From "Marvelous Momentum" to Health Care for All

January 23, 2007

by Paul Farmer

Paul Farmer is associate chief of the Division of Social Medicine and Health Inequalities at Brigham and Women's Hospital; Presley Professor of Medical Anthropology at Harvard Medical School; and a founding director of Partners In Health, an international charity that provides health care to and undertakes research and advocacy on behalf of the sick and poor.

The last quarter of the twentieth century saw little investment in international health or in the health problems of the world's poor. Over the past few years, as Laurie Garrett notes ("The Challenge of Global Health," January/February 2007), "driven by the HIV/AIDS pandemic, a marvelous momentum for health assistance has been built and shows no signs of abating." But after this upbeat introduction, Garrett proceeds to lay out the perils associated with this new momentum, chief among them that an influx of AIDS money has drawn attention away from other health problems of the poor, weakened public health systems, contributed to brain drain, and failed to reach those most in need.

I respond as a physician who has lived through the dry spell, seen the rains coming, and witnessed the burgeoning of the first sprouts of hope in a long time. Because many others who work in places such as rural Africa and Haiti — the examples used by Garrett in her essay — also see the threat of bad seed ruining the harvest, I will not dispute her argument about the disproportionate use of scarce health-care resources. In fact, I agree with most of her claims. I would rather focus on how the new enthusiasm about global health can be translated into efforts to close the widening "outcome gap" between rich and poor.

The stakes are enormous. It is well known in development circles that huge amounts of aid have often brought few improvements to the lives of the world's poorest. A first principle for the emerging global health movement, in fact, might well be: "Don't emulate the mainstream aid industry." That said, aid is not bad in itself, and if managed appropriately it can achieve impressive results. The end of the funding drought has been a tremendous boon, especially for the destitute and sick (and those who provide care to them).

It is worth comparing the situation in 2002, the year the Global Fund to fight AIDS, Tuberculosis, and Malaria made its first pledges, and today. Garrett is correct to remind us that AIDS is far from the only problem faced by the destitute sick in rural Africa, but it is the leading infectious cause of adult death there. At the beginning of the millennium, there was no real political will, and no money, to treat the poorest Africans with AIDS, in spite of declarations to the contrary. In 2007, on the other hand, there is some money for AIDS prevention and care, although it rarely makes it all the way to rural Africa. In 2002, there were almost no antiretroviral (ARV) medications in rural Africa, nor were there personnel to deliver them. In 2007, most African nations are working to make AIDS diagnosis and care "a public good for public health" — that is, a service paid for by the commonwealth or rich donors, rather than by individual AIDS sufferers and their families. Although these drugs are as yet reaching very few rural Africans, the past five years have seen significant investments, at the district if not the village level, to make AIDS therapy available for those who are able to walk or find other transportation to district hospitals.

The brain drain of health-care personnel from the developing world described by Garrett has not been reversed over the past five years, but the experience of Partners In Health (PIH) in Haiti and in Africa offers hope. As hospitals are refurbished and become something more than charnel houses, as medications are made available, some doctors and nurses are returning to the rural public-sector institutions in which we work. There is a growing awareness that not only doctors and nurses are needed to deliver medical care: Many are learning that proper "accompaniment" — closely supervised home-based therapy; social and psychological support; and help with everyday tasks, including feeding families — is what poor patients with AIDS need most of all, once the demand for coffins is replaced by the demand for a continuous supply of ARVs.

Garrett notes that "Guinea-Bissau has plenty of donated ARV supplies for its people, but the drugs are cooking in a hot dockside warehouse because the country lacks the doctors to distribute them." I would argue that in no country in the world are doctors effective as distributors of medication. PIH has instead trained community health workers called accompagnateurs, who have achieved AIDS treatment outcomes in rural areas of poor countries much better than those registered in what is today termed "inner-city" America. Accompagnateurs, not doctors or nurses, are the appropriate distributors of medications — which is why we have now imported the Haiti model to Boston.

In 2002, AIDS prevention and care were considered different and opposed activities, as experts and activists fought over scarce resources. In 2007, although this struggle continues, prevention and care have been integrated in some settings with excellent results. In 2002, experts advocated what could only be described as substandard care for poor Africans with AIDS, even if these recommendations were sometimes dressed in fancy-sounding names such as "home-based" or "palliative" care. In 2007, progress has been made, since some argue that while the "home-based" part of the formula is correct, the "care" component must include ARVs, and that "palliative care" — code for helping people to die with less pain — should not be used for a disease that strikes mostly young adults and children unless that disease is untreatable. It is true that sub-standard guidelines persist in 2007, but they are being challenged by many who seek to improve the quality of care available to the rural poor.

In 2005, PIH initiated, with the Rwandan Ministry of Health and the Clinton Foundation, a new rural AIDS initiative based on the Haiti model, and it is starting to have some success. More than 2000 people with AIDS are now receiving therapy within two health districts in eastern Rwanda — districts that, prior to 2005, were served by not a single doctor. Over 400,000 people live in these districts; 60 percent of them are resettled refugees or others displaced by war and genocide. PIH did bring in doctors at first, some of them Americans, but within months of our arrival, over 95 percent of our employees were African, most of them accompagnateurs. And most of what we do, in Rwanda as elsewhere, has more to do with primary health care than with AIDS. We also work within the public sector so that the doctors, nurses, and paraprofessionals who work with us are not part of the brain drain at all.

Unfortunately, such practices — and such results — are the exception rather than the rule. "By one reliable estimate," notes Garrett, "there are now more than 60,000 AIDS-related NGOs alone." Yet by 2006, after a global campaign to bring AIDS care to Africa, fewer than 25 percent of Africans who needed ARVs to survive were receiving them, with the fraction dwindling to less than 5% in rural areas. Worse, new infections continue apace. So what on earth, one might ask, are all these AIDS-focused NGOs doing? That is a very good question, and we should all be grateful to Garrett for posing it so provocatively.

As Garrett notes, it is not NGOs alone that suck up resources intended for the poor; corrupt governments divert many of these resources to the pockets of the non-poor, including a huge "helping class" that is quite international in flavor. Garrett cites a 2006 report by the World Bank estimating that "about half of all funds donated for health efforts in sub-Saharan Africa never reach the clinics and hospitals at the end of the line," and this is surely true. But it is important to add that the same international financial institutions issuing such reports are contributors to the situation — having for years suggested "capping" social expenditures in health and education and even made such restructuring of public budgets a pre-condition for access to the credits and assistance upon which poor governments depended for survival.

Garrett is correct to emphasize the importance of strengthening public sector health institutions and to criticize "vertical" or "stovepiped" approaches to health care. And she is to be lauded for describing the distortions that frequently ensue when large sums of money are introduced into cash-starved health systems. Our experiences at PIH, however, suggest that while her general thesis is right, Haiti is not a good example of it. Garrett claims that former U.S. President Bill Clinton is wrong to suggest that AIDS initiatives "end up helping all other health initiatives." "The experience of bringing ARV treatment to Haiti," she writes, "argues against Clinton's analysis. The past several years have witnessed the successful provision of antiretroviral treatment to more than 5,000 needy Haitians, and between 2002 and 2006, the prevalence of HIV in the country plummeted from six percent to three percent. But during the same period, Haiti actually went backward on every other health indicator."

There are three problems with these correlations and inferred claims of causality. First, are they true? The reduction of HIV prevalence has been well documented. But has Haiti actually gone backward "on every other health indicator" between 2002 and 2006? This might be true in the chaos of present-day Haiti, but the national level surveys that would provide such data have not been conducted, much less completed and analyzed.

Second, even were such a claim shown to be true, how would we know that the primary reason for such backsliding was too much AIDS funding rather than, say, the 2004 coup d'état, the country's 34th, an event that led to great political upheaval, attacks on hospitals and clinics, disruption of medical supply chains, and the effective dissolution of Haiti's national AIDS commission (which had been ably chaired by First Lady Mildred Aristide, one of the primary architects of Haiti's successful Global Fund application)?

Third, I am confident, even without the results of national surveys, that Garrett's stovepiping hypothesis, manifestly true in most countries mentioned, does not hold true in central Haiti, where close to half of the Global Fund grant went and where half of those 5,000 "needy Haitians" on ARVs live. There, as we have documented, the increased AIDS funds were spent exactly as Garrett advocates: to strengthen the public health system in general. Even if we measure, as she suggests, by maternal mortality and life expectancy at birth (rather than the "short-term numerical targets" she deplores), we see that "AIDS funds" may be used to reduce maternal mortality and increase life expectancy.

The data shown in Figures 1 and 2 below come from the first public clinics rehabilitated during the course of 2002-3, the very period Garrett discusses in referring to Haiti. They demonstrate that "AIDS money," when used as a means of strengthening health systems well beyond the stovepipes justly excoriated by Garrett, can indeed have a salutary and rapid impact on, say, provision of women's health care or uptake of vaccinations.

These results show that through careful program design, stovepiped intentions may be subverted or "horizontalized" in order to introduce new resources to the cash-starved public sector and disadvantaged rural regions in some of the poorest countries of the world.

PIH learned to do this decades ago. We found that it is simply not possible to have vertical programs in poor, rural areas, because people in those areas typically suffer from more than one disease at a time. In fact, the great majority of our patients in Africa and in Haiti do not have AIDS. And about half of our African AIDS patients also have tuberculosis infection. So how could we not link our AIDS and TB programs? Malaria kills far more African children than does HIV. Women's health must be comprehensive — from family planning to modern obstetrics to AIDS care — for prevention to be effective and ethical; it must be linked with efforts to make clean water available if pediatric HIV disease is ever to be eliminated. When you are the only hospital for miles around (because the other NGOs are in the city), and people come to you with pneumonia, broken limbs, and epilepsy, you cannot refer them to a local vertical program designed to treat pneumonia, broken limbs, and epilepsy — because such programs do not exist.

The influx of AIDS funding can indeed strangle primary care, distort public health budgets, and contribute to brain drain. But these untoward or "perverse" effects are not inevitable; they occur only when programs are poorly designed. When programs are properly designed to reflect patients' needs rather than the wishes of donors, AIDS funding can strengthen primary care. PIH has shown this throughout central Haiti, in eastern Rwanda, and in the mountains of Lesotho, and is going to use the same model in southern Malawi. In each of these settings, we work under the aegis of the Ministry of Health (and, in three of them, with the Clinton Foundation) in order to promote the notion of health as a human right. In some cases, programs have to be built from scratch; in others, it is necessary to rebuild public infrastructures damaged by war, neglect, or the misguided advice of outside experts.

Those concerned about global health must not only promote a commitment to social justice, but also teach our allies to make a careful analysis of how the global outcome gap came to be and why it continues to worsen in spite of many well-intentioned efforts to reverse it. Garrett's critique is welcome as a part of that analysis, but it should be directed primarily at the badly designed programs — lest casual observers incorrectly conclude that good results cannot be achieved, when in fact they can.