not fit into the scale-up program treatment entry criteria). The money he receives from the sale is enough to buy malaria treatment for his wife and to pay for some of his children’s needs. In this scenario, what are the health priorities of the family? At a broader level, what priority should the provision of ARV therapy have in relation to the many other competing interests that fall outside of the health care sector in developing countries?[33]

Although HIV/AIDS is an epidemic of unprecedented proportion, people in developing countries suffer from a wide range of devastating ailments, such as malaria and tuberculosis, for which there are not enough resources to provide treatment. Problems also exist at the basic public health level due to a lack of immunization programs and inadequate access to proper nutrition, clean drinking water, and sanitation.[34] In the recent past, more than 14 million people have been at risk of starvation in Zimbabwe, Zambia, Lesotho, Swaziland, Malawi, and Mozambique in Southern Africa alone.[35] Resources in those countries are urgently required for education, nutrition, roads, strengthening local agricultural systems, as well as for industrial and technological developments to compete in the global market.[36] The above hypothetical does not suggest that donors should stop funding AIDS programs; rather, donors need to explore how their programs can have synergy with other programs addressing other health and welfare needs and, more broadly, how funding for HIV/AIDS can improve health care infrastructure and development in the long-term. Approaches to expand access to ARV treatment should simultaneously strengthen health systems, interact with treatment, prevention, and health care services, and reach vulnerable groups.[37]

V.       Sustainability of Treatment Programs

The current initiatives to provide funding for treatment are laudable: nevertheless, there needs to be a frank discussion about the sustainability of donor-funded treatment programs. How should poor countries deal with the political question of dependency on rich countries for life sustaining treatment for their citizens? Is this a sustainable relationship in both political and ethical terms? The sustainability issue is not limited to donor funded programs, but also to national government funded initiatives that rely to a large extent on the political will of the day. For example, in February 2004, AIDS activists in Nigeria reported that a government plan to provide cheap AIDS drugs had failed, threatening the lives of people with HIV who had started taking the drugs two years ago. Launched in 2002, the Nigerian scheme was riddled with logistical, financial, and corruption problems that led to treatment centers handing out expired drugs or turning patients away.[38]  

Finally, at the institutional level, technical assistance should be offered to help health and humanitarian organizations and governments that are traditionally designed to deal with health problems on shoe-string budgets, to cope with the influx of huge amounts of donor money that could very well transform the nature of their operations.

Some Recommendations

Ethical problems will be encountered at every turn of the scale-up process. Furthermore, ethical decision making in many countries will be constrained by the prevailing political climate. It bears emphasizing that the manner in which these ethical problems are framed, debated, and decided upon will have serious consequences for the health of people. As the scale-up process unfolds, new ethical issues will emerge requiring guidance based on sound ethical principles. I however believe that while deliberation on ethical issues is extremely important, it need not and should not stop or slow down the provision of ARV therapy in resource constrained countries until some ideal or final ethical consensus is reached. Whereas there may be no clearly right or ideal solutions, there can be solutions that are clearly wrong. In that spirit, I believe that ethical decisions should be made in real time and should be further refined by a process of ongoing deliberation that is informed by, and informs, ongoing ARV treatment efforts.

On the question of allocation of treatment, I think it is necessary for donors and international aid agencies involved in providing resources for ARV scale up to engage with recipient countries to establish a minimum entry criterion grounded in internationally recognized human rights.  I believe that such a criterion is important to create a level of horizontal equity, that is, people who are similarly situated should receive similar treatment, and to prevent blatant discrimination against certain classes of persons. This common criterion would also provide some guidance for decision-making at the national/local level. It would include broad rights-based elements such as the right not to discriminate on defined grounds like gender, sexual orientation, religion and ethnic background. Countries would thus ideally be bound by these agreed set of norms, but would still have enough leeway to make ethically defensible decisions on who should get treatment within the set framework, with the full understanding that tragic  trade-offs will be made, and that decisions arrived at by the countries and communities, will vary.

At the national or local level, I think that recipient countries have an ethical imperative to create impartial decision-making bodies to carefully deliberate (in an open and transparent manner), the ethical problems arising in scale-up processes, provide reasoned guidance and make context specific decisions about how treatment should be distributed and administered.[39] I acknowledge that this is a potentially difficult proposition because such a body would have to have sufficient political leverage to get its decisions enforced, and at the same time, be independent enough to take actions that may contradict government policy. In addition, while the decision-making body might have a set criteria within which to select candidates at the Olympian level, at the human level, members of the decision-making body may find it very difficult to turn away persons who do not fit neatly into the criteria – and may even try to find ways to “make them eligible” in order to receive treatment.[40] I realize that even with all the best decision-making process in place, there will inevitably be problems surrounding representation on such bodies, legitimacy, corruption and sustainability of such processes—however, I believe that if created, such bodies would offer a decent start to solving some of these ethical dilemmas, and they would be able to refine their deliberative processes as ARV scale-up progresses.

Finally, as stated above, the provision of treatment should not be delayed in the name of perfecting the decision-making process or in the name of finding the best ethical theories to support pragmatic decisions. Even if a decision makes moral sense, it may be difficult or impossible to implement it on the ground. I strongly believe that making concrete attempts to anticipate and address ethical problems will make for more effective and equitable programs, as will a commitment to reconsidering and refining these ethical hurdles as the programs progress. However, while ethical principles and ethical deliberation are most important, they should not used as an excuse to delay  or prevent the implementation of lifesaving treatment—saving any life is better than saving none.

* Angela Amondi Wasunna LLB LLM is Associate for International Programs at the Hastings Center, New York. She is also an Advocate of  the High Court of Kenya.

[13] WHO/UNAIDS, “Commitment to Action for Expanded Access to HIV/AIDS Treatment” excerpt from International HIV Treatment Access Coalition,  http://www.who.int/entity/hiv/pub/prev_care/en (accessed 2003)

[15] Ibid.