Brand Page

A Note on the Statistics Cited in this Book

Throughout this book, I refer to various statistical entities, including “HIV rate,” “HIV prevalence,” and “HIV incidence.” “HIV rate” and “HIV prevalence” are synonyms meaning “the proportion of people in a given population who are HIV positive at a given time.” In Africa, these figures usually refer only to adults aged fifteen to forty-nine, the group assumed to be most sexually active. Thus, to say that “the HIV rate in Uganda was 18 percent in 1992” means that 18 percent of adults aged fifteen to forty-nine were HIV positive in 1992. “In 1992, HIV prevalence in Uganda was 18 percent” means exactly the same thing.

HIV incidence is a measure of the speed at which HIV spreads. It refers to the proportion of people in a given population who contract HIV during a given time interval, usually one year. Thus, to say that “HIV incidence was 4 percent in 1987” means that 4 percent of the adult population contracted HIV that year. “HIV incidence in Kagera fell by 75 percent between 1987 and 1993” means that 75 percent fewer people contracted HIV in Kagera in 1993 than in 1987.

Most of the statistics cited in this book come from UN documents, the U.S. Demographic and Health Surveys, or the scientific literature. Most figures are derived from two different types of surveys, antenatal clinic surveys and population-based surveys. These two types of surveys sometimes give different results. Antenatal surveys tend to lead to higher overall prevalence estimates because pregnant women are by definition at higher risk because they have recently had unprotected sex. Also, HIV infection rates in Africa tend to be higher among women than men. Why this is the case is not entirely known. Women are not physically more vulnerable to infection than uncircumcised men are, and women also have fewer partners then men. One possible explanation is that when long term concurrent partnerships are common, an individual’s risk of infection depends less upon her own behavior than on that of her partner.¹

ANTENATAL SURVEYS

Antenatal surveys are based on tests of blood samples drawn from pregnant women in antenatal (or prenatal) clinics. As part of routine antenatal care, pregnant women are tested for syphilis and other conditions. In Africa, selected hospital labs are designated “HIV surveillance sites.” At these labs, some of the women’s blood samples are randomly selected to be tested for HIV, and the fraction found to be positive provides a rough measure of the HIV infection rate among sexually active women of childbearing age in the catchment area of the hospital. These surveys are usually conducted each year in selected African hospitals. They thus provide information not only about the HIV rate at a given time, but also about changes in the epidemic from year to year.

The dire condition of health services in many African countries means most women deliver their babies in their homes, not in hospitals. However, this does not significantly reduce the accuracy of antenatal HIV surveys because the vast majority of pregnant women attend a health clinic for antenatal care at some point during their pregnancies, even in the poorest African countries.²

Antenatal surveys are completely anonymous. The blood samples that arrive in the lab are labeled only with the women’s ages and the hospital’s name. There is no way of tracing a sample to a given woman, so neither the woman nor doctors and nurses nor lab technicians nor anyone else knows whether she is positive or not. Her confidentiality is thus entirely protected. A growing number of clinics in Africa now offer HIV counseling and testing for women who wish to know their HIV status.

The prevalence of HIV infection in pregnant teenagers aged fifteen to nineteen is sometimes used as a rough measure of the incidence of new infections in a population. This is because few children below age fifteen are HIV positive (see below), so most cases of infection in pregnant teens are assumed to have occurred recently.

POPULATION SURVEYS

Population surveys measure the HIV rate in the general population—including men, nonpregnant women, children, and the elderly, as well as pregnant women. These surveys are usually conducted by teams of researchers who go from house to house in a given area collecting blood or saliva samples that are then tested for HIV. In most cases, participants are offered the chance to learn their results or are referred to voluntary testing and counseling services.

Population surveys provide information about which population groups are most vulnerable to HIV. For example, they have shown that in East and southern Africa, HIV rates are higher among women than among men, while the opposite is the case in the rest of the world. Sub-Saharan Africa is also home to 90 percent HIV-positive children. Without antiretroviral drug treatment roughly 25 percent of babies born to HIV-positive women inherit the virus. Most of these children die of AIDS by age five, although a growing number are now being kept alive longer with AIDS treatment. HIV rates among children aged six to fifteen are extremely low, but then rise rapidly among sexually active teens.³ Some population surveys are conducted annually, and this allows researchers to calculate the incidence of new infections in a given population.

HIV statistics are often regarded with skepticism. All statistics should be questioned, whether they refer to the chance of rain tomorrow, the proportion of dentists who recommend chewing gum, or the prevalence of HIV in Africa. However, HIV infection is one of the most accurately measured diseases in Africa. In many countries, we know far more about the HIV/AIDS situation than we do about malaria, measles, postpartum hemorrhage, or other serious public health problems.

But mistakes do happen. In 2004, demographers at UNAIDS discovered an error in the analysis of antenatal surveillance data that led to an overestimation of HIV in some rural areas in Africa. Since a large fraction of the population of sub-Saharan Africa resides in rural areas, this led to a downward revision of the UNAIDS HIV prevalence estimates in many countries by 25 to 40 percent. For example, in 1992, Uganda’s HIV rate was estimated to be 30 percent, but the correction put the figure at roughly 20 percent. The error arose from the fact that the surveillance clinics used to estimate HIV prevalence in rural areas were actually located in towns and trading centers that had a distinctly urban character. Since HIV rates tend to be higher in urban areas, this led to an overestimation of HIV prevalence in rural Africa. The problem was corrected and now UNAIDS’s estimates correspond well with those of population-based surveys, such as the U.S. Agency for International Development’s demographic and health surveys.⁴

THE FUTURE OF THE GLOBAL AIDS EPIDEMIC

Estimates of the future course of the AIDS epidemic vary widely and are far less reliable than estimates of current prevalence. Predicting the future course of any event relies on numerous assumptions, and agencies generally present a range of possible scenarios. For years, UNAIDS; the World Health Organization; the Global Fund for AIDS, Tuberculosis and Malaria; and other agencies claimed that such countries as India, China, and Russia, which have generally had relatively low HIV rates, were “on the brink” of major epidemics on the scale of East and southern Africa.⁵ For reasons described in this book, the spread of HIV in Asia and eastern Europe is almost entirely confined to clearly defined risk groups, including prostitutes, intravenous drug users, and their sexual partners, and it is very unlikely that HIV will spread widely in the general population in these regions.⁶

MEASURING SEXUAL BEHAVIOR

Declines in HIV infection rates are finally underway in various countries in Africa, Asia, the Caribbean, and Latin America. There is general agreement that these declines are due to behavior change and not simply to the deaths of high risk individuals.⁷ However, there is considerable debate about what accounts for these welcome trends, particularly the most proximate behavioral changes—abstinence, faithfulness, or condom use—that must underlie them.

Sources of data on behavior change include the USAID-funded Demographic and Health Surveys, other national surveys and local research studies conducted by academic groups at Columbia University, Johns Hopkins University, Imperial College London, and others. The interpretation of these data should be straightforward. However, discrepancies sometimes arise in the measurement, analysis, and interpretation of the results. For example, indicators of condom use include “ever used,” “used at last sex,” “used at last higher risk sex,” and “always use.” Indicators of multiple partnerships include “multiple partners in last year,” “multiple partners in past four weeks,” and “sex with a non-spousal, non-cohabitating partner in past year.” Affirmative answers to any of these questions could have very different implications for the interpretation of data and trends. For example, for years, the UNAIDS program used indicators that over-estimated both the fraction of people engaging in “risky sexual behavior” and the fraction of people using condoms effectively.

Measuring “risky sexual behavior” is especially complicated. For many years, it was assumed that “risky sex” meant sex outside of marriage or cohabitation. Thus, the many researchers measured “risky sex” with the indicator “number of people who had sex with a nonspousal, noncohabitating partner in past year” of “all those who had sex in the past year.” We now know that in Africa, much HIV transmission takes place in ongoing relationships, including marriage, and therefore this indicator doesn’t really capture all of the riskiest sexual encounters, especially for young people. For example, using the usual indicator, a very high rate of “risky” sex would result if 99 percent of all young people were abstinent, and 1 percent were faithful to a single partner whom they were neither married to nor living with. In that case, 100 percent of sexually active young people in such a society would be having “high risk sex”—even if they were all abstinent or faithful! More realistically, since men tend to marry late in Africa, most of the young men who have sex, have it out of wedlock, so even if they are faithful to their partners, they will show up in the “risky sex” category. In Rwanda, for example, the fraction of young men and women having so-called “high risk” sex in 2005 was 48 percent and 15 percent, respectively, (meaning with a “non-marital/non-cohabitating partner”—among all youth who had sex) but only 9 percent of ALL young men and 5 percent of ALL young women had premarital sex at all.

Since the late 1990s, the US-funded Demographic and Health Surveys have been reporting the much simpler and more informative indicator “fraction of people with multiple partners in past year,” and as of 2008, UNAIDS will begin doing so as well.

When it comes to condom use, many researchers, report only the indicator “ever used,” or “used at last higher risk sex” (meaning with a non-marital, noncohabitating partner). But most HIV transmission in Africa takes place in longer term relationships and in such cases, condoms are only protective when used during every sexual act.¹ Some studies report such “consistent use,” but the practice is rare. In 2007 a Ugandan newspaper ran a feature story about a couple that had managed to use condoms throughout their fourteen-year marriage. This was considered national news.

Preface

One morning in November 2001, two officials from a Kenyan AIDS organization picked me up from my hotel in Nairobi and took me on a drive. We drove and drove all day, over muddy tracks, through endless pineapple and coffee plantations, rural villages and slums, through all of Africa, it seemed, to arrive at a small field, perhaps half an acre, with some weeds growing in it and an old woman standing there with a hoe.

I had not expected this. I was reporting on AIDS programs for an American foundation, and most of the other projects I had visited were either medical programs, AIDS awareness campaigns using billboards, radio or television spots, or traveling roadshows designed to promote AIDS awareness or condoms or HIV testing. I was about to say something when one of my guides spoke first.

“We are very proud of this project.”

So I said nothing. About twenty women had saved up for two years to buy this land. All of them were supporting orphans whose parents had died of AIDS, and they hoped the land would produce enough food for about fifty people in all. On a nearby hill, one of Kenya’s vast corporate-owned coffee plantations loomed like the edge of the sea. The old woman kept glancing at it as though it might sweep her away. I was moved by what I saw, although I didn’t understand at the time how this project was supposed to fight AIDS. This book explains how I came to do so.

The worldwide AIDS epidemic is ruining families, villages, businesses, and armies and leaving behind an immense sadness that will linger for generations. The situation in East and southern Africa is uniquely severe. In 2005, roughly 40 percent of all those infected with HIV lived in just eleven countries in this region—home to less than 3 percent of the world’s population.¹ In Botswana, Lesotho, and Swaziland, roughly a quarter of adults were infected, a rate ten times higher than anywhere else in the world outside Africa. In other world regions, the AIDS epidemic is largely confined to gay men, intravenous drug users, commercial sex workers, and their sexual partners. But in East and southern Africa, the virus has spread widely in the general population, even among those who have never engaged in what health experts typically consider high-risk behavior and whose spouses have not done so either. Although there were predictions that HIV would soon spread widely in the general population in Asia and eastern Europe, this has yet to occur, even though the virus has been present in those regions for more than two decades. The UN AIDS Program now predicts it probably never will.²

Why is the epidemic in East and southern Africa so severe? And why has it been so difficult to control? I started thinking about this in 1993, when I quit a postdoctoral job in molecular biology at the University of California and went to Uganda to work on an AIDS vaccine project. My results, like those of many others, were disappointing.

For more than twenty years, scientists have been trying to make such a vaccine, and most experts predict it will take at least another decade.³ The editor of Britain’s prestigious medical journal The Lancet has even suggested that a truly effective AIDS vaccine may be a biological impossibility.⁴

I continued to work on AIDS as a writer and consultant for various development agencies after I left Uganda, and I continued to wonder about what might be done to arrest the epidemic, and whether some other device or program might substitute for a vaccine. In 1996, a combination of three antiretroviral drugs, taken for life, was found to dramatically relieve the symptoms and extend the lives of HIV-positive people. At the time, these drugs were patented and extremely expensive, and for years they were out of reach of the millions of poor African patients who needed them. Before long, a worldwide network of AIDS activists began to pressure pharmaceutical companies to cut the prices of these drugs and urged international donors to raise billions of dollars to fund AIDS treatment programs in developing countries. As a result, millions of Africans with HIV are now receiving treatment.

In this book, I do not deal at length with this extraordinary struggle, a story that has been ably covered by other writers, some of whom are activists themselves.⁵ While the humanitarian urgency of AIDS treatment programs is inarguable, these drugs will not halt the epidemic on their own. They are not a cure, they don’t work for everyone, and they can have severe side effects. In Africa, those most likely to spread the virus to others are often at an early stage of infection and are not in need of treatment. In many cases, their infections may not even be detectable by HIV tests.⁶ Because Africa’s health-care infrastructure is in such a dire state, treatment programs are expensive and difficult to administer, even when the drugs themselves are practically free. Those who do receive treatment can expect to gain, on average, only 6.6 years of life because the virus eventually develops resistance, necessitating second- and third-line treatment, presently all but unavailable in Africa.⁷ It is impossible to put a price on six years of anyone’s life, least of all that of an African mother whose children would otherwise be orphaned, so the international community must endeavor to expand the range of AIDS drugs available in Africa. However, it would be better by far if that mother had never become infected in the first place.⁸

To date, the closest thing to a vaccine to prevent HIV is male circumcision, which was shown in 2006 to reduce the risk of HIV transmission by roughly 70 percent.⁹ The widespread practice of male circumcision in the predominantly Muslim countries of West Africa may largely explain why the virus is so much less common there than it is along the eastern and southern rim of the continent. It is urgent that as many men as are willing to undergo the procedure have access to cheap, safe circumcision services. But it may take years to develop such services and in the meantime, millions of people will become infected. In any case, HIV infection rates may be quite high, even in West African cities where nearly all men are circumcised.¹⁰

As international concern about the epidemic has grown, along with foreign-aid budgets for programs to fight it, a global archipelago of governmental and nongovernmental agencies has emerged to channel money, consultants, condoms, and other commodities to AIDS programs all over the world. During the past decade, I have visited dozens of these programs and spoken to hundreds of people. I never found a panacea, but I did learn a great deal. I learned, for example, that AIDS is a social problem as much as it is a medical one; that the virus is of recent origin, but that its spread has been worsened by an explosive combination of historically rooted patterns of sexual behavior, the vicissitudes of post-colonial development, and economic globalization that has left millions of African people adrift in an increasingly unequal world. Their poverty and social dislocation have generated an earthquake in gender relations that has created wide-open channels for the spread of HIV. Most important, I came to understand that when it comes to saving lives, intangible things—the solidarity of ordinary people facing up to a shared calamity; the anger of activists, especially women; and new scientific ideas—can be just as important as medicine and technology.

Like many newcomers to Africa, I learned early on that the most successful AIDS projects tended to be conceived and run by Africans themselves or by missionaries and aid workers with long experience in Africa—in other words, by people who really knew the culture. The key to their success resided in something for which the public health field currently has no name or program. It is best described as a sense of solidarity, compassion, and mutual aid that brings people together to solve a common problem that individuals can’t solve on their own. The closest thing to it might be Harvard sociologist Felton Earls’s concept of “collective efficacy,” meaning the capacity of people to come together and help others they are not necessarily related to. Where missionaries and aid workers have, intentionally or not, suppressed this spirit, the results have been disappointing. Where they have built on these qualities, their efforts have often succeeded remarkably well.

It’s easy to be pessimistic about Africa. Headlines from the continent chronicle apparently endless war, tyranny, corruption, famine, and natural disaster, along with a few isolated nature reserves and other beauty spots. Certainly there are many war-torn countries in Africa and many poor, sick people who need assistance. But sometimes helplessness is in the eye of the beholder. There is also another Africa, characterized by a striking degree of reciprocity, solidarity, and ingenuity. Time and again, African people have relied on these qualities to save themselves—and at one time, the entire human family—from extinction. Now, faced with the scourge of AIDS, some of them, including the farmer I met in Kenya, are trying to do so again.

Most of the black Africans who now live in the region covered in this book are descended from Bantu farmers who began migrating from western Africa several thousand years ago, across the continent and then south.¹¹ On the way, some of them encountered other African population groups—the San and Khoi of southern Africa and the Nilotes of the Sahel, for example—with whom they exchanged aspects of language and culture and with whom they sometimes intermarried. Subgroups splintered off from each other and adapted to local circumstances.

Their story is, with some exceptions, not about the accumulation of great personal fortunes and the founding of cities with palaces, cathedrals, and libraries. It is a story of relatively small groups banding together to survive on a harsh and dangerous frontier, of natural disasters and political and economic crises.

Survival was not inevitable. The ancient, infertile soils of Africa could not sustain large permanent farming settlements, and the development of towns was further prevented by infectious diseases that spread rapidly as soon as populations reached a certain threshold. When farmers cleared large tracts of land to grow crops, malaria bloomed in the sunlit mud; as herds expanded, the animals succumbed to tuberculosis and sleeping sickness, which spread to their owners.

Faced with such a mutable, dangerous world, the people of East and southern Africa developed a genius for local improvisation, adapting to life in forests, deserts, or lakesides. Cut off by the Sahara from the developing technologies of Europe and Asia, they were forced to innovate and developed their own methods of agriculture, iron smelting, and mining. In a world without the apparent consolations of property and bureaucratic institutions, a powerful sense of spirituality provided moral order and solace to the suffering.¹² Few groups developed writing, but they relied on drumming, the patterns woven into cloth and beadwork, and their prodigious memories to transmit information and an ever-changing repertoire of stories and myths.

On the harsh African frontier, you were nowhere without other people, and this is still the case, even though the crises facing the continent are very different and constantly changing. It is almost impossible to be truly alone in Africa, and this has a profound effect on how people see the world and act in it. In remote villages, the poorest families will invite strangers into their houses and won’t let them leave until they have eaten an enormous meal. Most Africans I know live in households that swarm with a vast and changing cast of inhabitants, including grown offspring, nieces, nephews, poor relations, aged aunts and uncles, and innumerable children. You would need a spreadsheet to establish who is related to whom and how.

These societies, wrote the historian Basil Davidson, “enclosed relations between people within a moral framework of intimately binding force…. an intense and daily interdependence that we in our day seldom recognize, except in moments of postprandial afflatus or national catastrophe. The good of the individual was a function of the good of the community, not the reverse.”¹³

This sense of solidarity has a downside when it contributes to tribalism and social rigidity, but it can also be a source of power and creativity, and it has been at the heart of the region’s most successful responses to AIDS.

What I didn’t know when I was in Uganda in the early 1990s was that something remarkable was happening there. During the 1990s the HIV infection rate fell by some 60 percent in the arc of territory along the northern and western shores of Lake Victoria, an area comprising southern Uganda and the remote Kagera region of Tanzania. This success, unique on the continent at the time, saved perhaps a million lives. It was not attributable to a pill or a vaccine or any particular public health program, but to a social movement in which everyone—politicians, preachers, women’s rights activists, local and international health officials, ordinary farmers, and slum dwellers—was extraordinarily pragmatic and candid about the disaster unfolding in their midst. This response was similar to the spontaneous, compassionate, and angry AIDS activism of gay men in Western countries during the 1980s, when HIV incidence in this group also fell steeply.¹⁴ Why has such a response been so slow to emerge elsewhere? The complete answer may never be known, but in this book, I suggest that outside of Uganda and Kagera, health officials misunderstood the nature of the AIDS epidemic in this region, in particular why the virus was spreading so rapidly in the general population. As a result, the programs they introduced were less effective than they might have been and may have inadvertently reinforced the stigma, shame, and prejudice surrounding the disease.

Much of the stigma and confusion surrounding AIDS has to do with its common association with perceived “irresponsible” or “immoral” sexual behavior. However, what many people—from policy-makers, to public health experts, to ordinary African people at risk—did not realize is that most HIV transmission in this region results from normative sexual behavior, practiced by large numbers of people. It’s not that African people have more sexual partners, over a lifetime, than people in Western countries do—in fact, they generally have fewer.¹⁵ However, in many African communities, both men and women are more likely than people in other world regions to have more than one—perhaps two or three—overlapping or “concurrent” long-term partnerships at a time. A man may have two wives, or a wife and girlfriend, and one of those women may have another regular partner, who may in turn have one or more other partners and so on. This “long-term concurrency” differs from the “serial monogamy” more common in western countries, and the casual and commercial “one-off” sexual encounters that occur everywhere. But long-term overlapping relationships are far more dangerous than serial monogamy, because they link people into a giant network that creates a virtual superhighway for HIV.¹⁶

Concurrent sexual partnerships have strong cultural, social and economic roots in East and Southern Africa, and this has made fighting HIV very difficult. Fifteen years of vigorous condom promotion in many African towns and cities has had little effect on the epidemic, probably because most transmission occurs in long term relationships in which condoms are seldom used. As family planning experts have known for decades, people use condoms mainly in casual and commercial relationships, and inconsistent condom use offers poor protection against either pregnancy or STD transmission in long term relationships.

Urging African people to abstain or be faithful has its limitations too, because most people are faithful already, if not to one person, then to two or three. Many of those at highest risk of infection are the faithful partners of men or women with only one other trusted long-term partner; others are in mutually faithful relationships, in which one partner had concurrent partners in the past.

In this book, I argue that the key to the success of Uganda’s early AIDS campaigns was that Ugandans, like the gay men of San Francisco and New York and like the Thai men of Bangkok, knew where their risks were coming from, and this made it easier for them to respond pragmatically. Although Ugandan public health officials didn’t know the word “concurrency” they did know that the virus was spreading in relatively ordinary families and relatively ordinary relationships and that it wasn’t just a problem for “promiscuous” people alone. This made it possible for everyone to speak more openly about how HIV had devastated their lives and to regard those affected by the disease with greater compassion. This openness in turn led people to take obvious steps to protect themselves.

Outside of Uganda, most AIDS prevention campaigns in Africa were for years aimed almost exclusively at so-called “high risk groups”—meaning prostitutes, truckers, soldiers, and other rootless, migrant men. Such an approach made sense in the rest of the world, where HIV remains largely concentrated in “high risk groups.” But in East and Southern Africa, the concurrency “superhighway” puts virtually everyone at risk, from cabinet ministers to the women selling tomatoes on the street, even if they are not typically “promiscuous.”

Much has been written about the lessons of Uganda’s early success against AIDS, not all of it in agreement, either with the interpretation advanced here or with other interpretations. This is partly because the evidence for what happened in Uganda, and what it meant for billions of dollars in AIDS prevention funding, soon became the object of a tug-of-war between researchers and policy makers on the left and right of the political spectrum. Health officials on both sides interpreted the data in their own ways, and used their conclusions to support programs they favored. We will never know whether lives could have been saved had these policies been based on the evidence for what actually occurred in Uganda, but it is possible that this partly explains why fighting AIDS in Africa has been so difficult, and it may also suggest a way forward.

The AIDS epidemic is finally beginning to subside in many African countries, owing to increasing awareness and commonsense changes in sexual behavior. This is heartening, but it is possible that many lives might have been spared had policymakers better understood the nature of the epidemic early on.

Much of this book is concerned with donor-funded AIDS programs that failed in some way, beginning with my own vaccine project. I tell these stories not with a sense of satisfaction. I could not have done better myself at the time. But in science, failures are often as important as successes, because they tell us where the limits are. Only by looking honestly at our mistakes can we hope to overcome them. When it comes to fighting AIDS, our greatest mistake may have been to overlook the fact that, in spite of everything, African people often know best how to solve their own problems. They have been doing so throughout human history. Had they not succeeded, I would not be here to write these words, nor would you be here to read them.

HELEN EPSTEIN

The Invisible Cure

PENGUIN BOOKS

Contents