Final Report for the
AIDSCAP Program in Brazil: Country Program Description

This report comprehensively summarizes the FHI/AIDSCAP program in Brazil (1992-1997). The report discusses program accomplishments and constraints, as well as providing behavioral outcome data.

Country Context

Cases of AIDS were first reported in Brazil in 1982, although a number of cases were identified retrospectively as far back as 1980, suggesting that HIV infection probably began to take place in the country sometime during the mid- to late-1970s. Since the reporting of these early cases, the size of the epidemic has continued to grow rapidly and steadily. With more than 150 million people, the extent of HIV/AIDS infection in Brazil tends to be masked. In reality Brazil has the second highest reported number of AIDS cases in the world (UNAIDS, 1996). However, if the per capita incidence is considered, Brazil occupies a much lower position, between the 40^th and 50^th (MOH, 1996). The number of HIV cases for the whole country is between 330,000 and 448,000. From 1980 until August 1997 the cumulative number of AIDS cases was 116,389 (of which 73 percent were reported in the Southeast region), a percentage of 65.5 percent were reported in the southeastern states of São Paulo and Rio de Janeiro.

*Total of the southeast region includes the states of São Paulo (59,242),Rio de Janeiro (17,081), Minas Gerais (7,670) and Espirito Santo (1,025).

A gradual increase in reported cases of AIDS has been taking place in all regions of the country, as well as in the rural and interior districts. However, from the 5,000 municipalities of the country only 1,740 reported 1 or more cases and 427 reported 5 or more cases. This means that the epidemic continues to be concentrated in highly-populated urban areas such as São Paulo, Rio de Janeiro, Porto Alegre, Santos and Belo Horizonte (MOH, 1996). Major cities and state capitals are the epicenters of the HIV/AIDS epidemic in Brazil. In such centers, the accumulated incidence of AIDS was equivalent to the most severe epidemic in other countries. In the city of Santos, for example, the incidence rate was 432.4 per 100,000 inhabitants, a rate which exceeded the concurrent incidence rates found in San Francisco and New York (Castilho and Chequer, 1995).

Table II shows the ten municipalities most affected by the HIV/AIDS epidemic in reported number of cases in Brazil and Rio de Janeiro and São Paulo.

* Although Rio de Janeiro and São Paulo are not among the current ten most-affected municipalities, they were included because they were part of the AIDSCAP target area.

Modes of Transmission

Over the years an important decline on the homosexual and bisexual transmission can be observed. In 1988, 50 percent of all reported cases were infected homosexual or bisexual transmission, while in 1996-97 this proportion has dropped to 23 percent. The opposite phenomenon occurred with heterosexual transmissions, which have been increasing. They were 4.5 percent in 1998, 10 percent in 1991, 23 percent in 1993 and 31 percent in 1996/1997.

As a result of the heterosexual transmission increase, the male-to-female ratio in adults has changed dramatically, moving from 9:1 in 1987 to 4:1 in 1994. In some urban centers, the ratio is now 3:1. Results from seroprevalence studies conducted in Brazil during the life of the project showed relevant variation in the levels of HIV infection both between states and within states, based on comparisons of urban core groups and sentinel studies conducted among pregnant women in several Brazilian state capitals. These studies showed that the levels of HIV infection among pregnant women varied from 0 percent in the North of Brazil to 2.6 percent in Porto Alegre, South of Brazil. Moreover, considering the data available from seroprevalence studies examined, multiple epidemics occurring at the same time. In a country with the size of Brazil there are relevant differences in seroprevalence data between the risk groups and among pregnant women. Thus, future programs need to continue taking into consideration the geographic variation of HIV levels among core groups and the general populations.

Brazil's blood supply appears to be relatively safe, and blood transfusion does not represent a significant route of HIV transmission. At this time, injected drug use (IDU) is considered the main bridge to heterosexual transmission of HIV. Infection linked to injected drug use has risen from 0.8 percent of the total reported cases in 1984 to 19 percent in 1997. The number of AIDS cases related to injected drug use is highly concentrated in the state of São Paulo, Santa Catarina, Paraná and Mato Grosso do Sul, producing important and relevant differences in the epidemiological profile of the epidemic in these states when compared to the rest of the country. Seroprevalence studies among injecting drug users shows prevalence as high as 64.8 percent in São Paulo (HIV/AIDS Surveillance Data base/ US Bureau of Census 1993-1997).

Following the rapid increase in the number of AIDS cases among women, it has been observed that a significant increase due to vertical transmission in the number of AIDS cases. It has increased from 0.57 percent in 1985 to 3.2 percent in 1997. Prenatal transmission is the dominant cause of pediatric HIV infection, with many cases originating in IDU mothers. The number of pediatric AIDS cases is expected to increase in the years to come due to the increase of HIV infection among women during their reproductive years, especially in the Southeast region.

Groups known to be at high risk of HIV infection such as commercial sex workers (CSWs) show accelerated HIV transmission rates. In 1991 a study showed that HIV prevalence among CSWs was 27 percent in Santos, 12 percent in São Paulo and 14 percent in Campinas (Lurie et al, 1995). In the city of São Paulo, seroprevalence study among 112 transvestites showed a prevalence rate of 60.7 percent in 1992. As elsewhere in the world, the highest concentration of HIV infection has been found among lower socioeconomic status CSWs.

Pauperization

The Brazilian MOH does not use socioeconomic variables to report AIDS cases; instead the level of education has been used as an indicator of the socioeconomic status. There is a trend indicating declining levels of education in reported AIDS cases. In 1985, 76 percent of the cases had a high education level (high school or university) and the remaining proportion only had primary school education levels or were illiterate. This proportion dropped to 31 percent in 1994 (MOH, 1996), increasing the incidence of HIV/AIDS among those with little formal schooling.

It has been observed that since 1987 two-thirds of women who reported HIV/AIDS cases had low- or no-education levels. In 1994, they were 78 percent of the AIDS cases. Among the cumulative cases reported, 3.5 percent of the men and 6.1 percent of the women were illiterate.

Age

Since the beginning of the epidemic, the age group most affected by AIDS has been 20- to 39-years-old (counting 71 percent of the cases reported until August 1996). However it has been observed that there is a progressive increase of infection of younger people, aged 20-29. This trend seems to be stabilized since 1991.

Sexually Transmitted Infections

STI incidence in Brazil is virtually unknown. Notification of STIs are not compulsory except for congenital syphilis and AIDS. STI case definitions are not standardized, and the current surveillance system is dysfunctional. In addition, the infrastructure for diagnosis and management of STI is poorly developed. Some available data from clinics treating women indicate high prevalence of all forms of STI, including those most closely linked to AIDS transmission. Studies among female prostitutes in the State of São Paulo showed STI prevalence rates of 69 percent in Santos, 63 percent in Campinas and 66 percent in São Paulo (Lurie et al., 1995). Among chancroid patients in the city of São Paulo, almost all were infected with HIV (Siqueira et al., 1993). Men rely on self treatment through antibiotics obtained at pharmacies (estimated at 50-70 percent of STI occurrence among men) (Lima et al., 1995). Since STIs usually remain asymptomatic in women, they are left undiagnosed and untreated.

Policy and Other Donors

During the period from 1992 to 1997 there were important policy changes made to tackle the HIV/AIDS epidemic in the country. The Federal Republic of Brazil, through its Ministry of Health, requested a loan from the World Bank of US$ 160 million for use from 1993 to 1997. The government of Brazil's counter-funding was equivalent to US$ 90 million. The total budget for the AIDS and STI control project was US$ 250 million, out of which 41 percent was for prevention (US$ 102.7 million); 34 percent for treatment services (US$ 84.6 million); 7 percent of total costs to surveillance (US$ 16.2 million) and 18 percent for Institutional Development (US$ 46.4 million). When using AIDSCAP intervention categories, levels of World Bank funding indicate: $18.5 million for condoms, $8.3 million for behavior interventions, $51.5 million for STI prevention and control (of which $25 million was used for the procurement of STI drugs at 300 existing STI clinics); and $15 million was for information, education, and communication. In the AIDSCAP target areas during the life of the project, the World Bank committed $33 million to São Paulo and $10.4 million to Rio de Janeiro. Other donors included World Bank endemic project; US$ 9 million, UNDCP US$ 2.5 million, EEC US$ 2.0 million, PAHO US$ 1.6 million and the French Government US$ 1.3 million. The Foundation for the Support of Research in the State of São Paulo (FAPESP) provided US$ 0.5 million. There are also other smaller donors including the Ford Foundation, the McArthur Foundation, the Levi-Strauss Foundation and the National Research Center in Brazil.

The high level of investment in AIDS prevention has required most NGOs to work in close association with the MOH. This has expanded many small and community-centered programs all over the country. A great production of BCC materials, training courses and mass media campaigns have been produced under the auspices of the MOH/World Bank loan. As a result many AIDSCAP sub-projects were developed in close association or replicated with the MOH support. After the Conference of Vancouver, the MOH decided to universalize the use of the HIV "cocktail" and committed 750 million for the procurement of this program for the year of 1996/97. So far, the MOH has bought the cocktail for all patients. The program was disrupted and distribution problems in the logistic system were identified. Presently it is a question whether or not government support of the "cocktail" will be maintained next year. Up to this time, a total of 750 million was committed for the procurement of the cocktail (Folha de São Paulo, 24/10/1997, interview with the PNDST/AIDS Director).