Jump to ContentJump to Main Navigation
AIDS and the Ecology of Poverty$

Eileen Stillwaggon

Print publication date: 2005

Print ISBN-13: 9780195169270

Published to Oxford Scholarship Online: February 2006

DOI: 10.1093/0195169271.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 25 September 2018



(p.3) 1 Perspective
AIDS and the Ecology of Poverty

Eileen Stillwaggon (Contributor Webpage)

Oxford University Press

Abstract and Keywords

This chapter discusses the modes of transmission, worldwide distribution, and environmental conditions necessary for the spread of HIV. It explains the epidemiological framework for understanding disease spread, which includes characteristics of the pathogen, the host, and the environment. The appendix to this chapter presents cross-national survey data on sexual behavior. The data demonstrate that there is no correlation between higher national prevalence of HIV and countries with higher rates of early initiation of sex, premarital sex, or multipartnered sex.

Keywords:   HIV transmission, HIV distribution, epidemiology, pathogen, host, environment, sexual behavior

Global AIDS policy has failed to stem the epidemic spread of HIV.1 This book explains why HIV/AIDS exacts such a devastating toll in sub-Saharan Africa and why the epidemic continues to spread in Latin America, the Caribbean, Eastern Europe, the former Soviet republics, and South, Southeast, and East Asia. This book also explains the failure of international AIDS policy that is built on flawed analysis and inadequate tools. Policy makers and researchers working on HIV/AIDS have asked the wrong questions about the causes—not of AIDS itself, but of its epidemic spread. Global AIDS policies attempt to stop HIV transmission at the last possible moment, instead of grappling with the underlying causes of the epidemic.

The grim facts are numbingly familiar. Almost 40 million people are infected worldwide with HIV, more than 90 percent of them in developing countries. That includes almost 30 million people in sub-Saharan Africa, more than 6 million people in South and Southeast Asia, 2 million in Latin America and the Caribbean, almost 2 million in Eastern Europe and Central (p.4) Asia, and 1 million reported in East Asia. At least 20 million people have already died from the opportunistic infections that constitute the syndrome. There are 15 million AIDS orphans, 12 million in sub-Saharan Africa alone. While the epidemic may be slowing in a few countries, prevalence still appears to be growing in most of Africa, Asia, Latin America, Eastern Europe, and the former Soviet Union (UNAIDS, 2004).

HIV/AIDS epidemics emerged in distinct forms in industrialized countries and in the developing world. In affluent countries, national rates of HIV prevalence remain below 1 percent of the adult population. In sub-Saharan Africa, prevalence of HIV exceeds 2 percent of the adult population in almost all countries; in 21 African countries, more than 5 percent of adults are infected. In South Africa and Namibia, more than 20 percent of adults are infected with HIV, in Zimbabwe, 25 percent, and in Botswana, 37 percent. In Asia, HIV has increased very rapidly in some areas since 1995, and 4 to 5 million people in India (1 percent of the adult population) are now infected. In Latin America, most national rates are below 1 percent, but in some cities in the region, HIV prevalence exceeds 20 percent among men who have sex with men and 60 percent among injecting drug users. Several Central American and Caribbean countries have the highest national rates in the Americas, from 2 to 6 percent of adults. The fastest rate of increase in the world is reported in the former Soviet republics, in particular Ukraine, Russia, Latvia, and Estonia (UNAIDS, 2004).

Transmission of HIV

To understand the causes of epidemic HIV, we must begin with an examination of the ways that the virus is transmitted. In affluent countries, HIV affects almost exclusively men who have sex with men, people who share needles for injecting drugs, female partners of bisexual men and needle sharers and their infants, and hemophiliacs. To the surprise of epidemiologists, HIV has made few inroads in the rest of the population in affluent countries, in spite of heterosexual epidemics of other sexually transmitted diseases. In Latin America and Southeast Asia, epidemics that began among drug users and men who have sex with men are spreading through heterosexual contact and from mother to child. In very poor countries, exposure through contaminated blood and medical equipment is also a much greater risk than in rich countries (UNAIDS, 2004). China and some parts of sub-Saharan Africa have been especially affected by iatrogenic infections. In the former Soviet republics, transmission is thought to be primarily through the sharing of contaminated drug equipment. In sub-Saharan Africa and in other very poor regions, the primary modes of transmission of HIV infection are heterosexual intercourse and vertical transmission from mother to child in pregnancy and lactation (UNAIDS, 2004), although unsterile medical procedures are an important cause of (p.5) HIV spread (Gisselquist et al., 2003), and sex between men certainly also plays some role.

The efficiency of transmission of HIV (the probability that an individual will be infected through direct contact with an infected person or through blood products or instruments) varies, depending on the mode of contact. Blood transfusions are the most efficient mode of transmission, with 90 to 100 percent of persons transfused with contaminated blood becoming infected. Needle-sharing is also highly efficient, with the result that local epidemics among drug users have spread very rapidly. Vertical transmission varies in efficiency among regions. Rates of transmission, before the introduction of prophylactic treatment of mothers, ranged from 14 to 40 percent of infants of HIV-infected mothers. Among men who have sex with men, unprotected receptive anal intercourse is conservatively estimated to result in 5 to 30 infections per 1,000 contacts; nonreceptive anal intercourse has a lower risk. Among medical personnel, the risk of infection from accidental needle pricks is estimated to be about 3 per 1,000 accidents, without adjusting for any other possible sources of infection, such as drug use or unprotected sex (World Bank, 1997).

Unprotected heterosexual intercourse between otherwise healthy adults is a relatively inefficient mode of HIV transmission. The transmission of HIV between healthy persons in industrialized countries is relatively rare and has been insufficient to maintain a heterosexual epidemic in the absence of other factors. In the United States and Western Europe, transmission from an otherwise healthy HIV-infected adult female to a healthy adult male will occur in about one out of 1,000 contacts. Transmission from an otherwise healthy HIV-infected adult male to a healthy adult female occurs about once in 500 contacts (World Bank, 1997). (For statistical purposes; not for individual use.) This book is primarily about the heterosexual and vertical epidemics of HIV/AIDS. It is in those epidemics that the greatest difference in rates of infection between rich and poor countries occurs.

Epidemiology of HIV

While the immediate cause of much HIV transmission in sub-Saharan Africa and other very poor regions is heterosexual intercourse, the individual transmission and epidemic spread of HIV are not simply mathematical functions of sexual behavior. HIV is an infectious disease. Epidemiological, clinical, and laboratory evidence show that HIV infection is influenced by the same factors that promote transmission of other infectious diseases. An established literature in public health and a century of clinical practice demonstrate that persons with nutritional deficiencies, with parasitic diseases, whose general health is poor, who have little access to health-care services, or who are otherwise economically disadvantaged have greater susceptibility to infectious diseases, whether they are transmitted sexually or by food, (p.6) water, air, or other means. Regarding the conditions necessary for the transmission of infectious disease, Louis Pasteur is reported to have said: “The microbe is nothing, the terrain everything.”

Living conditions in poor regions provide a very different terrain for the propagation of infectious disease from that of industrialized countries. The real annual income of the average person in the United States is more than 60 times that of the average Tanzanian or Malawian (UNDP, 2004). Calorie intake per capita in sub-Saharan Africa has not increased since 1970 and is still only 70 percent of the consumption level of industrialized countries. Public and private spending on health services in Canada is about US$2,800 per person per year; in Nigeria, it is US$31 per person per year; and in Ethiopia, it is US$14 per person per year (UNDP, 2004). In affluent countries, virtually everyone has access to biologically safe piped water and sanitation systems, such as sewer connection, cesspool, or outhouse. In developing countries, many people do not have access to clean water, and for sanitation many do not even have latrines. In Haiti, only 28 percent of the people have any form of improved sanitation, and only 46 percent have access to an improved water source (any form of piped water). In Cambodia, only 17 percent of the population have adequate disposal facilities for human wastes, and only 30 percent have access to clean water. In most developed countries, more than 95 percent of the population have sustainable access to affordable essential medicines, but that is true for less than 50 percent of the population in Brazil, Paraguay, Ecuador, Haiti, Georgia, Tajikistan, India, and many countries in sub-Saharan Africa, Central America, and Southeast Asia (UNDP, 2004). If we look at HIV as we do other infectious diseases, we should expect the HIV epidemic to develop differently in regions with widespread malnutrition and other sanitary and health-service deficiencies known to contribute to disease susceptibility.

The question of how, and to whom, HIV has spread in different parts of the world is not just an interesting puzzle for epidemiologists. The design of effective prevention strategies hinges on an understanding of the causes of the pandemic and its regional epidemics. Early explanations of the syndrome were chiefly virological, seeing HIV as a virus that barges in and clobbers the host's immune system until the person dies from one or more opportunist infections. Since HIV appeared to kill all those infected, the focus of the first decade of research was on the virus itself. Yet, the standard epidemiological model of disease causation recognizes the importance of host factors in disease transmission and progression. Drawing on that tradition, more recent work has focused on the immune response to HIV—that is, on the interaction of HIV with the host's own immune system (Whitaker, 1997).2 (p.7) Host susceptibility is considered a key to the emergence of the new and resurgent infectious diseases, of which HIV/AIDS is only the most notorious (Morris and Potter, 1997).

The importance of a person's immune response to HIV was first noted in the industrialized countries in the more rapid transition from HIV to AIDS in malnourished persons (Baum and Shor-Posner, 1998; Baum et al., 1997; Semba et al., 1993; Fawzi and Hunter, 1998). Another indicator of the importance of host characteristics (in both mother and child) is the significant difference between rich and poor countries in the rate of transmission of HIV from mother to child (vertical transmission). Before medical intervention was available, 14 percent of infants born to HIV-infected mothers in Europe became infected, and in the United States, 17 to 25 percent were infected without intervention. In Africa, vertical transmission occurred in 25 to 40 percent of births to untreated mothers with HIV or AIDS (Fowler and Rogers, 1996). In rich countries, where antiretroviral prophylaxis and delivery by Caesarean section are available, mother-to-child transmission is now rare. But in most poor populations, the likelihood of transmission from mother to child is still at least 30 percent (UNAIDS, 2004).

Greater attention to host factors also helps to explain the extremely high rates of heterosexual HIV transmission in sub-Saharan Africa and increasingly in other poor regions. Two factors that we find across sub-Saharan Africa and among poor populations in Asia and Latin America that are known to undermine immune system response are malnutrition and parasite infection. Throughout the world, nutrition is the most important determinant of susceptibility to disease because of its impact on both the protective barriers (skin and mucous membranes) and on immune response at the cellular level (Morris and Potter, 1997). Parasite infection is also widespread in poor populations and produces chronic immune response to the foreign bodies (the parasites), leading to immune system exhaustion (Bentwich et al., 1995). (The roles of both nutrition and parasite infection are explained in chapters 2 and 3.)

Malnutrition and parasite infection contribute to greater susceptibility to any infectious disease, including those transmitted sexually. The conditions of poverty increase HIV susceptibility, not only to opportunist diseases after HIV infection but also to HIV transmission itself, just as they increase susceptibility to other infectious diseases. There are also certain diseases prevalent among poor populations in Africa, Asia, and Latin America, but rare in the rest of the world, that sharply increase the probability of transmission of HIV in particular (see chapter 3).

Ecology of disease transmission

The enormity and speed of the AIDS pandemic are menacing. It is understandable that observers initially tended to lose perspective, unable to ex (p.8) plain how such a pandemic could come about, even though the conventional model of infectious disease is quite adequate for interpreting HIV spread. Some 20 years into the pandemic, it is harder to understand why the failure of global AIDS policy to halt the spread has led to so little reflection on its underlying causes. What is lacking in HIV/AIDS policy is both context and historical perspective.

The living and working conditions of some 2 to 3 billion people in the world promote epidemic and endemic diseases, both infectious and parasitic. People live on garbage dumps and survive by scavenging. Others take their drinking water from streams, rivers, and lakes contaminated by household and industrial waste. Residential crowding contributes to transmission of many diseases, including tuberculosis and meningitis. More than half of early childhood deaths around the world are attributable to malnutrition and its synergies with infectious disease. Malnutrition in young girls and malnutrition in pregnancy are important contributing causes of maternal mortality. In at least 20 African countries and in numerous countries in Southeast, South, and Southwest Asia, as well as in Latin America, a woman still has a much greater chance of dying in childbirth than from AIDS (derived from WHO, 2000b, and UNAIDS, 2004). And those deaths occur not just in the hinterland of the poorest countries but also in the urban slums and close-by suburbs of Buenos Aires, Rio de Janeiro, and other elegant cities of the middle-income countries. Mortality tells only part of the story of human suffering and wasted potential. Nutritional deficiencies, aggravated by parasite infection, also lead to physical and mental disability and thereby perpetuate a cycle of unproductive labor, poverty, and impairment (Stillwaggon, 1998).

We already know a great deal about why epidemics spread. Pathogens abound, but they do not always cause disease in a particular individual. Nor does the presence of disease in some members of a population always cause an epidemic. Random introductions of pathogens into human populations occur continually, but they rarely lead to epidemics or pandemics. Propitious conditions are necessary for a microbe to make a person sick or for the disease to spread throughout a population.

Throughout history it has been clear that the epidemic spread of a disease requires favorable conditions. Rats (or soldiers) aboard a ship from an eastern port carried plague-infected fleas into Italy in 1348 and sparked the epidemic spread of plague in Europe, wiping out one-third of the population in most of the continent. This introduction was a random event, but it was certainly not Western Europe's only exposure to rats or plague. In 1348, plague entered a continent weakened by 30 years of falling per capita food consumption and increasing immiseration of the peasantry due to increased feudal demands. The population of Europe had already been falling in the decades leading up to 1348, and a series of disastrous harvests exacerbated the effects of war (Pounds, 1994). Even though many nobles and townspeople perished in the Black Death, the ecologic context for the (p.9) epidemic was the worsening economic situation of the peasantry: “The generation born in this age of crisis was so debilitated by hunger, disease, exploitation, war and disorder that a few years later it succumbed to a still greater catastrophe, the worst in world history” (Fischer, 1996, 41). That depiction of Europe in the early 1300s could describe parts of sub-Saharan Africa, Asia, and the Americas from the 1970s to today.

Plague is not merely a historical curiosity. It is endemic among rodents in the southwestern part of the United States, and there are about 10 to 15 human infections every year, with a case fatality rate of about 14 percent (CDC, 2003). Yet that low but steady transmission has not provoked a plague epidemic in the United States. When an epidemic of plague did break out in India in 1996, the presence of eight rats for every person in that country (that is, 8 billion rats) was perhaps a contributing factor, but it was the filthy conditions of the slums of the port city of Surat that allowed the disease to spread (Badhwar, 1994).

One way that pathogens are introduced into previously unexposed human populations is through zoonoses, animal viruses that become human diseases through mutation. Influenza is a zoonosis, and every year a new strain of influenza emerges from the pork and fowl ecology of southern China. The flu virus that raged through the world in 1918 might have been a particularly virulent strain, but it also found propitious terrain in populations ravaged by World War I, when more people died of hunger and disease, even before the influenza pandemic, than from combat. The increased mobility of populations—of troops and refugees—also facilitated its spread.

For 100 years there was no cholera outbreak in the Americas, although cholera vibrio were introduced into coastal waters throughout the region, from the United States to Argentina, on countless occasions over the twentieth century by trade with Asia. In 1991, however, the discharge of cholera vibrio in the bilge water of a ship in Lima harbor caused an outbreak that reestablished cholera as an endemic disease in the Americas. The years leading up to 1991 had seen rapid growth of the slums of Lima, without the necessary expansion of water lines and sanitation systems. The government did not invest in new infrastructure in the poor settlements, or pueblos jóvenes, and, even if it had been willing, it faced the continuing destruction of power plants, clinics, and other facilities by Sendero Luminoso (the Shining Path) guerrillas.

A relatively high dose of contaminated food or water is necessary to make a person sick with cholera compared to many other pathogens. The return of cholera to the hemisphere followed substantial deterioration in sanitary conditions throughout Latin America during the “lost decade” of the 1980s, a period of economic crisis, decreased government social expenditures, falling incomes, and increasing inequality. From one squalid slum to another, cholera spread north and south from Peru across Latin America, and it remains today an endemic disease throughout the region.

(p.10) HIV is thought to be a zoonosis deriving from simian immunodeficiency viruses (SIVs) that infected humans, perhaps from hunting or butchering injuries that allowed blood contact. (Even if that origin is disputed by some, what remains true is that some virus was introduced into human populations, regardless of where and how that first occurred.) SIVs were introduced no fewer than seven times into human populations, although HIV only became epidemic after 1980. A similar virus might have been introduced on numerous other occasions but did not have epidemic results. The discoverers of the simian viruses observe that their findings “illustrate the classic maxim that the epidemiology of an infectious disease reflects complex interactions between the infectious agent, the host, and the environment” (Hahn et al., 2000, 612). The introduction of SIV was random, but the continuation and spread of HIV are not.

When HIV came on the scene in the early 1980s, it was barely noticed in some countries because of the routine enormity of suffering. Even today, HIV is far from the only threat to poor people. A study of the workplace impact of AIDS in 15 large firms in Tanzania in 1998 reported that 50 percent of employee deaths were AIDS-related (Baruch and Clancy, 2000). It is likely that workers in large firms would have steady employment and relatively higher wages than the average for Tanzanians. Since Tanzania has a relatively mature epidemic (at least 15 years since the first reported infections of AIDS) and high HIV prevalence, one would also expect that AIDS deaths would be a high proportion of all deaths of working-age people. Even allowing for underreporting due to the stigma of AIDS, it is remarkable that only half of employee deaths were from AIDS. Even in that relatively privileged population (for Tanzania), other maladies are killing employees at roughly the same rate as AIDS.

Another workplace study of a plantation in Malawi reported that there were, on average, three deaths per month from AIDS, certainly a terrible toll. But there were six non-AIDS deaths of adult workers per month, and 15 deaths per month of workers' dependents (Morris, 2003). Malawi has a very serious AIDS epidemic (adult prevalence 14 percent at the end of 2003), but only one-third of the workers' deaths were from AIDS. AIDS flourishes where people are dying of other diseases. Nowhere in the affluent countries is the death toll among working-age people and children so high, nor is the toll from AIDS comparable. That is not mere coincidence.

Limitations of global AIDS policy

International AIDS policy seems to exhibit amnesia regarding everything that epidemiologists know about disease transmission, in poor populations in particular, and everything economists and other social scientists know about the environment of poverty in the developing world. Most policy documents written since the Thirteenth International AIDS Conference in (p.11) Durban, South Africa, in 2000 begin with the acknowledgment that HIV disporportionately afflicts poor countries. Rarely is that statement accompanied by an adequate analysis of how poverty contributes to the spread of AIDS, and then only to suggest that poverty can provide the impetus to risky behaviors, which is certainly true, but it is only part of the story. Even when the economic and social context of AIDS is noted, that is not matched by policies that address the environment of poverty in which HIV is spreading.

Why did people formulating AIDS policy forget almost everything they knew about the requirements of disease transmission? Sex is part of it. People get distracted by sex, and because HIV is sexually transmitted many people ignore the fact that even sexually transmitted diseases require the combination of pathogen, host, and environmental factors. Syphilis provoked the same stigmatizing and unscientific reaction in earlier decades. Moreover, HIV has had its most devastating spread thus far in Africa, a part of the world about which most Europeans and North Americans have little correct information and plenty of erroneous notions, and this fact has had a crucial role in the misunderstanding of the AIDS epidemics. Economists, other social scientists, and policy makers from international organizations, all of whom should understand the complexity of African poverty, made little use of their knowledge of poor populations and relied on cultural explanations of the African epidemic. Deep-seated Western stereotypes of African sexuality hijacked the AIDS-in-Africa discourse, and, consequently, a behavioral bias continues to limit AIDS analysis and policy. HIV seemed to confirm Western notions of African hypersexuality and derailed investigation of other causes of the epidemic spread of HIV.

The narrow individual focus of what I call the behavioral paradigm was exacerbated by the methodologies employed in contemporary epidemiology and economics, as well as the current trend in public health to favor individual, curative interventions over “upstream” prevention. Consequently, prevention policy is restricted, almost exclusively, to interventions aimed at influencing individual sexual or needle-sharing behavior, rather than addressing the social and biological context in which those behaviors occur.

The limitations imposed on AIDS policy by the African stereotype and by the methodologies employed in the social and health sciences were only part of the reason the HIV/AIDS epidemics have been misunderstood. The grave impact of the epidemic has made policy makers reluctant to consider more remote causes, and the immediacy of the epidemic's human toll seems to have a paralyzing effect on devising creative solutions. The impact of the epidemic will continue to be devastating for decades to come. In several countries it is predicted that 25 percent or more of the adult population will die 30 to 40 years before normal life expectancy. Disease causes disruption of social and economic networks, loss of income for families, and loss of output for the economy. Children are taken out of school to care for sick parents and to earn incomes. Orphans generally do not continue (p.12) in school after the death of a mother, and so investment in human capital for the country as a whole shrinks. Agricultural production is falling in severely affected countries, and in many cases surviving family members switch to production of less labor-intensive, but also less nutritious, crops. Widows and orphans lose rights to land and sometimes are compelled to enter commercial sex work to survive, compounding and accelerating the epidemic's effects. (For a thorough recent treatment of the impact of AIDS, see Barnett and Whiteside, 2002.) Policy makers must shift out of panic mode and create long-term policies for this now-permanent crisis.

A few economic analyses of the impact of AIDS have argued, correctly, that forecasts comparing a future “with AIDS” and “without AIDS” are no longer relevant (Loewenson and Whiteside, 2002). It is not possible to talk meaningfully about the future of sub-Saharan Africa, or any other region seriously affected by AIDS, without considering the impact of the epidemic. One paper suggested the useful concept of endogenizing AIDS: that is, recognizing AIDS as an endogenous variable in growth models for sub-Saharan Africa. Any predictions about the region would have to incorporate the AIDS variable (McPherson et al., 2000). That recognition is important, but it is only part of the answer.

Along with AIDS, numerous other afflictions, including malaria, tuberculosis, worms, and malnutrition, are, as they always have been, endogenous variables for growth projections for Africa and other very poor regions. Those endemic problems and myriad others provide an important, but neglected, part of the reason why people in Africa remain poor and why rates of HIV in the region are many times higher than in more prosperous regions.

Not only is AIDS an endogenous variable in future growth, but the AIDS epidemic was itself endogenously determined. The conditions of poverty, malnutrition, parasite infection, war, economic disruption, and rapid urbanization were essential elements of the epidemic spread of HIV. Consequently, stopping AIDS requires attacking the underlying causes. Models that fail to recognize the interactions among those endogenous variables that produced the AIDS epidemic cannot fully incorporate AIDS into future projections. Without discerning the complex causes of the AIDS epidemic, they conclude that “[u]nder present circumstances, African countries will remain in a downward spiral so long as behavior patterns do not change” (McPherson et al., 2000, 25). Behavior change is important, but the downward spiral can only be stopped if there is substantial change in the health profile of sub-Saharan Africa and other poor regions.

The answer cannot be to treat poverty eradication as an expensive diversion of funds from AIDS prevention, as some have suggested, but to strengthen poverty reduction as a necessary component of health promotion. The enormity and urgency of AIDS have led to pressure to concentrate all health spending on HIV prevention, instead of expanding comprehensive (p.13) primary health care and health education. Mainstreaming AIDS is interpreted to mean that AIDS-specific programs take a bite out of every other budget, such as having the transport ministry paint AIDS ribbons on rail cars or having agricultural extension agents lecture farmers about sexual behavior. It is far more effective to find ways for each sector—whether it be trucking, commerce, agriculture, health care, government, or mining—to combat the conditions that produced the epidemic, whether they be biological, social, behavioral, economic, or environmental.

Recognizing underlying causes is especially important now that resources are finally becoming available for AIDS programs. The momentum that has been building since 2000 to allocate more funds to AIDS prevention and treatment has run up against a new obstacle, the supposed lack of absorptive capacity for the amount of money that is now offered. There is some truth in the statement: it would be a waste to send $3 billion worth of antiretroviral medicines if they spoil on African wharves for lack of transport, or to deliver them to Latin American governments only to find that corrupt administrators appropriate $5 billion of illegal profits, with very little improvement in the health of the population. Certainly, putting all the money into treatment or narrowly conceived condom and peer-education programs is a bad idea. But there is plenty of scope for using US$3 billion, or even US$15 billion, right now in ways that will help prevent HIV transmission immediately and for the long run.

AIDS policy currently ignores what we know about epidemics and why they spread. It lacks historical perspective. It disregards the social context of disease, except to the extent that the social context is used to explain individuals' sexual or drug-using behavior. Global AIDS policy is based on a model of HIV transmission that assumes “one risk fits all” in regions of vastly different economic and ecological conditions. It attempts to address those very varied risks with an extraordinarily limited set of solutions that hardly go beyond the provision of condoms, peer education, needle exchange, and voluntary counseling and testing. In 2003 WHO made an urgent call for a minimum of 24 billion condoms a year, particularly for Southeast Asia, to solve the AIDS crisis (Kaisernetwork, 2003b). But spending billions of dollars leaving Southeast Asia ankle deep in used (and unused) condoms would change little in the region's health profile or the economic, social, and ecological setting that promotes epidemic disease.

Global AIDS policy is missing the big picture. The pieces of the puzzle are all around, and numerous scientists, social scientists, and other observers in the past two decades have made some of the arguments in this book. In spite of the evidence, however, AIDS policy remains narrow, short-sighted, and even counterproductive. Documents of policy organizations report the failure of, and even resistance to, AIDS-prevention strategies, and yet policy makers respond by spending more money on the same failed strategies. They fail to see that the epidemic spread of HIV derives from the same context (p.14) as the propagation of so many other afflictions. AIDS policy has asked very little about that context and so has generated strategies that cannot solve the problem of poor health, of which HIV is only a part.

Structure of the book

This book explains the rapid spread of HIV in the context of poverty, malnutrition, and parasitic and infectious disease. It describes the health conditions that derive from the social and economic context in three regions of developing and transition economies. It critiques some of the methodological limitations in the social and health sciences that encumber research on HIV and AIDS in developing countries, and it criticizes the policies that derive from those limited and limiting methodologies. Finally, it offers pragmatic solutions to social, economic, and biological factors that promote disease transmission, including the spread of HIV.

The appendix to this chapter presents survey data on sexual behavior in many countries. AIDS discourse is colored by the explicit or implicit premise that differences in national prevalence of HIV reflect differences in sexual behavior between populations. Assertions to that effect rely on anecdotal evidence, or no evidence at all (see chapter 7). The data presented in the appendix demonstrate that there is no correlation between higher rates of HIV and countries with higher rates of early initiation of sex, premarital sex, or multipartnered sex.

Chapter 2 brings together the results of numerous scientific studies on the biology of immune function and disease. It presents the biomedical evidence that malnutrition and parasite infection contribute to greater susceptibility to any infectious disease, including those which are transmitted sexually. The conditions of poverty increase HIV susceptibility, not only to opportunist diseases after HIV infection but also to HIV transmission, just as they increase susceptibility to other infectious diseases. Chapter 3 explains the biological synergies of malnutrition, parasitic and infectious diseases, and immune response that are specific to HIV transmission and that are widespread among poor populations in Africa, Asia, and Latin America, and increasingly in the former socialist countries.

Chapters 4, 5, and 6 describe the economic conditions in three world regions that provide the context in which HIV is spreading most rapidly. Chapter 4 examines the biomedical effects of economic conditions in Africa that contribute to high rates of HIV transmission. Economic and health variables distinguish sub-Saharan Africa from affluent countries with lower rates of HIV. This economic/biomedical conclusion implies a broad policy response for confronting HIV/AIDS in Africa, Asia, and Latin America.

Chapter 5 applies the lessons of HIV in sub-Saharan Africa to Latin America and the Caribbean. Economic and biological factors are important determinants of HIV transmission there, as they are in Africa. Statistical (p.15) analysis shows high correlation between HIV prevalence and GDP per capita, international migration, urbanization, and calorie supply. The results reflect the dualism of the Latin American and Caribbean economies and their dual HIV epidemics and suggest the reasons why HIV is spreading fastest in lower-income groups.

Chapter 6 examines the health profile of the countries of Central and Eastern Europe and the former Soviet Union as the context for an emerging HIV epidemic. The HIV epidemics in the transition economies of Europe are apparently fueled by needle-sharing and prostitution, unlike the epidemics in most of the developing world. The emphasis on the behavioral conditions, however, masks the important role played by falling living standards and the collapse of public health services. This chapter broadens the analysis of declining health in the former socialist economies and situates the study of HIV epidemics there within a biomedical analysis of disease vulnerability, rather than the reigning behavioral model.

The conventional model of epidemiological inquiry should have spotlighted the conditions detailed in chapters 4 through 6, informed by the recognized synergies discussed in chapters 2 and 3. Since global AIDS policy does not reflect most of what is known about the regions of the world in which that policy is promoted, it is necessary to investigate what prevented social scientists and health scientists from using the information at hand. Chapter 7 examines Western preconceptions regarding African sexuality that distorted early research on the social context of AIDS in Africa and still limit the scope of preventive policies. Key works cited repeatedly in the social science and policy literature constructed a hypersexualized pan-African culture as the main reason for the high prevalence of HIV in sub-Saharan Africa. Africans were portrayed as the social “Other” in works marked by sweeping generalizations and innuendo, rather than useful comparative data on sexual behavior. Chapter 7 offers a critical analysis of the discourse on African “exceptionalism” and the historical and philosophical origins of treating Africa as a special case.

Chapter 8 discusses the methodologies employed by health scientists and social scientists that prevent them from seeing the interactions that produce epidemic spread of HIV, from measuring those interactions, and from using that information in prevention policy. It examines the methodological limitations in the fields of public health, epidemiology, and economics that are impediments to understanding health in the developing world and specifically obstruct prevention of HIV/AIDS.

Chapter 9 examines the impact of misguided development policy and inadequate epidemiology on the choice of interventions for HIV and other health problems in poor countries. Since 2000, there is increasing recognition that HIV/AIDS is a development issue, but policy prescriptions still do not reflect a development agenda. HIV/AIDS prevention plans are narrowly focused on provision of condoms, behavior modification, and treating cofactor STDs (sexually transmitted diseases). From a development per (p.16) spective, those interventions come very late in the process. Chapter 10 examines workplace HIV/AIDS programs as an example of the limitations of global AIDS policy.

Chapter 11 uses the economic concept of externalities to evaluate the positive spillovers that exist in health interventions. It argues for mainstreaming AIDS prevention by addressing a broad array of development problems, rather than by employing a just-in-time approach to HIV intervention. The lesson from the biomedical literature on disease interactions is that the best investments in health require us to look beyond the immediate problem. There are serious conceptual and practical difficulties with the use of interventions that are narrowly focused on HIV and other STDs, including the divorce of sexuality from other aspects of a healthy lifestyle and the waste of health services infrastructure, squandering opportunities for economies of scale and scope. Fundamentally, HIV-only programs do nothing to change the environment that produces unhealthy people, unhealthy behaviors, and high transmission rates.

What this book is not

This book is not part of the debate about whether HIV causes AIDS. This work accepts that HIV is a necessary causal factor in AIDS. Of course, it is important to continue to evaluate anomalies in the HIV-AIDS link because that work may yield important information about HIV and AIDS and about other diseases and conditions. It is not useful, however—and, in fact, is irresponsible—to assert that there are sufficient anomalies in the data to deprioritize safe sex. I accept the mainstream explanation for HIV and AIDS because it is plausible and it has been substantiated adequately to pursue research and policy on the basis of causality.

Regrettably, this book does not address the epidemic in the United States or other affluent countries. That is the result of time constraints and my particular expertise. Many of the arguments of this work also apply to poor and marginalized populations in the developed countries. There is also not very much in this book about the Asian epidemics, although they are very important. I preferred to concentrate my analysis on the parts of the world in which I have lived and worked over the past 30 years—sub-Saharan Africa, Latin America and the Caribbean, and Eastern Europe and the former Soviet Union—which are also the parts of the world with the highest prevalence and the fastest reported spread of HIV. The lessons of this book, however, apply to the Asian epidemics as well, and at least to some extent to marginalized populations in rich countries.

This work also deliberately avoids ideology as much as possible. I know what the sides are in each of the methodological and policy disputes discussed here, and I do not think it furthers the cause of health to encamp with one side or the other in the pages of this book. Dogma only narrows (p.17) the range of options for solving concrete problems. Even in the current political and economic context, there are pragmatic, affordable, and feasible solutions that will immediately improve the lives of poor people and stem the spread of HIV and other plagues of poverty.

This book also does not take a side in the debate over what has been labeled “public health nihilism” (Fairchild and Oppenheimer, 1998): the argument between changing only the socioeconomic context or only the curative inputs. Dichotomies such as prevention versus cure do not promote good health. Clearly, this book supports a comprehensive approach to public health. But I would not squander all the wonderful, life-saving innovations that we in this era are privileged to enjoy and wait for economic change to bring about good health for poor people.

This book does not dichotomize the debate on treatment versus prevention. I heartily support both treatment and prevention (including the preventive interventions that I repeatedly refer to as end-game, last-minute, and even paternalistic). We should use every means we have to save lives. But the means that have been tried thus far are not enough. The prevention of disease and other forms of oppression needs to be moved upstream as well as being continued in the behavioral forms that it now takes. Although I repeatedly emphasize that global AIDS policy depends too heavily on condom provision and behavior-change communication, I do not propose abandoning those tactics. They must be part of the wider strategy to bring the enjoyment of good health to more people.

In this book I cannot say all that must be said about gender roles. From birth to death, those roles impinge on every aspect of the lives of women and men. That is as true in the affluent countries as it is in Africa, Asia, and Latin America. Radical change in the status of women and girls worldwide is necessary.

Finally, this book is not pessimistic. There are myriad known solutions to the problems of poverty and poor health. Many of them are simple and low-cost. Some would even raise incomes and profits. And all of them provide positive spillovers in economic and human development. The means are at hand for preventing the spread of HIV/AIDS and for improving lives. The only problem is convincing people that those two objectives are one and the same.

Appendix: sex, everywhere

Since HIV can be transmitted sexually, and most HIV is transmitted sexually in poor countries (and increasingly in transition countries), it is reasonable to include sexual behavior as one of the determinants of HIV prevalence. Nevertheless, differences in sexual behavior between different populations do not adequately explain differences in HIV prevalence.

In much of the popular and professional discourse, there is a presump (p.18) tion—generally unstated—that populations with high rates of HIV must have higher rates of early initiation of sex, multipartnered sex, and commercial sex than countries with low rates of HIV. Below are a few examples of assertions published by a multilateral organization and two academic publishers that sexual behavior is the key determinant of rates of HIV in different populations. These overt statements, together with less-explicit versions, are examples of the behavioral paradigm. It is a paradigm, in Kuhn's (1970) sense, because it determines what questions can be asked and what solutions can be sought regarding HIV/AIDS.

HIV transmission has been much faster in societies where there are high levels of unprotected sexual interaction with prostitutes, who have very many sexual partners, than in societies where the majority of people have few or moderate numbers of partners in their lifetime. (Ford, 1994, 88) [Note: this statement was not followed by supporting evidence.]

In contrast to Americans, who usually view sex morally and think that people who have multiple partners (even if unmarried) are immoral and unfaithful, most Africans do not judge sexual behavior in such terms at all. They experience little guilt about sex, and they enter into sex more casually and have more sexual partners than Westerners do. (Rushing, 1995, 62) [Note: there are no citations for these statements in the original.]

“Promiscuity, and the Primacy of Cultural Factors: A Lethal Mixture in Africa” was the title of the largest of several text boxes in the AIDS Update 1999 published by United Nations Population Fund (UNFPA, 1999, 6).

Chapter 7 addresses, at length, important and representative works in the AIDS discourse that unequivocally posit a behavioral explanation for differences in HIV prevalence or imply a behavioral explanation by presenting non-Western societies as exotic and different from the West, even in matters quite mundane. Chapters 9 and 10 discuss the policies that originate in an acceptance of the behavioral paradigm.

Since the “lots of sex” assumption plays consciously or unconsciously in discussion of HIV/AIDS and in AIDS policy, it interferes with seeing the epidemic in any other way. I present here comparative data on sexual behavior in various countries that demonstrate that there is no empirical basis for a presumption that high rates of sexual activity correlate with high HIV prevalence. Once we recognize that differences in sexual behavior cannot explain differences in HIV rates between countries, we can proceed with the subject of this book: what is causing the high and rising prevalence of HIV in developing and transition-economy countries, and what can be done to prevent further spread of HIV/AIDS and other crippling diseases.

The data on sexual initiation, premarital sex, multipartnering, and com (p.19) mercial sex tell us very little about the distribution of HIV around the world. What follows are the findings of a number of studies on sexual behavior from UNAIDS, WHO, UNDP, and scholarly journals. In some cases I have added comments to situate the findings within the HIV discourse.


UNAIDS. 1999a. Fact Sheet on Differences in HIV Spread in African Cities. Geneva: UNAIDS.

This study surveyed sexual behavior in four African cities in 1997 and 1998 and found significant variation in sexual behavior and in HIV prevalence, but it found no correlation between the two. Adult HIV prevalence in the four cities was: Cotonou (Benin), 3 percent in both men and women; Yaoundé (Cameroon), 4 percent in men and 3 percent in women; Kisumu (Kenya), 20 percent in men and 30 percent in women; Ndola (Zambia), 23 percent in men and 32 percent in women. Among the findings were: the highest rate of partner change was found in Yaoundé, where men reported more than twice as many lifetime partners as in the other sites, and women reported slightly more. Fewer married men reported extramarital relationships in the high-HIV sites. Very few married women (1–3 percent) reported extramarital relationships, except in Yaoundé (12 percent).

The researchers found that in the high-prevalence sites the age at which women became sexually active was slightly lower, there was a higher prevalence of herpes-2, and there was a much lower rate of male circumcision. Rates of partner change, concurrent sexual relationships, and contacts with sex workers were not correlated with rates of HIV. None of the factors taken into consideration could account for the high prevalence among teenage girls, soon after first sex and with few exposures. They state that, “In fact, behavioural differences seem to be outweighed by differences in HIV transmission probability,” although it is unclear if they would consider factors beyond those they mentioned as contributing to that difference in probability.

Comment: This study provided a clear empirical challenge to the behavioral assumption. Unfortunately, UNAIDS has not followed up on unanswered questions, in particular why young women in Kisumu have such high rates of HIV (see chapters 3 and 4 of this book for further discussion).

Africa, Asia, South America

John Cleland and Benoît Ferry (eds.). 1995. Sexual Behaviour and AIDS in the Developing World. London: Taylor and Francis for the World Health Organization.

(p.20) This book reports on 16 surveys of heterosexual behavior in 15 countries (nine in Africa; four in Asia; Brazil and Mauritius) conducted from 1989 to 1991 under the auspices of the WHO Global Programme on AIDS (WHO/GPA). Almost all of the surveys were national, and an attempt was made to standardize the methodology because the stated intention of the surveys was to collect empirical data on behavior to understand the AIDS epidemic. The authors lamented: “The dearth of empirical studies that directly address sexual behaviours reflects a lack of interest from anthropologists. This is evident from reviews of the literature on human sexuality in sub-Saharan Africa (Standing and Kisseka, 1989; Larson, 1989; Caldwell, Caldwell, and Quiggin, 1989; Caldwell, Orubuloye, and Caldwell, 1992) and in Asia (Sittitrai and Barry, 1990) where most of the evidence about sexuality is anecdotal” (Caraël, 1995, 76).

The Cleland et al. summary of the WHO/GPA surveys contains useful data on premarital sex and multipartnering that the authors themselves found surprising: “Unexpectedly large proportions of men and women reported no sex during the last month, more so in Africa than elsewhere” (Caraël, 1995, 104). The surveys also contained questions intended to evaluate people's sense of subjective efficacy, or the belief that one's own actions can influence a situation. They found that in all the African surveys a very high proportion believed that they could change their behavior and their life situation, a finding which they felt contradicted the assertions that superstition and magical beliefs about transmission were more common in Africa than elsewhere.

Finally, as stated in the “Summary and Conclusions,” written by Cleland, Ferry, and Caraël: “To return to the key point, the majority of men in all sites reported no sexual associates outside of regular partnerships in the year preceding the survey. … WHO/GPA surveys are thus immensely valuable in correcting wildly inaccurate perceptions of sexual behaviour, that were based on guesswork or small unrepresentative studies” (Cleland et al., 1995, 211).

Africa, Latin America and the Caribbean, Asia

Ann K. Blanc and Ann A. Way. 1998. “Sexual Behavior and Contraceptive Knowledge and Use among Adolescents in Developing Countries,” Studies in Family Planning 29(2):106–116.

This article reports data from the Demographic and Health Surveys, which are large-sample, nationally representative surveys of population, health, and nutrition in developing countries, sponsored by USAID and carried out by Macro International. Among the findings reported in this article: from the late 1960s to the early 1990s, the percentage of women who had sex by age 18 fell in 12 out of 17 African countries surveyed (in East, West, and Southern Africa) and increased by only 0.1 percentage point (p.21)

Table 1.1 Percentage of 15- to 19-year-olds, never married, male and female, who have had intercourse (selected countries with year of survey and adult HIV prevalence)


Male %

Female %

Adult HIV prevalence %

Great Britain (1991)




Brazil (1996)




United States (1995)




Dominican Republic (1996)




Haiti (1994–1995)




Tanzania (1996)




Mali (1995–1996)




Ghana (1993)




Zimbabwe (1994)




Source: Data for the first and second columns are provided by S. Singh. The third column data are from UNAIDS, 2004.

in another. It also decreased in the two Asian countries for which data were available, and in four of the eight Latin American and Caribbean countries surveyed.

Comment: The hypothesis that a sexual revolution (at least as regards earlier initiation of sexual activity) accompanied urbanization and modernization in the developing world is not supported by these data.

Africa, Caribbean, United States, Great Britain

Susheela Singh, Deirdre Wulf, Renee Samara, and Yvette P. Cuca. 2000. “Gender Differences in the Timing of First Intercourse: Data from 14 Countries,” International Family Planning Perspectives 26(1):21–28, 43.

Tables 1.1 and 1.2 are derived from this article and from data provided to me by S. Singh. HIV prevalence data are from UNAIDS (2004). The point is not that sexual behavior is unimportant; it is that behavior does not explain differences in rates of HIV between different populations. Early initiation of sex is also frequently mentioned as one of the factors contributing to high rates of HIV in developing countries. But the United States has the highest rate of early initiation as seen in Table 1.2. In Mali, early marriage is common for women and is the reason for the high percentage of sexually experienced 17-year-old women there.


Table 1.2 Percentage of 20- to 24-year-olds sexually active before age 17, and median age at first intercourse (selected countries, sorted by male sexual activity percentage)

Sexually active by age 17 (%)

Median age at first intercourse






United States (1995)





Brazil (1996)





Dominican Republic (1996)





Great Britain (1991)





Haiti (1994–1995)





Tanzania (1996)





Ghana (1993)





Zimbabwe (1994)





Mali (1995–1996)





Source: Derived from S. Singh et al., 2000.

Latin America


Nancy J. Murray, Lauri S. Zabin, Virginia Toledo-Dreves, and Ximena Luengo-Charath. 1998. “Gender Differences in Factors Influencing First Intercourse among Urban Students in Chile,” International Family Planning Perspectives 24(3):139–144, 152.

This study surveyed 4,248 urban Chilean students, aged 11 to 19 years. “Results: Overall, 21% of the young women and 36% of the young men had ever had sex, with the median ages of first intercourse being 15 years and 14 years, respectively” (Murray et al., 1998, 139).


Mahler, K. 1997. “Increased Risk of STD Infection among Peruvian Women Linked to Their Partners' Sexual Practices,” International Family Planning Perspectives 23(1):39–40.

In Peru, in a random sample of preemployment health screenings, 59 percent of men reported sex with a sex worker, and only 30 percent reported always using a condom with sex workers; 13 percent of the men reported having had urethritis, 19 percent reported genital ulcers, and 3 percent had had syphilis. The mean age of the men was 25, mean age for first sex for men was 16 years, and 21 percent had their first sexual experience with a sex worker. Those young men reported a lifetime mean of 11 partners, and 56 percent had multiple partners in the previous year. Some 60 percent of men with a regular partner also reported sex with a casual partner; 27 (p.23) percent of men with a regular partner reported having sex with a prostitute in the year prior to the interview. None of the figures on women's behavior came anywhere close to the men's, and yet 33 percent of women and 12 percent of men had one or more STDs.

Dominican Republic

Abel, David. 1999. “Aids Linked to Infidelity in Dominican Republic,” Boston Globe, December 28, pp. A2, A4.

The article states that Dominican Health Ministry studies report more than 50 percent of men have extramarital affairs, and 44 percent of boys and 36 percent of girls have had sexual intercourse by age 14.

United States

Susheela Singh and Jacqueline E. Darroch. 1999. “Trends in Sexual Activity among Adolescent American Women: 1982–1995,” Family Planning Perspectives 31(5):212–219.

In surveys from 1982, 1988, and 1995 of U.S. women, aged 15 to 19: “In each of the surveys, about 40% of all 15–19-year-olds [female] had had sexual intercourse in the last [previous] three months” (Singh and Darroch, 1999, 212).

Olga A. Grinstead, Bonnie Faigeles, Diane Binson, and Rani Eversley. 1993. “Sexual Risk for Human Immunodeficiency Virus Infection among Women in High-Risk Cities,” Family Planning Perspectives 25(6):252–256, 277.

This article contains data from 3,482 women in 23 cities in the United States who were 18 to 49 years old. The study showed that 15 percent of women engaged in sexual behavior that put them at risk of acquiring HIV, including “having multiple sexual partners, having a risky main sexual partner or having both multiple partners and a risky main partner. An additional 17% of women with no other risk factor report that they do not know their main partner's HIV risk status. … In general, women with a risky main partner are the least likely to use condoms consistently” (Grinstead et al., 1993, 252).

John O. G. Billy, Koray Tanfer, William R. Grady, and Daniel H. Klepinger. 1993. “The Sexual Behavior of Men in the United States,” Family Planning Perspectives 25(2):52–60.

This article reports on a 1991 nationally representative survey of U.S. men aged 20–39, of whom 95 percent had had vaginal intercourse. Of those, 23 percent reported 20 or more female partners. “About one-fifth of never-married and formerly married men had had four or more partners over a recent 18-month period” (Billy et al., 1993, 52).


June M. Reinisch, Stephanie A. Sanders, Craig A. Hill, and Mary Ziemba-Davis. 1992. “High-Risk Sexual Behavior among Heterosexual Undergraduates at a Midwestern University,” Family Planning Perspectives 24(3):116–121, 145.

This article reports on a 1998 survey of undergraduate students (generally the age group 18–22) at a large Midwestern state university in the United States. About 80 percent reported having ever had intercourse. Mean age at first intercourse was 17. Average number of sexual partners for women was six, and men reported an average of 11 partners.

Tom W. Smith. 1991. “Adult Sexual Behavior in 1989: Number of Partners, Frequency of Intercourse and Risk of AIDS,” Family Planning Perspectives 23(3):102–107.

This article reports on a 1988/1989 survey of U.S. adults. The average reported number of partners since age 18 was seven, with men claiming many more partners than women.

Kathryn Kost and Jacqueline Darroch Forrest. 1992. “American Women's Sexual Behavior and Exposure to Risk of Sexually Transmitted Diseases,” Family Planning Perspectives 24(6):244–254.

This article reports on surveys in 1988 and 1989 of U.S. women aged 15 to 44. The percentages of women and their reported number of partners are as follows:

  • 67% of sexually experienced women had more than one partner

  • 41% had four or more partners

  • 23% had six or more partners

  • 8% had more than 10 partners

  • 71% had one or more nonmarital partners

They also reported that 27 to 39 percent of women aged 18–44 who were sexually active had had contact with more than one sexual partner, directly or indirectly, during the previous year.


N. MacDonald, G. Wells, W. Fisher, et al. 1990. “High-Risk STD/HIV Behavior among College Students,” Journal of the American Medical Association 263(23):3155–3159.

This article reports on a 1988 survey of 5,514 first-year students at Canadian community colleges and universities. Of the men, 74 percent were sexually active, and of the women, 69 percent. “Among the 21.3% of the men and 8.6% of the women with 10 or more partners, regular condom use was reported in only 21% and 7.5%, respectively” (3155). “40% of the (p.25) men and 25.2% of the women reported at least 5 different partners” (3156). “Knowledge per se [of safe sex methods] was not typically translated into safer behavior” (3158).



M. Breslin. 1998. “Abortion Rate among Young Romanians Declines: Those Not in Union Report Rise in Contraceptive Use,” International Family Planning Perspectives 24(3):150–152.

This article reports on a survey of Romanian 15- to 24-year-olds. Some 20 percent of women and 41 percent of men in the 15–19 age group had had sexual intercourse. Of sexually active men aged 15–24, 73 percent had initiated sexual activity by age 17. Of men who were sexually experienced, aged 15–24, 99 percent had had sex before marriage. Of sexually experienced men, 60 percent had had four or more partners, and 13 percent reported only a single partner.

France, Great Britain

R. Turner. 1993. “Landmark French and British Studies Examine Sexual Behavior, Including Multiple Partners, Homosexuality,” Family Planning Perspectives 25(2):91–92.

This article reports on surveys in France (1991–1992) and Great Britain (1990–1991). In France, the average number of sexual partners reported by men was 11, and by women, three. In urban areas, 18 percent of men and 10 percent of women had two or more sexual partners in the previous year. Of those, the percentage whose multiple partnerships were concurrent increased with age group. In Britain, the proportion in each age group, by gender, reporting 10 or more lifetime partners was as shown in Table 1.3.

Table 1.3 Britons reporting 10 or more lifetime sexual partners (by age group and gender)


Male (%)

Female (%)













Source: Derived from Turner, 1993.

(p.26) Poland

Zbigniew Izdebski. 2002. Selected Aspects of Evaluation of the National HIV/AIDS Prevention Program. Warsaw: National AIDS Center, UNDP.

This report presents the findings of national surveys in Poland in 1997 and 2001. Of all people interviewed, 15–49 years of age, 13 percent of women and 25 percent of men reported having had a sexual partner other than spouse or regular partner, while in the relationship. Of those interviewed who were sexually active, 15 percent of women and 29 percent of men had had another sexual partner while in a relationship.

From 1997 to 2001 the percentage of those who reported having had sex with a person other than their regular partner in the previous 12 months had dropped from 17 percent to 14 percent. Of those having partners other than their regular partner, half had one additional partner, 26 percent had two additional partners, and 28 percent had three or more additional partners. Most irregular contacts were persons known to the other, but 14 percent of irregular sexual contacts in 2001 and 12 percent in 1997 were met for the first time at the time of sexual contact. In 1997, 9 percent of all men reported paying for sexual services, and 12 percent reported using a commercial sex worker in 2001. From 1997 to 2001 the proportion of people going to a clinic or hospital in response to symptoms of STDs dropped from 39 percent to 26 percent, an increase of 3 percent went to a private doctor, and an increase of 10 percent asked advice of a friend.


The survey data presented here demonstrate that everywhere people have sex, everywhere some people have lots of sex, and everywhere most people do not. The numerous articles, books, and monographs asserting or assuming that differences in sexual behavior can explain variations in HIV prevalence rely on anecdotal evidence, not on survey data. Anecdotal evidence (such as in Caldwell and Caldwell, 1987, and Caldwell et al., 1989, discussed in chapter 7) cannot present an accurate assessment of broad patterns of sexual behavior. Only carefully randomized sampling techniques designed to ensure that respondents fairly represent the populations being studied can prevent sampling error. The use of anecdotal evidence in the attempt to understand the wide variation in the prevalence of HIV among different populations is fatally undermined by sampling bias in the study of sexual behavior.

Anecdotal evidence can be an important source of information and can illuminate and illustrate what would otherwise remain sterile and abstract observations. Nevertheless, anecdotal evidence cannot give us any information about the frequency of a behavior or of a condition in a population. The behavioral paradigm asserts or assumes widely different frequencies of (p.27) certain types of sexual behaviors in different populations, and anecdotal evidence is useless in verifying that assertion or assumption. Even though survey data have well-known potential flaws (undersampling of certain subpopulations despite the best efforts of those who construct the sample design, inappropriately constructed interview schedules that generate misleading results, and poorly trained interviewers, for example), it is the only appropriate instrument for generating statements about frequencies in different populations.

The survey data that do exist show that differences in sexual behavior cannot explain differences in HIV rates between countries. We need to look at what are the important differences between populations that affect their vulnerability to disease. Chapters 2 through 6 explore those issues. (p.28)


(1.) AIDS is the acquired immune deficiency syndrome, the umbrella term that indicates illness with one or more of the opportunistic infections associated with infection with HIV, the human immunodeficiency virus. The term HIV/AIDS conflates two distinct phenomena, a virus and the illnesses from which a person might suffer as a result of being infected with the HIV virus.

(2.) The host here is the person becoming infected—that is, serving as host to the infecting pathogen. In other contexts, the term host means the carrier, or the person already infected.