The Status and Trends of the Global HIV/AIDS Pandemic Satellite Symposium of the XI International Conference on AIDS was jointly organized by the AIDS Control and Prevention (AIDSCAP) Project of Family Health International, the François-Xavier Bagnoud Center for Health and Human Rights of the Harvard School of Public Health and the Joint United Nations Programme on HIV/AIDS (UNAIDS). This two-day Symposium, held on 5-6 July 1996 in Vancouver, Canada, resulted from an extraordinary response to The Status and Trends of the HIV/AIDS Epidemics in Africa Workshop held in Kampala, Uganda, in December 1995 prior to the IX International Conference on AIDS and STD in Africa. With specific recommendations for urgent action to prevent the further spread of HIV in Africa, the Kampala workshop report raised issues that cut across HIV/AIDS epidemics in developing countries globally and provided a successful model for future action-oriented international HIV/AIDS symposia. The workshop report, released during the Kampala conference and quickly disseminated internationally by the workshop organizers, became available for immediate programmatic reorientation, planning and implementation by policy makers and program managers working on HIV/AIDS-related issues in Africa and around the world. Seizing on the impact of the Kampala initiative, the workshop's co-chairs, Daniel Tarantola of Harvard's François-Xavier Bagnoud Center and Peter Lamptey of AIDSCAP, quickly convened a larger Symposium on The Status and Trends of the Global HIV/AIDS Pandemic at the XI International Conference on AIDS in Vancouver, Canada, which they also co-chaired. The objectives for the Vancouver Symposium included: reviewing the current status and trends of the global HIV/AIDS pandemic, including the epidemiological and behavioral patterns; identifying the specific data needs for monitoring and forecasting; and producing a consensus report on the global pandemic and on current as well as projected trends for the epidemics in various parts of the world. Ten regional teams composed of 50 leading epidemiologists, public health and development specialists (see List of Participants at end of this report), each nominated by an international steering committee, were chosen to collect and analyze data and information on the status and trends of HIV/AIDS in their region prior to the Symposium. Regional working group sessions held by each team on site provided the basis for summary presentations to the Symposium participants. Finally, a plenary discussion was held on each region at which the Symposium participants reached conclusions and made their final recommendations for this report. Released within 24 hours of the Symposium's conclusion, the provisional Vancouver report was amended during the Vancouver conference to the version herein. A day later, the Vancouver Symposium organizers began planning the next Status and Trends of the Global HIV/AIDS Pandemic Symposium to be convened at the 4th International Conference on AIDS in Asia and the Pacific, to be held in Manila, Philippines, in October 1997. Globally, the HIV/AIDS pandemic continues to sweep across continents: the number of estimated adult HIV infections worldwide has more than doubled since 1990 from 10 million to a mid-1996 total of 25.5 million. Composed of distinct epidemics, each with its own features and force, the pandemic is disproportionately impacting the developing world. HIV infections, however, are leveling off and even decreasing in some populations. HIV incidence has declined in young women in Uganda, young men in Thailand and in gay men in the U.S., Australia, Canada and western Europe. From the beginning of the pandemic until mid-1996, an estimated 27.9 million people worldwide were infected with HIV. Of these, 14.9 million were men (58 percent) and 10.5 million were women (42 percent). The majority of HIV infections -- 26 million (93 percent) -- have occurred in developing countries. The largest numbers of HIV-infected individuals were in sub-Saharan Africa, totaling 19 million (68 percent of the global total), and in South and Southeast Asia, totaling 5 million (18 percent of the global total). The number of HIV-infected people in South and Southeast Asia is now more than twice the total number of those infected in the entire industrialized world. Worldwide, 5.8 million people (4.5 million adults and 1.3 million children), 75 percent of all those with AIDS, are estimated to have died from AIDS. Sub-Saharan Africa, representing about 60 percent of the world's total HIV infections, accounts for almost 90 percent of the current 13.3 million HIV infections in adults and adolescents in Africa. The rates of newly acquired HIV infections are highest in the 15- to 24-year-old group among both females and males in most of sub-Saharan Africa. Of the 3 million HIV-infected infants born in the world with HIV infection since the beginning of the pandemic, over 90 percent have been born in Africa. Many of these children typically develop AIDS and die within a few years. Eighteen countries in the region have at least 100,000 people living with HIV. Central and East African countries have 37 percent of all current HIV infections on the continent. A second group of countries in southern Africa contributes about 15 percent to the total number of adults and adolescents living with HIV in the region. In other sub-Saharan countries -- mostly in west and central Africa -- HIV epidemics are currently passing through their intermediate stage where between 1 and 10 percent of women attending urban antenatal clinics are HIV-infected. In contrast to the increasing spread of HIV-1, the prevalence of HIV-2 has remained rather stable in West Africa, perhaps the result of the higher transmissibility of HIV-1 compared to HIV-2. Urban and trading centers generally show substantially higher prevalence of HIV infection than rural areas; however, rates of HIV infection in some rural populations have increased steadily. Open conflicts, environmental degradation, natural disasters and low-intensity wars also have led millions of Africans to leave their homes and, in some situations, to turn to survival strategies that have increased the practice of unsafe sex. Migration and urbanization also have led to high concentrations of predominantly male communities and increased participation in commercial sex. Demographic surveys in several countries have already noted significant increases in infant and child mortality. Projections for Zambia and Zimbabwe indicate that AIDS may increase child mortality rates nearly threefold by the year 2010. Due to high levels of fertility, populations will generally continue to grow, but critical deficits will affect the economically active age groups. Asia, home to more than 60 percent of the world's adult population, presents epidemiology and HIV prevalence estimates that are extremely diverse, ranging from countries with low prevalence (Mongolia, DPR Korea) to countries with high prevalence (Cambodia, Myanmar and Thailand). HIV is spreading mostly through heterosexual contact. Infected men probably outnumber infected women by a factor of 3 to 1 or more, and gender inequality and the frequent practice of men visiting sex workers have strongly influenced spread of HIV. Sharing of needles among injecting drug users (IDUs) also played a significant role early in the epidemics, particularly in the Golden Triangle region (from Thailand and Vietnam, across southern China, to Myanmar and Manipur State in India) and in northern Malaysia. Thailand has an estimated three-quarters of a million people living with HIV. Yet there is evidence that Thailand's active multi-sectoral prevention efforts are taking effect, as HIV infection levels in military conscripts have dropped from 3.6 percent in 1993 to 2.5 percent in 1995. Pediatric HIV infection is difficult to assess regionally, but an estimated 6,400 children are infected annually in Thailand, making up approximately 10 percent of the new infections of HIV. Evidence in India suggests rapid, extensive and uncontrolled spread of HIV in many parts of the country, and HIV seroprevalence is high in the South and West. In Bombay, prevalence went from 2 to 3 percent in sexually transmitted disease (STD) clinic attendees before 1990 to 36 percent in 1994. Injecting drug use has been a problem in Manipur State, where prevalence reached 60 percent by 1992. Evidence suggests an estimated 2 to 5 million HIV infections nationwide in mid-1996. Low use of condoms and high rates of sexually transmitted disease continue to be a major problem in India, threatening to multiply exponentially the spread of HIV. In Cambodia data indicate that the current extensive HIV epidemic started during the late 1980s or early 1990s and is predominantly occurring among heterosexuals with multiple sex partners. Myanmar has one of the most serious epidemics in the region, with an estimated half a million people infected with HIV by 1996. In Malaysia HIV infection levels in IDUs have grown rapidly from 0.1 percent in 1988 to 20 percent in 1994. In Vietnam, evidence shows the HIV epidemic to be growing rapidly, with high levels in IDUs in treatment (32 percent in 1992-95), and increasing levels among young men and women in the south. China, because of its size and rapid changes in social and sexual behavior, represents another major potential focus of the epidemic in Asia. The majority (about 70 percent) of reported HIV infections and AIDS cases have been among IDUs in Yunnan Province, but infections are believed to be increasing among heterosexuals in southern China, especially in the areas surrounding Hong Kong. An estimated 10,000 persons were infected with HIV in China at the end of 1993, growing to 100,000 by the end of 1995. HIV transmission may be starting in the heterosexual population in Laos. Bangladesh, Indonesia, Nepal and Sri Lanka show high levels of other STDs, implying a strong possibility for extensive HIV spread. In Hong Kong, Japan, Mongolia and the Republic of Korea, extensive spread has not been documented. No cases of AIDS or HIV have been reported in DPR Korea or Bhutan. In the Philippines the epidemic shows slower growth, and in Singapore, HIV infection levels in sex workers have been growing quite slowly. In Latin America and the Caribbean the spread of HIV/AIDS has been slower than in other regions, but the pandemic is well-established and some Caribbean countries report AIDS incidence rates among the highest in the world. Dominant modes of transmission vary from one country to the next: mainly through homo- and bisexual contacts in some countries, to epidemics connected to injecting drug use in others, to still others primarily determined by heterosexual transmission. Epidemiological evidence signals a rapid shift of new infections to younger ages, particularly to young people between 15 and 24 years old. The number of new HIV infections in Mexico, Guatemala, Belize, El Salvador, Honduras, Nicaragua, Costa Rica, Panama, Cuba, Dominican Republic, Haiti and Puerto Rico continues to rise, reflecting increasing HIV/AIDS incidence and accelerated heterosexual transmission. Haiti is of particular importance because, perhaps alone in the region, it represents a relatively mature epidemic. HIV prevalence is particularly high among sex workers, STD clinic attendees and tuberculosis (TB) patients. High rates of HIV prevalence are found among pregnant women aged 14 to 24. In the English-speaking Caribbean, the male-to-female ratio of new AIDS cases has fallen dramatically over the past 10 years, under 2 men to 1 woman in 1994. Women aged 15 to 19 now have higher annual incidence rates than men of the same age. AIDS is the leading cause of death among young men in some Caribbean countries, while pediatric AIDS cases have been steadily rising and now account for 5 percent of all new cases. The extremely low incidence of HIV infection through contaminated blood represents a partial success story for the Caribbean region. HIV infections and AIDS cases in South America are rising steadily. Brazil accounts for 75 percent of AIDS cases reported and is followed by the Andean Region (15 percent) and the Southern Cone (10 percent). Sexual transmission of HIV accounts for 74 percent of all infections (51 percent homo/bisexual and 23 percent heterosexual), injecting drug use 19 percent (although recent data in Brazil suggests that the HIV transmission through injecting drug use seems to be leveling off) and 7 percent for blood and vertical transmission and undocumented cases. The impact of HIV/AIDS on morbidity and mortality is being seen in major urban centers, such as in São Paulo, Brazil, where AIDS is now the leading cause of death in women of reproductive age. North America has seen the HIV epidemic slow in recent years as new infections start to level off, largely due to the decline in sexual transmission between men as a result of behavior change. Nevertheless, HIV prevalence in gay men remains high on the continent. In the United States, HIV prevalence among IDUs has decreased. AIDS cases related to heterosexual contact represent an increasing proportion of newly diagnosed cases in North America. Since the start of the epidemic from 1 to 1.5 million cumulative HIV infections have occurred in North America, and HIV infection has been one of the major causes of death for individuals between the ages of 25 and 44. Among men in this age group, it was the leading cause of death in the U.S. and the second leading cause of death in Canada in 1994. In the same year, HIV infection was the third leading cause of death among 25- to 44-year-old women in the U.S. An estimated 12,000 children in the U.S. are living with HIV, although AIDS incidence among children under 13 has declined annually since 1990. Although there has been an overall slowing in AIDS incidence, there has been substantive shift in the populations affected. In 1995 AIDS incidence was 6.5 times greater for blacks and 4 times greater for Hispanics than for whites, 20 percent of persons diagnosed with AIDS were women, and 15 percent were infected heterosexually. AIDS among prisoners was 7 times the rate of the non-incarcerated population, and AIDS was the second leading cause of death among prisoners. In Europe an estimated 450,000 adults were living with HIV in western Europe at the end of 1993, with an annual incidence of around 40,000 since 1990. Over the past 2 to 3 years, AIDS incidence appears to have stabilized in several countries in northwestern Europe and condom use increased markedly, particularly for the most sexually active population groups. In contrast, countries in southwestern Europe show no indication of AIDS leveling off. Transmission of HIV through injecting drug use continues to play a major role in the dynamics of the epidemic. It is responsible for the majority of AIDS cases in some of the western countries with highest incidence (Spain and Italy) and is strongly associated with AIDS cases occurring among heterosexual adults and among children in these countries. In central and eastern Europe (with the exception of Romania) and central Asia, the HIV/AIDS epidemic is much more recent and AIDS incidence much lower than in western Europe. In some countries, a rapid spread of HIV is indicated, mainly linked with injecting drug use. In Poland and the Federal Republic of Yugoslavia, where IDUs account for the largest proportion of cases, the incidence of AIDS is rising rapidly. Before 1990, most AIDS cases were diagnosed in men who have sex with men (MSM). Since 1990, however, IDUs account for the highest proportion of yearly diagnosed cases in the region (43 percent of adult and adolescent cases in 1995). The shift in transmission patterns is accompanied by an increase in the proportion of female cases, which rose from 11 percent in 1986 to 20 percent in 1995. The vast majority of children have been infected through mother-to-child transmission, and HIV prevalence in pregnant women has been much higher in urban than in rural areas. The epidemic among children is dominated by the epidemic in Romanian hospitals, which was detected in 1989 and accounts for over 50 percent of the 6,060 pediatric cases reported in the European region. Another, though much smaller, epidemic among children in hospitals occurred in the Russian Federation in the late 1980s. The most worrisome information coming from STD surveillance arose recently from the independent republics of the former Soviet Union. Substantial increases in syphilis rates have been seen since 1990 in several of these states. In the North and South Pacific around 7,400 cases of AIDS had been reported by the end of 1995, with over 7,000 of them in Australia and New Zealand, where the major pathway of HIV transmission (about 85 percent of HIV infections) has been through sexual contact between men. This pattern also has been reflected in the French Territories of New Caledonia and French Polynesia. The HIV epidemic in Papua New Guinea (PNG) has developed more recently and appears to involve a major component of heterosexual transmission; by the end of 1994 PNG had an estimated 4,000 adults living with HIV, overtaking Australia on a per-capita basis to give the highest prevalence in the region. The incidence of AIDS has reached a plateau in Australia, and appears to be declining in New Zealand, due to the drop in the rate of sexual transmission of HIV between men that mostly occurred ten years earlier. In Australia, and to a lesser extent New Zealand, high rates of STDs other than HIV in indigenous people have led to mounting concern about the potential for a major heterosexual epidemic of HIV infection in these populations, among whom the rate of HIV diagnosis has increased in the past six years. In conclusion, the HIV/AIDS pandemic is as powerful as ever: HIV continues to spread in the industrialized world, where, increasingly, it affects people who, for reasons of race, sex, behavior or social and economic status, have lesser access to services. From a global perspective, the pandemic disproportionately affects the developing world, where the needs for effective prevention and care are escalating. But the pandemic has now become immensely complex. It has become fragmented and is now a mosaic composed of a multitude of epidemics, which can be distinguished on the basis of: predominant modes of transmission; geographic focus; HIV sub-types; age, sex, socioeconomic or behavioral characteristics of the populations most affected; rapidity of or potential for HIV spread; stage of maturity and, in some communities and countries, declining HIV incidence. As the HIV epidemics pursue their course, the social, economic and demographic impacts of HIV/AIDS in particular in the developing world are likely to exacerbate the burden on individuals, communities and countries, which will challenge the stability of entire regions. Current evidence of the effectiveness of HIV prevention and recent progress achieved in the development of new therapies provide the scientific basis on which an expanded response to growing prevention and care needs can be built. Existing epidemiological surveillance systems are inadequate to monitor the HIV epidemics effectively. Creativity and sustained effort must be applied to collect and analyze data that better reflect and will help understand, predict and prevent the dynamic spread of HIV in vulnerable populations worldwide. In mid-July 1996, an estimated 21.8 million adults and children worldwide were living with HIV/AIDS, of whom 20.4 million (94 percent) were in the developing world. Of the adults, 12.2 million (58 percent) were male and 8.8 million (42 percent) were female.1 Close to 19 million adults and children (86 percent of the world total) were living with HIV/AIDS in sub-Saharan Africa and in South and Southeast Asia. Worldwide during 1995, 2.7 million adult HIV infections occurred in adults (averaging more than 7,000 new infections each day). Of these, about 1 million (an average of nearly 3,000 new infections per day) occurred in Southeast Asia and 1.4 million infections (close to 4,000 new infections per day) were in sub-Saharan Africa. The industrialized world accounted for about 55,000 new HIV infections in 1995 (nearly 150 new infections per day; about 2 percent of the global total). In 1995 approximately 500,000 children were born with HIV infection (about 1,400 per day); of these children 67 percent were in sub-Saharan Africa, 30 percent in South and Southeast Asia, and over 2 percent in Latin America and the Caribbean. From the beginning of the pandemic until mid-1996, an estimated 27.9 million people worldwide have been infected with HIV. The largest numbers of individuals ever infected with HIV were in sub-Saharan Africa, totaling 19 million (68 percent of the global total), and in South and Southeast Asia, totaling 5 million (18 percent of the global total). Since the beginning of the pandemic, the majority of HIV infections-26 million (93 percent)- have occurred in the developing world. The number of HIV-infected people in South and Southeast Asia is now more than twice the total number of infected people in the entire industrialized world. The global cumulative number of HIV infections among adults has more than doubled since the beginning of the decade, from about 10 million in 1990 to almost 25.5 million by mid-1996. Of these, 14.9 million were men (58 percent) and 10.5 million were women (42 percent). More than 6 million adults have developed AIDS from the beginning of the pandemic to July 1996, and of these 4.5 million (close to 75 percent) were in sub-Saharan Africa; 0.4 million were in Latin America and the Caribbean (7 percent); and 0.75 million were in North America, Europe and North and South Pacific combined (12 percent). In South and Southeast Asia, where the pandemic gained intensity more recently, it is estimated that 330,000 adults have developed AIDS. Of the 1.6 million children with AIDS, the majority -- 1.4 million (85 percent) -- were in sub-Saharan Africa. By July 1996, 5.8 million people (4.5 million adults and 1.3 million children), 75 percent of all people with AIDS, are estimated to have died from AIDS worldwide. In summary, the HIV/AIDS pandemic is as powerful as ever. This report will show that the pandemic is now composed of distinct epidemics each with their own features and force, and disproportionately impacting on the developing world. The following sections of this report will show that as the HIV/AIDS epidemics within each region and country have become increasingly diverse and fragmented, they have created a multifaceted and devastating pandemic. The Scourge of AIDS Marches On Estimated Number of Persons Living with HIV/AIDS, July 1996 Click on this icon to see a detailed image (51K) Global Total: 21.8 million HIV-infected adults and children. Source: UNAIDS 6/96 1 For more details, see "The HIV/AIDS Situation in mid-1996, Global and Regional Highlights. Fact Sheet 1 July 1996," UNAIDS and WHO, Geneva, Switzerland. Sub-Saharan Africa By mid-1996, 13.3 million adults were living with HIV in sub-Saharan Africa, representing about 60 percent of the world's total. Three broadly defined geographic areas, which include countries with severe epidemics and others with epidemics at their intermediate stages, account for almost 90 percent of all current HIV infections in adults and adolescents in Africa. Within these three areas, 18 countries have at least 100,000 people living with HIV. In central/eastern Africa, Cameroon, Ethiopia, Kenya, Rwanda, Sudan, Uganda and Zaire have 37 percent of all current HIV infections on the continent. A similar proportion is contributed by a second group of countries in southern Africa: Botswana, Malawi, Mozambique, South Africa, Tanzania, Zambia and Zimbabwe. Finally, West African countries, including Burkina Faso, Côte d'Ivoire, Ghana and Nigeria, contribute about 15 percent to the total number of adults and adolescents living with HIV in Africa. In Kenya, Malawi, Rwanda, Tanzania, Uganda, Zambia, Zimbabwe (countries where HIV began to spread widely in the early 1980s), more than 10 percent of women attending antenatal clinics surveyed in urban areas have been found to be HIV-infected, with rates which may exceed 40 percent in some surveillance sites. In these populations, transmission of HIV occurs mainly through heterosexual contact, beginning in early teen years and peaking before age 25. Following similar patterns of spread and intensity, HIV epidemics have recently expanded in Botswana, Lesotho, Swaziland and South Africa. The observed high rates of HIV in women of reproductive age have resulted in high numbers of HIV-infected newborns. Of the 3 million HIV-infected infants born in the world with HIV infection since the beginning of the pandemic, over 90 percent have been born in Africa. Many of these children typically develop AIDS and die within a few years. In other sub-Saharan countries (mostly in west and central Africa) HIV epidemics are currently passing through their intermediate stage where between 1 and 10 percent of women attending urban antenatal clinics are HIV-infected. A few of these countries still have relatively low levels of HIV prevalence, but these have begun to rise in several countries such as Nigeria and Cameroon, which earlier had been relatively spared. HIV-2 is primarily found in West Africa, but HIV-2 infections also have been confirmed in African countries elsewhere, including Angola and Mozambique. The highest prevalence of HIV-2 infection is found in Guinea Bissau and in southern Senegal. In contrast to the increasing spread of HIV-1, the prevalence of HIV-2 has remained rather stable in West Africa. This is thought to be the result of the higher transmissibility of HIV-1 compared to HIV-2. The likelihood of transmission of HIV-1 through heterosexual intercourse is estimated to be about three times higher per exposure than for HIV-2. In addition, perinatal transmission rates of HIV-2 are reported significantly lower (less than 4 percent for HIV-2 compared with 25 to 35 percent for HIV-1). Under circumstances that are not yet fully understood, epidemics may suddenly explode, with rates of infection increasing several fold within only a few years, as has been observed recently in Botswana and South Africa. HIV prevalence in pregnant women in South Africa has grown dramatically. From 1993 to 1995, HIV prevalence increased from 4.3 to 11 percent, and from 9.6 to 18 percent, in the provinces of Free State and Kwazulu/Natal, respectively. Population mobility, patterns of sexual behavior, and societal factors are likely to influence the potential for such explosions. The presence of sexually transmitted diseases (STDs) suggests a marked risk of concurrent HIV infection for at least two reasons: (1) the modes of transmission of HIV and other STDs are similar; and (2) the role of STDs in facilitating the transmission of HIV has been clearly established. STDs are affecting young adults, especially women, with direct serious consequences. For women, these consequences include pelvic inflammatory disease, cervical cancer, infertility and postpartum endometritis. For infants, maternal STDs may lead to low birth weight, neonatal syphilis and gonococcal opthalmia. The lack of circumcision in males has been shown to add to the risk of acquiring STDs. The World Health Organization estimates that in 1995, 65 million new cases of curable STDs occurred in Africa. Populations Affected The transmission of HIV in adults and young people in sub-Saharan Africa occurs essentially through heterosexual contact. Rates of HIV infection among sex workers are now found as high as 80 percent in Nairobi and 55 percent in Abidjan. HIV antibody testing of blood donations remains incomplete in most countries in sub-Saharan Africa. Transfusions continue to play a role in the spread of HIV to those most likely to receive them: women of reproductive age and children. Within each country, HIV epidemics have progressed with different velocity in various population groups. Early in the evolution of the epidemics, urban populations and rural communities located along highways were more rapidly affected. Among them, young adults with multiple sexual partners acquired high rates of infection. Urban and trading centers generally continue to have substantially higher prevalence of HIV infection than rural areas. But, this pattern is by no means universal: population displacement, armed conflicts, proximity to highways or intense migration and population mobility for economic reasons influence strongly the spread of HIV. As a result of a combination of these factors, some rural communities of Kenya, Tanzania and Uganda have higher infection rates than those observed in neighboring urban populations. In some countries where HIV epidemics were initially found in urban areas, rates of HIV infection in some rural populations have increased steadily over recent years. In other countries, perhaps with poorer transport networks, this has not been the case. As epidemics mature, they tend to spread into younger populations, especially young women. The rates of newly acquired HIV infections are highest in the 15- to 24-year-old group among both females and males in most of sub-Saharan Africa. The peak of new infections occurs several years earlier in young women than in young men. In Masaka, Uganda, for example, HIV prevalence in 13- to 19-year-old females is over 20 times higher than in males of the same age. Most of the infections in 15- to 19-year-olds are in females, although as young men get older, their prevalence increases as well. Apart from possible biological factors, there are at least two reasons for the disproportionate risk of young women acquiring HIV infection early, including: (1) an earlier age of sexual initiation of girls (in Masaka, the median age at first sexual intercourse is 15 for females and 17 for males); and (2) the patterns of sexual mixing, wherein young women tend to have sex with older men in the context of marriage or in exchange for money or advantages, whereas young men tend to have sex with young women. But for many women, the major risk factor for HIV is the behavior of their spouse or regular sexual partner. Monogamous women are at a disadvantage in protecting themselves against HIV when spouses are engaged in high-risk behavior. Populations on the Move Major political, social and demographic changes have occurred in sub-Saharan Africa over the last few decades and have resulted in important population displacement, migration and rapid urbanization. The improvement of transportation and communication networks, the increased exchange of goods, and the creation of large-scale development programs have stimulated the movement of young men and women within and across national boundaries. Open conflicts, environmental degradation, natural disasters and low-intensity wars have also led millions of Africans to leave their places of residence and, in some situations, to turn to survival strategies that have increased the practice of unsafe sex. Consequences of political and civil unrest and subsequent population displacement have led to an increased spread in HIV transmission; populations displaced from Ethiopia, Mozambique, Rwanda and Liberia are examples. In addition, the movement of troops from West Africa to Angola and Mozambique has been linked to the spread of HIV-2 to these countries. Migration within countries, across borders, and urbanization (e.g., from rural areas to urban centers or industrial sites) have led to high concentrations of predominantly male communities and increased participation in commercial sex. Professional groups characterized by mobility, for instance, truck drivers, traders and military personnel, have also been associated with a higher risk of HIV infection. Population mobility facilitates the spread of STDs, including HIV. Economic development programs (the construction of highways, dams, and the creation of new industries or agriculture projects, for example) need to include an initial appraisal of the potential impact of these projects on the vulnerability of the labor force and the local population to HIV infection and other STDs. Measures to minimize this impact, such as reducing gender imbalance in the labor force, enabling workers to be joined by their families, allowing for regular contacts with distant spouses, and incorporating HIV/STD programs in development schemes, need to be built into the project design. But even with such initiatives, the sheer dynamic of transition towards increasingly urbanized society brings with it changing behavior mores that create new needs and present new opportunities for HIV transmission. All of these social and demographic changes need to be addressed by well-designed national and inter-country HIV/STD prevention programs based on epidemiological, behavioral and social determinants research. Burden of Disease Although the constantly growing HIV/AIDS care needs have already overwhelmed the coping capacity of urban health systems in hard-hit countries, demands for care will increasingly fall on poorly equipped and under-funded rural services, households and individuals. Already, 80 percent of hospital beds in an infectious disease hospital in Abidjan, Côte d'Ivoire, and 50 percent in a hospital in Kampala, Uganda, are occupied by people with HIV. Demographic surveys in several countries have already noted significant increases in infant and child mortality. Projections for Zambia and Zimbabwe indicate that AIDS may increase child mortality rates nearly threefold by the year 2010. Other estimates point to a more modest impact. In either case, due to high levels of fertility, populations will generally continue to grow, but critical deficits will affect the economically active ages. Studies in areas where 8 percent of the adult population is HIV-infected have measured a doubling of mortality due to HIV and a decrease of 5 years in life expectancy. Some HIV epidemics will have severe effects on the population age structure, indenting the population pyramid in young adults, the main contributors to social and economic development, but this may not occur in all areas. Successes in Prevention To date in sub-Saharan Africa, there has been a lack of rigorous evaluation of intervention strategies, especially for the behavioral interventions designed to reduce the sexual transmission of HIV. Without good behavioral, social and contextual data, however, it is difficult to attribute observed changes in HIV prevalence rates to specific program efforts. STD control programs, through early diagnosis, treatment and the promotion of safer sexual behavior, have been shown to reduce significantly the rates of STD infections. Successful programs have been documented in Zambia and Zimbabwe. In a research study in Mwanza, Tanzania, early treatment of STDs in a rural population has been associated with a 42 percent decline in the rate of newly acquired HIV infections. Emerging data also show substantial decline in some STDs. This important finding supports the revitalization of STD control programs benefiting from new approaches that have already been initiated in several sub-Saharan countries.
Hope that the number of new infections occurring may have decreased comes from studies of the epidemic in Uganda, a country with one of the older epidemics in Africa. A study of recent trends in HIV infection in women attending several antenatal clinics in Uganda shows significant declines in HIV prevalence. Between 1990-93 and 1994-95, overall HIV prevalence in pregnant women at sentinel sites decreased 29 percent (from 21 to 15 percent), and decreased 35 percent in both 15- to 19- and 20- to 24-year-olds. Since infection levels (prevalence) in this young age group reflect more recent patterns in new infections (incidence), these data suggest a substantial reduction in the incidence of HIV infection in young people over time.
Similar declines in HIV prevalence in young adults are reported from another study in the Masaka district in Uganda. These findings could indicate that the growth of the epidemic has been blunted, perhaps transiently, by behavioral changes resulting in decreased spread of HIV in younger age groups. Surveys of such populations suggest that behavior change might have led to these apparent declines; however, more rigorous qualitative and quantitative behavioral and social data will be required to help interpret these results. Notwithstanding these encouraging signs, new infections remain high, especially in young people. The combination of reductions in levels of infection and continuing evidence of new infections should provide additional impetus for enhancing prevention efforts. North Africa and the Middle East This region represents 22 countries ranging from Morocco in the west to Pakistan in the east. Information on HIV infection in the region is sparse. Information available from mandatory screening of blood donors indicates low HIV prevalence in these populations, except for Djibouti. The highest levels of HIV infection have been documented in Djibouti (9.3 percent in pregnant women and from 2 to 20 percent in STD patients). HIV prevalence among STD patients rose from 1.3 to 5 percent in Sudan; this pattern has also been seen in Yemen, Pakistan and the Syrian Arab Republic. Seventy-five percent of reported AIDS cases are from five countries in the region: Morocco, Sudan, Saudi Arabia, Tunisia and Djibouti. The future size and trends of the epidemic in this region are difficult to predict. There is anecdotal evidence of high STD rates and risk behaviors. The region is characterized by late introduction of the virus, the status of women in society, the highly stigmatizing nature of STDs, and the difficulty of conducting effective sexual health programs. This region includes Bangladesh, Bhutan, Brunei Darussalam, Cambodia, China, India, Indonesia, Hong Kong, Japan, DPR Korea, Republic of Korea, Laos, Malaysia, the Maldives, Mongolia, Myanmar, Nepal, the Philippines, Singapore, Sri Lanka, Thailand and Vietnam. It is home to over 60 percent of the world's adult population, hence what happens in the region will have a major impact on the global pandemic. The general epidemiology and estimated prevalence rates for these countries are extremely diverse, ranging from countries with low HIV prevalence rates (Mongolia, DPR Korea) to countries with high HIV prevalence (Cambodia, Myanmar and Thailand). There has been substantial variation in the timing and rate of growth of the epidemic. In some countries, e.g., Cambodia, India, Myanmar and Thailand, HIV spread has been extensive, with extremely rapid growth in some geographic areas. In others, e.g., DPR Korea and the Republic of Korea, the Philippines and Singapore, only limited spread has occurred to date and the rate of growth appears to be substantially lower. The epidemic in Thailand is among the best documented in the world, with an estimated three-quarter is of a million people living with HIV. Nationally, HIV prevalence among injecting drug users rose quickly in 1988 to approximately 35 percent. HIV among brothel-based sex workers rose from 3.5 percent in 1989 to 33 percent by late 1994. Infection levels in males at STD clinics grew from 0 percent to 8.6 percent over the same time period. HIV prevalence in women attending antenatal clinics has continued to rise steadily from 0 percent in 1989 to 2.3 percent in 1995. This trend is expected to continue for several years. However, there is evidence that prevention efforts are taking effect; HIV infection levels in military conscripts have dropped from 3.6 percent in 1993 to 2.5 percent in 1995. In India HIV seroprevalence is high in the south and west. For example, in Bombay prevalence went from 2 to 3 percent in STD clinic attendees before 1990 to 36 percent in 1994. HIV prevalence in sex workers rose from 1 to 51 percent between 1987 and 1993, and antenatal clinic women tested positive at a 2.5 percent rate in 1994. There is great geographical variation in India. HIV seroprevalence in the central, eastern and northern parts of the country are generally lower than in the rest of India. Studies among sex workers in Calcutta have shown a clear and consistently low prevalence of 1.2 percent. In Vellore rates among women attending antenatal clinics have been steady at 0.1 percent, although STD clinic rates there grew from 4 percent to 15 percent between 1993 and 1995. Injecting drug use has been a problem in Manipur State, with prevalence reaching 60 percent by 1992. This geographic variability and the size of the country have made estimation of the actual number of infections difficult. At the end of 1994, WHO estimated 1.75 million HIV infections, while evidence suggests an estimate of between 2 and 5 million in mid-1996.
In Cambodia the HIV/AIDS data indicate that the current extensive HIV epidemic started during the late 1980s or early 1990s and is predominantly occurring among heterosexuals with multiple sex partners. To date, there has been no evidence of a significant problem of injecting drugs in Cambodia. Levels among blood donors in Phnom Penh have risen from less than 0.1 percent in 1991 to about 10 percent in 1995. Dramatic rises have also been seen in sex workers, the police, the military, STD patients and pregnant women. The epidemic in Myanmar is one of the most serious in the region. There are an estimated half a million people with HIV in this country in 1996. The epidemic began with the infection of large numbers of injecting drug users in the late 1980s, with a prevalence of 60 to 70 percent since 1992. HIV prevalence in sex workers has steadily risen from 4.3 percent in March 1992 to 18 percent in March 1995. There is substantial geographic variability, with infection rates in pregnant women varying according to region between 0 and 12 percent in 1993. High levels of other STDs, low levels of condom use, the clandestine nature of commercial sex, and limited blood screening due to cost constraints are contributing factors to HIV spread. In Malaysia, HIV infection levels in IDUs have grown rapidly from 0.1 percent in 1988 to 20 percent in 1994. In female sex workers, rates have gone from 0.3 percent in 1989 to 10 percent in 1994. A behavioral study conducted nationwide in 1992 found that almost one in three sexually active men and one in ten married men reported having had casual sexual contact in the previous month. Reported condom use in commercial sex is low. This implies serious potential for heterosexual transmission. The rapid growth in prevalence in IDUs and sex workers in Malaysia in the last three years is similar to that seen in Thailand and Myanmar in the early stages of their epidemics. In Vietnam there is some evidence that the HIV epidemic is now growing rapidly. High levels have been demonstrated in IDUs in treatment (32 percent in 1992-95), and recent evidence suggests increasing levels among young men and women in the south of Vietnam. Rates in some sex worker populations rose from 9 to 38 percent between 1992 and 1994-95. In China the majority (about 70 percent) of reported HIV infections and AIDS cases have been among IDUs in Yunnan Province. HIV infections are believed to be increasing among heterosexual populations in southern China, especially in the areas surrounding Hong Kong. The Chinese Academy of Preventive Medicine has estimated that there were 10,000 HIV-infected persons in China as of the end of 1993, growing to 100,000 by the end of 1995. Limited HIV/AIDS data for Laos suggest that HIV transmission may be starting in the heterosexual population. Additional data are needed to confirm the beginning of an HIV epidemic in Laos. In Bangladesh, Indonesia, Nepal and Sri Lanka the situation must be assessed based upon relatively limited testing, low rates of HIV detection in most populations, and low numbers of reported HIV and AIDS cases. These limits to our knowledge of the situation make any estimates of total prevalence or incidence quite speculative. However, most of these countries appear to have high levels of other STDs in their populations, implying a strong potential for extensive HIV spread. In Hong Kong, Japan, Mongolia and the Republic of Korea, extensive spread has not been documented. In DPR Korea and Bhutan no AIDS cases or HIV infections have been reported, but only limited surveillance has been carried out. In the Philippines there appears to be slower growth of the epidemic, with much lower levels (less than 1 percent) of HIV among sex workers. Early AIDS cases indicated some spread of HIV among men having sex with men. The lower number of clients and more indirect nature of sex work in the Philippines may help to explain the more gradual evolution of the situation. In Singapore, infection levels in sex workers have been growing quite slowly. The rapid growth of HIV infection in sex workers seen elsewhere in the region has not been seen there, perhaps as a result of prevention efforts. Populations Affected The epidemics in Asia are predominantly spreading through heterosexual contact. On a regional basis, infected men probably outnumber infected women by a factor of 3 to 1 or more, since commercial sex clients, injecting drug users and men having sex with men have contributed most strongly to the rapid initial growth of the epidemic. This male/female ratio is expected to drop as the epidemic spreads into the general population through spread of HIV from clients of sex workers to their regular partners and spouses. The HIV/AIDS epidemics in Asian countries have been strongly influenced by gender inequality and the frequent practice of men visiting sex workers. Since sexual expression for females is typically more limited than for males, the small population of sex workers has large numbers of clients, and consequently high rates of other STDs, which enhance HIV transmission. As a result, most epidemics begin with rapid prevalence increases in sex workers and their clients (as seen through STD clinic data). This growth can be quite explosive. Annual incidences in sex workers as high as 25 percent and in clients of almost 10 percent have been seen in India. High growth rates have also been well documented in numerous studies in Thailand and Cambodia. Sharing of needles among injecting drug users, given its high efficacy for HIV transmission, has also played a significant role early in the epidemics, particularly in the Golden Triangle region (from Thailand and Vietnam, across southern China, to Myanmar, to Manipur) and in northern Malaysia. As the epidemics mature, transmission from sex worker clients and IDUs to their wives or girlfriends becomes the most important route of female infection, although this transmission occurs at slower rates than that between sex worker and client. The ultimate size to which the epidemic might grow in most countries is difficult to assess because few studies of risk behaviors in the general population are available. Only Hong Kong, Malaysia, the Philippines, Singapore and Thailand have done national studies of risk behavior. Indicating that the total number of men engaging in sexual risk behavior is lower in Hong Kong, the Philippines and Singapore than in Thailand and Malaysia, these studies may help to explain the slower growth of the epidemic in those countries. Pediatric HIV infection is also difficult to assess in this region, given the wide geographic variability in antenatal clinic infection levels. In Thailand it is now estimated that 6,400 children are infected annually, approximately 10 percent of total new HIV infections. Impact of Prevention Programs The extent of behavior change in the region has varied greatly from country to country. Thailand has best documented the most extensive behavioral change, the result of an active multi-sectoral national effort. In national surveys conducted in 1990 and 1993, the percentage of men visiting sex workers in the last year declined from 22 percent to 10 percent. Condom use in commercial sex transactions is now the norm. As a consequence of these behavioral changes STD rates have fallen precipitously, with reported cases dropping to one-fourth of their initial levels. Male HIV incidence is estimated to have fallen by an even greater factor. While there has been substantial success of HIV prevention in commercial sex trade, the situation in non-commercial casual sex remains of concern. Current levels of condom use between boyfriend/girlfriend or with other longer-term partners remains low, on the order of 10 percent. Another area in which there has been only limited success has been slowing HIV transmission within HIV-discordant married couples in which the husband is HIV-infected and the wife is not. As these women become infected, rates found in antenatal clinics continue to climb. The slow growth of the epidemic in Singapore may largely be attributable to general awareness and programs promoting condom use at STD clinics and in brothels. It is reported that condom use by sex workers has reached fairly high levels, although commercial sex by Singapore residents traveling overseas remains an important avenue of HIV transmission. Efforts to produce behavior change have been less effective in other countries of the region. In India, no formal studies have been done on the large-scale impact of prevention programs. From focus group discussions, however, it appears that fear of acquiring HIV has risen among the educated and the higher socioeconomic classes. This may lead to higher condom use in these populations, but this is not yet documented. Unfortunately, in the lower socioeconomic classes and rural areas there is still a gross lack of awareness and knowledge of HIV prevention methods, suggesting that behavioral change has probably been minimal. There still appears to be low use of condoms in many sex worker populations, especially among those who have many clients per day. Condoms continue to be the exception rather than the rule for most premarital and extramarital sex in India. Sexually transmitted diseases continue to be a major problem in this country, a fact not well recognized prior to the HIV epidemic. In the Philippines, behavioral surveys in 1990 and 1994 in Metro Manila have shown fairly constant levels of casual and commercial sex, implying little behavioral change during that time. The levels of condom use, while rising somewhat in Metro Manila, remain quite low. STD rates are lower than in many other countries of the region, but as mentioned earlier, are high in certain populations, including sex workers. Myanmar and Malaysia's effectiveness in inducing behavioral change is difficult to evaluate because no periodically collected data on risks is available there. However, extensive NGO efforts in Malaysia and grass-roots efforts in Myanmar may be reducing risk behaviors and increasing the use of condoms. For those countries in the early stages of HIV epidemics (e.g., Bangladesh, Bhutan, Brunei Darussalam, Indonesia, Nepal, the Maldives and Sri Lanka), national efforts at HIV control have been fairly limited and major nationwide behavioral change is unlikely to have yet occurred. Non-governmental organization and governmental program efforts targeted at commercial sex may have raised condom use somewhat in more heavily populated urban settings, e.g., Jakarta, Kathmandu and Colombo. Impact of Care Programs Because the epidemics in the region are comparatively young, many doctors fail to properly diagnose AIDS and, in addition, medical care is often difficult to access or limited in scope. As a result, what little data are available on issues of survival and the effect of care show somewhat shorter survival after diagnosis with AIDS than in the industrialized world. In one study in Thailand, median survival time after a diagnosis of AIDS was only 7 months, much shorter than in many industrialized countries, possibly because cases were only diagnosed when illness was quite advanced. In the Philippines, a small study following HIV-infected sex workers found survival times of one and a half years after the recognition that the immune system was seriously compromised. In Thailand, approximately one-fifth of children infected at birth were found to have developed AIDS after 6 months. However, the findings of these small preliminary studies can hardly be generalized. Studies of accessibility to and use of care, and its impacts on disease progression and survival are urgently needed throughout the region. AIDS: the Eruption in Asia It is critical to recognize the sheer numbers of people living in South and Southeast Asia, a region that contains more than 60 percent of the world's adult population. In particular the evidence gathered in India suggests rapid, extensive and uncontrolled spread in many parts of the country. There is an urgent need for a comprehensive synthesis of the state of the epidemic in India. It is clear that there is a critical need, in this country as elsewhere in the region, to gather more credible HIV/AIDS data on rural populations. China, too, because of its size and rapid changes in social and sexual behaviors, potentially represents a major focus of the epidemic in the region. The different rates of spread within and between countries must be acknowledged and better understood. For example, why is the spread of HIV in the Philippines and Indonesia apparently slower than in Malaysia and Thailand? Is it related to later introduction of the virus, to lack of reliable information or differences in behavior? Some governments (Hong Kong, Malaysia, Singapore, Thailand) have committed extensive resources to responding to the epidemic. However, the majority of governments in the region are relying heavily on external financial support to prevent epidemics occurring within their own borders. In addition, there continues to be a serious problem of denial and reticence about releasing surveillance and behavioral information by some governments in this region. Latin America and the Caribbean Latin America and the Caribbean region is heterogeneous and diverse, with a total of 44 countries and territories, an estimated population of 470 million people with a variety of ethnic backgrounds, and four main languages (English, Spanish, French and Portuguese). The rate of spread of HIV/AIDS has been slower than in other developing regions of the world, but the pandemic is well established and there is a wide variation in the level of HIV infection and the speed of the many epidemics among sub-regions and countries. The dominant modes of transmission vary from one country to the next, ranging from some epidemics that are predominantly related to homo- and bisexual behaviors, to epidemics connected to injecting drug use, and to others that are primarily determined by heterosexual transmission. In spite of this epidemiological diversity, sexual transmission of HIV/AIDS accounts for 80 percent of overall transmission in the region, ranging from 64 percent in Brazil to as high as 93 percent in the Andean sub-region (Bolivia, Colombia, Ecuador, Peru, Venezuela). Although data are limited and sometimes spotty, they reflect an increasing pandemic that is progressively affecting heterosexual populations and non-urban areas. As of June 10, 1996, Latin America and the Caribbean accounted for 26 percent (176,930) of the cumulative total of cases reported in the Americas to the Pan American Health Organization (PAHO) and 13.4 percent of the cases reported worldwide to the World Health Organization (WHO). It is estimated that 1.6 million people in the region have already been infected with HIV and that some countries are at particular risk of rapid dissemination of HIV from traditional "at-risk" groups (sex workers, men who have sex with men [MSM], men with multiple partners) and to other vulnerable groups in the general population (women, youth and children). Sexual behaviors across the region reflect patterns that place the population at risk for HIV. These behaviors include early onset of sexual behavior, cultural acceptability of multiple partners, especially for males, and low levels of condom use. In this region, however, despite the relatively high proportion of men who have sex with other men, patterns of homo- and bi-sexual behavior are still poorly understood. Bisexual behavior is more prevalent than exclusively homosexual behavior, while self-identification with a gay lifestyle or culture is not common. Consequently, targeting messages likely to reach MSM is difficult. The current epidemiological profile of HIV/AIDS in Latin America and the Caribbean is driven by high-risk situations favorable to a rapid spread of the HIV infection. Slowly but steadily the pandemic is taking hold of communities rendered doubly vulnerable due to their socioeconomic disadvantage and lack of information. Migration, both between countries and from rural to urban areas, contributes to the continued spread of HIV/AIDS and creates additional challenges to HIV prevention. The epidemiological evidence signals a rapid shift of new infections to younger ages, particularly toward people between 15 and 24 years old. In addition, there are marked tendencies for HIV infection to increase among the general population and among specific populations, in particular women, children, the poor, rural communities and, generally, those who have lower socioeconomic status and those who lack access to basic educational and health services. Mexico, The Isthmus of Central America and the Latin Caribbean The number of new HIV infections in Mexico, the Central American Isthmus (Guatemala, Belize, El Salvador, Honduras, Nicaragua, Costa Rica and Panama) and the Latin Caribbean (Cuba, Dominican Republic, Haiti and Puerto Rico) continues to rise. As of June 10, 1996, 60,564 cases were reported to PAHO. This represents 8.7 percent of the total number of cases reported for the Americas and 4.6 percent of the cases reported worldwide to WHO. However, the estimated "true" incidence of AIDS is substantially higher than the number of cases reported by 20 to 70 percent, with a one- to two-year lag in data collection. There is evidence of continued increasing HIV incidence among MSM in Mexico, although the rise is not as rapid as it was in the 1980s. Transfusion-associated HIV infection and AIDS cases have drastically declined in this country as in the rest of the region due to effective blood screening. In Mexico this has resulted in an apparent slowing of AIDS cases among women, but there is, in fact, a much younger epidemic of heterosexually acquired HIV infection emerging among women. Consequently, in this country two epidemics are observed: an urban epidemic, more mature and affecting predominantly MSM and an emerging rural epidemic, which is predominantly spreading through heterosexual transmission. The Central American Isthmus and the Latin Caribbean reflect epidemics with increasing HIV/AIDS incidence and accelerated heterosexual transmission. Honduras accounts for 57 percent of AIDS cases diagnosed in Central America, while it has only 17 percent of its population. HIV seroprevalence levels among sex workers in Honduras have reached as high as nearly 40 percent. Sentinel surveillance of pregnant women in the city of San Pedro Sula has documented prevalence of up to 4 percent. Commerce, migration patterns and communication within this sub-region suggest that HIV is spreading within each country in well-established local epidemics and, externally, across international borders. In the Latin Caribbean, Haiti is of particular importance because, perhaps alone in the region, it represents a case of a relatively mature epidemic. Due to social, economic and political instability, among other factors, HIV prevalence rose from 2 percent in 1989 to an estimated 5 percent of the adult population in rural areas in 1994. In urban areas the prevalence was estimated at 10 percent in 1994. HIV prevalence is particularly high among sex workers, STD clinic attendees and tuberculosis (TB) patients. High rates of HIV prevalence found by recent studies among pregnant women aged 14 to 24 are of particular concern. Within this sub-region, there is diversity in the structure and organization of commercial sex, ranging from informal networks to thriving sex industries. The latter involve countries from which sex workers in other countries within and outside this region originate and others that have organized sex tourism. In the Dominican Republic, HIV seroprevalence among Dominican population subsets reached levels up to 11 percent among sex workers, 5 to 8 percent among STD patients and, by 1993, 1.2 percent among women attending antenatal clinics. International and intra-regional travel, including tourism and employment seeking, also exert major influences on the dynamics of the epidemics in the Caribbean, enhancing the potential for spread of HIV. The English-Speaking Caribbean The predominant mode of transmission for HIV in the English-speaking Caribbean is heterosexual, but estimates of HIV transmitted through homo/bisexual contacts account for 14 percent of all new infections. Inter-country variation exists in AIDS incidence rates and in the underlying HIV infection levels but, in general, the number of cases is increasing in all countries. As of June 10, 1996, this region accounted for 4.6 percent (9,399) of the cumulative total of cases reported in the Americas to PAHO and 0.7 percent of the cases reported worldwide to WHO. The doubling time for the annual number of new AIDS cases in this sub-region is four to five years. Some Caribbean countries report AIDS incidence rates that are among the highest in the world. Among the many small countries of the Caribbean, the presence of countries with very high and very low rates of HIV incidence indicates that there are many different epidemics and not one regional pattern. The male-to-female ratio of incident AIDS cases has fallen dramatically over the past 10 years, standing at just less than 2 men to 1 woman in 1994. Women aged 15 to 19 now have higher annual incidence rates than men of the same age. Pediatric AIDS cases have been steadily rising and now account for 5 percent of all incident cases. AIDS has become the leading cause of death among young adult men in some Caribbean countries. There is an urgent need for increased surveillance of behavioral risk factors for AIDS and HIV infection, although the small size of most Caribbean countries makes the confidentiality issue an important obstacle to data collection and analysis. Among the heterosexual population in the Caribbean, increasing numbers of persons from marginalized groups are becoming infected, including migrant workers, sex workers and users of crack cocaine. The extremely low incidence of HIV infection through contaminated blood represents a partial success story for the Caribbean region. Available data from sentinel surveillance indicates increasing HIV prevalence rates among pregnant women, sex workers, applicants for visas to the U.S. and migrant farm workers in some Caribbean countries. South America The number of HIV infections and AIDS cases in South America is rising steadily. As of June 10, 1996, South America accounted for 15.5 percent (106,841) of the cumulative total of cases reported in the Americas to PAHO and 8.2 percent of the cases reported worldwide to WHO. However, as in other sub-regions, the true incidence of AIDS is believed to be substantially higher due to under reporting and difficulties in data collection. Within this specific region, Brazil accounts for 75 percent of all cases of AIDS reported to PAHO/WHO, followed by the Andean Region (15 percent) and the Southern Cone (10 percent). Sexual transmission of HIV accounts for 74 percent of all reported AIDS cases (51 percent homo/bisexual and 23 percent heterosexual), injecting drug use for 19 percent and blood and vertical transmission and undocumented cases, 7 percent. The HIV/AIDS epidemics in the region are at differing levels of maturity, but are well established in most countries. There is considerable transmission of HIV/AIDS due to injecting drug use in Brazil (27 percent) and the Southern Cone countries of Chile, Argentina, Uruguay and Paraguay (30 percent), although recent data in Brazil suggests that the HIV transmission through injecting drug use seems to be leveling off. The pandemic in this region has progressed since the early 1980s from one predominated by homo/bisexual transmission to one with accelerated heterosexual transmission. In addition, there is an emerging transition from epidemics centered in major urban areas to increasing involvement of smaller urban centers and rural areas. Epidemics are increasingly taking hold in specific population subsets, including adolescents, marginalized communities, and others characterized by low socioeconomic status and lack of basic socioeconomic, educational and health services. High HIV seroprevalence levels have been reported among specific South American populations: 27 percent among sex workers in Santos City, Brazil; 30 to 60 percent in several studies of urban IDUs in Brazil and Argentina, 23 percent in MSM in Rio de Janeiro, and 1 percent to 3 percent among pregnant women in Santos City, Brazil. The impact of HIV/AIDS on morbidity and mortality is already seen in major urban centers in Latin America and the Caribbean. In the city of São Paulo, Brazil, for example, AIDS deaths are now the leading cause of mortality in women of reproductive age. Challenges for Prevention A significant increase in knowledge, attitudes, practices and behaviors (KAPB) about HIV/AIDS has occurred in the region in the past ten years. Behavior changes are most visible among sex workers, MSM and health care providers involved in AIDS management. The behavior changes observed invariably coincide with prevention interventions. However, in spite of these trends, knowledge of the relationship between HIV and AIDS and of asymptomatic transmission is still very limited in the region as a whole. Knowledge of sexually transmitted diseases and their relationship to HIV is limited, too. This is further compounded by the fact that although awareness of HIV/AIDS has substantially increased to levels over 80 percent in many countries, there are still many misconceptions regarding the transmission of HIV through casual contact. Surveys on knowledge, attitudes, practices and behavior (KAPB) have documented the coexistence of high levels of knowledge of HIV/AIDS in many populations with myths and misconceptions, unsafe practices and low self-perception of risk. Immediate and targeted attention to specific population subsets (women, adolescents and children) is needed as these populations are expected to become most vulnerable in the next phase of the epidemic. While attention has been given effectively to partner reduction, non-penetrative sex, and the increase and correct use of condoms, programs have not fully capitalized on and need to be complemented with realistic prevention messages addressing abstinence, delayed sexual initiation and monogamy. In brief, as the pandemic escalates in Latin America and the Caribbean, affecting larger segments of the population, the social, economic and demographic impacts of HIV/AIDS are likely to exacerbate the burden on individuals, communities and countries, threatening the development and stability of the region as a whole. Hence, the need for continued and increased support and an expansion of HIV/AIDS prevention and control programs is critical to effectively combat the pandemic in this region. The growth of the AIDS epidemic in North America has slowed in recent years and is approaching stable incidence, largely due to the decline in sexual transmission between men. However, current AIDS incidence is at an unacceptably high level and it must be recognized that this leveling off should in no way be considered reason for complacency. AIDS data do not reflect current HIV infections, and HIV infection continues to occur at an alarming rate in a number of sub-populations and geographic areas. The characteristics of persons with HIV infection and AIDS continue to change, reflecting the evolving patterns of transmission. Populations Affected Estimates from a statistical model show that in 1992 in the United States about 750,000 persons were living with HIV, and that year about 60,000 persons became infected with HIV. In Canada an estimated 34,000 adults were living with HIV in 1994, and 2,500 to 3,000 persons were newly infected with HIV each year in the period from 1990 to 1994. Recent estimates based on surveys of childbearing women indicate that approximately 3.2 per 10,000 children born in Canada and 15.1 per 10,000 children born in the United States carried HIV antibodies. In the U.S., an estimated 12,000 children are currently living with HIV. Since the start of the epidemic, from 1 to 1.5 million cumulative HIV infections have occurred in North America. HIV infection has become one of the major causes of death for individuals between the ages of 25 and 44. Among men in this age group, it was the leading cause of death in the U.S. and the second leading cause of death in Canada in 1994. In that same year, HIV infection was the third leading cause of death among 25- to 44-year-old women in the U.S. Through December 1995, 513,486 persons had been reported with AIDS in the United States; 13,291 had been reported through March 31, 1996 in Canada. Overall AIDS incidence in North America has been slowing progressively. Although there were large increases in the number of persons annually diagnosed with AIDS-related opportunistic illnesses (AIDS-OIs) through the early 1990s, the annual increase since 1993 has been less than 5 percent. In 1995, after adjustment for delays in reporting, approximately 62,000 persons were diagnosed with AIDS-OIs (29 per 100,000 population) in the United States and 2,166 in Canada (9 per 100,000 population). In North America, although there has been an overall slowing in the increase in AIDS incidence, there has been substantive variation in the populations affected. For example, in the United States, the increase in AIDS incidence in the 1990s has been greatest for women compared to men, blacks and Hispanics compared to whites, and persons infected through heterosexual contact compared to those infected through other modes of transmission. As a result of these trends, AIDS incidence in 1995 was 6.5 times greater for blacks and 4 times greater for Hispanics than for whites, 20 percent of persons diagnosed with AIDS were women, and 15 percent were infected heterosexually. The HIV infection rates are also high among certain groups, such as incarcerated persons. In 1994, 2.3 percent of nearly 1 million prisoners in the United States were known to be infected with HIV, the rate of AIDS among prisoners was 7 times the rate of the non-incarcerated population, and AIDS was the second leading cause of death among prisoners. Among Canadian prisoners, HIV prevalence is higher in women, between 2 and 10 percent versus 1 to 4 percent for men; for both sexes, transmission is primarily related to injecting drug use. In the United States, AIDS incidence among children less than 13 years of age has declined annually. For example, while there were 938 cases in 1992, there were approximately 600 cases in 1995. Only 21 Canadian children were diagnosed with AIDS in 1995. This decline may well reflect such factors as lower conception rates in women diagnosed with HIV and the possible impact of maternal and neonatal zidovudine therapy on HIV transmission. In North America, syphilis incidence has declined, yet 1994 rates in the U.S. were 60 times greater for blacks than for whites. The incidence of gonorrhea has also declined. In 1994, the U.S. rate was 168 per 100,000 and the Canadian rate was 21 per 100,000. Estimates from statistical models and data from several cohort studies demonstrate that HIV transmission among men who have sex with men (MSM) has declined from the very high levels of the early 1980s. In Canada, HIV incidence among MSM has dropped from about 5 to 10 percent per year in the early 1980s to an estimated 1 to 2 percent per year in the early 1990s. The HIV seroprevalence rate among MSM attending STD clinics in the U.S. fell from over 30 percent in the late 1980s to 24 percent in 1995. However, the prevalence of HIV infection among MSM remains high in almost all areas of North America. The declining trends in HIV infection and morbidity among MSM are consistent with trends in STD surveillance data, which show large decreases in the rates of syphilis and rectal gonorrhea. These declines are also consistent with behavioral survey results | ||||||||||||||||||