UNAIDS/WHO AIDS epidemic update: December 2005

LATIN AMERICA

HIV and AIDS statistics and features, in 2003 and 2005

The number of people living with HIV in Latin America has risen to an estimated 1.8 million [1.4 million–2.4 million]. In 2005, approximately 66 000 [52 000–86 000] people died of AIDS, and 200 000 were newly-infected [130 000–360 000]. Among young people 15–24 years of age, an estimated 0.4% [0.3–0.8%] of women and 0.6% [0.4–1.1%] of men were living with HIV in 2005.

Primarily due to their large populations, the South American countries of Argentina, Brazil and Colombia are home to the biggest epidemics in this region. Brazil alone accounts for more than one third of the estimated 1.8 million people living with HIV in Latin America. The highest HIV prevalence, however, is found in the smaller countries of Belize, Guatemala and Honduras—where approximately 1% of adults or more were infected with HIV at the end of 2003.

The region’s epidemics are being fuelled by varying combinations of unsafe sex (both between men, and men and women) and injecting drug use.

The region’s epidemics are being fuelled by varying combinations of unsafe sex (both between men, and men and women) and injecting drug use, with the role of sex between men in HIV transmission a more prominent factor than is commonly acknowledged. In nearly all the Latin American countries, the highest levels of HIV infection are being found among men who have sex with men. The second-highest HIV levels are found among female sex workers, according to one recent collection of cross-sectional studies. HIV prevalence among men who have sex with men ranged between 2% and 28% in different areas, while prevalence among female sex workers ranged between 0% and 6.3% (Montano to account for 25–35% of reported AIDS cases in countries such as Argentina, Bolivia, Brazil, Guatemala and Peru.

By far the largest and most populous country in the region, Brazil harbours a diverse epidemic which has penetrated all 26 states in the country. Although national HIV prevalence among pregnant women has remained below 1%, a growing share of new HIV infections are among women, and those living in deprived circumstances appear to be disproportionately at risk of infection (Marins et al., 2003). In some states, high prevalence has been found among pregnant women, for example up to 3–6% in sites in the southern Rio Grande do Sul state (UNAIDS/WHO, 2003).

There is some evidence that sexual behaviour among young Brazilians might be changing, with a trend toward earlier sexual activity. According to a 2004 survey, more young people are having sex at earlier ages and with more partners. More than one third (36%) of 15–24 year-olds had had sex before their 15th birthday (compared with 21% of those 25–39 years of age), 20% said they had had sex with more than ten partners thus far in their lives, and 7% had had sex with more than five partners in the previous year. Yet knowledge of HIV was poor. Only 62% of 15–24 year-olds knew how HIV was transmitted. Less-educated youth knew the least about the epidemic. On the other hand, the percentage of young people who reported using condoms the first time they had sex increased from less than 10% in 1986 to more than 60% in 2003 (Ministerio da Saude do Brasil, 2005). The latter trend may be tempering the effect on HIV transmission of risky behaviour reported earlier. HIV prevalence among military conscripts has remained consistently low (0.08% in 2002, the same as in 1998) (Ministerio da Saude do Brasil, 2005).

Meanwhile, in Brazil’s cities, the contribution of injecting drug use to HIV transmission appears to have declined. Some of this success could be attributed to harm reduction programmes (see AIDS epidemic update 2004 for more details). Official estimates derived from the national HIV surveillance system show that three quarters of the estimated 200 000 drug injectors in Brazil now do not use non-sterile syringes. In some areas, though, drug injectors still comprise at least half of all AIDS cases. Available data indicate relatively low levels of HIV infection among female sex workers, with HIV prevalence of 6.1% among the almost 3000 sex workers who participated in one major survey (Chequer, 2005).

In Argentina, HIV initially circulated mainly among male injecting drug users and their sexual partners, as well as among men who have sex with men. This gradually changed as more HIV-infected men passed the virus on to wives and girlfriends. Most new infections have been occurring during unprotected heterosexual intercourse, with increasing numbers of women acquiring HIV. The male-to-female ratio among reported AIDS cases shrank from 15:1 in 1988 to 3:1 in 2004. Among new infections, that ratio was 1.5:1 in 2004, with new infections disproportionately occurring in poor urban areas (Ministerio de Salud de Argentina, 2004). Meanwhile, injecting drug use and sex between men continue to provide impetus to the spread of HIV, especially in the urban areas of Buenos Aires, Cordoba and Santa Fe provinces, where an estimated 80% of AIDS cases have occurred. When tested in the city of Buenos Aires, some 44% of drug injectors were HIV-positive, for example, while HIV prevalence of 7–15% has been found among men who have sex with men in various studies (Weissenbacher et al., 2003; Pando et al., 2003; Segura et al., 2005; Montano et al., 2005; Bautista et al., 2004).

In Chile and Uruguay, most HIV infections are concentrated in urban areas (National AIDS Commission Chile, 2003; National AIDS Program Uruguay, 2005). About three quarters of Uruguay’s reported HIV cases have been in and around the capital, Montevideo, while Antofagasta, Santiago, Tarapaca and Valparaiso are the most affected parts of Chile. In contrast, HIV has penetrated rural parts of Paraguay, especially along the borders with Argentina and Brazil (National AIDS Program Paraguay, 2005). Injecting drug use and sex between men appear to be the most prominent factors in Uruguay’s epidemic, where about one quarter of reported HIV cases have been in drug injectors (many of them younger than 25 years) and one third of HIV diagnoses have been attributed to sex between men (Osimani, 2003).

In the Andean area, unprotected commercial sex and sex between men serve as HIV’s main pathways. As more men pass the virus on to their wives and girlfriends, however, HIV transmission routes are becoming more diverse. Bolivia’s epidemic remains small (much of it concentrated in urban areas such as La Paz and Santa Cruz) and appears to be driven largely by commercial sex and sex between men (Ministry of Health Bolivia and PAHO/WHO, 2003; Khalsa et al., 2003). While HIV prevalence among women attending antenatal clinics has remained under 1%, infection levels in groups of men who have sex with men have reached 15% in La Paz and almost 24% in Santa Cruz (Montano et al., 2005). The vulnerability of male and female street children to HIV infection in Bolivia has escaped notice until recently. A study in the city of Cochabamba has shown that 3.5% of the surveyed street youths were living with HIV; most had been infected sexually (Lambert et al., 2005). Female sex workers, by and large, have managed to avoid infection: in Cochabamba, Oruro and Tarija, for instance, prevalence was 1% in 2002, while it was 0.5% in La Paz (where health authorities estimate 70% of sex workers use condoms regularly) (Carcamo, 2004). However, those infection levels were recorded among (mainly brothel-based) sex workers who regularly visit sexually transmitted infection clinics for check-ups. Patterns elsewhere in Latin America suggest that sex workers are at higher risk of infection. As of yet, though, there are few studies tracking the possible spread of HIV among sex workers—not only in Bolivia, but in Latin America generally.

It is vital that preventive activities, as well as HIV diagnosis and treatment services, are expanded among men who have sex with men.

While it is necessary to maintain prevention efforts among sex workers, it is vital that preventive activities, as well as HIV diagnosis and treatment services, are expanded among men who have sex with men, who continue to face social stigma and discrimination. Sex between men appears to be a prominent factor also in Ecuador’s growing epidemic, where new reports of HIV cases have almost doubled since 2001, reaching 573 in 2004 (Ministerio de Salud de Ecuador, 2005). More than two thirds of reported HIV cases have been in men, and HIV prevalence of 17% and 23% has been found in Quito Pichincha and Guayaquil Guayas, respectively, among men who have sex with men. Prevalence among female sex workers was low (under 2%) (Ministerio de Salud de Ecuador, 2005). It appears that a large number of women with HIV have been infected by their husbands or regular partners who had acquired the virus during sex with another man.

In Colombia, HIV initially affected mostly men, so much so that they comprise 83% of all AIDS cases reported to the national health authorities to date. As in several other countries in the region, HIV prevalence recorded in groups of men who have sex with men (as high as 20% in Bogotá) has surpassed prevalence found among female sex workers in Colombia (0.8% in 2001-2002 in Bogotá, for instance) (Montano et al., 2005; Khalsa et al., 2003; Mejía et al., 2002). However, a significant proportion of men who have sex with men also maintain sexual relationships with women. As a result, the epidemic’s pattern has been changing, with increasing numbers of women becoming infected—especially along the Caribbean coast and in the north-east of the country. Most will have acquired HIV from infected husbands or boyfriends who had had unsafe sex with men and/or women. Thus, of women testing HIV-positive at projects aimed at preventing the transmission of HIV from mother-to-child, 72% were in stable relationships and 90% described themselves as ‘housewives’ (García et al., 2005).

There are recent signs of significant HIV spread in Colombia: prevalence found among young people (aged 15–24 years) and pregnant women ranged from 1.2–1.3% in Santander and Valle, to 2.4% in Atlantico (Prieto et al., 2004). In addition to pervasive gender inequalities, it is possible that forced displacement caused by the country’s long-running civil conflict could be increasing women’s risks of acquiring HIV (García, 2005). A positive development has been the country’s national initiative to reduce transmission of HIV from mothers to children, which ranks among the strongest components of its AIDS programme. According to one study, it has reduced the risk of perinatal HIV transmission from 40% to under 4% (García R et al., 2005).

Nationally, HIV infection levels among pregnant women in Peru have remained low (0.2% in 2002, according to the latest available data) (Ministerio de Salud de Peru, 2004). However, much higher HIV prevalence has been recorded among men who have sex with men—6–12% in the cities of Arequipa, Iquitos, Pucallpa and Sullana, and as high as 23% in Lima in a 2002 study (Ministerio de Salud de Peru, 2005). In all but the first two cities, infection levels had risen since 2000. Since a very large proportion of men who have sex with men (more than three quarters, according to some urban studies) also have sex with women, the possibility of HIV transmission to their wives and girlfriends is high (Guanira et al., 2004). Commercial sex is another prominent factor. Almost half the men (44%) aged 18–29 years in 24 urban areas said they had bought sex, and condom use was generally erratic.

Where antiretroviral treatment access has improved notably, AIDS mortality appears to be reducing.

HIV in Central America is spreading both amongst the most vulnerable groups and, in a number of countries, across the wider population. The virus is being transmitted primarily during unprotected sex (between men and women, and also between men). While comprehensive HIV surveillance information remains incomplete, the available data show epidemics that are concentrated mainly in and around major urban areas and transport routes, with some important exceptions.

HIV has acquired a varied but firm presence in Honduras, which has approximately one third of the people living with HIV in this subregion. With estimated national adult HIV prevalence of just under 2%, AIDS is the leading cause of death for Honduran women and is believed to be the second-biggest cause of hospitalization and death overall in the country (UNAIDS/WHO, 2004). As far back as 1999 already, median HIV infection levels among pregnant women ranged from 2.9% in urban areas to 3.6% in some rural areas—indicating a relatively mature epidemic, with HIV circulating in the wider population. Commercial sex and sex between men, however, remain major drivers of the epidemic. Among female sex workers in Tegucigalpa, median HIV prevalence of 8–9% was measured in 2001, while in San Pedro Sula prevalence of 13% has been found. Among men who have sex with men in major urban areas, 12% were found to be living with HIV in 2002 (UNAIDS/WHO, 2004; Secretaria de Salud de Honduras, 2003). Such findings, along with high rates of other sexually transmitted infections, suggest that conditions favour the epidemic’s continued growth in Honduras.

Central America’s other epidemics are also strongly related to commercial sex and sex between men. Although HIV infection levels among sex workers vary considerably, they are consistently higher among those who are street-based and who are therefore more difficult to reach with HIV prevention services. In San Salvador and Puerto de Acajutla (in El Salvador), for instance, 16% of street-based sex workers have been found to be HIV-positive (Ministerio de Salud Pública y Asistencia Social de El Salvador, 2003). Since condom use tends to be lower among regular sexual partners, the male clients of sex workers are likely to pass on the virus to their wives and girlfriends. The same applies to men who have sex with both men and women. For women, fidelity offers little protection against infection, as one study in Chinandegas (Nicaragua) has illustrated. There, married women were twice as likely as sex workers to be living with HIV (UNAIDS/WHO, 2004).

Guatemala’s epidemic rivals that of Honduras in size, but HIV data gathering has been too intermittent to enable a confident assessment of recent trends. Available information suggests uneven HIV spread, with the highlands possibly least-affected. Most HIV cases appear to be concentrated in urban areas and along the transport and commercial routes that cross the country. HIV surveillance in some sexually transmitted infection clinics has revealed no HIV cases; in others, prevalence as high as 9% has been found (in Izabal, for instance). Among pregnant women, HIV levels have varied from close to 0% to over 1% (in Retalhuleu and San Marcos) (UNAIDS/WHO, 2004; Ministerio de Salud Pública y Asistencia Social de Guatemala, 2003). Here, too, sex work seems to play a prominent role: HIV prevalence as high as 15% has been recorded among street-based female sex workers. However, sex between men could be a bigger factor than is commonly assumed (Ministerio de Salud Pública y Asistencia Social de Guatemala, 2003). HIV infection levels of almost 12% have been found in Guatemala City among men who have sex with men, one in five of whom also had regular sexual relations with women (UNAIDS/WHO, 2004).

Figure 19

Sex between men features also in the smaller epidemics of El Salvador, Nicaragua and Panama, where HIV prevalence of 18%, 9% and 11%, respectively, have been found among men who have sex with men. In Costa Rica, sex between men is clearly the driving factor in the country’s epidemic; there, men who have sex with men comprise more than two thirds of all reported AIDS cases (UNAIDS/WHO, 2004).

Adult national HIV prevalence in Mexico has remained well below 1%, but the epidemic shows varied patterns across this large country. Almost 90% of officially recorded AIDS cases have been attributed to unprotected sex, about half of them attributable to sex between men (CENSIDA, 2003). An estimated 160 000 adults were living with HIV at the end of 2003; two thirds of them are men who are believed to have been infected during sex with another man (UNAIDS, 2004; Magis-Rodríguez et al., 2002). That pattern possibly accounts for the increasing signs of heterosexual transmission of HIV that have been noted, as more women become infected during sex with (bisexual) male partners (Magis et al, 2000). The role of injecting drug use in Mexico’s epidemic is difficult to determine, but an association with injecting drug use has been observed in cities along the border with the United States of America (Minichielloa et al., 2002; Magis-Rodríguez et al., 1997).

The possible role of migration in Mexico’s epidemic has been the subject of much speculation, but confident conclusions cannot yet be drawn. One recent study has shown that male international migrants on average had almost twice as many sexual partners in the previous year, compared with their non-migrant counterparts. At the same time, condom use was found to be substantially more frequent among international migrants (Magis-Rodríguez et al., 2004). Nevertheless, the higher AIDS case load in some rural areas of Mexico hints at a possible link between HIV infection and migration to the United States of America (Magis-Rodríguez C et al., 2004).

Access to antiretroviral therapy in Latin America has expanded considerably, although Brazil’s achievements on this front remain unique. Under Brazil’s policy of providing antiretroviral drugs to all in need, people with advanced HIV infection are eligible for antiretroviral drugs via the country’s national health system. The number of Brazilians on antiretroviral therapy has continued to increase and reached approximately 170 000 in September 2005. Treatment adherence rate among patients on antiretroviral therapy has been estimated at 75%. Treatment coverage is high also in Argentina, Chile, Cuba, Mexico, Uruguay and Venezuela (PAHO, 2005), though the terms under which it is provided are not as favourable as those in Brazil. In Costa Rica and Panama, where antiretroviral treatment access has improved notably, AIDS mortality appears to be reducing. But elsewhere, especially in the poorer countries of Central America and the Andean region of South America, progress has been slower. Fewer than 1000 Ecuadoreans were receiving antiretroviral treatment in 2004 (Ministry of Health Ecuador, 2004), for example, while treatment roll-out efforts in El Salvador, Guatemala, Honduras, Nicaragua and Paraguay continue to lag (PAHO, 2005).