9.14 Acquired immunodeficiency syndrome
Oxford Textbook of Public Health
Acquired immunodeficiency syndrome
Wiput Phoolcharoen and Roger Detels
History of the HIV/AIDS epidemic
Natural history of HIV infection and disease
Strains of virus
Transmission of HIV
Political and economic factors
The global epidemiology of HIV
Monitoring the epidemic
The spectrum of clinical manifestations
Treatment in the developing countries
The impact of HIV/AIDS
The impact on life expectancy and health
The economic impact
The societal impact
The role of stigmatization
Ethical and human rights issues
Intervention and prevention strategies
Policy and structure for HIV/AIDS prevention
Sexual behaviour change and condom use
Preventing vertical transmission
Control of sexually transmitted diseases
Universal precautions and postexposure prophylaxis
Summary and forecast
History of the HIV/AIDS epidemic
Acquired immunodeficiency syndrome (AIDS) has been the most dramatic disease event of the second half of the twentieth century. It has been estimated that, by December 2000, 57.9 million individuals had been infected by the human immunodeficiency virus (HIV), including over 21.8 million who had died from AIDS (UNAIDS 2000). The majority of those infected have been in the most productive years of their life, causing a secondary economic impact beyond the cost of caring for them and creating a new cohort of orphans. The total number of deaths and expected deaths from HIV already exceeds the total killed in all the major wars of the twentieth century (Fig. 1). And the epidemic is still spreading! Its impact is particularly tragic because 95 per cent of HIV infections occur in developing countries that are least prepared to absorb the impact of the health and economic consequences.
Fig. 1 The toll of AIDS (data from 1999). (Reproduced from Global Health Council 2000.)
AIDS was first recognized and reported in 1980 by an alert young physician, Michael Gottlieb, at the University of California Los Angeles School of Medicine, who realized that the unexplained immune deficiency that he observed in three young men represented a new disease (Gottlieb et al. 1981). Gottlieb reported his findings to the United States Centers for Disease Control and Prevention and published them in the Morbidity and Mortality Weekly Report, which resulted in the recognition by other physicians in San Francisco and New York that they were seeing the same unexplained immune deficiency. It soon became apparent that the disease occurred primarily in two groups—men who had sex with men and injecting drug users—but the infection soon spread to haemophiliacs.
The reasons for the sudden appearance of HIV infection and AIDS are not completely understood, but it is clear that both biological and sociological factors were involved (Gao et al. 1999). A report by Beatrice Hahn suggests that the biological parent of HIV-1 was present in chimpanzees. She gathered evidence that the virus spread to humans through exposure to chimps as pets and through butchering of infected animals for food. However, the virus has low infectivity and it is doubtful that it could have spread if there had not been a major change in sexual behavior towards greater acceptance of multiple sexual partners in the second half of the twentieth century. The change may have been due, in part, to the introduction of the contraceptive pill, which reduced the fear of pregnancy resulting from sexual intercourse, a major inhibitor of multiple partners and early sex in many parts of the world.
Although the disease was first recognized in the United States, that country has not been the primary focus of the epidemic. For the most part the early epidemic in the United States and the other developed countries was confined to men who have sex with men and the injecting drug-using population, who represent a relatively small proportion of the total population. In the developing world, however, it soon became apparent that the major mode of spread was through heterosexual intercourse. The countries of sub-Saharan Africa were the initial focus of the epidemic and this region continues to have the highest number of new infections. The epidemic soon spread to other countries as well, including Latin America, and in the late 1980s and early 1990s to South and Southeast Asia, currently the second most heavily involved region of the world. Although the prevalence of HIV is still low in both India and China, their large populations, representing 38 per cent of the world’s population, and the increasing prevalence of sexual mixing and sexually transmitted diseases (STDs) suggest that it is only a matter of time before the epidemic establishes a firm stronghold in both these countries. The estimated number of HIV-infected individuals in India already exceeds the total estimated for all the rest of South and Southeast Asia. Clearly, dramatic steps must be initiated to prevent the further spread of HIV. However, the high proportion of silent infections and the stigma associated both with belonging to the most severely affected risk groups and being infected, combined with the reluctance of most societies to discuss sexual behaviours openly, make the implementation of effective intervention strategies very difficult, but not impossible; Thailand has successfully reduced the spread of HIV (Phoolcharoen et al. 1998).
Natural history of HIV infection and disease
AIDS is caused by HIV, a retrovirus with ribonucleic acid (RNA) as its genetic material. HIV is presented schematically in Fig. 2. Because the natural history of HIV infection is unusual and determines the strategies which can be used for intervention and treatment, it is useful to understand it.
Fig. 2 HIV-1 virus.
HIV has a relatively simple structure that can be divided into two major structural components—the core and the envelope. The core is made up of several proteins of which the major one has a molecular weight of 24 000 kilodaltons (kDa) and is referred to as p24. Within the core are contained the RNA and the enzymes which are collectively known as reverse transcriptase. The envelope is composed of lipid material into which is inserted a glycoprotein with a molecular weight of 41 000 kDa (known as gp41). Loosely attached to the gp41 is a second glycoprotein with a molecular weight of 120 000 kDa (known as gp120) which is the part of the virus that initially attaches itself to the target cell.
The target cell of HIV is primarily the CD4+ T lymphocyte, one of the major cells of the immune system, which is involved not only in the production of cellular immunity but also influences the production of antibodies. HIV can attach to any cell that has the CD4+ receptor. Although these receptors are found primarily on the CD4+ lymphocytes, they are also found on 40 per cent of mononuclear cells, including macrophages, 5 per cent of B cells, probably on mature CD8+ cells, and the counterparts of these cells in the central nervous system.
The natural history of HIV infection of the CD4+ cell is pictured in Fig. 3. The process of HIV replication begins with the attachment of the virus to the cell, which is accomplished primarily by attachment to two cell receptors: the CD4+ receptor and either the CCR5 or the CRCX4 receptor. Early in the natural history of HIV infection, HIV will use the CCR5 receptor, but as the HIV undergoes changes within the infected individual from the so-called non-syncytium-inducing forms, found early in the course of disease, to the syncytium-inducing form in advanced disease, it also utilizes the CRCX4 receptor. There are other receptors that also influence the ability of HIV to attach to the cell, but the biology of these receptors is still under investigation.
Fig. 3 Evolutions: retroviruses.
Once HIV has successfully attached to the cell, the core of the virus is intruded into the cytoplasm of the cell and the viral RNA is transcribed into viral deoxyribonucleic acid (DNA) by the enzymes known collectively as reverse transcriptase. The transcription process, however, is imperfect and, on average, at least one mutation occurs per replication cycle. This is important because the human immune response to the virus is absolutely specific. Thus the immune system will not recognize the mutation and mounts a new immune response to the progeny virus, a process which may take several weeks, during which the HIV continues to infect other cells which produce more mutated virus. The ‘errors’ in this step are a major reason why HIV is able to escape the immune system and persist. Reverse transcription is blocked by the group of drugs known collectively as reverse transcriptase inhibitors.
Once transcription of the genetic material to viral DNA is completed, the viral DNA is integrated into the host cell DNA with the assistance of a viral enzyme known as integrase. After the viral DNA has been incorporated into the host DNA it is indistinguishable from the host DNA and is referred to as the ‘provirus’. Each time the cell divides the viral DNA will also be passed on to the progeny cells. This is known as the ‘latent’ stage of the HIV replication cycle. The virus may remain in the latent stage for years.
The process of viral production begins with the production of two types of RNA: genomic RNA, which will be incorporated into the HIV that are produced by the cell, and messenger RNA, which directs the production of the viral proteins such as p24 and gp120. The viral proteins are produced in long chains that must be broken down into smaller components in order to be incorporated into the progeny HIV. This process is accomplished by the enzyme protease, which is also thought to be involved in the final maturation stage of the progeny HIV immediately after they are released from the cell. A second group of drugs, known collectively as the protease inhibitors, block this stage of the replication process. As the new HIV is extruded from the host cell, lipid from the cell wall is incorporated onto the virus and forms the envelope of the progeny virus. A single CD4+ cell may produce hundreds of new HIV progeny.
It is important to realize that neither group of drugs, reverse transcriptase inhibitors nor protease inhibitors, actually result in the elimination of the virus. The viral DNA that has been incorporated into the host cell DNA will be unaffected and thus the infected individual is not ‘cured’.
Studies by David Ho and George Shaw have suggested that the immune system, rather than being deficient in the initial stage of infection, actually produces as many as 2 × 109 CD4+ cells per day in response to the virus (Ho et al. 1995; Wei et al. 1995). However, the production of more CD4+ cells provides more opportunities for HIV to replicate and as many as 108 to 109 viruses may be produced per day. For a surprisingly long period the host immune system can maintain this huge production of CD4+ cells, but eventually HIV production outpaces the productive power of the immune system. A number of factors, including infection with other viruses and sexually transmitted diseases, can alter this dynamic equilibrium allowing HIV to destroy the immune system.
The natural history of HIV infection in the host, as opposed to infection of the cells as described above, is pictured in Fig. 4. Successful infection of the host is actually a rare event, occurring, for example, in only approximately 1 per 1000 episodes of vaginal intercourse. The type of intercourse and the concurrent presence of other infections, especially sexually transmitted infections, however, can increase the probability of HIV infection. Once the individual is infected there is huge production of HIV, which persists for several weeks until the immune system is mobilized. The major component of the immune system that challenges the continued replication of the virus are the CD8+ cytotoxic cells, which kill HIV-infected cells. The induction of sufficient numbers of cytoxic CD8+ cells, however, can take weeks, in some cases months. About the time that the immune system is responding, many infected individuals will develop an influenza-like illness characterized by fever, chills, malaise, and weakness. The illness usually resolves within about 2 weeks. Once the immune response is in place, production of antibodies to HIV results in a positive response to the standard HIV tests that measure antibodies to HIV, not the presence of the virus itself. This period, during which there are high levels of HIV production, before the antibody test becomes positive, is known as the ‘window period’ and is of great concern, especially for effective screening of the blood supply. The window period is shortening as the sensitivity of the antibody test improves and tests for HIV itself are developed and used. The level of viral production during the initial 6 to 12 months has been found to play a major role in determining the future course of HIV disease in the individual (Mellors et al. 1997).
Fig. 4 Natural history of HIV infection.
Once the immune response has been established and the antibody test has turned positive, the individual remains completely asymptomatic on average for 8 to 9 years although the asymptomatic interval may vary from as little as 1 year to decades. It is this long asymptomatic period that makes recognition and control of the epidemic so difficult. The only way to identify infected individuals during this period is to test them for the presence of the virus or the antibodies to the virus. About 1 year prior to the diagnosis of AIDS the individual begins to have symptoms of HIV infection. These may include weight loss, low-grade fevers, night sweats, and frequent fungal infections. Finally, either an opportunistic infection or malignancy occurs which is diagnostic of AIDS. The exceptions to this are the occurrence of AIDS dementia and wasting syndrome, both of which are diagnostic of AIDS, but are not apparently caused by opportunistic agents. The diagnostic criteria for AIDS, as proposed by the United States Centers for Disease Control and Prevention in 1987 and revised in 1993, are given in Table 1 and Table 2. The designation of less than 200 CD4+ cells is not always used by researchers and others in the field, but is useful in determining the need for medical assistance. Death in the absence of treatment usually occurs within 6 months to 2 years depending upon the opportunistic infections and malignancies that occur in the infected individual. However, with the recent development of effective treatment strategies, this period has been greatly extended.
Table 1 Classification of HIV disease (1987)
Table 2 Revised classification system for HIV infection and expanded AIDS surveillance case definition for adolescents and adults (1993)
In developed countries the proportion of infected individuals progressing to AIDS based on studies in homosexual men and blood-transfused cases was 5 to 6 per cent per year (Jaffe et al. 1985; Goedert et al. 1989; Lee et al. 1989). This incubation range corresponded to a median AIDS-free period of 9 to 10 years or longer. More recent studies suggest that, in the absence of treatment, half of all infected people will develop AIDS within 9 years after infection (Muñoz et al. 1989, 1997). Few individuals develop AIDS during the first 2 years after infection. Thereafter, the annual rate of developing disease increases from about 3 per cent in the third year of infection to 9 per cent in the seventh year. Among those AIDS-free for 10 years the probability of developing AIDS during the eleventh year may be as high as 15 per cent. However, some infected individuals have remained apparently healthy for more than 15 years (Lifson et al. 1991).
Survival rates among adults in the industrialized world have gradually improved. In the late 1980s the median survival after diagnosis was approximately 18 months (Lemp et al. 1990; Whitmore-Overton et al. 1993). The slowing disease progression and mortality was concurrent with the increased availability and use of prophylactic and therapeutic agents, in particular antiviral drugs and improved prevention and treatment of Pneumocystis carinii pneumonia (Gardner et al. 1992; N’Galy et al. 1998).
In Africa, progression rates to AIDS have been reported to be similar to rates in industrialized countries (Hira et al. 1990; Lindan et al. 1992; Bulterys et al. 1993; Anzala et al. 1995). However, two African studies suggest a more rapid progression. In a study of women sex workers in Kenya, the median incubation period from seroconversion to AIDS diagnosis was only 3.8 years (Naglekerke et al. 1990). Computer simulations based on this data suggest that progression to symptomatic disease was considerably more rapid than in developed countries (Simonsen et al. 1990). Poor survival among women sex workers in this population may be related to unknown lifestyle factors and frequent re-exposure to HIV (Mulder et al. 1994a). In a prospective cohort study of the population of 15 villages in Uganda, the mortality rate among HIV-seropositive adults was 11.6 per cent per year during a 2-year period (Mulder et al. 1994b; Subhash et al. 1998). The 2-year mortality figure of 23.2 per cent corresponds to a median progression from seropositivity to death of 5.25 years. Moreover, a substantial proportion of those who died progressed from asymptomatic infection or mild disease to death within 6 months.
The two studies mentioned above suggest that both the incubation and symptomatic survival periods may be shorter, in some circumstances, corresponding to a median survival from infection to death of only 5 to 6 years. The proportion of HIV-infected individuals who will eventually develop AIDS is, however, over 90 per cent in both developed and developing countries.
The long average incubation period greatly complicates the control of HIV infection, particularly during the early stages of the epidemic in a country or region. A model, summarized in Table 3, demonstrates the difficulty. This model, based loosely on the experience in Thailand, assumes that the incubation period from infection with HIV to diagnosis of AIDS is exactly 10 years. The first column of the table gives the number of new HIV infections per year (incidence), the second column the prevalence (or existing infections) per year, the third column the number of new AIDS cases per year, the fourth column the prevalence of AIDS, and the fifth column the number of deaths due to AIDS per year. It can be seen that by the time of the first occurrence of AIDS at 10 years there are already 15 000 HIV-infected individuals. Thus, even as the first cases of AIDS are appearing, the epidemic is already firmly established in the population. Although the assumption that the incubation period is exactly 10 years is incorrect, there are other factors, including the interval before AIDS is recognized and diagnosed by health professionals, that make this model useful in understanding why it is essential to take action against the epidemic even when only a few cases of AIDS have been recognized.
Table 3 Incidence, prevalence, and mortality of HIV/AIDS (example)
Strains of virus
There are at least three major strains of HIV: HIV-0, HIV-1, and HIV-2. However, the majority of the infections worldwide are associated with HIV-1. The other two major strains occur primarily in limited areas of Africa and have not established a foothold in other continents. Within the HIV-1 group, there are subtypes which are designated by letters A, B, C, etc. The significance of these subtypes is not well understood as yet. Initially, they occurred in different regions of the world.
Transmission of HIV
HIV can be transmitted only through exposure to infected blood and/or genital secretions, and vertically from mother to infant. However, there are certain groups that are more likely to be exposed. These include both heterosexual and homosexual individuals who have many different sexual partners, recipients of infected blood or blood products, including injecting drug users who share injecting equipment, haemophiliacs, health personnel, and infants born to infected mothers. The relative efficiency of transmission among these risk groups is given in Table 4.
Table 4 Activity-specific HIV risk in the United States
Sexual transmission currently accounts for more than 75 per cent of infections worldwide. The vast majority of sexual transmission is heterosexual, but male-to-male transmission, the first route of transmission recognized in the United States, is still the predominant mode of transmission in many of the developed countries and accounted for the introduction of HIV in many countries in which the major mode of transmission subsequently became heterosexual or drug-related. The type of intercourse (vaginal, anal, or oral–genital), the subtype of HIV, the presence of other STDs in either the infected or susceptible partner, the stage of infection of the infected partner, and whether the male is circumcised all influence the likelihood of HIV transmission occurring sexually.
Blood transfusion and blood products were a major mode of transmission early in the epidemic, but with the implementation of blood-donation screening and the treating of blood products in many countries, blood donation no longer accounts for a major portion of infections. The risk from blood transfusion in the United States is now estimated to be about 1 per 440 000 to 600 000 transfused units (Chamberland 1998). Blood-borne infection occurred primarily through contamination of the blood supply and through infected factor VIII, derived from plasma. However, some epidemics were reported among plasma and blood donors themselves, caused by inadequate sterilization of equipment between donors (Wu et al. 2001). The spread of HIV through sharing of needles, syringes, and drug paraphernalia among injecting drug users unfortunately still persists. Explosive epidemics have occurred in Europe, the United States, and Asia. Despite many efforts at education and intervention among drug users, injecting drug users are still the major infected risk group in several countries, including Vietnam and China. However, heterosexual transmission is rapidly replacing drug-user transmission in China and Vietnam. Another source of infection is accidents resulting in exposure to HIV-infected blood among health and laboratory personnel.
Vertical transmission of HIV in the absence of treatment occurs in an average of one-third of infants born to HIV-infected mothers, but the probability of transmission varies depending on the stage of HIV disease of the mother, the duration of labour, the viral load in the mother, the level of CD4+ cells, the phenotype (non-syncytium-inducing form/syncytium-inducing form) of the virus, and other factors. Mothers are most likely to transmit disease during the earliest stage of HIV infection and after becoming symptomatic, periods during which the HIV viral load is particularly high. With the advent of treatment of the mother the risk of transmission of HIV to the infant has been dramatically reduced and is now estimated to be about 2 per cent for infants whose mothers are treated with the newest drug strategies. Unfortunately, mothers in most developing countries do not have access to treatment, particularly the newest drug regimens. The original studies of treatment done in the United States initiated treatment in the second trimester of pregnancy and also treated infants for the first month of life (Connor et al. 1994). Recently, drug trials in developing countries have demonstrated that even a short course of monotherapy initiated at 36 weeks of gestation, or even at onset of labour in the mother and immediately after delivery in the infant, can reduce the risk of transmission to infants by as much as 50 per cent (Guay et al. 1999; Shaffer et al. 1999; Wiktor et al. 1999).
The rapid spread of HIV in the second half of the twentieth century is due in large part to rapidly changing sexual behaviours and attitudes that either permitted or condoned the practice of having many sexual partners. The high rate of sexual mixing (having intercourse with many different partners) in the homosexual population of many of the developed countries promoted the transmission of a virus, HIV, which has a relatively low infectivity rate. High rates of sexual mixing in Africa probably accounted for the rapid spread there. In Asia and Latin America it is part of the cultural expectation that men have multiple sexual partners both before and after marriage. However, a woman is expected to have only one sexual partner during her lifetime—her husband. To accommodate the men a small group of women become commercial sex workers servicing the large male population. In some countries of Asia and Africa commercial sex workers may have three to ten partners per night. Thus most sex workers in these countries become infected within 1 year of initiating commercial sex work and therefore become a reservoir of HIV infection. Their clients become the bridge population to the general population, including wives and children. In many countries that depend upon trucks for transport of goods, truck drivers who use the services of commercial sex workers become a vehicle for transmission of HIV along their routes of travel. Rapid urbanization, with migration of males from rural areas to seek jobs, also promotes spread of HIV. The temporary migrants rarely bring their wives and therefore turn to the services of commercial sex workers while they are in the cities, transmitting HIV to their families when they return home. For these reasons, changing the social roles of women and men in these societies will be an important factor in combating the epidemic.
The relatively low rates of heterosexual transmission of HIV in developed countries may be due to the relatively greater frequency of serial monogamy rather than multiple concurrent partners. In the United States and other developed countries, for example, men and women are likely to remain with one partner for months to years rather than having multiple partners concurrently. Thus the rate of sexual mixing is much lower in these countries.
Negative attitudes towards groups with high rates of HIV can deter effective surveillance to identify the reservoirs of infection in a population and may, for example, force homosexual men into marriage. Thus these men become an unwilling bridge between the heterosexual and homosexual populations. The fear of stigmatization forces these risk groups to hide both their risk practices and their illness once infected. Control of spread among homosexual men and injecting drug users becomes complicated when these groups are not identifiable within the population, but still may be spreading HIV.
Cultural and social stigmatization also deters voluntary testing for HIV among individuals who have been exposed to HIV. Knowing one’s HIV status is essential, not only for seeking early treatment but also to allow the infected individual to protect others, including his or her family. Studies have demonstrated that most individuals do modify their risk behaviour after learning that they are infected (Fox et al. 1987). Thus testing can be a very effective intervention tool, but is not used enough. ‘Risk-free testing’ is now technically possible. Individuals can be provided with test kits which will give results within 10 to 20 minutes in the privacy of their homes, but up to the year 2000 political factors have prevented the licensing and sale of such kits. The concept that an individual has a right to know their HIV status without risk of disclosure and social stigmatization has not yet achieved a political consensus in any country.
Political and economic factors
Maldistribution of resources promotes transmission of HIV because:
poor men and women are forced to turn to commercial sex work to survive
affluent men have the means to avail themselves of the services of commercial sex workers
the poor have limited access to adequate health care, including care of sexually transmitted diseases
many segments of the population cannot read or write, making intervention through education very difficult.
Political stability plays a very important role in the transmission of HIV. Countries with stable governments can devote resources to HIV/AIDS intervention and can provide consistent programs. An example of this is the success of Thailand in slowing the epidemic. War, on the other hand, promotes the transmission of HIV by inhibiting social restraints, especially among soldiers, and forcing poverty and separation of families. Many of the countries that have experienced the highest rates of HIV, such as Rwanda and Uganda in Africa, and Cambodia in Asia, have suffered from civil unrest. It is interesting that, with the onset of peace and stability, Uganda has been one of the few countries to report some success in control of HIV spread.
The global epidemiology of HIV
The epidemic of HIV started at different times in different regions and countries and has involved different risk groups. Thus in a sense the HIV pandemic is really a series of mini-epidemics, with different features depending on the social, cultural, geographic, and behavioural characteristics of the country, city, or region.
In the developed countries of North America, Western Europe, Australia, and New Zealand the epidemic began first either in the group of men who have sex with men or in injecting drug users, but has spread only slowly, if at all, into the mainstream heterosexual population. Within these countries, however, there may be differences between areas. Thus the predominant mode of spread in Los Angeles is among men who have sex with men, whereas in the New York metropolitan area the major mode of spread is through injecting drug users. Retrospective studies have indicated that the epidemic among men who have sex with men probably began in the 1970s. In most of the countries of these regions the epidemic has now reached its peak and is declining, although this is less true for some subpopulations, such as African-Americans in the United States. The reasons for the decline are probably due both to the efforts of the major risk groups (gay men and injecting drug users) and the exhaustion of susceptibles. The decline preceded the availability of highly active antiretroviral therapy (HAART). An organized government response probably played only a small role in the decline in the United States.
The epidemic in the developing countries of the world, in which 95 per cent of HIV infection is currently occurring, varies from region to region. In Africa, where the epidemic first became apparent among developing countries, transmission has been almost exclusively due to heterosexual transmission. Testing of stored sera from earlier decades has indicated that HIV was occurring in Africa as early as the 1950s, but did not become widespread until the 1970s. Although contamination of the blood supply and the practice of piercing the skin for cosmetic reasons and for female infibulation is not uncommon, it is unlikely that these modes have played a major role in the spread of HIV in the African countries. However, the entire continent of Africa has not been uniformly involved. The countries of eastern sub-Saharan Africa were first involved and only in the last decade has the epidemic spread to southern Africa, including South Africa, Zimbabwe, and Botswana, where it has involved 10 to 35 per cent of the adult population. The spread of HIV has been so rapid and complete that in some parts of sub-Saharan Africa the typical nuclear family consists of grandparents and grandchildren. The parent generation has succumbed to HIV.
The countries of northern Africa, as well as the Eastern Mediterranean region, report relatively low rates of HIV and AIDS. The extent to which this is due to a combination of poor reporting systems and cultural or religious taboos against sex outside marriage, making identification of infected individuals difficult, is hard to assess. There is some evidence that HIV is circulating in these countries; returning Filipino overseas workers, a high proportion of whom work in Eastern Mediterranean countries, are one of the groups with the highest prevalence of HIV in the Philippines.
The epidemic in the Latin American countries began at different times. The epidemic in the Caribbean countries seems to have begun about the same time as the epidemic in Africa and the United States and has the same characteristics as the epidemic in sub-Saharan Africa, i.e. spread is primarily through heterosexual activities. The epidemic in the South American countries began in the middle to late 1980s and was first identified in Brazil. The first groups involved in Brazil were men who have sex with men, bisexual men, and injecting drug users, although the epidemic has now spread to the heterosexual population. The epidemic in the southern cone (Argentina, Chile, Uruguay, and Paraguay) resembles the epidemic in the developed countries, involving primarily men who have sex with men and, to a lesser extent, injecting drug users.
The first cases of AIDS in Southeast Asia were reported in the middle to late 1980s, usually among foreigners or returning residents. These individuals were usually men who had sex with other men. By 1988–1989 the epidemic had spread to injecting drug users in northern Thailand and within 1 to 2 years involved commercial sex workers, both direct (brothel-based) and indirect (based in entertainment establishments), and subsequently their clients and the wives of their clients, many of whom were identified through testing in antenatal clinics (Fig. 5). The region around northern Thailand, often known as the Golden Triangle because of the production of narcotics in the region, became an early focus for the epidemic. Injecting drug users from southern China (Yunnan Province), Myanmar, and the northeast states of India became infected at high rates. Initially, men from the hill tribes, which are common to these countries, were infected, but over time the majority populations of these areas also became infected. The epidemic soon spread to the commercial sex workers, especially in Myanmar (Burma) and Thailand. In other areas, such as Mumbai (Bombay) and Chennai (Madras) in India and in Cambodia, the epidemic began among commercial sex workers and rates in injecting drug users have been low. On the other hand, the epidemic in Vietnam has now spread widely among injecting drug users over the last 5 years with recent spread to commercial sex workers. In Indonesia and the Philippines cases were reported in the middle to late 1980s but the epidemic has not taken off. Studies of commercial sex workers in the Philippines have suggested that low rates of sexual mixing may account, in part, for the slow progression of the epidemic in those countries.
Fig. 5 HIV Prevalence among high-risk groups, Thailand, 1989–1995: IDU, injecting drug users; CSW, commercial sex workers; STD, sexually transmitted diseases. (Reproduced from Phoolcharoen et al. 1996.)
HIV was first recognized in China among injecting drug users in southern Yunnan Province in the late 1980s, at about the same time that infected injecting drug users were being recognized in Myanmar, Manipur, and northern Thailand. For several years the epidemic seemed to be confined to injecting drug users in southern Yunnan, but now infected individuals are reported in all 31 provinces of China and are being seen in other groups, including blood and plasma donors and commercial sex workers. It is estimated that by 2000 there will be 1.2 million HIV-infected individuals in China (Watanabe 1999).
Although many haemophiliacs were infected in Japan through contaminated factor VIII produced in the United States in the early 1980s, the spread of HIV in Japan, as well as in Korea and Mongolia, seems to be very slow.
The distribution of HIV in the Pacific islands has varied as well. The epidemic has established a firm foothold in Fiji, French Polynesia, and Papua New Guinea, but only sporadic HIV infections are reported from most of the other island nations. This may be due to their relative isolation and the small populations inhabiting them.
China and India together include 38 per cent of the world’s population. Already there are more HIV-infected individuals in India than in all the rest of Southeast Asia combined, and the same is also probably true for the number of infected individuals in China versus the rest of the Western Pacific region. Therefore the potential for an even worse catastrophe exists if the epidemic continues to progress at its current rate in these two countries.
Specific subtypes of HIV-1 infection have been identified and labelled according to the alphabet. The major strain in the United States and many of the developed countries is subtype B (sometimes known as ‘clade’), whereas subtypes A and C are the major subtypes in Africa. The distribution of subtypes worldwide is shown in Fig. 6. Some of the subtypes are recombinants of two other subtypes. At one time it was thought that some subtypes were associated primarily with vaginal intercourse and others primarily with anal intercourse and injecting drug use, but these distinctions appear to be breaking down (if they existed at all). Some investigators feel that it is important to include in vaccines under development subtypes specific to the region in which the vaccines will be used, but whether protective immunity is associated with subtype has yet to be established. However, periodic evaluation of subtypes in regions can provide clues about the global routes of transmission. For example, subtype C, the major subtype in India, was first identified in Africa. The spread of HIV among injecting drug users in the countries of Southeast Asia was initially associated with subtype ‘Thai B’, but the major subtype among injecting drug users in Thailand in the late 1990s was primarily subtype E, which was initially found primarily among individuals infected through vaginal intercourse. Thus the analysis of subtypes provides information both about the spread of HIV between regions and about spread within a region.
Fig. 6 Global breakdown of HIV subtypes. (Reproduced from UNAIDS 2000.)
Monitoring the epidemic
The HIV epidemic has spread rapidly in many areas, but much more slowly in others. Because of the long incubation period from infection to clinical disease, HIV can be spreading rapidly in the population with little awareness on the part of public health officials. For example, in the early part of the epidemic in Thailand it was estimated that there were 8000 to 10 000 infected individuals for every diagnosed case of AIDS (Weniger et al. 1991). For this reason it is important for countries at risk of HIV to establish a system to identify the introduction of HIV into their population, to identify the first risk groups involved, and to monitor the spread of the epidemic so that effective, targeted intervention strategies can be implemented. Thus surveillance systems for HIV/AIDS need to be introduced, ideally before the epidemic has gained a foothold in the population. Although it is far easier to introduce surveillance for clinical AIDS, the information from surveillance for AIDS will reflect the HIV epidemic 8 to 10 years earlier and so will have little relevance for control of the current epidemic. For this reason it is necessary to implement surveillance for HIV, a much more difficult process requiring testing of blood, saliva, or urine. There are many surveillance strategies that might be used for monitoring the epidemic, but because of the unusual dynamics of HIV in the population only some are useful.
Surveillance has been defined by Alexander Langmuir, the founder of the American Epidemiologic Intelligence Service, as ‘The continued watchfulness over the distribution and trends of incidence through the systematic collection, consolidation and evaluation of morbidity and mortality reports and other relevant data together with dissemination to those who need to know.’ Somewhat more colourfully, Donald A. Henderson, the director of the Smallpox Eradication Program, has said ‘Surveillance serves as the brain and nervous system for programs to prevent and control disease.’ The key elements of surveillance are as follows:
the collection of health data expressly for use in health planning, disease control/prevention, and/or health promotion
ongoing collection of data
understandable presentation of data
dissemination of results
action based on results
periodic evaluation to assure that the system is still providing relevant information.
Surveillance has been used to establish the extent and distribution of HIV infection in the population/country, to monitor the epidemic, to establish appropriate programme priorities and resource allocation, and to evaluate intervention strategies. It is important to emphasize that the accuracy of the sentinel surveillance system is of less importance than the speed with which the data is collected, processed, and presented to decision-makers and the public.
Surveillance is very expensive and can be afforded by few developing countries. Therefore many developing countries have utilized a strategy known as ‘sentinel surveillance’. The objective of sentinel surveillance is to provide an early warning system that HIV infection is occurring in the population. To achieve this objective it is necessary to identify those groups in the population who are most likely to engage in the activities that expose them to infected blood and genital secretions. Examples of such subgroups are injecting drug users, men who have sex with men, commercial sex workers, people attending sexually transmitted disease clinics, health workers, and laboratory workers. The specific risk groups in which HIV infection may be introduced will vary among different populations. Surveillance for HIV infection is set up in these groups to warn when HIV has been introduced into the population. In the initial stages of the epidemic it is important to select individuals within subgroups who are most likely to be exposed first to HIV. Once the initial reservoirs of HIV infection have been identified it is more useful to select representative samples of the subgroups to more accurately reflect the true prevalence of HIV infection among them. Knowledge of the reservoirs of infection for the particular population provides essential information for the development of focused, effective intervention strategies early in the course of the epidemic.
An example of the type of information that can be generated from a good sentinel surveillance system is shown in Fig. 5, which shows the increasing prevalence of HIV in each of the sentinel groups in Thailand. The Thai sentinel surveillance system was initiated in 1989 in injecting drug users, direct commercial sex workers, indirect commercial sex workers, men who attend sexually transmitted disease clinics, and women attending antenatal clinics. By the time the sentinel surveillance programme was initiated the epidemic had already become well established among injecting drug users, but the spread from them to direct commercial sex workers and ultimately to women attending antenatal clinics can be easily seen in Fig. 5. Armed with this knowledge the Thais implemented effective intervention strategies directed initially to the brothels. The intervention programme in Thailand, which is based on knowledge generated from their excellent sentinel surveillance system, has resulted in a decline in the epidemic. The key elements of this intervention programme are discussed later in this chapter.
The spectrum of clinical manifestations
The spectrum of AIDS-defining illnesses includes viral, bacterial, and fungal infections, and malignancies. In addition, two conditions can apparently be caused by HIV alone: AIDS dementia and wasting syndrome. A list of AIDS-defining conditions has been published by the Centers for Disease Control and Prevention in the United States, and is reproduced in Table 5. In the developed world, before the advent of effective prophylaxis and antiretroviral treatment, the most common AIDS-defining condition was Pneumocystis carinii pneumonia, followed by Kaposi’s sarcoma, which occurred primarily among men who have sex with men. With the advent of HAART the frequency of non-Hodgkin’s lymphoma and other malignancies has increased as survival with AIDS and HIV has increased, and the frequency of Pneumocystis carinii pneumonia has decreased. Wasting syndrome, first reported in Africa, has increasingly been reported in developed countries as well, especially among injecting drug users.
Table 5 Conditions included in the 1993 AIDS surveillance case definition (Centers for Disease Control and Prevention)
The development of opportunistic infections and malignancy in people with HIV is clearly correlated with the severity of immune deficiency. The vast majority of AIDS-related opportunistic infections occur when immunodeficiency is advanced. Although the level of immunodeficiency defines the risk of development of HIV-related opportunistic infection and malignancy, a major determinant of the clinical spectrum in a particular setting is the environmental and human microbial flora. A high background prevalence of Mycobacterium tuberculosis infection enables both reactivation of latent infection as immunodeficiency develops among people with prior exposure, and primary infection, which is often progressive, in those previously unexposed.
The spectrum of HIV-related conditions among people living with HIV in the developing world is poorly outlined. Clinical spectrums are derived predominantly from hospital-based clinic sites and some routine AIDS surveillance. There are several limitations to both these forms of clinical spectrum data collection. First, although presumptive diagnosis is possible for several AIDS-defining illnesses, definitive diagnosis of many conditions requires a level of laboratory and other diagnostic services not available in most developing countries. Second, referral bias is always a potential limitation to clinical series derived from tertiary referral hospitals, the clinical site for most of these studies. Thus the clinical spectrum may be somewhat different in a population-based setting. Finally, both AIDS surveillance and many clinical series are based on the initial AIDS-defining illness alone; therefore conditions which occur at advanced stages of immunodeficiency will be under-represented (Subhash et al. 1998).
Tuberculosis is the major AIDS-defining illness in most developing countries. Fungal infections, in particular oral and oesophageal candidiasis and cryptococcal disease, are relatively common in both developed and developing settings; penicilliosis, infection with the fungus Pennicillium marneffei, is an exception, being described in series from Thailand and Hong Kong only (Chesseman et al. 1995).
Both cytomegalovirus disease and Mycobacterium avium complex infection occur when immune deficiency is very advanced, with the majority of cases presenting subsequent to AIDS diagnosis and associated with a CD4+ cell count less than 50/mm3. The relative absence of both these conditions associated with very advanced immune deficiency from the initial clinical spectrum series from Thailand, India, and Papua New Guinea may indicate either a lack of diagnostic services, more rapid progression after AIDS diagnosis, or a different natural history of HIV disease. If people with HIV infection in these countries are dying from other opportunistic infections, such as tuberculosis, prior to development of very advanced immune deficiency, then the prevalence of conditions such as cytomegalovirus disease and Mycobacterium avium complex infection will be relatively low. However, a recent study in Thailand has found an 18 per cent prevalence of Mycobacterium avium complex infection among persons with HIV and unexplained prolonged fever (Sathapatayavong et al. 1997). Furthermore, an AIDS autopsy study from Mumbai, India, found evidence of gastrointestinal infection in 27 per cent of cases, although no cases had associated mucosal disease (Lanjewar et al. 1996). These studies support the theory that lack of diagnostic capability may be largely responsible for the apparently low prevalence of these diseases in developing countries. Recent improvements in HIV management in Thailand and India, including introduction of antiretroviral therapy and opportunistic infection prophylaxis, may also lead to longer survival with consequent increased development of very advanced immune deficiency.
The development of an effective treatment, much less a cure, has been very difficult. The virus makes at least one transcription error each time it mutates and thus rapidly develops resistance to individual antiretroviral agents. Treatment approaches have been based on interrupting the replication cycle at vulnerable points in the process.
The first antiretroviral drug to show an effect in humans was azidothymidine, currently more commonly known as zidovudine. Azidothymidine interrupts the transcription of viral RNA to viral DNA by blocking the action of the reverse transcriptase enzymes (Ho and Hitchcock 1989). Azidothymidine, a nucleoside reverse transcriptase inhibitor, became available in 1987. Because azidothymidine had no effect on cells that were already infected with HIV and did not completely block the process of replication, it did not result in elimination of the virus and thus was not a cure. By the late 1980s additional nucleoside reverse transcriptase inhibitors were produced and in the early 1990s non-nucleoside reverse transcriptase inhibitors had been developed. Although each of these drugs had an impact on viral production none of them remained effective for more than a few months to a maximum of 2 years. Further, they were associated with severe side-effects, especially at the doses necessary to produce an effect when not given in combination with other drugs (Richman et al. 1987). Although these drugs were initially given only after the diagnosis of AIDS, by the early 1990s many HIV-infected individuals began treatment before the onset of AIDS. This resulted in delaying the onset of AIDS, but the duration of survival was the same whether the drugs were initiated prior to or after the diagnosis of AIDS (Detels et al. 1997).
By the early 1990s nucleoside reverse transcriptase inhibitors and non-nucleoside reverse transcriptase inhibitors were being given in various combinations. The philosophy behind combination therapy was that doses of individual drugs given in combination could be lowered, reducing the probability of side-effects, while the probability of developing resistance to two different drugs was much lower than the probability of developing resistance to a single drug given alone.
The big treatment breakthrough came with the introduction of protease inhibitors in 1995 (Lewis et al. 1997). The recommended therapy became the use of three different drugs. This regimen was called highly active antiretroviral therapy (HAART) (Detels et al. 1998). The protease inhibitors block the assembly of the progeny HIV within the CD4+ cell. The changing use of the different treatment regimens in homosexual men in the Multicenter AIDS Cohort Study (a large cohort study initiated in 1984 in the United States) is shown in Fig. 7. The addition of protease inhibitors to combination therapy resulted in a dramatic decline in the incidence of both AIDS and death as demonstrated in Table 6 (Detels et al. 1998, 2001). The efficacy of HAART has now been demonstrated in most groups, including women, injecting drug users, heterosexuals, and men who have sex with men. As a result of HAART the AIDS mortality rate among males 25 to 44 years in the United States has dropped from the leading cause of death in the late 1980s to number five in 1997. Nonetheless, HAART is not the final answer. Side-effects and intolerance to the drug combinations do occur and the efficacy of specific combinations of drugs does decline over time. Further, HAART requires taking 15 or more pills per day on different time schedules and with different restrictions. Thus compliance becomes a major issue. Skipping of drugs or irregular use of drugs increases the probability of drug resistance developing. Highly active antiretroviral therapy has been a major advance, but more drugs with fewer side-effects and improved ease of administration need to be developed. Thus the search for new drugs continues, especially drugs acting at different stages of the replication cycle. In the foreseeable future it is unlikely that a cure will be developed, but it is reasonable to hope that HIV/AIDS in the individual may be controlled much as high blood pressure and diabetes are controlled by effective medication.
Fig. 7 Use of antiretroviral therapy by Multicentre AIDS Cohort Study seroconverters while free of AIDS. (Reproduced from Detels et al. 1998.)
Table 6 Antiretroviral therapy and relative AIDS-free and survival times by calendar periods
Treatment in the developing countries
In the industrialized world, new treatment approaches can find their place with relative ease into the existing health-care systems, reaching most, although not all, potential beneficiaries. To make these therapies available to people living with HIV/AIDS in the developing world will require a major improvement in the number of qualified health-care providers and a willingness on the part of the drug manufacturers to use a different pricing strategy for developing countries or to allow developing countries to manufacture the drugs at cost. Most developing countries are only able to provide treatment or prophylaxis for the more common opportunistic infections and supportive care for symptomatic and dying patients. The clear need for improvement of HIV/AIDS care in developing nations, however, may create the opportunity and provide the impetus needed to reconsider resource allocation and use in the economic and social development sectors as well as to redress the insufficiencies of existing health systems.
Historically, care for HIV-infected people has not been viewed as playing a preventive role in reducing HIV incidence. People with HIV must be seen as important partners in prevention; the virus can only be transmitted through them. However, if they feel abandoned by care services, they are less likely to understand the need for prevention and to be motivated to protect others (MacNeil and Anderson 1998).
Regardless of the effectiveness of prevention efforts being made today and new advances in treatments, the numbers of people becoming ill as a result of HIV infection will dramatically increase over the next few decades. In addition, increasing demand for, and access to, counselling and testing services will increase the number of HIV-infected persons who know their HIV status and will demand treatment. As the number of infections increases, demands for care will mount globally. As these demands mount, increasing consideration needs to be given to the mitigating effect care and social support can have on the epidemic, including its effect on prevention. While few health planners and providers would dispute the fact that both HIV prevention and care are necessary, the challenge confronting many developing countries is how to provide both in an environment with limited resources. Both developing and developed countries must be concerned with this issue.
HIV/AIDS intervention programmes need to recognize and understand the dynamics and existing strengths and weaknesses of household and community arrangements for care. In particular, home-based care programmes need to be made gender sensitive and designed so that they relieve the excessive caregiving burden on women. The emotional state of the women and their need to address their feelings of resentment and hatred should be recognized and given due importance in HIV/AIDS counselling. This is especially important when they are blamed for their partners’ health condition and are culturally expected to provide care in sickness. At the community level, the social climate for those affected by HIV/AIDS needs to be improved. In the developing countries, the capacities of households and communities need to be strengthened to better manage care for their infected members.
One of the most tragic aspects of the HIV/AIDS pandemic has been the infection of infants. It is estimated that 30 to 50 per cent of HIV-infected pregnant women will give birth to an infected baby in the absence of treatment. The probability that the infant will be infected depends on the level of HIV, the stage of HIV infection, and the current phenotype of HIV (syncytium-inducing, non-syncytium-inducing) in the mother, her immune competence, premature birth, and chorio-amnionitis (Newell et al. 1997). Of infected infants, 30 to 50 per cent are infected during the birth process and another 15 to 20 per cent through breast feeding. Prior to the development of effective therapies, prevention depended on reducing the prevalence of HIV in women, reducing the exposure of the infant to blood during the birth process, not breast feeding the infant unless the mother was unable to obtain clean water and prepare safe formula, vaginal washing, topical and oral treatment of the infant, treatment with specific HIV-immune globulin, and delivery by Caesarean section (Bryson 1996). None of these efforts had a dramatic effect on reducing the risk of HIV infection to the infant. The most important determinant of maternal–fetal transmission is the level of HIV in the mother (Dickover et al. 1998).
Infants infected from their mothers follow one of two courses: rapid progression with early death within months of birth, or a slow course lasting years and in some cases into adolescence. Infants infected in utero tend to experience rapid progression whereas those infected during the birth process or infected through breast milk tend to experience a slower course, extending in many cases well beyond 2 years. The level of HIV in the infant during the first weeks to months is a better predictor of survival than CD4+ level. There is also a very interesting subset of infants from whom HIV can be identified shortly after birth, but who apparently clear the virus and are no longer infected (Bryson 1995; Bryson et al. 1995). This last group is particularly interesting because identification of the mechanisms allowing the infant to clear the virus may provide clues to reversing infection in the majority of infants who cannot otherwise clear it.
In 1994 the results of a drug trial in which azidothymidine was given to infected women from the second trimester of pregnancy through delivery and to the infant for 1 month following birth demonstrated that the proportion of infants being infected could be reduced by three-quarters (Connor et al. 1994). Although this was good news to women in developed countries, the high cost of the long course of treatment was beyond the economic capacity of most developing countries. A 1998 study of azidothymidine given orally to mothers from 34 weeks of gestation through delivery and another study of nevirapine alone given orally at onset of labour both demonstrated a 50 per cent reduction in the prevalence of HIV among the infants (Guay et al. 1999; Shaffer et al. 1999; Wiktor et al. 1999). The cost of giving a single dose of nevirapine to the mother as well as the infant is US$4 to US$8. With the introduction of HAART given in the first trimester, the proportion of infants infected has been reduced to 5 per cent, and many investigators feel that the risk can be further reduced to 2 per cent (Bryson 1996).
The impact of HIV/AIDS
HIV/AIDS causes incalculable human suffering and social and cultural disruption, and has huge economic costs. The effects pervade the industrial, agricultural, and health-care sectors and permeate society, affecting individuals, families, and communities. The World Health Report 1999 (WHO 1999) indicates that heart disease, strokes, and acute lower respiratory infections—typical causes of death in old age—are the only causes of death to surpass HIV/AIDS. HIV/AIDS is now the primary disease burden in developing countries, affecting primarily younger and middle aged people.
The impact on life expectancy and health
The most obvious impact of AIDS is on life expectancy. Measuring and predicting the impact on a society or nation, however, is difficult, not only because of the lack of quality data, but also because the magnitude of the impact depends on many factors, including the rate of spread and success in fighting other health problems. In the most severely affected countries, AIDS threatens to reverse a century of progress in the fight against infectious disease. Elsewhere, it is likely to account for an increased share of the infectious disease burden. But AIDS is only one of many health problems confronting people in developing countries (World Bank Policy Research Report 1997).
Life expectancy is a basic measure of human welfare and of the impact of AIDS. From 1900 to 1990, dramatic successes in the fight against infectious disease raised life expectancy from 40 to 64 years in developing countries, narrowing the gap between these countries and the industrial countries from 25 to 13 years. AIDS has slowed, and in some countries reversed, this trend. Life expectancy in Burkina Faso, a mere 46 years, is 11 years shorter than it would have been in the absence of AIDS. In several other hard-hit countries, life expectancy also has been pushed back to levels of more than a decade ago. AIDS accounted for about 1 per cent of all deaths worldwide in 1990; this proportion is likely to rise to 2 per cent of all deaths in 2020 (Murray and Lopez 1996). However, the proportion of total deaths caused by a disease is an imperfect representation of its burden on society, because it ignores illness and does not distinguish among the deaths of people of different ages. The cost of diseases can be estimated in terms of disability-adjusted life years (DALYs). DALYs include the disability as well as the mortality effects of disease and use age weights to discount the importance of infant and elderly deaths. In 1990, poor health in developing countries resulted in the loss of 265 DALYs per thousand persons per year, almost twice the 124 DALYs per thousand per year lost in industrial countries. Because HIV/AIDS causes substantial disability before death and disproportionately strikes productive-age adults, it has a larger impact on health measured as DALYs than as a share of total deaths.
It was projected that HIV/AIDS would account for almost 3 per cent of all DALYs lost in developing countries in the year 2020, up from 0.8 per cent in 1990 (United States Bureau of the Census 1997). One reason that HIV/AIDS does not account for a larger percentage of lost DALYs is that other causes of death in developing countries also cause substantial disability and premature death. Further, some of the increased impact of HIV/AIDS is offset by the decreasing proportion of productive-age adults in the population associated with the demographic transition from high proportions of very young to greater proportions of elderly.
The economic impact
The economic impact of HIV/AIDS can be classified into macroeconomic and microeconomic effects. Macroeconomics describes economic systems and normally refers to a country’s economy, while microeconomics examines individuals, groups, and firms.
Economic analysis during the first years of the HIV epidemic was speculative, and assumed that AIDS would adversely affect economies at all levels, from national to household. In 1988, the first international conference on the global impact of AIDS considered the impact on food production as well as on families and development (Fleming et al. 1988). It was recognized that AIDS would have a major impact since the infection is concentrated in the 20 to 40 age group, the age group which carries an enormous economic burden in societies in which nearly 50 per cent of the population is under 15 years of age.
In the early 1990s, studies began to demonstrate how AIDS might affect the economies of developing countries. The World Bank, using DALYs, demonstated enormous regional and global inequalities in health status (World Bank 1993). In sub-Saharan Africa in 1990, ill health and accidents led to the loss of 575 DALYs per 1000 population, of which HIV and sexually transmitted infections accounted for 8.8 per cent, whereas in industrialized countries only 117 DALYs were lost per 1000 population and HIV accounted for only 3.4 per cent of these. In extending the concept of DALYs to prevention and control, the study suggested that HIV in developing countries can be substantially reduced by interventions costing less than US$100 per DALY saved.
In sub-Saharan Africa, the macroeconomic impact of HIV/AIDS has been demonstrated in two ways (Over 1992). First, it has an impact on human capital, the skill base of an economy. Economic growth is closely correlated with increases in urban skilled populations. An adverse affect to these populations has a disproportionately negative impact on economic growth. Second, HIV/AIDS affects savings, which are essential for economic growth, as they allow new investment. Funds for AIDS treatment is taken from savings, so that investment and future growth suffer. Thus AIDS has the potential to slow economic growth significantly.
The effect of AIDS is felt first at the microeconomic, or household level. The next step occurs when economic activities, such as agriculture, transport, and mining are affected. Later, impacts on the national economy are detected.
The societal impact
In the period from 1981 to 1986, when the precise dimensions of the AIDS epidemic in the United States remained uncertain and the extent to which populations not initially infected might be exposed was unclear, analysts predicted that no geographical region and no aspect of social life would remain untouched. By 1985 the prospect of widespread heterosexual transmission of HIV in the United States caused anxiety. In the early 1990s, after more than a decade of living with AIDS, a very different, more subtle picture had emerged (Bayer 1996). The combination of a conservative national administration, and the fact that HIV occurred primarily in socially marginalized groups, shaped the impact of AIDS on the United States. The response to AIDS has been most successful where politics permitted gay men and their allies to press for changes, to elicit sympathetic responses, and to wrest, sometimes grudging, concessions.
The AIDS epidemic has had the unexpected impact of incorporating gay (homosexual) organizations into the pluralistic fabric of American political life (Altman 1988). Thus it was possible for gay groups to thwart widespread HIV testing without consent and to win the support of the Centers for Disease Control and Prevention for antidiscrimination and confidentiality (Bayer 1991) measures supporting both gay men and individuals with HIV/AIDS. Gay men have established a remarkable network of political, social, and service organizations to meet their needs and protect their interests.
In 1993, the National Research Council stated that the limited response to the AIDS epidemic could be explained because the absolute number of infected individuals, relative to the United States population, was not overbearing and because social institutions in the United States were strong, complex, and resilient. More important, the concentration of the epidemic was in socially marginalized groups and had only slowly spread to the mainstream heterosexual population. Thus those who continued to be affected were socially invisible, beyond the sight and attention of the majority population (National Research Council 1993).
HIV/AIDS has also impacted the cultures of Asia, but a different social impact was identified. Among the many factors influencing the ways in which households respond to HIV/AIDS in both India and Thailand are the economic situation of affected households and communities, prevailing gender relations and women’s social status, local health beliefs, local health-care practices, and the stigma attached to HIV/AIDS (Bharat et al. 1998). While sociocultural contexts shape the ways in which households and communities respond to the epidemic, the epidemic itself is beginning to influence local practices and perceptions. This was evident from changes reported in funeral rituals and the ways in which illnesses are perceived as less or more threatening, depending on the severity of visible bodily symptoms. Gender biases were evident at the household level, where men received greater care and acceptance than women. This is not a surprising observation in the male-dominated, patriarchal culture of Asia where men’s needs are considered more positively than those of women.
A major problem has been the apprehension of affected individuals and households of being stigmatized by their neighbours and relatives. Consequently, there is no openness about HIV status, resulting in an inability to tap the traditional sources of support and counselling such as village elders and extended family. Misconceptions about the epidemic have led to fear of discrimination and sometimes to suicide. The issue of stigmatization must be addressed if persons at risk of HIV are to be treated as individuals who need understanding, compassion, and access to treatment.
The global scope and extent of the AIDS-induced orphan situation is now better appreciated. Many of the countries in which the effects of HIV/AIDS have been most severe also have high adult mortality rates from other causes, including civil conflict and war. As a result, these countries already had many orphans even in the absence of HIV/AIDS. Since HIV/AIDS affects mainly young and middle-aged adults, AIDS causes a marked increase in the number of orphans, in some developing countries as much as 50 per cent (Ainsworth and Over 1992). The number of children who had been orphaned by HIV/AIDS by the year 2000 was estimated to be 13.2 million (UNAIDS 2000).
Three general observations about these orphans provide a basis for analysis and planning (Levine et al. 1996). First, most orphans are not infected with HIV. Second, the orphans create special stress on family and community resources, and third, they are highly vulnerable because families affected by HIV/AIDS usually have more than one ill or dying member. In African countries the HIV/AIDS pandemic has added a new group of orphans to the already serious problem of children who are parentless, but in industrialized countries the phenomenon of large numbers of children left parentless from a single disease is unprecedented. In Asia the rapid spread of HIV infection, especially among young women, presages a crisis around orphans.
Traditional family stuctures of people in developing countries, already stressed by poverty, poor health, and increased burdens of care, are in many instances reaching the breaking point. People with HIV/AIDS have received limited care by family members because of poverty, other responsibilities, and stigma. Some adult patients were cared for by children because there were no other adults in the home (Seeley et al. 1993). For those orphans, material assistance, such as food, bedding, medicine, and school fees, is needed to make up for lost income, but more than money is needed to overcome the burden of AIDS on families. Although the general ‘safety net’ for people with HIV/AIDS has gaps, the net for dependent orphaned children after the death of a parent is even looser. While many families have absorbed orphaned children out of love, custom, or moral obligation, they may not be able to do so indefinitely. An independent assessment is needed to determine the level of assistance required to promote family function. Assistance to orphans should not be viewed as a short-term project but rather as a long-term commitment to support local sustainable initiatives. The well being of children orphaned by the HIV epidemic, like children in distress from other causes, is the test of our future commitment to social stability, economic development, and human rights.
The role of stigmatization
One of the most difficult problems in controlling HIV/AIDS has been stigmatization of both groups at highest risk of HIV and of individuals who are infected. In the United States the home of a young haemophiliac infected with HIV was burned down in rural Florida because the youngster insisted on attending school. Sentinel surveillance, as mentioned above, plays an important role in providing information on which effective intervention programmes can be based. Stigmatization reduces the willingness of risk groups to be identified, making surveillance difficult, if not impossible. Thus the epidemic may rapidly increase in these groups and spread from them into the general population without the knowledge of public health professionals and policy-makers. The reluctance to diagnose AIDS on the part of physicians may also cause them to avoid reporting a diagnosis of AIDS in order to spare their patients embarrassment and stigmatization. But if the presence of AIDS in the community is hidden then public health officials are unable to take the necessary steps to prevent the further spread of HIV.
One of the most effective tools for altering risk behaviour is to provide individuals with the knowledge of their HIV infection status (Kalichman 1998). Thus HIV testing is an important intervention strategy. Unfortunately, many people are reluctant to be tested, fearing that they may test positive and have their HIV status revealed, subjecting them to discrimination, job loss, and loss of family. Further, many health officials and politicians feel that they must protect the HIV-infected person even at the expense of the uninfected. Even implementing such a simple strategy as premarital testing for HIV in countries with high levels of HIV infection in young persons has met with resistance from public health officials who seem to have forgotten that the most important task of public health is to protect the uninfected, albeit while doing the minimum amount of harm to the infected. Surely a young woman entering into marriage has a right to know that her husband will not kill her and their yet-to-be-born infants by infecting them with HIV (and vice versa). Rapid tests which can be interpreted by anyone are now available, but are blocked in many countries for political reasons. Making these ‘risk-free’ test kits available to individuals so that they can learn their HIV status without risk of disclosure should be a priority.
Because of the complication for control of HIV caused by stigmatization, a major effort should be directed towards reducing and eliminating stigmatization. This process, often called ‘normalization of HIV’, emphasizes the need to treat HIV/AIDS as an ordinary chronic disease such as diabetes or hypertension. This concept is appropriate for developed countries where HIV/AIDS patients can receive treatment and return to a relatively normal lifestyle, but is more difficult to apply in developing countries where treatment and return to a normal lifestyle are not usually a possibility due to economic constraints. Nonetheless, HIV/AIDS patients in developing countries need to be treated with understanding and compassion. Although a return to ‘normalization’ is not possible for HIV/AIDS patients in developing countries, they still need to obtain care and the social support which is an essential part of the compassionate treatment of chronic, fatal illnesses to which all human beings are entitled. To promote an understanding of this need in developing countries is a major challenge, but one that must be continually addressed.
Ethical and human rights issues
From the beginning of the epidemic in the early 1980s, ethical and human rights issues have been the subject of vigorous debate. One of the first issues was the question of whether to systematically identify and exclude homosexual men as blood donors in the United States in 1983 (Institute of Medicine 1995). The ethics of protecting the uninfected who might receive infected blood was in conflict with the rights of homosexual men to privacy. This basic conflict between the rights of the uninfected and the ethical public health obligation to protect the infected has persisted with the debate over whether to test pregnant women and/or infants for HIV infection and whether to require premarital testing in areas of high HIV endemicity.
Testing has been another area in which there has been considerable debate. Testing and counselling have been demonstrated to be an effective strategy for inducing behavioural change in people at risk for HIV infection (Fox et al. 1987; Kalichman 1998). Many public health professionals believe that individuals who test positive for HIV should be reported to the public health authorities by name to provide accurate information on which to determine the prevalence of infection in specific communities. Among those groups with a high prevalence of HIV, however, the concern has been that confidentiality cannot be maintained and naming could therefore result in disclosure to insurance companies and others who might take actions harmful to the individual. The issue of testing has now become even more complicated by the development of an effective treatment regimen that can delay onset of AIDS and extend life. For the treatment regimen to be most effective, however, treatment should begin before the onset of clinical AIDS. Thus the ethical rationale for promoting testing, especially to identify those who would benefit from treatment but do not suspect that they are infected, is even stronger than in the past.
The HIV/AIDS epidemic has also emphasized the ethical issues surrounding the rights of women to be able to protect themselves from HIV infection. The greatest HIV risk for many women, especially in developing countries, is their husband. The plight of the commercial sex worker not only raises the question of the forcing of women and men to become commercial sex workers in many societies (and in high-prevalence areas, assuring that they will become infected), but also the question of whether it is ethical to test commercial sex workers because the disclosure that they are infected causes an ethical dilemma for them. Should they stop working as a commercial sex workers and put themselves and their family back into poverty or continue to put their customers at risk?
The availability of an effective, but expensive treatment regimen raises the issue of the disparity between wealthy developed countries in which the prevalence of HIV/AIDS is low and the developing countries in which the vast majority of HIV/AIDS cases occur. HIV-infected individuals in the wealthy countries have access to effective treatment, whereas infected individuals in developing countries, which have 95 per cent of the infected persons, do not. What is the ethical responsibility of the drug companies and the wealthy nations to provide both the drugs and the health professionals capable of administering the drug regimen correctly?
Among many health professionals in the field of HIV/AIDS the philosophy has developed that the correct approach is to give the highest priority to protecting the rights of the infected. This policy has been called the policy of ‘exceptionalism’ to emphasize the idea that HIV/AIDS is different from other diseases and therefore should be treated as an exception and thus not be subject to the usual public health strategies for control of transmission. Given the magnitude of the epidemic and its continued rapid spread, particularly in countries least equipped to control the epidemic, it is time to recognize that although HIV/AIDS is an exceptionally severe disease associated with stigmatization, the public health principle of protecting the uninfected while doing the least possible harm to the infected should be observed. Only in this way will it be possible to control the HIV/AIDS epidemic.
Intervention and prevention strategies
HIV/AIDS prevention has progressed along two independent but related tracks (Mann and Kay 1991). One track is biomedical and uses research into the mechanisms of HIV and the human immune response to develop therapeutic and preventive technologies. The second track centres on individual and collective human behaviours to prevent HIV transmission. Efforts to prevent and control the pandemic of HIV/AIDS began worldwide as early as 1981. The dynamics of the epidemic, scientific progress, the experience of different population groups, and the variety of responses by a wide range of organizations in order to cope with the pandemic have provided considerable information about prevention strategies. A central theme emerges within each of these areas: HIV prevention requires efforts at the individual, community, and societal levels. Only a combination of all three will be sufficiently powerful to combat the pandemic.
An individual’s capacity to change risky behaviour is strongly influenced by community and societal ‘norms’. Therefore prevention activities should recognize the need to involve the community. Traditional public health approaches provide a framework to deal with the individual and the community, but should include promoting social awareness and concrete action.
Policy and structure for HIV/AIDS prevention
HIV/AIDS prevention has evolved through a number of stages since the beginning of the epidemic. The initial stage was characterized by programmes that focused on individual risk and responsibility through provision of information, awareness-raising campaigns, and sometimes delivery of fear-invoking messages. As the epidemic progressed, it was clear that effective messages needed to focus on sustaining safe behaviour. Greater understanding of the cultural and social dimensions of behaviour were crucial to developing more sophisticated responses aimed at facilitating community support for changed behaviour (Malcolm and Dowsett 1998). According to Jonathan Mann’s conclusive study (Mann and Tarantola 1996), four policy approaches to HIV/AIDS prevention have evolved into preventive strategies, as outlined below.
Epidemiological information approaches
During the period of early response (1981–1984), individual risk reduction became codified as the central goal (WHO 1988). Epidemiological studies quickly identified the modes of HIV transmission even before the discovery of the virus, and identified the specific behaviours associated with increased risk of HIV spread (Tarantola 1995). In the early stages of the AIDS epidemic in the United States (Watney 1996), it was apparent to epidemiologists that gay and bisexual men were disproportionately affected. Accordingly, they were described as a risk group for AIDS together with other disproportionately affected people, including haemophiliacs and injecting drug users. The aim of defining such risk groups was to implement prevention activities, as well as to mobilize adequate and appropriate resources for those in greatest need.
Two factors initially led to the use of the term ‘risk groups’ in the early years of the epidemic. First, the mass media unfortunately used the phrase in such a way that named risk groups were seen to pose a risk to the rest of the population. This approach contributed to stigmatization and misunderstanding. Second, an exclusive focus on risk groups implied that HIV was not a risk for those who were not defined as members of risk groups. Public health professionals recognized that this promoted stigmatization and so they placed emphasis on risk behaviours facilitating the transmission of HIV. Thus the concept of risk behaviours became increasingly used in public health, and in HIV/AIDS education and prevention policies around the world. The shift from ‘risk groups’ to ‘risk behaviours’ suggested that everyone was at equal risk from HIV. However, the global epidemiology of HIV clearly demonstrates that HIV is not an ‘equal opportunity’ disease. Moreover, exclusive attention to risk behaviour increased the difficulty of prevention, since those at greatest risk could no longer be named, and therefore their entitlement to services and support was difficult to ascertain.
In the first period of response to HIV/AIDS, epidemiological information about risk formed the basis of informational campaigns, including some of the most aggressive and large-scale public information efforts ever undertaken in public health. Information campaigns based on risk activities were used worldwide as well as in developing countries.
The social intervention approach
A comprehensive programme to provide individuals with messages, education, and specific services that would help promote and support the process of individual behavioural change was initiated in the mid-1980s. Remarkable public health efforts, delivered through specific services and activities, were undertaken to support individual risk reduction. Overall this approach was guided by a three-part prevention programme model developed as part of the World Health Organization (WHO) Global AIDS Strategy (Mann and Tarantola 1996). Two of the three parts involved traditional public health ideas and one was innovative. Based on prior public health efforts to change various personal behaviours, HIV prevention programmes were designed to provide information and education, along with the health and social services needed to promote and support the recommended behavioural changes. These services included condom distribution, HIV testing and counselling, treatment for other STDs, needle exchange, treatment for injecting drug users, and the provision of safe blood and blood products.
A unique element in HIV prevention programmes was the recommendation that discrimination against persons with HIV/AIDS be prevented (WHO 1998). Field experience had demonstrated that the fear of profound personal and social consequences led those most likely to be infected to avoid participating in HIV prevention programmes, or even to be tested. Accordingly, stigmatization was identified as a tragic and counterproductive result of the epidemic that interfered with control efforts. Thus for the first time in public health’s history prevention of discrimination against infected people became an integral part of a strategy to control an epidemic of an infectious disease.
Many of those who became infected had been labelled as socially undesirable before the beginning of the HIV epidemic. They included homosexual men and injecting drug users in developed countries (Watney 1996). In the developing world, they were commercial sex workers, injecting drug users, homosexual men, street children, prisoners, illegal migrants, and indigenous minority groups. It has become evident that the almost exclusive focus on individual risk reduction was too narrow to deal concretely with the social realities of the wide group of individuals at risk worldwide.
Another approach has considered the social influences on personal behaviour. This effort has taken many forms, from incorporating social context into sexual behaviour research, to proposals that pay greater attention to the influence of social norms and human development on individual behaviour. The WHO guidelines (1988) focused on the individual, but called for pragmatic societal support at the national level. Most official AIDS programmes are organized within institutions with a public health mandate, but even non-governmental efforts emphasize information and education to change individual behaviour. Thus WHO called for the establishment of national control programmes in all the countries to provide policies, education campaigns, and health and social services, as well as a central co-ordination and technical support to all involved partners.
The empowerment approach
Vulnerability is the converse of empowerment (Mann and Tarantola 1996). Vulnerability itself implies the extent to which individuals are incapable of making and implementing free and informed decisions about their life. A person who is genuinely able to make free and informed decisions is least vulnerable and most empowered; the person who is ill-informed, or unable to make informed decisions freely and carry them out is most vulnerable.
Personal vulnerability (Harvard School of Public Health 1993) to HIV/AIDS involves both cognitive and behavioural dimensions. Cognitive factors involve informational needs about HIV/AIDS, sexuality, and services for reducing vulnerability to infection. Behavioural factors can be considered in two overlapping categories: personal characteristics and personal skills. Personal characteristics include emotional development, perception of risk, attitudes towards risk-taking, personal attitudes towards sex and sexuality, and history of sexual and substance abuse. Personal skills are required to cope with risky situations such as the need to negotiate safe sexual practices, seek treatment for substance abuse, and practice safe sexual and drug-injecting behaviours.
Programmatic vulnerability has been defined broadly in terms of three major prevention elements:
information and education
health and social services
non-discrimination towards people with HIV/AIDS.
Efforts to reduce pragmatic vulnerability have principally involved ensuring and strengthening the availability and success of key programme elements, the quality and content of each element, and the process through which the elements are designed, implemented, and evaluated.
Research and empirical observation (Catania et al. 1990) have demonstrated that societal factors profoundly influence and condition personal behaviour. To be effective it is necessary to influence the relevant societal factors (Frank 1995). For example, inferior economic and social status limits the ability of many women to refuse unwanted or unprotected sexual intercourse, regardless of how much they know about AIDS or wish to adopt recommended individual risk-reduction practices. Similarly, it has become evident that HIV/AIDS programmes are created within, and therefore are constrained by, the larger society. Thus in many countries, governmental refusal to inform the public about condom use for HIV prevention or allow harm-reduction strategies for injecting drug users severely handicaps their national AIDS programme.
The concept of societal vulnerability builds upon the insight that collective, societal factors strongly influence both personal vulnerability and pragmatic vulnerability. Societal vulnerability focuses directly on the contextual factors which define personal and pragmatic vulnerability, such as governmental structure, gender relationships, attitudes towards sexuality, religious beliefs, and poverty, all of which influence the capacity to reduce personal vulnerability to HIV, both directly and as mediated through programmes.
There have been several attempts to relate the causes of the HIV/AIDS epidemic to issues of income growth and distribution. The World Bank (1997) studied the influence of aggregate-level social conditions on the size of the epidemic in 72 developing countries. The analysis demonstrated that both low income and unequal distribution of income are strongly associated with high HIV infection rates. The authors concluded that the widespread poverty and unequal distribution of income that typify underdevelopment promote the spread of HIV. They therefore asserted that rapid and fair distribution of economic growth would slow the HIV epidemic. Gender-related factors, such as the ratio of males to females in urban centres and the gap between adult male and female literacy rates, also influence the epidemic. In addition, the accelerated labour migration, rapid urbanization, and cultural modernization that often accompany economic growth also facilitate the spread of HIV. For developing countries, the AIDS epidemic exacerbates the poverty and social inequality that promote the epidemic, thus creating a vicious cycle. Policy-makers who understand these links have the opportunity to break this cycle.
Empirical experience suggests that poverty is an important contextual issue in HIV/AIDS prevention; as poverty increases so does social vulnerability to HIV (O’Shaughnessy 1994). Poverty forces people, especially women, to engage in high-risk activities, creates a lack of disposable income for purchasing condoms, and limits access to health services and HIV prevention programmes. In addition, large gaps between the highest and lowest socio-economic strata (relative poverty) generate the conditions in which people adopt survival strategies which amplify their vulnerability to becoming HIV-infected (Panos Institute 1990). Such survival strategies involve providing unprotected sexual intercourse in return for money, lodging, food, or other necessities. The relationship of HIV/AIDS and poverty is a particularly strong example of the consistent global relationship between socio-economic status and health status (World Bank 1993). One of the most effective HIV prevention strategies would be reduction or elimination of poverty and the gap between the rich and the poor.
The ultimate aim of vulnerability reduction is to expand people’s capacity to exert control over their own health. By identifying the larger social issues that constrain or promote this ability, contextual analysis stresses the need for positive and synergistic interaction between the individual services, programmes and other collective initiatives, and the social environment.
The human rights approach
Insight into the relationship of human rights to HIV vulnerability has been obtained through two lines of evidence (Mann 1995). First, a meta-analysis of the evolving HIV epidemics in countries around the world has revealed that people who are marginalized, stigmatized, and discriminated against have become, over time, those at the highest risk of HIV infection. Thus HIV is primarily a problem of people who live at the margins of society.
The second source of insight about human rights and HIV prevention is the detailed analysis of limits and failures in prevention programmes. For example, married and monogamous women who receive the normal benefits of HIV prevention programmes (information, education, access to testing and counselling, and condom availability) may nevertheless still be at risk of HIV infection. The recommendation to reduce the number of sexual partners as part of a risk-reduction approach fails in the real world for several reasons. First, women’s risk is related to their male sexual partner’s behaviour. Second, multiple sexual partnerships may be the only route of access to resources (for school, financial credit, or jobs). Third, women often lack control over their sexual relationships. In marriage, without legal recourse or legal rights to property, the pervasive threat of physical violence or divorce may totally disempower a woman, even if she knows about AIDS, that condoms are available, and that her husband is HIV-infected.
Therefore the central problem for HIV infection among women cannot be solved with risk-reduction approaches such as information campaigns or condom distribution systems. Since the central issue is the inferior role and subordinate status of women, the disadvantage created by society cannot be redressed through information and education or HIV-specific health and social services alone. To the extent that women’s human rights and dignity are not respected, society creates and enhances their vulnerability to HIV and more generally to ill-health (Cook 1995). Many other groups also suffer vulnerability to HIV infection due to discrimination, including gay and lesbian people worldwide, commercial sex workers, and adolescents, whose competence and voice are rarely acknowledged in any meaningful way.
However, the concepts and language of modern human rights can be used to describe vulnerability and the societal meaning and impact of the epidemic’s evolution (Mann et al. 1994). Guaranteeing human rights offers an extremely powerful approach to reduce marginalization, stigmatization, and discrimination. Indeed, the relationship of human rights to public health has been evolving rapidly in the context of HIV/AIDS. The HIV/AIDS epidemic has emphasized to public health professionals that human rights are an integral part of human well being.
Prior to 1986, none of the industrialized nations funded AIDS prevention and care programmes in the developing world. Part of the responsibility for this delay lies with the United Nations system, which was slow to identify the AIDS pandemic as an important global problem (Finlay et al. 1992). However, the delay also reflected confusion within official development assistance agencies concerning the extent and seriousness of the AIDS pandemic, uncertainties about how to proceed in funding for AIDS prevention and care activities, and concern that AIDS would divert funding from existing health priorities and projects.
Global leadership and advocacy for HIV/AIDS prevention and control have been critical ingredients to progress. The WHO Global AIDS Strategy was established in 1986 to alert the world to HIV/AIDS. It was supported by official development assistance agencies, which transfer funds from the industrialized donor countries to international agencies and governmental and non-governmental organizations concerned with AIDS programmes in developing countries.
International funding for AIDS research, treatment, care, and programme management grew steadily from 1986 through the end of the decade. Total funding increased from US$200 000 in 1986, to about US$59 million in 1987, to over US$212 million in 1990 following the creation of the WHO Global Programme on AIDS. The total official development assistance funding assigned to AIDS increased rapidly through 1990. The annual rate of increase was 127 per cent between 1987 and 1988 but then declined to 4–11 per cent during 1990 to 1993 (Laws 1996).
The decline in international AIDS financing between 1991 and 1994 was explained by two main factors. First, the official development assistance agencies responded to increased competing demands for resources, which were compounded by economic recession. Second, donors’ commitment to subsidizing HIV/AIDS programmes began to erode, which perhaps reflected scepticism about the efficiency and impact of past efforts to control the global HIV/AIDS pandemic.
In the early stages of the global mobilization, donors channelled most of their resources through WHO. However, donors have become dissatisfied with this funding strategy, which relies largely on multilateral channels. Consequently, there has been a major shift towards bilateral funding since 1990. Industrialized countries provided grants to specific recipient countries.
From 1986 to 1994, the World Bank awarded an increasing number of loans to developing countries for HIV/AIDS prevention and care, totalling approximately US$565 million. By mid-1995, 49 World-Bank-supported projects devoted partly or wholly to HIV/AIDS were in operation in 35 countries. The apparent shift from government grant money to government and private loans for HIV/AIDS programmes reflects two main factors. First, developing countries are turning to the World Bank as grant money is both becoming more difficult to obtain from other sources and is not increasing parallel to the growth of HIV/AIDS prevention and care needs in severely affected areas. Second, the World Bank has recognized the negative impact of HIV/AIDS on developing economies and is responding more favourably to loan requests for HIV/AIDS programmes than in the late 1980s.
Since 1989, the overall demand for development assistance has increased significantly. The combination of a shrinking pool of donor countries, rising demands for aid, and frustration by official development assistance agencies about the ability of developing countries to demonstrate progress towards the goals of development have led to a complex state of donor fatigue. Clear evidence of donor fatigue was revealed when the summit of heads of governments on AIDS, held in Paris in December 1994 with the aim of mobilizing significant international funds in support of the Global AIDS Strategy, did not succeed in generating the expected resources.
The Joint United Nations Programme on HIV/AIDS (UNAIDS) was established in 1995 to consolidate six organizations of the United Nations system (UNAIDS 1995). UNAIDS serves as the global AIDS programme for its six cosponsors (United Nations Children’s Emergency Fund, United Nations Development Programme, United Nations Population Fund, United Nations Educational, Scientific, and Cultural Organization, the WHO, and the World Bank). At the country level, UNAIDS can be described as the joint action and collective resources of its cosponsoring organizations, co-ordinated by the UNAIDS central office in Geneva and UNAIDS country and regional staff.
The main focus of UNAIDS has been on strengthening national capacity for an expanded response. Thus UNAIDS works with government departments and ministries, people living with HIV, communities affected or threatened by the epidemic, non-governmental and community-based organizations, academic institutions, and the private sector, as well as bilateral and intergovernmental organizations. It has four mutually reinforcing roles: policy development and research, technical support, advocacy, and co-ordination.
However, while the HIV epidemic in most developing countries continues unabated, international funding for support of a global AIDS strategy is shrinking. Advocacy alone cannot overcome existing poor health and social infrastructures in ultra-low gross domestic product countries. Daunted by the lack of resources and other structural impediments to organized care for HIV, many in positions of power still shy away from the topic of care. The increase in the number of people in need of care globally, the limited capacity of health services, and the recognition that care needs go well beyond clinical therapies to psychosocial support and community-based care for the burgeoning populations with HIV/AIDS, especially in sub-Saharan Africa, can and does frighten some donors and governments.
Sexual behaviour change and condom use
The sexual behaviour of populations is still largely unknown in most of the countries in the world. The levels and forms of risk behaviour must be assessed in order to understand the potential for widespread HIV transmission, to monitor changes in behaviour, and to determine the magnitude of ongoing prevention needs. In only about 40 countries (UNAIDS/WHO 1998) have there been general population surveys which looked at sexual behaviour changes over time. Many factors contribute to the lack of behavioural data, including cost, reluctance to conduct such surveys for political or religious reasons, lack of personnel or technical resources, and concerns about the reliability and validity of data that might be gathered.
Without a complete picture of behavioural risk in a population it is difficult to prioritize prevention activities. Certain situations contribute disproportionately to HIV transmission, for example, commercial sex. Knowledge of important variations in sexual behaviours within populations is rare, so prevention efforts are not necessarily focused on those with the greatest risk behaviour. Populations, such as men having sex with men and sexually active youth, may be missed or under-represented, and other vulnerable groups such as factory workers, military personnel, fishermen, and truck drivers may not be included (Mills et al. 1998).
HIV risk reduction ranges from abstinence to the practice of safer sex. Both abstinence and safer-sex intervention reduce HIV sexual risk behaviours, but safer-sex interventions may be especially effective with sexually experienced adolescents and may have a longer-lasting impact (Jemmott et al. 1998). Abstinence intervention implies delaying sexual intercourse or reducing its frequency. Safer-sex intervention always stresses condom use. In the reproductive age population (Polacsek et al. 1999) the use of a condom with regular sex partners depends on their partners’ reaction to condoms, perceived partner risk, length of relationship, sterility, cohabitation, perceived vulnerability to HIV infection, and perceived peer norms about condom use. When information, skills training, and services are made available to young people, they are often more likely to make use of them than their elders. Since sexual activity in youth is high, it is necessary to ensure that teenagers can exercise the right to protect themselves from risky sex.
Thailand has demonstrated a successful prevention programme, evolving from information through a public campaign (Phoolcharoen et al. 1998). A national survey of sexual behaviour conducted in 1990 found that a high proportion of men had sex outside marriage. The results of this survey were widely publicized, and government officials and the general public were made aware that their country was vulnerable to rapid spread of HIV. Although prostitution is illegal in Thailand, the state worked to set up partnerships with brothel owners to address the problem. With government support, brothel owners and sex workers enforced a policy of 100 per cent condom use in brothels to reduce transmission in what the evidence suggested was a focal point for infection in Thailand. In parallel, mass media campaigns encouraged respect for women and discouraged men from engaging in commercial sex, while young women were offered better educational and vocational opportunities to keep them out of the sex industry. These energetic campaigns have resulted in a reduction in risky sexual behaviours and decreased spread of the virus (Nelson et al. 1996).
In AIDS-endemic areas, it is complicated to promote safer sex by HIV-infected persons who have uninfected partners. The specific intervention (Norman et al. 1998) for these persons should emphasize communication skills and rehearsal of serostatus disclosure as well as risk-reduction discussions. Voluntary HIV testing with counselling in the community turned out to be a key strategy to support safer sex among those at risk, and has been made available in many countries.
The importance of injecting drug use in many epidemics of HIV in many countries has been sporadically recognized. Although many discrete and interconnected epidemics of HIV infection among injecting drug users in various parts of the world have been well described, their role as epicentres for the wider epidemic, and their impact on whole communities, have undergone little formal investigation. Accordingly, public health responses to the specific issue of HIV among injecting drug users in developing countries have been slow to emerge. This issue is seen as secondary to the reduction of drug demand and the control of supply (Crofts et al. 1998).
For drug users in most parts of Asia, the move from smoking or inhaling to injecting of heroin, once established, is rarely seen to revert (Wu et al. 1996). The major factor for the transition to injecting is thought to be economic, but has not been deeply investigated. Among injecting populations, the sharing and reuse of injecting equipment is subject to many pressures, including scarcity and relative cost of equipment, social organization of drug use, and custom. A substantial proportion of injecting drug users have shared their needles and syringes with little effective cleaning beforehand. In many countries needles and syringes can be purchased from pharmacies, but carrying a needle can be seen as incriminating evidence of drug use and injecting drug users may therefore be reluctant to procure clean injecting equipment from a pharmacy.
The prevalences of HIV infection among injecting drug users in the countries of Southeast Asia are among the highest reported anywhere. Few signs exist of any major decrease in this epidemic. Various countries of this area have experienced the most rapid diffusion of HIV infection among injecting drug users found anywhere in the world. Within a period of approximately 12 months many areas reached a prevalence of HIV infection among injecting drug users of 40 per cent or greater (Sarkar et al. 1993).
The current consensus is that there is an urgent need to reduce the HIV risks associated with drug injecting to the users themselves, to their sex partners and children, and to society. One way to do this is through a comprehensive prevention programme based on the principle of ‘harm reduction’. Just as in the case of sexual HIV transmission, the prevention of transmission through drug injecting calls for a package whose components operate simultaneously. A recent comparison of cities with high and low HIV prevalences in injecting drug users showed that those with success in averting a drug-user epidemic had three features in common. First, they used community outreach or peer education to reach and educate drug users, including those who would not otherwise receive HIV/AIDS information. Second, they ensured that drug users had cheap and easy access to sterile syringes through pharmacies or needle-exchange programmes. Third, they all started their prevention programmes early on, before HIV prevalence had risen past a critical point (UNAIDS/WHO 1998).
The best way for an individual to avoid HIV infection through drug injecting is to stop injecting. Many countries have responded to the HIV epidemic among injecting drug users by expanding drug treatment services. Evaluation of studies on the effectiveness of psychosocial interventions in reducing the risk of HIV infection in injecting drug users has showed that the existing programmes were not acceptably effective (Gibson et al. 1998).
Needle-exchange programmes, when part of a comprehensive harm-reduction approach, represent another way of combating the sharing of equipment. These programmes provide drug users with clean needles and syringes in exchange for used ones. HIV infection rates have been shown in various studies to be over three times lower in injecting drug users who participate in needle-exchange programmes than in those who do not (Des Jarlais 1993). The success of needle-exchange programmes is by no means limited to industrialized countries. There are currently programmes in the cities of Santos and Salvador in Brazil, the Nepalese capital of Kathmandu, and the Akha hill-tribe communities of northern Thailand, to name but a few. Most of these remain relatively small scale, however. Only a handful of countries, such as Australia and a few nations in Northern Europe, come close to meeting the demand.
Preventing vertical transmission
The overwhelming majority of children who die of AIDS acquired the infection from their mothers before or around the time of birth, or through breast milk. Most of these children with HIV live in the developing world. There are four major reasons for the difference between developed and developing countries. These are the higher proportion of HIV transmitted heterosexually, breast-feeding practices, limited access to drugs for reducing mother-to-child transmission, and the health infrastructure in each country.
Since it first became clear that HIV could be transmitted through breast milk, very few infected women in industrialized countries have chosen to breast feed their children, and so transmission of infection at the nipple is negligible. In developing countries, however, between one-third and one-half of all HIV infections in young children are acquired through breast milk. There are several reasons for this. First, more than nine out of ten HIV-positive women in developing countries do not know their HIV status. Secondly, a woman may choose to breast feed even if she knows about her infection, and knows she might pass it on through breast milk. Breast feeding protects the infant against a range of other infections, is convenient, is approved by most cultures, and is free. By choosing artificial feeding a woman may avoid passing on HIV, but where the water supply is unsafe she may expose her child to other deadly diseases. There are some governments, such as in Thailand, which distribute free or subsidized artificial milk to such women. But in many countries the critical first step remains to provide counselling, voluntary HIV testing, and information about safe infant feeding to all pregnant women.
In 1994, a study (Connor et al. 1994) demonstrated that giving an antiretroviral drug to women during pregnancy and delivery and to the infant after birth could reduce HIV transmission from mother to child by as much as two-thirds. This quickly became common practice in industrialized countries, but is not currently feasible in developing countries. The regimen is difficult to administer, involving regular drug taking over several months and an intravenous drip during delivery. The cost is around US$1000 per pregnancy.
A trial in Thailand (Shaffer et al. 1999) recently demonstrated that a short course of antiretroviral pills given to pregnant women during the last weeks prior to and during labour successfully reduced the rate of vertical transmission during pregnancy and delivery by half. The women were also given safe alternatives to breast milk and did not breast feed. A recent study that gave nevirapine at a cost of US$4 to US$8 at onset of labour and to the infants at birth also resulted in a 50 per cent reduction in HIV transmission to the infant (Guay et al. 1999).
The cost of treating HIV-positive women with a short course of antiretrovirals in late pregnancy and around the time of delivery compares well with that of many other health interventions. The cost-effectiveness varies according to the level of infection in a country. A study in Tanzania suggested that counselling, testing, and short-course antiretrovirals for pregnant women would cost under US$600 per HIV infection that is averted. This translates into around US$30 per healthy year of life gained. In high-prevalence areas of Thailand the cost per infection averted is around US$2800. With the use of nevirapine, the cost is reduced much further.
In Western countries it was shown that a positive attitude among health workers towards HIV testing contributes to lower anxiety among their pregnant patients (Boyd et al. 1999). However, in most developing countries the poorly developed health infrastructure has limited counselling for HIV-infected pregnant women. In some countries (Cartoux et al. 1999) this has been replaced by group pre- and post-test counselling. At the public health level, group pretest counselling can be easily integrated into existing sessions of antenatal care counselling routinely performed by the current clinic staff.
Control of sexually transmitted diseases
The presence of STD has repeatedly been shown to enhance the probability of HIV infection. Thus it seems reasonable that effective treatment of STDs would reduce the incidence of HIV, especially in very sexually active populations. Studies on the efficacy of STD treatment programmes on reducing the incidence of HIV have been conflicting. Grosskurth et al. (1995), working in Tanzania where the prevalence of HIV was still relatively low at the time of the study, found that improved treatment of STDs, including an adequate supply of drugs, lowered the incidence of HIV although not the prevalence of STDs. Presumably the period during which individuals were actively infected with an STD was shortened. On the other hand, a study in the Rakai district of Uganda, a district with a very high prevalence of HIV, found that periodic mass treatment of all residents for STDs every 10 months had no impact on the incidence of HIV (Wawer et al. 1999). Presumably the higher prevalence of HIV in the Rakai district, which increased the likelihood of exposure to HIV for sexually active people, lessened the impact of concurrent STDs on risk of HIV infection, although the difference in the therapeutic approaches may also have played a role. Nonetheless, at the individual level, diagnosis and treatment of STDs probably does reduce the risk of HIV infection.
Universal precautions and postexposure prophylaxis
Health workers are at risk for HIV transmission. Therefore intensive surveillance for health facilities began in some countries in 1985 (Weiss et al. 1985). According to studies of HIV exposure in health workers, the risk of HIV infection is 0.32 per cent per percutaneous exposure, 0.09 per cent per mucous membrane exposure, and zero per intact skin exposure (Bell 1997).
According to a collaborative study including the United States, France, Italy, and United Kingdom, the three main determinants for percutaneous infection are obvious exposure to infected blood, terminal illness in the source patient, and administration of antiretroviral postexposure prophylaxis.
The best prevention for health workers who may be exposed to HIV-infected specimens from clients is ‘universal precautions’. This strategy imposes the burden of the increased cost of disposable and protective equipment that many developing countries cannot afford. However, even with universal precautions, there may be some accidental exposure to HIV-contaminated material. Thus postexposure antiretroviral prophylaxis is recommended for exposed health providers (Centers for Disease Control and Prevention 1995). In Western countries, a combination of three antiretroviral drugs has been recommended as prophylaxis for HIV-exposed health workers, but in most developing countries such a regimen is not affordable (Centers for Disease Control and Prevention 1996). Currently, postexposure antiretroviral prophylaxis has been accepted in coping with other causes of HIV exposure, such as after rape (Katz and Gerberding 1997). However, it is not recommended by any public health authority. HIV nosocomial transmission to patients may be a concern since there has been at least one report of transmission of HIV from health providers to their patients (Centers for Disease Control and Prevention 1991). The development of a DNA-sequencing capacity has facilitated the identification of the source of HIV transmission.
Vaccines have been demonstrated to be the most effective tool for controlling many infectious diseases. Unfortunately, development of a vaccine against HIV has not been accomplished as of the year 2002. The basic problem is that no natural immunity following infection has been demonstrated. There are thousands of variants of the virus and evoking an effective immune response appears to be specific to each variant. Inducing protection against the thousands of HIV variants is not feasible. Further, recovery with elimination of the virus and development of immunity, as occurs in most diseases for which a vaccine has been developed, has not been demonstrated in HIV except, possibly, in a few children. On the other hand, resistance to infection has been demonstrated in a few individuals and provides hope that resistance to infection can be induced, although not necessarily using the classic strategies for vaccine development (Imagawa and Detels 1991; Detels et al. 1994, 1996a,b).
In the late 1980s and early 1990s the scientific community was discouraged from working on vaccine development because of the formidable problems that had to be overcome. By the middle to late 1990s the urgent need for a vaccine overcame the previous discouragement and the National Institute of Allergy and Infectious Diseases, and private foundations such as the Rockefeller Foundation, provided funds to spur development of a vaccine. As a result many groups are now working on the development of a vaccine, but it will probably be well into the twenty-first century before an effective vaccine becomes available.
Summary and forecast
Twenty years after the identification of AIDS and 16 years after the discovery of HIV, more is known about the virus and the natural history of the disease than for the majority of other agents discovered many decades ago. And progress has been made towards control of HIV/AIDS. The incidence is declining in most developed countries, and several developing countries, including Thailand, Cambodia, and Uganda, have been successful in slowing the spread of HIV. Although science has not discovered a cure, the prognosis for HIV-infected and AIDS patients has improved and the prospects are good that the disease will become a chronic disease allowing an infected individual to lead a normal life, albeit requiring continuing medication. There has been considerable progress in the search for an effective vaccine although much more needs to be done before a vaccine is a reality.
But the epidemic continues. HIV continues to spread rapidly in both India and China, which contain 38 per cent of the world’s population. In China there is evidence that HIV is now spreading to the heterosexual population. HIV continues to spread quickly in Africa and Asia, particularly in South and Southeast Asia, and has affected the longevity and productivity of the most severely affected countries in these regions. The challenge to control the epidemic is particularly great in the developing world where the epidemic is now concentrated, because of the lack of resources, health infrastructure, and trained manpower.
The epidemic can be controlled but to do so will require a better understanding of the sexual behaviour of populations at risk, improving the status of women (particularly in Asia and Latin America), increasing awareness of HIV/AIDS and prevention strategies, adoption of strategies to prevent transmission, including promotion of safer sex and condom use, decreasing the rate of sexual mixing, promoting ‘risk-free’ testing, and developing the political will to mandate more funds and a more effective health infrastructure to combat its spread. A key factor will be to reduce/eliminate stigmatization of groups at risk of HIV infection, HIV-infected individuals, and AIDS patients. The cycle of poverty promoting HIV transmission that in turn creates more poverty must be broken. Finally, the developed countries of the world need to recognize their responsibility to assist the developing countries to overcome this epidemic. If progress can be made in addressing these issues then the prospect for control of HIV/AIDS is good even in the absence of an effective vaccine.
Ainsworth, M. and Over, M. (1992). The economic impact of AIDS: shocks, responses and outcomes, Technical Working Paper 1, Report Number 11276, Population, Health and Nutrition Division, Africa Technical Department, World Bank, Washington DC.
Altman, D. (1988). Legitimacy through disaster. In AIDS: the burdens of history (ed. D. Fox and E. Fee). University of California Press, Berkeley, CA.
Anzala, A., Nagelkerke, N., Bwayo, J.J., et al. (1995). Rapid progression to disease in African sex workers with human immunodeficiency virus type I infection. Journal of Infectious Diseases, 171, 686–9.
Bayer, R. (1991). Public health policy and the AIDS epidemic: an end to HIV exceptionalism? New England Journal of Medicine, 324, 1500–4.
Bayer, R. (1996). Societal and political impact of HIV/AIDS. In AIDS in the world II (ed. J.Mann and D. Tarantola), pp. 117–28. Oxford University Press.
Bell, D. (1997). Occupational risk of human immunodeficiency virus infection in healthcare workers: an overview. American Journal of Medicine, 102 (Supplement 5B), 9–15.
Bennett, A. and Sharpe, A. (1996). AIDS fight is skewed by federal campaign exaggerating risks. Wall Street Journal, 1 May, pp. A1, A6.
Bharat, S., Singhanetra-Renard, A., and Angleton, P. (1998). Household and community response to HIV/AIDS in Asia: the case of Thailand and India. AIDS, 12 (Supplement B), S117–S122.
Boyd, F., Simpson, W., Johnstone, F., Goldberg, D., and Hart, G. (1999). Uptake and acceptability of antenatal HIV testing, British Journal of Midwifery, 7, 151–6.
Bryson, Y.J. (1995). HIV clearance in infants—a continuing saga. AIDS, 9, 1373–5.
Bryson, Y.J. (1996). Perinatal HIV-1 transmission: recent advances and therapeutic interventions. AIDS, 10 (Supplement 3), S33–S42.
Bryson, Y.J., Pang, S., Wei, L.S., Dickover, R., Diagne, A., and Chen, I.S. (1995). Clearance of HIV infection in a perinatally infected infant. New England Journal of Medicine, 332, 833–8.
Bulterys, M., Nzabihimana, E., Chao, A., Bugingo, G., Musekera, J., Saah, A., and Van de Perre, P. (1993). Long-term survival among HIV-1 infected prostitutes. AIDS, 7, 1269.
Cartoux, M., Sombie, I., Van de Perre, P., Meda, N., Tiendrebeogo, S., and Dabis, F. (1999). Evaluation of 2 techniques of HIV-pre-test counselling for pregnant women in West Africa. International Journal of STD and AIDS, 10, 199–201.
Catania, J., Kegeles, S., and Coates, T. (1990). Toward an understanding of risk behavior: an AIDS risk reduction model. Health Education Quarterly, 17, 53–72.
Centers for Disease Control and Prevention (1991). Transmission of HIV infection during an invasive dental procedure—Florida. Morbidity and Mortality Weekly Report, 40, 377–81.
Centers for Disease Control and Prevention (1992). 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. Morbidity and Mortality Weekly Report, 41, 1–17.
Centers for Disease Control and Prevention (1995). Case–control study of HIV seroconversion in health-care workers after percutaneous exposure to HIV-infected blood—France, United Kingdom, and United States, January 1988–August 1994. Morbidity and Mortality Weekly Report, 44, 929–33.
Centers for Disease Control and Prevention (1996). Provisional Public Health Service recommendations for chemoprophylaxis after occupational exposure to HIV. Morbidity and Mortality Weekly Report, 45, 468–72.
Chamberland, M.E. (1998). Surveillance for transfusion-transmitted viral infections in the United States. Biologicals, 26, 85–8.
Chesseman, S., Havlir, D., and McLaughlin, M. (1995). Phase I/II evaluation of nevirapine alone and in combination with zidovudine for infection with human immunodeficiency virus. Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology, 8, 141–51.
Connor, E.M., Sperling, R.S., Gelber, R., et al. (1994). Reduction of maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. Pediatric AIDS Clinical Trials Group Protocol 076 Study Group. New England Journal of Medicine, 331, 1173–80.
Cook, R. (1995). Gender, health and human rights. Health and Human Rights, 4, 350–4.
Crofts, N., Reid, G., and Deany, P. (1988). Injecting drug use and HIV infection in Asia. AIDS, 12 (Supplement Bf), S69–S78.
Des Jarlais, D. (1993). Protective effects of syringe exchange against blood-borne virus infection among injecting drug users. First National Conference on Human Retroviruses and Related Infections, p. 178.
Detels, R., Liu, Z., Hennessey, K., et al. (1994). Resistance to HIV-1 infection. Journal of Acquired Immune Deficiency Syndromes, 7, 1263–9.
Detels, R., Mann, D., Carrington, M., et al. (1996a). Persistently seronegative men from whom HIV-1 has been isolated are genetically and immunologically distinct. Immunology Letters, 51, 29–33.
Detels, R., Mann, D., Carrington, M., et al. (1996b). Resistance to HIV infection may be genetically mediated. AIDS, 10, 102–4.
Detels, R., Muñoz, A., Peng, Y., et al. (1997). Early versus deferred zidovudine monotherapy: impact on AIDS-free time and survival in the Multicenter AIDS Cohort Study. Antiviral Therapy, 2, 21–9.
Detels, R., Muñoz, A., McFarlane, G., et al. (1998). Effectiveness of potent antiretroviral therapy on time to AIDS and death in men with known HIV infection duration. Multicenter AIDS Cohort Study Investigators. Journal of the American Medical Association, 280, 1497–1503.
Detels, R., Tarwater, P., Phair, J.P., et al. (2001). Effectiveness of potent antiretroviral therapies on the incidence of opportunistic infections before and after AIDS diagnosis. AIDS, 15, 347–55.
Dickover, R.E., Dillon, M., Leung, K.M., et al. (1998). Early prognostic indicators in primary perinatal human immunodeficiency virus type 1 infection: importance of viral RNA and the timing of transmission on long-term outcome. Journal of Infectious Diseases, 178, 375–87.
Finlay, J., Mann, J., and Tarantola, D. (1992). Funding the Global AIDS Strategy. In AIDS in the world (ed. J. Mann, D.J.M. Tarantola, and T.W. Netter), pp. 511–35. Harvard University Press, Cambridge, MA.
Fleming, A., Carballo, M., Fitzsimmons, D., Bailey, M., and Mann, J. (1988). Global impact of AIDS. Alan R. Liss, New York.
Fox, R., Odaka, N.J., Brookmeyer, R., and Polk, B.F. (1987). Effect of HIV antibody disclosure on subsequent sexual activity in homosexual men. AIDS, 1, 241–6.
Frank, J. (1995). The determinants of health: a new synthesis. Current Issues in Public Health, 1, 42–65.
Gao, F., Bailes, E., Robertson, D.L., et al. (1999). Origin of HIV-1 in the chimpanzee Pan troglodytes. Nature, 397, 436–41.
Gardner, L., Brudnage, J., McNeil, J., et al. (1992). Predictors of HIV-1 disease progression in early and late-stages patients: the US Army natural history cohort. Journal of Acquired Immune Deficiency Syndromes, 5, 782–93.
Gibson, D., McCusker, J., and Chesney, M. (1998). Effectiveness of psychosocial interventions in preventing HIV risk behaviour in injecting drug users. AIDS, 12 (8), 919–29.
Global Health Council (2000). http://www.globalhealth.org/issues/hivaids.html
Goedert, J., Kessler, C., Aledort, L., et al. (1989). Prospective study of human immunodeficiency virus type 1 infection and the development of AIDS in subjects with haemophilia. New England Journal of Medicine, 321, 1141–8.
Gottlieb, M.S., Schroff, R., Schanker, H.M., Weisman, J.D., Fan, P.T., Wolf, R.A., and Saxon, A. (1981). Pneumocystis carinii pneumonia and mucosal candidiasis in previously healthy homosexual men: evidence of a new acquired cellular immunodeficiency. New England Journal of Medicine, 305, 1425–31.
Grosskurth, H., Mosha, F., Todd, J., et al. (1995). Impact of improved treatment of sexually transmitted diseases on HIV infection in rural Tanzania: randomised controlled trial. Lancet, 346, 530–6.
Guay, L.A., Musoke, P., Fleming, T., et al. (1999). Intrapartum and neonatal single-dose nevirapine compared with zidovudine for prevention of mother-to-child transmission of HIV-1 in Kampala, Uganda: HIVNET 012 randomised trial. Lancet, 354, 795–802.
Harvard School of Public Health (1993). Towards a new health strategy for AIDS: a report of the global AIDS policy coalition. Francois-Xavier Bagnoud Center for Health and Human Rights, Boston, MA.
Hira S.K, Ngandu, N., Wadhawan, D., et al. (1990). Clinical and epidemiological features of HIV infection at a referral clinic in Zambia. Journal of Acquired Immune Deficiency Syndromes, 3, 87–91.
Ho, D.D., Neumann, A.U., Perelson, A.S., Chen, W., Leonard, J.M., and Markowitz, M. (1995). Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature, 373, 123–6.
Ho, H.T. and Hitchcock, M.J. (1989). Cellular pharmacology of 2′, 3′-dideoxy-2′,3′-didehydrothymidine, a nucleoside analog active against human immunodeficiency virus. Antimicrobial Agents and Chemotherapy, 33, 844–9.
Imagawa, D. and Detels, R. (1991). HIV-1 in seronegative homosexual men. New England Journal of Medicine, 325, 1250–1.
Institute of Medicine (1995). HIV and the blood supply: an analysis of crisis decisionmaking (ed. L.B. Leveton, H.C. Sox, Jr, and M.A. Stoto). National Academy Press, Washington, DC.
Jaffe, H., Darrow, W., Echenberg, D., et al. (1985). Acquired immunodeficiency syndrome in a cohort of homosexual men: a six-year follow-up study. Annals of Internal Medicine, 103, 210–14.
Jemmott, J., Jemmott, L., and Fong, G. (1998). Abstinence and safer sex HIV risk-reduction interventions for African American adolescents: a randomized controlled trial. Journal of the American Medical Association, 279, 1529–36.
Kalichman, S.C. (1998). Influencing HIV transmission risk. Focus, 13, 1–4.
Katz, M. and Gerberding, J. (1997). Postexposure treatment of people exposed to the human immunodeficiency virus through sexual contact or injection-drug use. New England Journal of Medicine, 336, 1097–100.
Lanjewar, D.N., Anand, B.S., Genta, R., et al. (1996) Major differences in the spectrum of gastrointestinal infections associated with AIDS in India versus the West: an autopsy study. Clinical Infectious Diseases, 23, 482–5.
Laws, M. (1996). International funding of the Global AIDS Strategy: official development assistance. In AIDS in the world II (ed. J. Mann and D. Tarantola), pp. 375–89. Oxford University Press.
Lee, C., Phillips, A., Efford, J., Miller, E., Bofill, M., Griffiths, P., and Kernoff, P. (1989). Natural history of human immunodeficiency virus infection in a haemophilic cohort. British Journal of Haematology, 73, 228–34.
Lemp, G., Payne, S., Neal, D., Temelso, T., and Rutherford, G.W. (1990). Survival trends for patients with AIDS. Journal of the American Medical Association, 263, 402–6.
Levine, C., Michaels, D., and Back, S. (1996). Orphans of the HIV/AIDS pandemic. In AIDS in the World II (ed. J. Mann and D. Tarantola), pp. 278–86. Oxford University Press.
Lewis, J.S., Terriff, C.M., Coulston, D.R., and Garrison, M.W. (1997). Protease inhibitors: a therapeutic breakthrough for the treatment of patients with human immunodeficiency virus. Clinical Therapeutics, 19, 187–214.
Lifson, A., Buchbinder, S., Sheppard, J., Maule, A.C., and Miller, A.C. (1991). Long-term immunodeficiency virus infection in asymptomatic homosexual and bisexual men with normal CD4+ lymphocytes counts: immunologic and virologic characteristics. Journal of Infectious Diseases, 163, 959–65.
Lindan, C.P., Allen, S., Serufilira, A., et al. (1992). Predictors of mortality among HIV-infected women in Kigali, Rwanda. Annals of Internal Medicine, 116, 320–8.
MacNeil, J. and Anderson, S. (1998). Beyond the dichotomy: linking HIV prevention with care. AIDS, 12 (Supplement 2), S19–26.
Malcolm, A. and Dowsett, G. (1998). Prevention in practice: summation of guiding principles, partners in prevention: international cases studies of effective health promotion practice in HIV/AIDS, pp. 58–66. UNAIDS, Geneva.
Mann, J. (1995). Public health and human rights. Current Issues in Public Health, 1, 97–101.
Mann, J. and Kay, K. (1991). Confronting the pandemic: the World Health Organization’s global programme on AIDS, 1986–1989. AIDS, 5 (Supplement 2), S221–9.
Mann, J. and Tarantola, D. (1996). From epidemiology to vulnerablity to human rights. In AIDS in the world II (ed. J.Mann and D. Tarantola), pp. 427–76. Oxford University Press.
Mann, J., Gostin, L., Gruskin, S. et al. (1994). Health and human rights. Health and Human Rights, 1, 6–22.
Mellors, J.W., Muñoz, A., Giorgi, J.V., et al. (1997). Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV-1 infection. Annals of Internal Medicine, 126, 946–54.
Mills, S., Ungchusak, K., Srinivasan, V., Utomo, B., and Bennett, A. (1998). Assessing trends in HIV risk behaviors in Asia. AIDS, 12 (Supplement B), S79–86.
Mulder, D., Nunn, A., Kamali, A., Nakiyingi, J., Wagner, H.U., and Kengeya-Kayondo, J.F. (1994a). HIV-1 incidence and HIV-1 associated mortality in a Ugandan rural population. Lancet, 343, 989–90.
Mulder, D., Nunn, A., Wagner, H.U., Kamali, A., and Kengeya-Kayonda, J.F. (1994b). HIV-1 incidence and HIV-1 associated mortality in rural Ugandan population cohort. AIDS, 8, 87–92.
Muñoz, A., Wang, M.C., Bass S. et al. (1989). Acquired immunodeficiency virus type 1 (HIV-1) seroconversion in homosexual men. American Journal of Epidemiology, 130, 530–9.
Muñoz, A., Sabin, C.A., and Phillips, A.N. (1997). The incubation period of AIDS. AIDS, 11 (Supplement A), S69–76.
Murray, C. and Lopez, A. (1996). The global burden of disease. Global burden of disease and injury series, WHO and Harvard School of Public Health, Vol. 2. World Bank, Washington, DC.
Naglekerke, N., Plummer, F., Holton D., Anzala, A.O., Manjii, F., Ngugi, E.N., and Moses, S. (1990). Transition dynamics of HIV disease in a cohort of African prostitutes: a Markov model approach. AIDS, 4, 743–7.
N’Galy, B., Ryder, R., Bila, K., et al. (1998). Human immunodeficiency virus infection among employees in an African hospital. New England Journal of Medicine, 319, 1123–7.
National Research Council (1993). Monitoring the social impact of AIDS in the United States. National Academy Press, Washington, DC.
Nelson, K.E., Beyrer C., Eiumtrakol, S., Khamboonruang, C., and Celentano, D. (1996). Changes in sexual behavior and a decline in HIV infection amoung young men in Thailand. New England Journal of Medicine, 335, 297–303.
Newell, M.L., Gray, G., and Bryson, Y.J. (1997). Prevention of mother-to-child transmission of HIV-1 infection. AIDS, 11 (Supplement A), S165–S172.
Norman, L., Kennedy, M., and Parish, K. (1998). Close relationships and safer sex among HIV-infected men with haemophilia. AIDS Care, 10, 339–54.
Over, M. (1992). Macroeconomic impact of AIDS in sub-Saharan Africa. Technical Working Paper 3, Population, Health and Nutrition Division, Africa Technical Department, World Bank, Washington, DC.
O’Shaughnessy, T. (1994). Beyond the fragments: HIV/AIDS and poverty, issues in global development 1. World Vision, Australia Research and Policy Unit, Melbourne.
Panos Institute (1990). Triple jeopardy: women and AIDS. Panos Dossier, London.
Phoolcharoen, W., Ungchusak, K., Sittitrai, W., and Brown, T. (1998). Thailand: lessons from a strong national response to HIV/AIDS. AIDS, 12 (Supplement B), S123–35.
Polacsek, M., Celentano, D., O’Campo, P., and Santelli, J. (1999). Correlates of condom use stage of change: implications for intervention. AIDS Education Preview, 11, 38–52.
Richman, D.D., Fischl, M.A., Grieco, M.H., et al. (1987). The toxicity of azidothymidine (AZT) in the treatment of patients with AIDS and AIDS-related complex. A double-blind, placebo controlled trial. New England Journal of Medicine, 317, 192–7.
Sarkar, S., Das, N., Panda, S., et al. (1993). Rapid spread of HIV among injecting drug users in northeastern states of India. Bulletin on Narcotics, 16, 17–23.
Sathapatayavong, B., Tansuphaswadikul, S., Kantiphong, P., Pornochaipoolthavee, S., and Chuchottaworn, C. (1997). In Prevalence of disseminated MAC in Thai AIDS patients. Abstract 5281, 20th International Congress of Chemotherapy, Sydney.
Seeley, J., Kajura, E., Bachengana, C., et al. (1993). Extended family and support for people with AIDS in a rural population in southwest Uganda: a safety net with holes? AIDS Care, 5, 117–22.
Shaffer, N., Chuachoowong, R., Mock, P., et al. (1999). Short-course zidovudine for perinatal HIV-1 transmission in Bangkok, Thailand: a randomised controlled trial, Lancet, 353, 773–80.
Simonsen, J., Plummer, F., Ngugi, E., et al. (1990). HIV infection among lower socioeconomic strata prostitutes in Nairobi. AIDS, 4, 87–92.
Subhash, K.H., Gregory, D., and Sirisanthana, T. (1998). Clinical spectrum of HIV/AIDS in the Asia–Pacific region. AIDS, 12 (Supplement B), S145–54.
Tarantola, D. (1995). Structural and environmental influences on HIV risk behavior and vulnerability. Presented at USAID 3rd HIV/AIDS Prevention Conference, Washington, DC.
UNAIDS (1995). Joint United Nations Programmes on HIV/AIDS: strategic plan 1996–2000. Background document, Second Meeting of the Programme Coordinating Board. Document UNAIDS/PCB(2)/95.3, UNAIDS, Geneva.
UNAIDS (2000). AIDS epidemic update. http://www.unaids.org/hivaidsinfo/documents/html
UNAIDS/WHO (1998). Global HIV/AIDS epidemic update. http://www.unaids.org/hivaidsinfo/documents/html
United States Bureau of the Census (1997). Recent HIV seroprevalence levels by county. Research Note 23, Health Studies Branch, International Programs Center, Population Division, Washington DC.
Watanabe, M.E. (1999). China confronts AIDS: international help needed to stop the spread. Scientist, 13, 1–6.
Watney, S. (1996). ‘Risk groups’ or ‘risk behaviors’? In AIDS in the world II (ed. J. Mann and D. Tarantola), pp. 431–2. Oxford University Press.
Wawer, M.J., Sewankambo, N.K., Serwadda, D., et al. (1999). Control of sexually transmitted diseases for AIDS prevention in Uganda: a randomised community trial. Rakai Project Study Group. Lancet, 353, 525–35.
Wei, X., Ghosh, S.K., Taylor, M.E., et al. (1995). Viral dynamics in human immunodeficiency virus type 1 infection. Nature, 373, 117–21.
Weiss, S., Saxinger, W., Rechtman, D., et al. (1985). HTLV-III infection among health care workers: association with needle-stick injuries. Journal of the American Medical Association, 254, 2089–93.
Weniger, B.G., Limpakarnjanarat, K., Ungchusak, K., et al. (1991). The epidemiology of HIV infection and AIDS in Thailand. AIDS, 5 (Supplement 2), S71–85.
Whitmore-Overton, S., Tillett, H., Evans, B., and Allardice, G. (1993). Improved survival from diagnosis of AIDS in adult cases in the United Kingdom and bias due to reporting delays. AIDS, 7, 415–20.
Wiktor, S.Z., Ekpini, E., Karon, J.M., et al. (1999). Short-course oral zidovudine for prevention of mother-to-child transmission of HIV-1 in Abidjan, Cote d’Ivoire: a randomized trial. Lancet, 353, 781–5.
World Bank (1993). World development report. Investing in health. World Bank, Washington, DC.
Ainsworth, M. and Over, M. (1997). AIDS: a challenge for government. World Bank policy research report. In Confronting AIDS: public priorities in a global epidemic, pp. 27–32. Oxford University Press, New York.
WHO (1988). Guidelines for the development of national AIDS prevention and control programmes, Technical Series Document 1, Global Programme on AIDS, WHO, Geneva.
WHO (1998). Avoidance of discrimination against HIV-infected people and persons with AIDS. Resolution 41.24. 41st World Health Assembly, WHO, Geneva.
WHO (1999). World health report 1999. Making a difference. WHO, Geneva. (www.who.int/whr/1999/en/pdf/whr99.pdf)
Wu, Z., Detels, R., Zhang, J., et al. (1996). Risk factors for intravenous drug use and sharing equipment among young male drug users in southwest China. AIDS, 10, 1017–24.
Wu, Z., Rou, K., and Detels, R. (2001). Prevalence of HIV infection among former commercial plasma donors in rural eastern China. Health Policy and Planning, 16, 41–6.