Jean-Hervé Bradol
Paris, février 2010

This document is the result of a visit in July 2010 to the MSF Homa Bay project in Kenya’s Nyanza Province and part of a study launched in 2009 on the future of MSF AIDS projects. The past two years have seen tense debates among the teams regarding future directions. Sharp exchanges – and sometimes needless humiliation – have left their mark on the people involved, both in Kenya and at Paris headquarters. This atmosphere reflects the bitter reality of the discussions on the directions to take and the power issues seen in any institution. The visit was useful, however, thanks to the resolve of all those concerned to prevent a useful debate from turning into a series of squabbles. The visitor was thus able over a period of about ten days to familiarize himself with the principal work situations and issues faced by the teams. The author has been going to Kenya for some twenty years, and has been visiting the MSF clinic serving the population of Nairobi’s Mathare slum since 1998. The clinic’s activities now centre on HIV, tuberculosis, and women victims of sexual violence. The focus of attention on this visit was the Homa Bay project, which covers two districts (Homa Bay and Ndhiwa) in Nyanza Province. As the project grows and evolves, the question of HIV treatment coverage for all those who are medically eligible is posed.

The internal discussions on the future of HIV projects were woven together from a number of topics, which can be grouped into three sections: the setting (local, national, and transnational) in which MSF works, the role it would like to play in the fight against HIV, and how it manages its AIDS projects. Once back in Paris, the challenge was not to tackle all the issues, but rather to examine certain important points that keep coming up in discussions, while trying to find openings where the assumptions and formulation of questions in the debate wouldn’t lead to the same impasses.

A field project is a set of activities aimed at meeting objectives in a given place or institution (for example, an administrative department, a town, a camp, a hospital or any other medical or social institution).

Each field project is run by a national or international team under the authority of a field coordinator. At the national level, projects are overseen by a capital team, led by the Head of Mission. For each member of the field team there is a double chain of command, political: (Field Coordinator, Head of Mission, Desk Programme Manager, Operations Director, Executive Director, President, and Board of Directors) and professional: (for example, Doctor, Field Medical Coordinator, Capital Medical Coordinator, Desk Medical Advisor, and Medical Director).

The desk is a subdivision of the Operations Department under the authority of the Operations Director. It is responsible for steering the projects, choosing among the different possible activities, and monitoring them from headquarters. It is led by a Programme Manager and consists of several representatives from the professions needed for operations (medical technology, human resources, administration, and logistics). See the sheet on the Kenya project, 2010, in the appendices.

THE LOCAL, NATIONAL AND TRANSNATIONAL CONTEXT

THE PANDEMIC, AS SEEN FROM KENYA

The 2007 national survey already showed an average prevalence of just over 7% for Kenya’s adult population (ages 15 to 64). However, HIV prevalence was 15% in Nyanza Province, yet less than 1% in the North Eastern Province. In one province AIDS is a health catastrophe, in the other just another disease. In some places, the epidemic has never assumed tragic proportions. In Kenya, as in other countries, the situation is still disastrous for large segments of the adult population living in high prevalence zones and for groups of individuals whose practices not only put them at greater risk of infection, but make them less likely to receive care, due to the disapproval they elicit (men who have sex with other men, prostitutes, drug users, etc.). But the discrepancy between available resources and the large number of cases as well as the stigmatization of certain lifestyles are not the only reasons why some people are excluded from treatment. Patients with severe forms of the disease and children – whose needs differ from those of the majority of patients, who are adults – are penalized by care poorly-adapted to more complex and expensive protocols.

The term “pandemic” – an epidemic progressing within a very wide geographic area, over several continents – is more a social and political necessity for mobilizing epidemic response than it is a solid epidemiological concept. Should we measure it by the proportion of the world population that is infected? The resulting morbidity and mortality? In 2009, when asked to give a precise definition of the term “pandemic” during the debate over public expenditures for H1N1 flu outbreaks, medical experts – the WHO leading the way – were hard-pressed to come up with an answer. The way that politicians and activists use the word “pandemic” suggests there’s a single reality encompassing all the various epidemics; that there’s an invisible thread connecting San Francisco’s backroom regulars with the patrons of Kinshasa’s brothels. Politically, the feeling of a shared fate with respect to the disease has unquestionably paved the way for international solidarity. It has helped provide wider access to a treatment that was at first available only to a privileged few – patients in wealthy countries. But for all that, being infected by the same virus doesn’t create a common interest, or indeed a common destiny, between such geographically and socially disparate people. Even within the same country, the epidemiological profile and daily lives of IV drug users set them apart from the rest of their HIV-positive fellow citizens.

THE PACE OF KNOWLEDGE ACQUISITION AND MUTATIONS IN FUNDING

One argument used to convince donors to continue their efforts is that, despite the emergency situation, the response is slow and inadequate. The fact that the available drugs and protocols cannot stop the spread of the disease or treat all of the patients contributes to the feeling that research is not progressing. In reality, there has been rapid progress in knowledge and know-how since the defining of the new syndrome in 1981. The disease was quickly understood to be caused by a virus (1982) belonging to the family of retroviruses (1983). The virus was cloned, and its genome decoded in 1984. Soon the first laboratory test for confirming diagnosis became available (1985), and the first drug, zidovudine (AZT), was proven effective (1987). A treatment capable of extending life by several years (HAART) was identified in 1996. In roughly ten years (2001-2010), five million people were treated with antiretroviral drugs, despite the fact that the vast majority of them lived in low- or middle-income countries.

The past three decades have seen the burgeoning of prevention and treatment initiatives on a scale and pace unheard of in the history of public health. So should we assume that biomedicine is now so expert and so powerful that we can expect equally profound and rapid advances in coming decades? The fact that MSF programmes were using HAART barely five years after it was proven effective is remarkable. It usually takes longer; humanitarian doctors often have to wait twenty years or more before being able to use a new treatment. While major strides have been made, we still don’t have a vaccine to prevent infection, or a drug capable of completely eliminating the virus from the body. And while prevention and treatment tools reduce the number of new cases and slow the progress of the infection, they are still ill-suited to living conditions and care-provider practices in the hardest-hit countries. In Kenya, the simple fact that 270,000 patients have been put on antiretrovirals in just a few years illustrates the pace and scope of the changes – especially since management of chronic diseases was until then a relatively small part of Kenya’s public healthcare sector.

In July 2010, results from Kenya’s latest national HIV prevalence survey were just beginning to circulate. The United Nations held Kenya up as an example, because the HIV prevalence in young adults aged 15 to 24 had declined between 2007 and 2010, meeting the goals set by the donor States. The UN’s announcement of these encouraging results implied that Kenya was at the cutting edge of the fight against the pandemic. In an international context where the will of donors is flagging, clearly it was useful to show – using the example that a part of Kenya’s youth were adhering to preventive measures – that the worldwide campaign against HIV could indeed claim some victories.

Repeated requests made by the Campaign for Access to Essential Medicines to the teams in Kenya for an example of how reduced funding impacts patients in the field are further evidence that global strategy affects local reality. Actually, Kenya has not yet lacked funds – in fact, it is having trouble using the money that’s available and justifying the use of certain sums already spent. “Kenya’s three Round 2 grants have reached the end of their 5-year terms, with a total of 47% ($67 m.) of the agreed Phase 1+2 funding un-disbursed and no longer available….Audits at the end of 2004, 18 months into the Round 2 grants, showed that 99% of the $7 million disbursed by then was unaccounted for. In August 2008, the Global Fund still listed those audit queries as being unanswered.”Bernard Rivers, Aidspan, Problems with Kenya’s Global Fund Grants. This paper was presented at a retreat of the Kenya Country Coordinating Mechanism meeting, Naivasha, Feb. 19-21, 2009, pp. 1-2.Kenya is neither ahead nor behind in the battle against the pandemic, but simply in a place of its own, which cannot be reduced to a single point on a numeric scale of quantitative indicators for monitoring the global HIV strategy. When pressed, the Kenya teams managed to make their point of view known to Access Campaign headquarters, and released a public statement limited to the country but in relation to the facts. At the same time, MSF communications departments everywhere else rallied to condemn the donor states’ lack of mobilisation at the XVIII International AIDS Conference, organized by the International AIDS Society, in Vienna in July 2010. How were public health institutions, only recently described as extremely inadequate, supposed to absorb all that money without a hitch and transform it into quality care?

The non-use of some of the international funds allocated to fighting HIV in Kenya was a reminder of a reality that didn’t tally with MSF advocacy, and thus had to be avoided in general discourse aimed at describing the weaknesses in the global pandemic response. First, information slanted in this way is objectionable, because it reduces a complex problem to the single issue of whether foreign institutions in the country in question have “the political will to fund the fight against HIV.” While the reductive nature of the message has been brought up, it is still considered useful to the cause. Repeated often enough, these buzz words take on the appearance of truth – based upon which decisions are made and errors committed. Practitioners, however, know that healthcare is not made by money alone, and that new scientific knowledge and intense social and political mobilisation are critical to progress in public health.

UNDERSTANDING, AND BEING UNDERSTOOD BY PATIENTS

An anthropological studyVanja Kovacic, Access for more – Overcome barriers to access to HIV/AIDS care in Homa Bay district, Kenya. Phase 1, June 2010, Homa Bay, Kenya. A qualitative survey was conducted using in-depth interviews in the patients’ own language. The fifty or so patients interviewed were adults from the Homa Bay and Ndhiwa districts who had been followed during the previous year at the Homa Bay district hospital’s Clinic B (run by MSF). About thirty members of the patients’ entourage likely to have an influence on their decision were also interviewed, including practitioners of traditional medicine, religious healers, herbalists, traditional birth attendants, religious leaders, members of the patients’ families, and village chiefs.was conducted from April to June 2010 to try to explain why some patients were receiving treatment too late, when their immune systems were already badly compromised (CD4 count below 100/mm3, WHO Clinical Stages 3 and 4).

The study also had several other aims: to explain how the “Luo community” viewed the HIV epidemic, to determine the epidemic’s sociocultural impact, and to assess the elements of Luo culture that might favour HIV transmission. The study’s sponsors, MSF operations managers at all levels (headquarters, capital of the country, and Homa Bay project), wanted to learn about patient perceptions and motivations in order to influence attitude and behaviour changes.

In order to explore the reasons why patients wait “too long” to come in, we need to define the “right time” to start HAART. According to promoters of HIV treatment,Towards universal access: scaling up priority HIV/AIDS interventions in the health sector: progress report 2009, UNAIDS, UNICEF & WHO, Geneva, 2009. http://data.unaids.org/pub/Report/2009/20090930_tuapr_2009_en.pdf.ideally, all adults would be screened voluntarily once a year. If positive, the hope is that the patient would heed the instructions for preventing transmission of the disease to others, encourage those close to him to be tested, and start antiretroviral therapy “at the right time,” that is, before his immune defences collapse. There are several problems with this ideal.

The “right time” invoked by care providers has changed constantly over the past few years. When we began prescribing HAART in 2001, treatment was rare and reserved for those who would now be considered “too late.” The situation for some of the people who are now coming in "at the right time” is changing, since the CD4 threshold below which HAART must be prescribed has just been raised. Basically, patients are now being asked to come in sooner. While they are coming in “at the right time” now, they will soon be considered “late.” The “right time” depends on both the state of medical knowledge and the ability of the healthcare services to provide treatment, and both of these are in constant flux. As a result, the criteria for starting HAART has changed frequently over time; from the initial period when drugs were scarce and the most serious cases given priority, to the current period of greater availability, where care providers want to start HAART earlier. Aside from its negative connotations, the definition of “patients who come in late” is not sufficiently stable to prevent confusion. In 2010, French health officials estimated that a third of HIV patients came in “too late.” Even with its considerable resources, one of the world’s best healthcare systems cannot produce patients disciplined enough to come in exactly on time. In comparison, in the first quarter of 2010,Quarterly report on medical activities – Kenya Programmes First quarter 2010: January – March.56% of treatments in Homa Bay were started at “the right time,” that is, on patients who were at WHO Clinical Stages 1 and 2.

Clinical staging of new HIV patients

	N	Proportion (%)
Stage 1	152	30
Stage 2	133	26
Stage 3	192	38
Stage 4	33	6
Total	510	100

We should note – by way of reference regarding the limits of qualitative studies – the conclusions of a studyPierre Mazet, Motifs de participation et de non participation au dépistage du cancer. Enquête qualitative auprès des habitants de deux territoires de l’agglomération grenobloise, Étude n°33, Grenoble, Odenore, Dec. 2009.conducted in France investigating the reasons for participation and non-participation in cancer screening: “[translation] There are a number of types of justification and explanation that shed light on the choice to have, or not have, screening tests: the person’s relationship to health and their body, their opinion regarding the principle of prevention, the fact of having been affected by cancer, their trust in their doctor, the fear of knowing, the fact of having receiving an “invitation letter” from the anti-cancer organisation, etc. But the way in which individuals relate or position themselves with regard to these explanatory factors is always singular. And it is always this specific relationship that explains why individuals act or not, take up the prevention system, or not.”

It would be more appropriate for us to ask ourselves about the problems arising from the patient/care-provider relationship, without prejudging about who is early and who is late. To describe patient attitudes regarding available care, we can use an economic and sociological classification scheme Observatoire des Non-Recours aux Droits et Services http://odenore.mshalpes.fr/documents/WP1definition_typologies_non_recours.pdf.from a study on the non-use of social assistance by eligible individuals. This classification distinguishes three typical situations: individuals who are not aware that assistance exists or that they are eligible for it (not aware); individuals who are aware of the assistance but choose not to ask for it (not requested), and individuals who have requested the service, but haven’t received it (not received). An examination of the data from the anthropological study in Homa Bay suggests that the reasons for non-use of HIV care often fall into the “not received” category. Prompted by their symptoms, all patients had gone to see either public health service representatives 60%) or other practitioners (40%), the latter most often outside biomedicine. This doesn’t mean that “not requested” had no part in the non-use of care. But the fact is that all patients tried to consult. Neither of the two healthcare sectors – biomedical or non-biomedical – offered patients appropriate diagnosis and treatment in a reasonable amount of time, despite the prevalence of the disease and the severity of their condition. Looking at the care trajectory of the fifty or so patients studied, only half of the patients were diagnosed by their fourth visit, in a region where 20% of adults are infected. Just over one in ten patients had to make seven attempts before being diagnosed and started on treatment.

The anthropological study also looked at the relationship between Luo culture and HIV transmission. The Head of Mission practicing there between 2003 and 2007 had already broached the issue in her report: “Luos, an exception among Kenyan tribes, don’t circumcise, they are also polygamous (60% of families in Rusinga highland in 2001). These two HIV transmission factors combine with others such as the consequences of the ongoing women inheritance culture, a culture of early sexual relationships before marriage (traditionally tolerated in elopement and a behaviour currently accepted by the majority of Luo men) in addition to prostitution caused by modern life and poverty (young widows refusing to be inherited or unmarried young women seeking a livelihood). Prostitution is especially common along the lakeshore where young fishermen benefit from a regular income. These factors have made the local Luo society highly vulnerable to sexually transmissible diseases. This accumulation of risk factors explains why HIV prevalence is up to 40% among adults in some areas of Nyanza.”

Tribal membership and identity continue to have considerable importance in the social and political life of contemporary Kenya. Until the most recent elections, many Luo didn’t recognise themselves in the country’s government. Homa Bay is a majority Luo region, where HIV prevalence is three times the national average. In this context, the natural tendency of those fighting HIV is to claim a link between the severity of infection in this region and what they consider the relevant sexual practices in Luo culture. Two of these practices, one of them levirate marriage (where a man must marry his brother’s widow), are mentioned by AIDS actors as favouring transmission of the virus. Both are connected with resumption of sexual relations by a widow, or her remarriage. While there may be a particular link between Luo culture and virus transmission, simply blaming certain aspects of the tradition governing sexual activities in widowhood is not a convincing argument. Culturalism is not the only explanation of why Luo culture is being blamed. There is also a widespread belief within AIDS organizations that gender inequality contributes to virus transmission. Epidemiological data on HIV prevalence by gender in Kenya confirms the view that women are at a disadvantage in negotiating sex and its attendant particulars (e.g., condom use). But there is nothing to indicate that if widows resumed sexual activity outside the rules of Luo culture they would be less likely to transmit the virus. The HIV epidemic can sometimes seem like a boon to those who would like to change tribal traditions and gender relations. Indeed, it’s hard for activists to resist the temptation to justify a "modernizing" or feminist position by claiming that it’s better for health.

Freed from the assumptions about patient "lateness” and the harmful health effects of Luo culture and gender relations, MSF’s anthropological efforts could examine the patient-care provider relationship in greater depth by drawing on the already-plentiful literature and data from Homa Bay. The goal is to identify the parameters that determine the success of lifelong treatment of a disease that is fatal if treatment is interrupted or ineffective. Note that the literature review should not be limited to the AIDS literature, but should also cover medical specialties in which monitoring chronic patients is part of everyday practice.

Patients’ experiences with the disease and its treatment are changing. In the late 1990s, patients starting HAART experienced long periods of painful and disabling symptoms. They were aware of being close to death, and had often accompanied a loved one during the terminal phase of the illness. In the initial phase of treatment, patients experienced relief from their symptoms and a return to normality.

Now, in Kenya as elsewhere, thanks to medical progress, the majority of patients are offered HAART based on lab tests, at a point when they have only minor clinical signs of the disease, or none at all. Increasingly, patients are being asked to follow a treatment protocol not to reduce pain and restore their independence, but to satisfy the abstract notion that if they do not correct an abnormal lab result, they risk suffering from the effects of a disease – AIDS – at some time in the future and which might kill them a few years down the road.

Moreover, the first patients received innovative, specialized care by care-provider teams inspired by a pioneering spirit conducive to an empathetic approach to the patient. Patient education was afforded considerable importance in those pilot programmes, combining a warm welcome, medical information, and enjoinders to follow treatment protocols carefully. This patient education explains, in large part, why care-provider teams in Africa achieved compliance and survival results on a par with those in more privileged environments (North America and western Europe, for example) with many more resources. Indeed, HAART’s success is not simply that of a drug cocktail, but also of a medical practice that takes the time to explain the treatment process to the patient, encourages him to ask questions, and verifies, through this exchange, that the patient and someone close to him have truly grasped the concepts essential to the success of the treatment protocol.

Extending treatment meant involving personnel from health ministries with fewer resources than the organizations that had launched the first pilot projects. Patient education has been one of the first victims of this personnel shortage, even though, due to the changing patient profile, it is now more essential to success than ever. We should note that in Kenya, the word used in both society and healthcare institutions to designate patients is “clients.” During a visit to a clinic taking part in the decentralization of treatment from the district hospital to peripheral health centres, a remark by the foreign visitor to a Kenyan colleague – “Here, this is not HIV friendly” – prompted an unequivocal response: “Kenya is not client friendly.” Yet one of the main issues in chronic disease management is avoiding disruptions in the continuity of care.

Why would a patient remain loyal to a practitioner who bullies him? Observing consultations outside the district hospital’s Clinic B, in peripheral care centres (e.g., Marindi, Ndhiwa and Rangwe) helps to see what discourage patients. Waiting times are long. Forget confidentiality – there is no guarantee of even minimal discretion, and patient files are easily accessible. The visit is basically just a set of administrative acts, during which other members of staff frequently open the door to the examination room for one reason or another. Patients rarely receive a physical exam, and there is as little time for discussion as there is staff training to provide pertinent information. Patient interviews conducted at the MSF-run clinic (the district hospital’s Clinic B), on the other hand, emphasized the importance of courtesy, discretion and patient education.

The anthropological study showed that patients did not receive care when they requested it. It also showed that the reasons for not requesting care, aside from the fear of death, were the fear of loss of reputation and of treatment side effects. How, then, will the even worse consultation conditions in a decentralized system help reassure patients? MSF, upon whom this point has not been lost, is trying to support peripheral health centre teams, using a mobile team (clinician, nurse and counsellor) to improve patient care. While this support is useful, its impact is limited by existing constraints – in particular, the instability of MoH staff and the lack of suitable facilities.

The lack of dedicated space shows the extent to which chronic disease management has never been an important objective in the countries where MSF works. A good illustration of the relationship between treatment objective and architecture is Malawi’s Chiradzulu district hospital. The layout of the new European Union-funded district hospital, opened in 2005, makes no provision for HIV/AIDS activities, despite the fact that nearly one in five adults in the district are HIV-positive, and 30,000 HIV patients are now being treated there. Some of the HIV activities in the Chiradzulu hospital are conducted in rooms intended for trauma care. The Homa Bay district hospital needed new buildings to be able to provide HIV and TB care.

Many of the patients are poor farmers whose financial problems are exacerbated by their illness, as their income and borrowing ability decline and expenses rise. Patients are not asked to help pay for antiretroviral drugs, consultations, or some lab tests. But that doesn’t mean they don’t have to pay anything – there are still transportation costs and bills for lab tests and medications not included in the “free” care.

There is yet another complicating factor. While it is true that a large percentage of patients are now started on treatment before their immune systems become too damaged, this doesn’t mean there are no more complicated cases requiring specialized medical care. If clinicians fail to respond to requests for medical help when the disease worsens because of the complexity and cost of care, this can’t help but impact treatment adherence and patient loyalty. Why trust practitioners who abandon “clients” just when their condition has deteriorated to the point where care is critical to their short-term survival? Abandoning terminal patients to deprivation (food, hygiene and treatment), pain, and isolation after their years of loyalty to the care system can only make those who follow even more reluctant to follow medical recommendations. Another constraint for care services is to be able to contact patients at home when they stop coming in for appointments. The need to extend care beyond the walls of healthcare facilities is also an issue in the management of terminal patients who are no longer mobile. Extending the care delivery network beyond the institution is essential to finding patients who have stopped coming in. Already, monitoring of patients over several years shows a disquieting percentage “lost to follow-up”. Of the 510 new patients at the district hospital in the first quarter of 2010, 494 were lost to follow-up.

New patients and outcomes during the first quarter 2010 – Homa Bay programme

	N
New patients total	510
New starting HAART	346
Deaths	14
Lost to follow-up	494
Transfer out	139

These data are consistent with data collected in other settings:Traitement du VIH/sida en Afrique: la gratuité pour limiter les coûts, Bernard Taverne, UMR 145 “VIH/sida et maladies associées”, Institut de Recherche pour le Développement (IRD), Université de Montpellier I, Centre régional de recherche et de formation à la prise en charge clinique (CRCF), Service des Maladies Infectieuses, C H N U Fann, Dakar, Sénégal. Global Health Promotion, 1757-9759, vol. 17 (3), pp. 89–91; 375177.

“[translation] Recent epidemiological studies describe the default, treatment failure, and viral resistance rates faced by treatment programmes in Africa:

• In many countries, a high percentage of patients stop their ART early (25% of patients stopped after 12 months, and 33% stopped after 24 months) Tassie J-M, Baijal P, Vitoria MA, Alisalad A, Crowley SP, Souteyrand Y. Trends in retention on antiretroviral therapy in national programmes in low-income and middle-income countries. J. Acquir Immune Defic Syndr. 2010; in press.
• These interruptions are due, in part, to early patient death – deaths related to delayed diagnosis of HIV infection – but also, in almost half of cases, to the patient discontinuing all medical follow-up.
• A study of five African countries (Botswana, Malawi, Uganda, South Africa and Cameroon) found that 15 to 25% of patients were in virological failure after 12 months or more of treatment Harries A, Zachariah R, van Oosterhout J, Reid S, Hosseinipour M, Arendt V, et al. Diagnosis and management of antiretroviral-therapy failure in resource limited settings in sub-Saharan Africa: challenges and perspectives. Lancet Infect Dis. 2010; 10 (1): 60–65.
• A study conducted in Cameroon reported that 16.9% of patients showed viral resistance after 2 years of treatment Kouanfack C, Montavon C, Laurent C, Aghokeng A, Kenfack A, Bourgeois A, et al. Low levels of antiretroviral-resistant HIV infection in a routine clinic in Cameroon that uses the World Health Organization (WHO) public health approach to monitor antiretroviral treatment and adequacy with the WHO recommendation for second-line treatment. Clin Infect Dis. 2009; 48 (9): 1318–22.; a broader study showed that nearly 90% of patients in virological failure were carrying viruses resistant to at least one of the three classes of antiretroviral drugs Gupta R, Hill A, Sawyer A, Cozzi-Lepri A, von Wyl V, Yerly S, et al. Virological monitoring and resistance to first-line highly active antiretroviral therapy in adults infected with HIV-1 treated under WHO guidelines: a systematic review and meta-analysis. Lancet Infect Dis. 2009; 9 (7): 409–417..

These high early default, therapeutic failure and viral resistance rates reflect the current shortcomings and limits of care and treatment mechanisms. Treatment defaults are extremely damaging to both the individual (increased morbidity, early death) and the community (in particular, by favouring the emergence and spread of viral resistance). They increase costs – the money spent actively searching for these patients, and then for their subsequent medical care – especially since the second-line treatments needed for patients with resistant viruses currently cost $1,200 per person per year (eighteen times more than the initial treatment). The number of people who need second-line treatment is already very high.”

Even leaving aside issues related to drug efficacy and side effects, a quick glance at the conditions that help to establish the relationship needed for successful treatment indicates that such conditions are far from the reality at public institutions. Kenya has started 270,000 people on treatment in just the past few years, and there are hundreds of thousands more waiting. It is plain that public healthcare services are being given a task that they cannot accomplish in high-prevalence regions. This means that other care-delivery models are needed. This tension between the lack of available care and the number of people waiting for treatment has existed since 1996 and the response has primarily been a combined approach that is both economic (lowering prices and increasing public funding) and medical (offering empathy and medical advice, simplifying the treatment protocols, and applying the subsidiarity principle According to the subsidiarity principle, people at a higher level should refrain from doing what someone at a lower level can do.when choosing personnel). There is still room for further reducing the patient’s share of the costs. Avenues for improving treatment efficacy, reducing side effects, and simplifying storage and dosing are being explored. HIV and immune status (CD4 count) can already be assessed by rapid tests; the same will soon be true for viral load. To sum up, the limits to expanding and adjusting available care are related primarily to the state of public healthcare facilities.

MSF’S ROLE

DISCOUNTING MSF’S HISTORY AND THE ILLUSION OF POWER

The draft report from the July 2010 mise à plat (MAP) The MAP is a meeting held twice a years, at which the managers from a given country and headquarters get together to examine the results and define the objectives for field programmes.on the Kenya projects (see Appendices), written in telegraphic style to allow rapid distribution to the field, summed up a decade-long policy: “Late 90’s – refusal to get involved and nothing to offer. Early 2000s –involvement and significant early successes (treatment and advocacy)”.

It is a common misconception that the Association’s involvement in HIV began with HAART’s arrival in the field in the early 2000s. In the minds of many colleagues, the three decades of rapid progress described earlier are reduced to only one – the last. In reality, the number of HIV projects prior to the 1997 arrival of HAART was already on the same order of magnitude as today. MSF teams were actively working in the main epidemic foci in Africa (western, eastern and central) and Asia, seeing patients, distributing information about the disease and how it is spread, preventing transmission during medical and nursing procedures, fighting for access to care, and treating opportunistic infections. Before the late 90s, the possibility of stopping the virus from multiplying in the body was virtually nonexistent, and the immune system’s decline irreversible. Hope at that time rested on prevention campaigns, which to many meant encouraging people to change their sexual behaviour. Those who had doubts about widespread change in sexual behaviour or whether it was appropriate for MSF – a foreign humanitarian organization – to participate in such campaigns were seen by their detractors as refusing to get involved in the fight against AIDS. These controversies and the Association’s limited participation in the public debate over how to fight HIV is remembered in the collective conscience of its members as an absence, or even a refusal to mobilize. In contrast, the early 2000s is seen as a period of heroic engagement with a series of rapid changes leading to the use of HAART in MSF-supported projects.

At meetings in Kenya with representatives from other institutions, MSF is given little credit for having been among the first to use and demonstrate the effectiveness of a protocol that now allows millions of patients to be treated. The initiative of which MSF is most proud – having been the first to prescribe antiretrovirals, at no cost to patients, in a public hospital in Kenya – has been forgotten. On the other hand, our Association is admired – even feared – for the weight its words carry in the public debate. While MSF’s mobilization against HIV is often described as belated, the impact of its advocacy campaigns on decisions by multinational pharmaceutical firms, the World Health Organisation (WHO), and donor countries is overestimated. Governments have been committed to fighting infectious diseases for as long as there have been state institutions, because epidemics threaten their ability to ensure public safety, and have significant economic consequences. The primary goal of governments, however, is not to provide care to individuals, but to ensure public safety and avoid economic losses and political instability.

In modern times, the exemplary model for fighting infectious diseases is the campaign to eradicate smallpox: an initial investment (the vaccination campaign) and a decisive result (eradication of the disease). The strategy is to eliminate the cause (the disease), not manage its consequences (the patients). How, then, to explain that, in the case of AIDS, donor states have agreed to spend several billion euros a year on treating patients with no prospects for eradicating the disease? The threat to public safety and the ubiquity of HIV, an extraordinary level of social mobilization, the potential major economic repercussions, rapid scientific progress and countless other factors have helped make AIDS – and the responses to it – unique. But to understand why donor states have found it in their interest to depart from their usual position and get involved in treating millions of patients without any prospect for ending the disease, we have to look at the impact that the public debate on pharmaceutical intellectual property issues has had on political decision-making.

In the late 1990s, when extending the use of antiretrovirals beyond high-income countries became an issue, one of the main themes of World Trade Organization (WTO) efforts was the globalization of intellectual property rules applying to trade. The high price of certain drugs – like antiretrovirals in the late 1990s – can be explained, in large part, by the trade monopoly conferred on pharmaceutical firms by patents. The combination of a public health disaster and prohibitive drug prices disproportionate to the costs of production raises the question of whether intellectual property rules are compatible with public safety – in this case public health. This was a major challenge. In today’s world, wealth and power are less often derived from ownership of physical objects, and increasingly based on intangible ownership of the knowledge (legally formalized by patents) that allows the production of objects and services. Reducing the economic advantages conferred by patents means reducing the value of that portion of capital represented by intangible objects.

The too high price of AIDS drugs in the late 1990s – several thousand dollars per patient per year – created friction between two political objectives: private ownership and public safety. The moral dimension of the conflict added a dramatic effect, enhancing its public visibility. The tension between the two imperatives – private ownership and public safety – exposed the fragility of an economic system that relies, in large part, on the belief that ownership of knowledge essential to public health can remain in the hands of a small number of individuals and institutions for long periods without any significant conflict. The major economic powers urgently needed to offer a series of concessions on access to drugs, before the "our lives versus your profits" debate compromised their ability to extend intellectual property rules to all global trade – especially since, at the time this was happening, in 2001, negotiations on other WTO issues were deadlocked and the institution was weakened. In that particular context, the major economic powers, led by the United States, supported a softening of the intellectual property rules – for the pharmaceutical market only – vis-à-vis public health institutions.

Basically, the fight against AIDS benefitted from an extraordinary combination of economic and political circumstances. Circumstances so conducive to rapid change in public health policy cannot last forever, nor be reproduced at will through advocacy. When they do occur, however, they create political openings and offer those capable of seizing it an opportunity to change long-established power relationships.

MSF AND DEVELOPMENT POLICIES

After the crisis caused by the lack of antiretrovirals in the countries hardest hit by HIV was resolved, the question of how and for how long MSF should be involved in the fight against HIV became the focus of internal debate once again. Questions about involvement in so-called “development” activities have been debated since the Association’s very first General Assembly, in 1972: “[translation] There are two opposing lines of thought: the first calls for volunteer medical aid that can be mobilized rapidly for short missions […]. The second, supported by volunteers returning from Bangladesh and Upper Volta, argues for the other urgency – the chronic lack of medical care in the Third World”.Anne Vallaeys, Médecins Sans Frontières: La biographie, Paris, Fayard, 2004. Nearly forty years later, the debate continues in the report from the June 2010 Kenya MAP, which refers to health development-type involvement, whose pertinence is questioned: “[translation] Mid-2000s, starting to question the pertinence of projects shared by field and headquarters, the latter also being concerned with the cost of the projects and the fact that they might not be consistent with MSF’s role.” Similarly, in late 2008, the Operations Department Politique Opérationnelle 2009-2011. Bilan et perspectives 2008-2009, Centre Opérationnel Paris.in Paris considered transferring a large portion of HIV activities, in Kenya and elsewhere, to other organizations: “[translation] This effort should start with an examination of our international, national and local work environments, to identify possibilities for transferring the responsibility for care and supplying drugs to some portion of the ‘stable’ patients that we are now treating directly to other actors (NGOs, Ministries of Health, peer groups, etc.).”

To better understand the changing opinions on “development” policies within MSF, we need to recall the state of international relations at the end of WWII. The Cold War had been intensifying since the late 1940s, and the so-called "underdeveloped” countries were the focus of a struggle for influence between the two sides. Contributing to the development of poor countries was one of the four points of President Harry S. Truman’s 1949 inaugural address:http://www.trumanlibrary.org/calendar/viewpapers.php?pid=1030“We must embark on a bold new programme for making the benefits of our scientific advances and industrial progress available for the improvement and growth of underdeveloped areas. More than half the people of the world are living in conditions approaching misery. Their food is inadequate. They are victims of disease. Their economic life is primitive and stagnant. Their poverty is a handicap and a threat both to them and to more prosperous areas.” In the early 1950s, French demographer Alfred Sauvy coined the expression “Third World” in reference to the Third Estate – under the French Ancien Régime, the voters (and their representatives) who did not belong to either the nobility or the clergy. In reality, the goal of developing the third part of the world – the one outside of NATO and the Warsaw Pact – transcended many political divisions, as did, in other times, Christianisation or the concept of the civilizing mission of colonization.It should be stressed that outside of this international consensus there were currents of thought opposed to these “civilizing” enterprises on both the right and the left, though obviously they employed different lines of argument.The dominant ideology of the time held that science, industrialization and economic growth would soon allow “backward” nations to close the gap, provided the “advanced” nations supported the project by allocating funds, adopting new trade rules, sharing scientific knowledge, and transferring technology.

It was in this context that a group of Third World nations launched the Non-Aligned Movement, introduced at the Bandung Conference (1955) and the Belgrade Conference (1961) as an alternative to the declining colonial powers and U.S. and Soviet hegemony. The membership of Cuba, North Vietnam, Yugoslavia and the Republic of China, however, was an indication that the Non-Aligned Movement would be unable to completely escape confrontations between the two opposing economic and political systems.

Traces of these typical Cold War confrontations can be found in internal discussions and MSF publications on the direction of international development aid. Third-worldism, development, poor nations’ debt, famines, and international health issues are at the heart of a colloquium conducted by the Liberté Sans Frontières foundation, an MSF satellite from 1982 to 1989. In 1986, as part of that discussion, MSF President Rony Brauman wrote: “[translation] The main demands of the “new order”In 1974, the UN General Assembly adopted the Declaration on the Establishment of a New International Economic Order (NIEO); one of its objectives was to respond to development issues related to the price of raw materials.desired by the entire third-worldist movement have this in common: they pursue perfectly laudable goals using methods that will inevitably destroy them.”Le Tiers-mondisme en question. Rony Brauman (ed.), Paris, Olivier Orban, 1986.Brauman specified two important points. First, the critique “came from within the third-worldist movement itself.” Second, it was aimed primarily at a “leftist” idea of development that he considered dominant in France, while “the right” was nowhere to be seen. In LSF’s view, third-worldism – the love child of “Leninism and Christian socialism” – represented “a kind of extension of traditional social morality on an international scale.” By the late 1980s, development had become an inalienable human right, thus taking on an aura of complete unreality: “The right to development is an inalienable human right by virtue of which every human being and all peoples are entitled to participate in, contribute to, and enjoy economic, social, cultural and political development, in which all human rights and fundamental freedoms can be fully realized.”United Nations General Assembly Resolution 41/128, Dec. 4, 1986.

LSF’s public positions were defined in opposition to third-worldism’s then-considerable influence in the health domain. The International Conference on Primary Healthcare was held in the USSR, in the town of Alma-Ata, Kazakhstan in 1978. Its final declaration accorded an important role to community health workers, modelled on Maoist China’s “barefoot doctors.”

LSF’s hostility to communist-influenced health policies wasn’t based solely on a rejection of the deadly, freedom-killing attempts at social and public health engineering, which it saw as the death throes of a totalitarian thought process whose sentimentalist third-worldist tone failed to mask its dangers. The rejection of totalitarianism was, in fact, reinforced by experience in the field. It was operations to assist Third World refugees during MSF’s first twenty years that had the greatest impact on the Association’s political and technical choices. These refugees were Vietnamese, Cambodian, Laotian, Afghan, and Ethiopian, proving, “by their very existence, the failure of communism”. Third World “popular democracies” were producing nearly 90% of the world’s refugees.Populations ín danger, François Jean (ed.), London, John Libbey & Co., 1992.In the 1980s, however, while MSF was formulating its criticisms of state authoritarianism, the neoliberalism of Ronald Reagan and Margaret Thatcher was winning to such an extent that it became a threat to public health in poor countries, and to all socially progressive public policy. The structural adjustment policy imposed by the international financial institutions was leading to significant budget cuts in “developing” countries, while the Cold War was on the wane. In reality, LSF did not take into account this shift in international power relations when it opposed third-worldism.

The fact remains, however, that at the same time it was positioning itself in opposition to third-worldism, MSF was participating in long-term third-worldist-inspired public health development activities within broad coalitions of actors (states, international organizations, and NGOs). While this was true of the French section, the first in the Sans Frontières movement, it is even more obvious in the history of the Belgian, Swiss, Dutch and Spanish sections, which until the early 1990s were involved primarily in public health development projects.Éric Goemaere, “Une ONG au ministère”, in Utopies Sanitaires, Rony Brauman (dir.), Paris, Le Pommier, 2000.This is evidenced by the fact that both in practice, through its field interventions, and by its public positions, MSF has supported every new, major international health initiative: the Expanded Programme on Immunization (1974); the promotion of primary healthcare following the Alma-Ata (1978) and Bamako (1987) conferences; the health-related Millennium Development Goals (2000), and the goals of the Global Fund to Fight AIDS, Tuberculosis and Malaria (2002).

From the earliest editions of MSF’s Clinical Guidelines,Médecins Sans Frontières, Clinical Guidelines, Paris, Hatier, 1988.the introduction places the Association’s medical action squarely in line with the movement for universal primary healthcare launched at Alma-Ata. “[translation] Curative care is only one component of the healthcare programmes adapted to the needs of developing countries. It is important to remember that other priority programmes also need to be developed. These are the foundation of all action aimed at improving the health status of a community. These high priority programmes focus on nutrition, water, hygiene, sanitation, the expanded programme on immunization and mother-child health.” Admittedly, MSF did criticise the final Alma-Ata Declaration to some extent. The first had to do with the definition of “Health for all by the year 2000,” which seemed completely unrealistic: “The Conference strongly reaffirms that health, which is a state of complete physical, mental and social wellbeing, and not merely the absence of disease or infirmity, is a fundamental human right and that the attainment of the highest possible level of health is a most important worldwide social goal whose realization requires the action of many other social and economic sectors in addition to the health sector.” The second criticism was aimed at the role to be played by community health workers in promoting primary healthcare. The magnitude of the responsibilities entrusted to a category of personnel with no medical or allied healthcare skills did not bode well for success, especially since the training, supervision and material resources to support their activities were lacking. That didn’t, however, stop the Association from distributing David Werner’s handbook, Where There is No Doctor – the veritable medical bible for community health workers – to every MSF field project library, where it remained until the mid-1990s.

In 1992, the Association’s internal debate over participation in development policies went to a new level. In 1985, the opinion of LSF members was expressed in the conclusion of the Third-worldism in question colloquium: “[translation] Third-worldism is an obstacle to development.” By the early 1990s, the problem was no longer third-worldism, but development itself. In the foreword to the first collectionPopulations ín danger, François Jean (ed.), London, John Libbey & Co., 1992.written since the suspension of Fondation Liberté Sans Frontières in 1989, Rony Brauman, then MSF International Council president, proposed a new definition of humanitarian action: “First, let us hazard a minimum definition. Humanitarian action aims to preserve life and human dignity and to restore people's ability to choose. To accept such a definition is to say that, in contrast to other areas of international solidarity, humanitarian aid does not aim to transform society but to help its members get through a crisis period, in other words when there has been a break with a previous balance.” What this new definition did was free humanitarian action from the obligation to participate in a global enterprise to transform the world, where the combined benefits of science and economic growth would ensure that “underdeveloped” countries could catch up, provided they followed the recommendations of “more advanced” countries. For humanitarian medicine – whose origins are, in part, linked to historic experiments (colonial military medicine, missionary and social reform medicine) in the march toward a new world – the impact of this break was considerable. It could now distance itself from a humanitarian action that might contribute to universalizing the civilization process begun in Europe and gradually spread to the rest of the world.

The influence on MSF members of this updated definition of humanitarian action by Rony Brauman in 1992 explains why the wording to the effect that the Association intends to respond to crises, on an emergency basis if possible, crops up again and again in the internal debate. “At the end of the 1990s, Médecins Sans Frontières (MSF) got involved in HIV/AIDS because we viewed it as an emergency: today, MSF still believes this is a crisis requiring an exceptional response.”No Time To Quit: HIV/AIDS Treatment Gap Widening in Africa, MSF, May 2010. http://www.msf.fr/drive/69bcb2a7e6c55b47888bd5f5204ce6fc.pdf.Similarly, the June 2010 Kenya MAP report says, “2009, recommitment to MSF’s role in the fight against HIV, which is again seen as an emergency….” But what should be our attitude in response to crises whose resolution involves long-term participation in a process of social transformation, like setting up better infectious disease monitoring, for example? AIDS projects and Access Campaign advocacy are examples of where we go beyond Rony Brauman’s 1992 definition. AIDS is a crisis that has already lasted several decades, and the introduction of HAART in poor countries is an obvious social transformation. While Brauman’s definition might free humanitarian action from the constraints of belief in the ineluctable march of progress guided by law and driven by the combined effects of science and economics, it can also preclude potential engagements by explaining that humanitarian aid “does not aim to transform society but to help its members get through a crisis period, in other words when there has been a break with a previous balance.” But is it possible to help a population get through a crisis while choosing not to participate in transforming the society in which the crisis occurs?

In MSF’s internal debates, the position that claims to give priority to the incontestable “save lives here and now” is offering an argumentum ad verecundiam. It expresses fidelity to the supposed origins of the Association, though in fact there is no evidence that there was ever a clear-cut decision, but on the contrary, evidence of a contradictory discussion that began at the first MSF General Assembly, and continues to this day. Emergency response is often held up – wrongly, as history proves – as the activity to which we must always remain loyal. The contradiction between the claim that MSF did not originally participate in development activities and the facts is partially hidden from the collective imagination, thanks to the organization’s promotional communications, which tend to emphasize the image of a humanitarian medical organization heroically responding to emergencies, above the political fray. The debate is reduced to an ontological argument, thus avoiding the political dilemmas. The hardening of identity in the arguments of those who are now against involvement in so-called health development activities is no minor detail to their opponents, who appear to disagree with them to the point of no longer being considered to possess the “MSF identity.” They aren’t just dissidents – they’re outsiders to the association they joined. A Kenya team leader who came out in favour of joint health development projects with public institutions thought she heard a representative from headquarters reply that she “had the wrong organization.”

Yet the discussion on the limits of MSF’s role shouldn’t end in a decisive victory by one side or another. MSF isn’t a Red Cross satellite or a UN agency, and it certainly isn’t the appointed operator of some former colonial power’s international cooperation department. The association defines for itself the scope of its mandate, offering its members the freedom to explore limits while giving meaning to their volunteer service, transforming it into something more than just a form of daily self-sacrifice. The authors of the MSF Charter used the expression “populations in distress”; twenty years on, Rony Brauman spoke of “populations in crisis.” These expressions can be thought of as malleable, or able to convey multiple meanings. They are inclusive – sometimes too inclusive – and ongoing critical reflection becomes essential. In this sense, the definition adopted in 1992 was a salutary attempt to narrow the range of possible meanings. But MSF’s history (its relief operations, its position among aid organizations, and its institutional growth) also demonstrates how having a flexible definition of its role has been extremely productive. The goal is not to settle the debate, but rather to keep it alive, so we can take advantage of the flexibility of the expressions “populations in distress” and “populations in crisis.” Internally, hardened identity references can only stifle critical debate. Definitively ending the debate over how to define the scope and methods of humanitarian action would narrow a priori the field of intervention. MSF would thus be depriving itself of the freedom to respond, other than with dogmatic pronouncements, to new and constructive questions by the teams at work in unique and changing fields of action.

BENEVOLENT AND SUSTAINABLE PUBLIC INSTITUTIONS?

The internal debate is not simply a discussion on the validity of MSF’s participation in transnational public health initiatives. It is obvious to those who support this approach that action necessarily involves the Association’s participation in building public healthcare institutions. “As seen above, the need for integration of activities with the MoH to scale up treatment and progressively envisage coverage of ART needs has been clearly stated.” This assertion is taken from the report written by the Head of Mission for Kenya from 2003 to 2007. Yet the idea that public healthcare institutions offer the most solid guarantee for sustainable public health action is debatable – not just politically, but also based on MSF's practical experience. Regarding AIDS, these same public institutions helped spread the virus (via injections, transfusions, etc.). Then they increasingly refused care as the disease became an everyday part of their clinical activities. And then, when an effective treatment (HAART) became available, the health ministries had enormous difficulty setting up administrative procedures for procuring it, although they could have obtained the drugs from the Global Fund without having to pay for them. Ten years on, few of these countries have made any financial effort to support HIV activities – and let’s not even go into the misappropriation of international funds. The ambivalent (to say the least) view of the public health employees’ integrity is illustrated by the pharmacy building constructed by MSF’s Homa Bay team. The MSF AIDS project pharmacy was built on the Association's privately owned site, not in the compound of the district hospital where the project’s medications are prescribed. Yet the building was constructed with a view to the long term, as its thermal insulation shows. The insulation was designed so that electric air conditioners would not be required to protect the drugs from high temperatures. The public sector is entrusted with the responsibility of covering the health needs of the population, but not to the extent of entrusting its employees with the keys to the pharmacy.

Whether to work in the public or private sector is a tactical choice, since neither offers a decisive advantage by its nature and in all situations. In Africa for example, church-affiliated healthcare centres often offer better quality, more reliable care than their public sector counterparts. In the Homa Bay district, a visit to the Catholic Asumbi Health Centre (35 inpatient beds, with ART and TB drug dispensing in addition to its standard preventive and curative activities) offered a good example of reliable, high quality care. In contrast, the history of public health institutions since the end of WWII in the countries where we work shows anything but stability. The history in African countries is a telling example: colonial medicine followed by independence and international cooperation (often with the former colonizer); collapse of public services in the 1980s, exacerbated by the structural adjustment policies of the international financial institutions; and then timid resumption in the late 1990s. This picture hardly suggests a clear relationship between the public nature of the healthcare institutions and the sustainability of steadily progressing healthcare activity.

PROJECT MANAGEMENT

During the visit, activities at Homa Bay were entering a new phase in their development: an increase and reinforcement of HAART dispensing sites outside the district hospital, in peripheral healthcare centres. This was the resumption of an effort to decentralize ARV dispensing initiated in 2003 in Ndhiwa. Paris headquarters, the capital team in Nairobi, and the field team all agreed on the direction to take. The latter was laid out, in broad strokes, taking both MSF objectives and local circumstances into consideration. The goal in Homa Bay – a high-prevalence area where available resources are insufficient – is to start treating as many patients and as early as possible. To achieve this, the plan is to increase the number of ARV dispensing sites in the Homa Bay and Ndhiwa districts, while ensuring that a large percentage of the patients being treated maintain the lowest possible viral load over time.

Despite the consensus on the objective, there was still some discomfort over how to concretely manage such an undertaking. The field team felt they were being asked for too detailed an action plan – the exact number of new dispensing sites in a given time period for a specific, detailed budget. The team pointed out that it could not provide that kind of planning data, because too little was known about the factors that would determine activities to construct a reliable timetable. To them, how to manage this new direction proposed by headquarters was just the first in a long list of hurdles to be overcome. Doing the project itself would be something of an adventure, since no one had ever done such a thing under social and epidemiological circumstances like those in Homa Bay. Moreover, it would mean maintaining high quality care for many years. The method usually used for monitoring projects seemed ill-suited to a totally new experience; due to their inherent risks, innovative enterprises require a special type of management. For example (and perhaps forcing the point slightly), it could be said that HIV care in district hospitals has become so familiar that this kind of project can be managed using a precise, pre-established plan. But overall, it’s better to assume that each field project has social, cultural, economic and political circumstances unique enough to create uncertainty to justify a management that measures the work accomplished other than by merely looking at progress along a series of pre-defined steps.

MANAGING UNCERTAINTY

Basic research institutions and industry draw up management models appropriate to innovative projects, i.e., those with a high degree of uncertainty. By way of example, Bruno Latour’s paper “L’impossible métier de l’innovation technique [The impossible job of technological innovation]Bruno Latour, “L’impossible métier de l’innovation technique”, in Encyclopédie de l’innovation, Philippe Mustar et Hervé Penan (ed.), Paris, Economica, 2003, pp. 9-26.provides a framework for monitoring these kinds of projects, defined by the author as ever more costly experiments conducted by uncertain researchers and decision-makers who explore various degrees of uncertainty while capitalizing information, in the hopes of generating support for the proposed object among potential consumers and citizens. How, according to him, is such an undertaking managed?

“[translation] Calculations cannot be used to assess the chances of an even slightly radical innovation, because the world in which it will be introduced isn't stable enough to get reliable numbers; yet it would be futile to trust natural selection, since evolution isn’t guided by any sense of efficiency. So do we just give up, extol the perils and greatness of research “that no one knows how to manage,” and support random projects while hoping for the best? This approach – while it might be gratifying to researchers – usually ends up being a colossal waste. The issue is to know whether we can evaluate without calculations. Things that can’t be calculated can still be described. But how do we give a good description of an innovation that doesn’t yet exist? The way research projects are usually introduced makes evaluation nearly impossible. The researcher always tends to present his discovery as the eighth wonder of the world. Without flaws, opposition or competition, it shines – according to him – with the combined light of scientific truth, technical efficiency, economic profitability, and perhaps even social justice – not to mention the inevitable progress. To hear him, shareholders, venture capitalists, colleagues and consumers need only pull out their chequebooks. This is only human…but it isn’t assessable.

Now let’s suppose that someone asks the innovator to describe his project not as an absolute necessity, but as a perilous adventure that might well fail. We ask him to name the competitors whose products currently occupy the niche he wants to fill; we ask him to spell out the alternatives his project will have to settle for if it fails to convince; we want to know how it can be modified to incorporate opponents’ objections, and so on. Instead of making his presentation watertight, we ask him to describe the risks. Why, you might ask, would that kind of description allow a better assessment than the impossible calculation? If we can’t, in all fairness, ask the promoter of a radical innovation to calculate his project's chances, we certainly can’t ask him to know the answers to all of these questions about the ecology of an innovation yet to come.

Nor does the evaluator’s judgment apply to in-depth knowledge; to a nascent innovation we can only expect a nascent response. The evaluation is based not on a thorough knowledge of the project’s environment, but only on the increasing richness of the innovator’s description. The inventor can’t know the future; he might fail; he might be wrong; he is feeling his way in the dark; we can’t rely on any expert to judge him; we can't trust unfair natural selection. While all of this is true, there is only one thing doesn't lie – Ariadne’s thread remains solidly in our grasp – is the description of the project’s future world richer and more detailed now, after the project has gone through testing, than when the innovator and evaluator last met? What the evaluator can measure with some small degree of certainty is the “learning delta”, which makes it possible, between two tests or two meetings, to improve the description of the project, making it both more easily articulated and more negotiable. “Negotiable? Take it or leave it!” cries the indignant innovator. If that’s the case, don’t give him a penny – let the project languish on the shelf with all the other brilliant but unworkable inventions. What you’re looking at is not the next great thing, but a white elephant, a labyrinthine contraption. In order to exist in ten or twenty years, the project has to be able to fit into an ecology as fragile as an Amazonian jungle; either the innovator tries to understand the environment with you, and you have to support him through his testing, or he’s only interested in his project and not its ecology, and his project has no chance whatsoever of becoming reality. Demanding the description, you’ll get savings the calculation wouldn’t get you – and that beats counting on Darwin.”

Along with the advice above – written for an audience from the world of industry – are sixteen indicators, divided into four categories. Each is aimed at preventing what the author calls “innovation pathologies”:

• the belief that an innovative project can be ballistic, in the sense that its initially defined trajectory will not change;
• paranoia that manifests itself as hostility and contempt for any approach critical of the project;
• manipulation so that the project is judged by unrepresentative experts and by non-relevant testing;
• the lack of a plan, as signalled by the absence of any foreseeable object likely to reconcile the different environments and contradictory interests revealed by the early phases of the project.

Latour’s indicators are presented briefly to underline the fact that there are specific methods for managing projects characterized by uncertainty in other professional environments – for example, basic research and industry. The stakes are not small, since – as the author points out – laxity has its price: “[translation] As one often hears at business dinners, research and innovation are the surest (but definitely one of the most pleasant) ways to go broke.” Given the risks inherent in innovation, the proposed management method consists of evaluating the “learning delta” measured from one step to the next as the project goes along. As each step is evaluated, resources are allocated to the project according to the knowledge and know-how acquired in the preceding steps. This way, allocated resources are transformed into new information about the different states of the world in which the innovation is deployed, and about the contours of the innovation-inthe-making that will have to find a place in it.

ADVOCATING BEHAVIOUR CHANGE: AN ADJUSTMENT VARIABLE FOR PUBLIC HEALTH POLICY?

The process of innovation in public health can be represented schematically using three variables:

• the state of science and technology;
• the political will;
• the required behavioural changes.

When scientific and technical knowledge exists but is inadequate (e.g., AIDS after HIV testing came on the market but before the discovery of the effectiveness of HAART), and political will emerges out of fear of widespread epidemics among heterosexuals (governments begin to view HIV as a threat to their security, economy, and political stability), then the demand for behavioural change is at its highest. Everyone on the planet is asked to restrict the number of their sexual partners and use a condom every time they have sex.

When the scientific and technical capital (e.g. development of HAART) and political will (resolutions by the G8 in 2000, and by the WTO and UN General Assembly in 2001) increase, the behavioural changes demanded are more moderate. HAART is introduced in the form of a twice-a-day tablet in poor and middle-income countries. The patient is not asked to help pay for the medication and is given information about the disease and his treatment. He then becomes subject to the constraints of managing a chronic disease, i.e., regular attendance at consultations and treatment adherence.

This suggests that the demands for behaviour modification required by the implementation of a public health policy increase in proportion to the need to compensate for inadequate scientific knowledge or lack of political will. This can be helpful when assessing the early potential of a public health proposal.

	1985-1996: Use prevention to limit the spread of epidemics to broad populations	2000: Treat patients with antiretrovirals in all countries	2010: Reduce disease transmission through use of antiretrovirals and other methods
Science and technology	+	+++	+
Political will	++	+++	++
Behavioural change	+++	+	+++

How would we describe the current situation? Science is marking time. The prospect of a vaccine is becoming more remote, and the other preventive measures are unable to prevent the roughly two-and-a-half million new cases every year. No new antiretroviral drugs have come along to radically simplify treatment. The lab tests needed to follow patients are becoming more accessible, thanks to techniques that can be used in the resource-poor settings of our medical practice. A set of prevention measures and the widespread and ever-earlier use of antiretrovirals offer a glimmer of hope that HIV incidence, mortality rates, and ultimately, perhaps, prevalence, will decline. Donor states’ political will is also marking time, as evidenced by the slower growth in HIV funding and assertion of other priorities, such as maternal and child health. Earlier we saw why treating patients without seeing their numbers decline was not part of the strategies deemed as pertinent by donor states. Yet the current officially-stated goal of governments and HIV coalitions is an ambitious one (universal access to treatment),Universal access to treatment is understood to mean coverage of at least 80% of HIV-positive patients with already damaged immune systems. which would mean treating more patients.

In 2010, the desire for universal access to HIV treatment in spite of gaps in the science and the diminishing political will is creating a situation much like the one in 1985-1996. Back then, knowledge had advanced rapidly to the point where an HIV test and monoand bitherapies were available, and the political will among donor states to respond to the epidemic was on the rise. Those who had relatively ineffective tools but who nevertheless hoped to stop the spread of HIV asked for maximal behaviour change – fewer sexual relations and partners, and systematic condom use. Today, scientific progress is slowing and the political will is weakening, as witnessed by the slower growth of available funds. The extremely ambitious goal of universal treatment means once again believing in the possibility of rapid, profound changes in the behaviour of patients, care providers, and healthcare institutions.

RADICALIZING THE FIGHT AGAINST HIV?

No treatment can completely eliminate the HIV virus in people who are already infected; thus, while success in providing universal access to treatment would lower mortality, transmission would still not be slowed enough to prevent a continuous increase in the number of carriers. This would mean having to solicit donors for more every year, with no prospect of putting an end to either the disease or the increase in costs. We showed above that the donors’ commitment to treating millions of patients despite the lack of any prospect of eradicating the disease or reducing expenses was the result of a specific set of circumstances in the late 1990s. Today, the vast coalition of actors involved in social and political mobilization is expressing its concern about the weakening of government will.

In response, the “test and treat” strategy that promises to reduce transmission and costs in just a few years is gaining popularity. At the start of the Kenya visit, on arriving in Nairobi in July 2010, the goal of the first meeting was to decide whether MSF should sign an appeal by a coalition of Kenyan organizations entitled “HIV: eradication is within reach.” The text called for a financial effort by foreign donors and the government in order to make Kenya an “HIV transmission-free zone.” It also made reference to the position of authors from the two departments responsible for AIDS and tuberculosis at the WHO in favour of a test and treat-type strategy whose expected result would be the elimination of HIV epidemics. Some WHO experts expressed this position as individuals in an article in the Lancet.R. M. Granich, C. F. Gilks, C. Dye, K. M. De Cock, B. G. Williams, “Universal voluntary HIV testing with immediate antiretroviral therapy as a strategy for elimination of HIV transmission: a mathematical model”, The Lancet, 373, 2009, pp. 48–57.In practical terms, they proposed testing everyone over the age of 15 every year, and starting everyone who tests positive on HAART, whether they have clinical signs of the disease or not. According to a case study by the authors – a mathematical model of a simplified South African situation (heterosexual transmission only) – there is reason to hope that the HIV prevalence could drop from 10% to 1% in this virtual South Africa in just a few years.

What this article offers is a mathematical model that satisfies donor demands to use the funds in a way that reduces the number of patients, and thus costs. Every individual would be expected to volunteer for a screening test at least once a year, and would agree to take medication every day for the rest of his life – not to take care of himself, but to contribute to the collective good by reducing the spread of the disease. Is this realistic? Data from the anthropological studySee in this work, Vanya Kovacic, Social Navigation.confirms that the discipline required of patients by public health policies runs up against the complexity of their lives.

What is most surprising about the “test and treat” proposal is the wide and rapid following it has attracted barely a year after its publication, despite the ethical dilemmas and considerable practical obstacles. What will happen to each patient’s right to accept or refuse prevention? How can a proposal that at least doubles the number of people on ART become a reality within fragile public health systems where staff is in short supply, and the drugs have limited effectiveness and significant side effects? In reality, this proposal – more economic and political than medical – fits the situation that developed following the WTO Doha Agreement (2001) and the creation of the Global Fund (2002) perfectly. In those very specific circumstances, the States agreed to invest in treating patients despite the lack of prospects for eliminating the disease. Once that extraordinary and advantageous set of circumstances ceased to exist, talk of eradicating the disease started up again among HIV researchers and activists. What does it matter if there has been little progress in developing prevention tools, especially a vaccine; maintaining contact with donors means having to find a way to prevent them from losing impetus for economic reasons. Indeed, the slower growth of HIV funding is beginning to be felt. When asked about “test and treat,” a high-level officialThe names of those interviewed will not be cited in this text. They were encouraged to be forthright in their comments on condition of anonymity.in Kenya's Ministry of Health confirmed his interest. In his view, it’s simply a matter of financial resources. Ideally, This MwalimuA reference to Julius Nyerere, president of Tanzania from 1964 to 1985, nicknamed Mwalimu (the teacher), a fierce supporter of independent development policies vis-à-vis the former colonial powers.supporter would prefer that his country provide the funding itself, because to him that seems the best guarantee of sustainable development. But in the absence of a national solution, he wants to see donors increase their funding sufficiently to cover treatment for everyone who is HIV positive. Yet Kenya is one of the countries where in previous years healthcare institutions weren’t even able to put all of the international funding to good use. The interview with this high-level MoH official gave the impression – often experienced in other contexts – of an attitude that transnational funding of a public health initiative is equivalent to guaranteed income for administration officials. This feeling was to be confirmed by an interview with an officer from the National AIDS Control Council (NACC), then in full fundraising mode, and by the outside but involved view of an American Centers for Disease Control and Prevention (CDC) officer stationed in Kenya.

Testing an entire population every year and treating all positives with antiretrovirals – whether they need treatment or not – in order to stop transmission might be an appropriate approach in certain extraordinary situations. It is an experiment worth trying in high-prevalence areas where the epidemic cannot be controlled using other strategies and its impact is devastating. But as a list of the prerequisites (high prevalence, acceptability, and feasibility) makes clear, only rarely are all of these met. The Homa Bay region has twice the HIV prevalence of South Africa, the example for the “test and treat” model. Developing a similar strategy in this region would require mobilization and completely new forms of social organization; so little evidence of these is seen in MSF’s day-to-day work that the goal of eradicating the epidemic in a few years seems highly unlikely. Given where HIV-fighting tools are today, reducing the disease’s pressure on society depends more on the ability of political and social actors to come up with entirely new forms of medical and social mobilization than on medical technology itself.

THE RETURN OF THE COMMUNITY HEALTH WORKER – FOOT SOLDIER AND SPECTRE OF PUBLIC HEALTH

The universal access to care and “test and treat” proposals agree on one point: that the resources needed to achieve the objective are to be found in community mobilization (a set of new behaviours by patients and those close to them who are not members of the medical or allied health professions). And so re-emerge the community health worker and community mobilization at the heart of the push for universal primary care in 2000 – a push that began at the 1978 Alma-Ata Conference. More than thirty years on, the supposed virtues of these community health workers are still at the heart of transnational public health discussions, as illustrated by an excerpt from an Institute for Development StudiesInstitute for Development Studies (IDS). “The UK Department for International Development (DFID) is the Institute’s largest funder. IDS also receives funds from the European Union, various UN agencies, and a wide range of aid agencies, trusts and foundations”. See http://www.eldis.org/download.cfm?do wnloadfile=5A2F29 50-A58D-9352-CB5B5E0F1684A2 D6&typename=dm File&fieldname=fil ename.document: “Human resources are at the very heart of a health system. Health systems cannot function effectively without sufficient numbers of skilled, motivated and supported health workers; yet estimates suggest that there is a shortage of 4.2 million health workers worldwide. The shortage is most severe in sub-Saharan Africa. In countries where formal health workers are too few, Community Health Workers (CHWs), local level volunteers, have an important role to play in providing services to the poorest and most vulnerable communities. As CHWs are members of the communities where they work, they know and understand the health needs of those around them. Moreover, they can be trained and deployed quickly, and are unlikely to emigrate.”

But meta-analyses of existing data – in particular data from randomized controlled trials, or RCTs – offer some nuance useful for understanding the impact to be expected from mobilizing the now re-named Lay Health Workers (LHWs). One of the largest literature reviews available S.A. Lewin, S.M. Babigumira, X. Bosch-Capblanch, G. Aja, B. van Wyk, C. Glenton, I. Scheel, M. Zwarenstein, K. Daniels, Lay health workers in primary and community healthcare: A systematic review of trials. A policy brief prepared for the International Dialogue on Evidence-Informed Action to Achieve Health Goals in Developing Countries (IDEAHealth). [http://www.who.int/rpc/meetings/LHW_review.pdf].sums up the current state of knowledge: "Conclusions. The use of LHWs in health programmes shows promising benefits, compared to usual care, in promoting immunization and breastfeeding uptake; in reducing mortality and morbidity from common childhood illnesses; and in improving TB treatment outcomes. Little evidence is available regarding the effectiveness of substituting LHWs for health professionals or the effectiveness of alternative training strategies for LHWs.” The current state of knowledge is itself a subject that warrants some clarification. Many studies focused on high-income environments; randomized trials on Lay Health Workers are poorly referenced in the scientific databases; not all of the studies in the literature review include a comparison between professionals and LHWs providing the same services; the studies comparing medical and allied health professionals to LHWs do not allow to draw a conclusion as to which category is better; the periods under review are too brief to give an idea of the effects of LHW activities beyond the short term; the differences in the amount of staff supervision between the studies and routine care needs to be accounted for in order to assess the reproducibility of the effects noted; and overall, the available data do not distinguish well between failures due to poor implementation of the activities and those due to the chosen activities’ lack of effectiveness.

In the case of Homa Bay, Kenya, several NGOs were given responsibility and funding by the government and international donor institutions to act at the community level. According to Kenya’s strategic plan for HIV/AIDS, an estimated 60% of services at the local level must be provided by community service organisations (CSOs), members of the Kenya AIDS NGOs Consortium (KANCO). Created in 1990 by seven organizations, the consortium now has more than a thousand member organizations. Women Fighting AIDS in Kenya, or WOFAK, created in 1994, is one example of an organization working in a variety of areas: prevention and education for the most at-risk groups, advice for people with HIV, medical care, and advice on following the treatment. WOFAK runs ten centres, three of them in Nairobi, and estimates that it serves 6,000 women and 2,000 children. WOFAK representatives travel to sites and organize meetings with HIV-infected people in the villages. When questioned, heads of WOFAK in the Homa Bay district made no secret of the lack of resources and organization for doing their work. Thus, while organizations of this kind intend to act through people living with HIV or their close friends and family, the reality is that they do not have the material means to maintain a regular relationship with the people charged with conducting public health activities outside health centres and hospitals. What kept coming up, not just in the interview with WOFAK but also with representatives of other NGOs (the African Medical and Research Foundation, for example), was MSF’s reluctance to pay for the contacts (transportation and food) between these organizations’ representatives and those of the “communities” targeted by their action. In fact, in order to decentralize care from the district hospital to peripheral health centres, MSF was counting on the help of these very organizations who in turn asked MSF for help to be able to carry out their work.

The above give reason to doubt the results of a national and global strategy – universal access to treatment – that expects to achieve a highly ambitious goal using community mobilization as a cure-all to remedy the lack of appropriate technology and policies. That doubt is reinforced by the memory of similar situations with regard to health policy in general (e.g. the attempt to achieve universal access to primary healthcare), and to HIV in particular (early hopes of eradicating the spread of the epidemic through rapid, widespread adoption of new sexual behaviours).

SO WHAT DO WE DO NOW?

CAST ASIDE OUR ILLUSIONS AND KNOW OUR WEAKNESSES

There is nothing in MSF documentation or history to indicate that long-term public health responsibilities should be ruled out, a priori, because they are incompatible with the “MSF identity.” We should, however, learn from past experience. The first lesson is that we should distance ourselves from fantasies such as universal access to treatment and “test and treat.” Why talk about distance, and not about refusal? First, because these grand schemes for the future convey new health norms that legitimize resistance to unacceptable aspects of the present reality. These major transnational mobilization efforts deserve credit for putting issues, about which they were until only recently hopelessly apathetic, on the agenda of the world’s most powerful. The new political dynamics offer us the option of joining a collective mobilization while maintaining a critical approach. But critical engagement is only advantageous if we can maintain real distance from the illusion at the core of the proposed policy. For example, believing it is possible to eradicate HIV with the tools now available implies that, while we might need to replace them, it is not essential. But in a context of chronically scarce public health resources, that which is needed but not essential quickly becomes superfluous. Thus, small advances as crucial as new diagnostic tests and drugs become less likely.

The second major lesson our experience should teach us has to do with our own limits. We do not have all the qualities required by certain aspects of health policy. In many African and Asian countries, in particular, pursuing “healthcare for all by the year 2000,” or hoping to restore access to care after war, we have tried to improve the management of some public healthcare institutions by convincing the Ministries of Health to delegate to us the administration of some of them. To put it bluntly, while interventions of this type conducted from the early 1980s to the mid-1990s were educational, they were also resounding failures. Either the institutions were not at a stage where our objectives were realistic (free clinics following the Bamako Initiative, in an environment of structural adjustment), or MSF expertise and staff were a poor fit for their institution-building needs (the Expanded Programmes for Immunization). In addition to the unfavourable financial context of reduced public spending in the countries in question, we need to be clearheaded enough to see our own weaknesses: a private organization trying to consolidate public institutions, being foreign and the constant turnover in expatriate staff, many of whom didn’t have the necessary qualifications for the task.

DESCRIBE AND UNDERSTAND WHAT IT IS WE’RE FIGHTING

First and foremost, the epidemic itself needs to be described at the level on which we intend to fight it – in this case, in the Homa Bay and Ndhiwa districts. The internal literature is full of provincial, national, and global information. We have relatively detailed information on the patient cohorts being treated. But one level is missing – the collective reality in which we practice. Here, this means the reality of the people in two rural districts in Kenya.

Medical teams should be paying attention to patients who are lost to follow-up – and more generally, to the people who aren’t using medical services for which they are eligible. We should be adapting our services to the patients’ culture and living conditions – not the reverse. Sociology and medical anthropology should be given greater autonomy in formulating the questions to be studied. While these questions are supplied exclusively by operations managers as terms of reference reflecting their own concerns and prejudices, social science studies are very likely to reinforce the opinions already widespread within the organisation, rather than shed new light on the problems encountered. Our current prevention and treatment methods are still very restrictive. To what extent, and in what proportions, can they be carried out? Only a combination of medical and sociological observation can answer this question.

INVENTORY AVAILABLE STRENGTHS AND KNOWLEDGE

The recent hire of a Kenyan manager in charge of partnerships with the other HIV actors in Homa Bay should enable us to get an overall view of the resources available in the two districts. Perhaps MSF should take the initiative by proposing a general meeting of everyone working on HIV in Homa Bay. In addition to the available resources, this would allow us to inventory the convergences and divergences between actors, and probably gain a better understanding of the patient associations' point of view.

In any case, once we have listed the resources and possible synergies, there will still be a large discrepancy between HIV prevalence and the means for combating it. MSF has the resources, particularly financial, which would allow us to recruit the personnel that are lacking. Similarly, we would be capable of covering the geographic area with community health workers deployed outside healthcare institutions. In the specific area of AIDS, past studies Jaffar S, Amuron B, Foster S, et al., “Rates of virological failure in patients treated in a home-based versus a facility-based HIV-care model in Jinja, southeast Uganda: a cluster-randomised equivalence trial”, Lancet 2009 Dec. 19; 374(9707): 2080-9.have shown that such personnel can play an important role in patient follow-up. But the current literature on Lay Health Workers is quite clear about a potential systemic effect. There are no examples suggesting that the use of LHWs constitutes a decisive advantage in developing a sustainable healthcare system.

Fundamentally, doesn’t an approach where tasks are allowed to slide down the chain of command (from specialist to generalist, from doctor to medical assistant or nurse, from nurse to lay health worker) ultimately delegate much of the responsibility for treatment to the patient and the people around him (family, friends, patient associations)? In a number of countries – Burundi, for example – the large percentage of patients being treated in nonprofit networks confirms this. The reasons for starting ART as early as possible are becoming clear. It’s not only about the need for good individual care, but also the need to prevent an infected patient from transmitting the virus to too many other people. As a result, the number of individuals theoretically eligible for ART is growing, and is becoming increasingly difficult for the healthcare facilities to handle. More and more patients who are recruited or come in on their own will have an abnormal lab result, a detectable viral load, and no major clinical manifestations. Delegating a significant part of treatment responsibility to them, with support from the people around them, is probably the best compromise for ramping up the HIV response in places where the epidemic is most deadly, without excessive expansion of the public healthcare facilities to treat just one single disease.

The tools needed to implement this plan – even if only the antiretrovirals and lab tests – are not currently available. The need for a new fixed-dose antiretroviral combination – at a price where over ten million treatments a year worldwide would be feasible – is becoming clearer every day. It will have to be more robust against resistance, tolerant of irregular dosing, relatively insensitive to weight variations, and have less severe side effects. The number of lab tests that can be done rapidly in resource-poor settings is growing quickly. The HIV antibody test and CD4 count are already available in this form, and the viral load soon will be. Imagine patients going to a walk-in counter near their home or work to be screened, starting treatment based on lab criteria before their immune systems are too damaged, being monitored for treatment efficacy using simplified lab tests (rapid CD4 count and viral load) and, if there are no major abnormalities, getting their ART prescriptions renewed without seeing a doctor or nurse. The only piece missing from this primarily laboratory-based system for monitoring HIV-infected people before they show clinical signs of the disease is a lab test able to tell whether the patient is following the treatment. This is one specification that could be included when developing a new fixed-dose combination to take over from the current first-line treatment. Wouldn’t it be possible for the new FDC to include a marker that would indicate, via a simple lab test, good treatment compliance? But laboratory tools will only be useful if they are in line with, as much as possible, patients’ wishes. No matter how sophisticated the available tools, care-provider teams and patients will have to decide together on a treatment protocol. A change in the first-line treatment – which will be unavoidable due to treatment failures and side effects – offers an opportunity to examine what patients, and the clinicians who follow them, want from the new treatment.

Testing new approaches requires an appropriate working environment, which a partnership with KEMRI“The Kenya Medical Research Institute (KEMRI) is a state corporation established through the Science and Technology (Amendment) Act of 1979, as the national body responsible for carrying out health research in Kenya.”– the national medical research institute – could provide. The fact that KEMRI and MSF are both founding members of the Drugs for Neglected Diseases initiative bolsters the idea of a partnership with enough autonomy to develop alternative approaches outside, or on the fringes of, the public system’s treatment protocols. The hope is that, having been studied by the national medical research institute, these alternatives can some day be adopted by MoH healthcare centres.

Obtaining better tools (drugs and lab tests) is part of the transnational dimension of the fight against HIV. The specific needs of medical research for patients treated in resource-poor settings, the ins and outs of medical product marketing, and public health financing remain central issues. Ten years ago, thanks to the legitimacy it gained with its early treatment successes in places where they had been thought impossible, MSF helped advance these issues. It enabled the Association to state the problem it was facing in simple terms: “Our patients aren’t dying due to the virulence of infectious agents, but because of the unjustifiably high prices of the drugs that could save them.” It’s easy to understand the power of this public statement, and its devastating effect on the positions of the interest groups involved – for example, the Big Pharma and the donor states. The contemporary equivalent of this powerful lever would be a demonstration that the epidemic can be controlled at the local level. Then, drawing on this new expertise, MSF would again be justified in demanding publicly the resources needed to control HIV in the places where it is taking the greatest toll.