HomeLista de Teses >  [CARACTERISTICAS SOCIODEMOGRAFICAS, COMPORTAMENTAIS...]


 

Souza, Claudia Teresa Vieira de . Características sóciodemográficas, comportamentais e vulnerabilidade à infecção pelo vírus da imunodeficiência humana em homens que fazem sexo com homens do "projeto Rio". [Doutorado] Fundação Oswaldo Cruz, Escola Nacional de Saúde Pública; 2001. 50 p.

 

4 - COLETÂNEA DE ARTIGOS

 

4.1 - Incidência do HIV no Rio de Janeiro - I. Estratégias de recrutamento e dados sociodemográficos de uma coorte de homossexuais e bissexuais masculinos soronegativos para o HIV no Rio de Janeiro, Brasil. (The Rio de Janeiro HIV vaccine Site - I. Recruitment Strategies and Sócio-demographic Data of a HIV Negative Homosexual and Bisexual Male Cohort in Rio de Janeiro, Brazil).

(Mem Inst Oswaldo Cruz - 92(1):39-46,1997)

Homosexual Cohort Recruitment and Socio-Demographic Data. F Sutmoller et al. Mem IOC Vol 92, 1997;000-000 3177

The Rio de Janeiro HIV Vaccine Site

I. Recruitment Strategies and Socio-demographic Data of a HIV Negative Homosexual and Bisexual Male Cohort in Rio de Janeiro, Brazil

Frits Sutmoller+, Claudia Teresa Vieira de Souza, José Carlos Monteiro, Therezinha Penna

Hospital Evandro Chagas, Instituto Oswaldo Cruz, Av. Brasil 4365, 21045-900 Rio de Janeiro, RJ, Brasil

The initial effort of the Brazilian Ministry of Health to be an active partner in the world effort in the preparation of future accurate human immune deficiency virus (HIV) efficacy trials was the establishment of a multi-centered cohort of homosexual and bisexual men. An open cohort was established to determine the HIV incidence and the socio-behavioral aspects involved in Rio de Janeiro. A total of 318 potential participants, originated from multiple sources (health units, public information, snowball recruitment), were screened and recruitment became effective through the direct involvement of target communities [with the support of Non Governmental Organizations (NGOs)] and the population. Among this group, seropositivity for sexually transmitted diseases (STD) was high with 23, 32 and 46% for HIV, syphilis and hepatitis B, respectively. The socio-demographic data from the first 200 participants of this HIV negative cohort suggests that the cohort volunteers are an appropriate sample of the general male population of the State of Rio de Janeiro.

Key words: human imunodeficiency virus - syphilis - hepatitis B - homosexual - bisexual - male - prevalence - cohort - socio-demographic aspects

_____________________________________________________________

Financial support was provided by the National STD/AIDS Program of the Brazilian Ministry of Health, the Global Programme on AIDS / WHO, Conselho Nacional de Pesquisa (CNPq) and the PAPES Program of the FIOCRUZ.

+ Corresponding author. Fax: +55-21-260.9749

Received 16 May 1996

Accepted 19 August 1996

_____________________________________________________________

In 1992, Brazil was selected by the World Health Organization, through its Global Programme on AIDS (GPA/AIDS), as one of four countries, along with Thailand, Uganda and Rwanda, to receive technical and financial support to prepare for future human immune deficiency virus(HIV) vaccine efficacy (Phase III) trials. A multi-centered study group located in three states (Rio de Janeiro, Minas Gerais and São Paulo) was established with the purpose of obtaining accurate estimate of HIV incidence necessary for vaccine trial sample size determination. The Brazilian Ministry of Health, together with the government of the three states decided to give priority to an open cohort of homosexual and bisexual men as they supposed that high incidence rates would be found in these groups and that the relation with social and behavioral aspects needed to be investigated.

The potential study populations, epidemiological methods, and the ethical consideration have been previously described (Heyward et al. 1994, Greco 1994, Sutmoller et al. 1994, 1996). The State of Rio de Janeiro had a cummulative number of reported AIDS cases, to the end of 1995, of 11433 with a cumulative coefficient index of 89.0 per 100,000 inhabitants in the State (Ministério da Saúde 1995).

In this paper we will discuss recruitment strategies and socio-demographic data obtained from the first 200 cohort volunteers considered pertinent to the establishment of a long term (three years) open cohort among a human immunodeficiency virus (HIV) negative homosexual and bisexual men. We also report on findings about sexually transmitted diseases (STD) seroprevalence in the potential study participants as it reflects risk behavior for HIV infection. This report covers the period of January 1995 to April 1996.

MATERIALS AND METHODS

This study was approved by the Oswaldo Cruz Research Ethics Committee and volunteers were admitted after signing an informed consent form.

Located on the campus of the Oswaldo Cruz Foundation, the Evandro Chagas Hospital (HEC) was selected as the main site for the study because of its high reputation in bio-medical research. A specific area was assigned in order to offer an adequate environment for this type of investigation (e.g. privacy and volunteer comfort). A limited number of volunteers was interviewed at the Servidores do Estado Hospital (HSE) of Rio de Janeiro in downtown and at the Sexually Transmitted Diseases Clinic, Fluminense Federal University (STD/UFF), located in Niterói, State of Rio de Janeiro, with the aim of decentralizing study activities.

The State of Rio de Janeiro has a population of 12.8 million inhabitants with a male population of 6.2 million. The male urban population is 5.8 million, 78.4% of which (4.9 million) live in the Metropolitan area where this study was conducted (FIBGE 1991).

Study population - Homosexual or bisexual men, ranging in age from 18 to 50 years, who live in the State of Rio de Janeiro.

Cohort entry criteria - Similar to the study population but with a recent (<1 month) HIVnegative test.

Recruitment - Volunteers were recruited from (1) Municipal Health Centers: three health units - Centro Municipal de Saúde (CMS) Oswaldo Cruz, CMS João Barros Barreto, Posto de Assistência Médica (PAM) 13 de Maio - were initially selected according to their location, concentration of free public HIV testing and the centralization of outpatient treatment for HIV positive patients. Since then HIV testing has changed and is now done in Anonymous Testing Centers and health units, requiring new alternative approaches which are being implemented; (2) Public Information - Information to the public about STD/AIDS Prevention Project consisted of a combination of media coverage (interviews with staff members in newspapers, radio, and television) and production, and distribution of pamphlets. Paid advertisements were printed during two months (early 1995), in newspapers close to the sex related activities classified section (O Globo, O Dia and Jornal do Brasil). A community radio in Nova Iguaçu aired information and announcements about the Project during a two month period (March and April 1996); (3) Community activities - Active information was disseminated at the homosexual night clubs and bars by staff- related personnel with the direct participation of "grass root" non governmental organizations (NGOs) - primarily the "Movimento de Emancipação Homossexual" (ATOBÁ) and the "Grupo 28 de Junho". Field work, carried out by qualified recruiters, with groups of male prostitutes and transvesties, was also a source of volunteers which started in 1996; (4) Volunteers - The recruiting activities performed by the volunteers proved to be effective with an cumulative effect (also called "snowball" or "net working" recruitment) and were accelerated, from October 1995, by the establishment of an expense reimbursement for specially trained volunteers, working as recruiters.

Study procedures - Potential volunteers were seen by the study staff and received detailed information about the study objectives and procedures; they also received pre-test counseling and an explanation of what was expected of a volunteer. A study entry form was then completed together with both a questionnaire to provide information on opinions and atitudes about health and HIV and psychological tests pertinent to sexual risk behavior. A blood specimen was drawn (for HIV, syphilis and hepatitis B markers and for a complete blood count) and sent to the laboratory under confidential specifications.

Candidates returned aproximately one week later for the laboratory results. All candidates, independent of HIV status, received post-test counseling. Those with a positive HIV serology were referred to a health unit near their residence or were included in the HEC clinical investigations of the asymptomatic HIV positive cohort if they fitted the entry criteria. Those with a negative HIV serology were invited to participate in the Program and after final discussions were asked to sign the informed consent form.

The admission questionnaire provided identification, address and procedures for future contact; it is a confidential document accessible only to investigators. Upon admission to the cohort the volunteer received a six digit number which identified him in all other study forms. A social and behavioral questionnaire composed of 100 question items was then administered by one of the study psychologists. A medical examination was part of the admission procedures.

Data collection - The study team which was previously trained to obtain standardized answers was in charge of data collection. The data referred to in this study was obtained from (1) initial visit form - Reasons for coming to the study, referral to the study, previous knowledge about the project, previous serology test results; (2) psychological profile -Health locus of control, interpersonal sensitivity (SCL-90) and impulse control (NEO Personality Profile) combined with other knowledge, attitude, beliefs and practices (KABP) questions. The psychological profile is a supplementary study and investigated exclusively at the Rio de Janeiro cohort site; (3) social behavioral questionnaire -Includes social and demographic data (age, marital status, living conditions, employment status, salary and number of people living on the family salary, religion), HIV/AIDS knowledge, sexual attitudes and present sexual practices as well as the volunteer’s beliefs on health and AIDS-related subjects. This will be discussed in detail in a separate paper.

Laboratory exams - Venous blood was collected for viral serology obtained via serum Vacutainer tubes (10 ml) and a specimen was obtained for a complete blood count with an EDTA Vacutainer (5 ml) tube. The laboratory tests were done by the Clinical Laboratory (Head Immunology Laboratory: Ingebourg Georg) of he HEC and the hepatitis test by the Virology Department (Head: Dr. Clara Yoshida).

The HIV test was done with two different enzyme linked immunoassay (ELISA) HIV Uniform-Form II (Organon Teknika, Boxtel, Netherlands) and Enzygnost anti-HIV 1/2 Plus (Behring, Marburg, Germany). All specimens were tested by indirect immunoflourescence (IF) produced by Bio-Manguinhos (Department of Technological Development of FIOCRUZ).

Syphilis serology was done for the Treponema pallidum by the Venereal Disease Research Laboratory (VDRL) method and the haemaglutination absorption (TPHA) test (Randox Laboratories, Antrim, U.K.).

Hepatitis markers were done using in-house enzyme immunoassays (Bio-Manguinhos - FIOCRUZ) with controlled specificity and sensitivity according to available commercial kits: HBsAg, anti-HBs and anti-HBc. No confirmation test for HBsAg was done.

Complete blood count was performed according to the standard procedures.

Data analysis - A data set was developed for the purpose of data analysis in FoxPro (version 2.0) and evaluated (frequency tables and Chi-square test for p-value calculations) with EPIINFO (version 5.0).

RESULTS

Recruitment - Sources of potential cohort volunteers (Table I) changed as recruitment efforts varied during the study. One of the study objectives was to determine which recruiting method should be used with this population. During 1994 most efforts and financial commitment were concentrated on the health sector (health units, hospitals, private physicians, etc.); they produced 45% of the first 56 candidates. Collaboration with the NGOs was still limited and the study pamphlet distribution only started in the last trimester of 1994 making the recruitment initially slow and more difficult than originally anticipated. During the first trimester of 1995, NGOs efforts increased with an active pamphlet dissemination. The efficacy of both newspaper paid announcements and of media coverage was evaluated. These actions were important for community information but poor for volunteer recruitment. During the second and third trimester of 1995, due to lack of continued financial support, recruitment activities were reduced and subsequently the number of people coming to the unit diminished. With renewed support we were able to stimulate volunteer snowball and NGO based recruitment. In the last trimester of 1995 and early 1996, over 90% of the recruitment came from the NGOs and volunteers.

For the first 318 volunteers, the main motivation to participate in the study was to have a reliable HIV test, rapid test results, access to free condoms and the wish to help in a study related to AIDS prevention.

Seroprevalence data - Three STD infections - HIV, syphilis and hepatitis B - were monitored serologically at the initial visit (Table II). The overall HIV seropositivity was 23.6%. Syphilis positive serology accounted for 32.8% of potential study participants and evidence of hepatitis B infection was shown in 47.9% of men. The first trimester of 1996 showed an increase of seropositivity due to the recruitment of commercial sexual workers ("michês").

The seroprevalence of syphilis and hepatitis B was significantly different among the potential study participants depending on HIV status. HIV carriers were positive for syphilis (47.7%), hepatitis B markers (79.7%), and HBsAg (9.1%). The HIV negative have shown results of 27.5%, 36.7% and 3.7%, respectively. Both groups have higher prevalence of these infections than the general population.

Social and demographic data - The age distribution of the first 200 volunteers enrolled in the cohort study (Table III) is similar to the male 18-50 year male population in the State of Rio de Janeiro, except for the 20-24 age group, which in our study represented 40% of the volunteers. The ethnic distribution, cited by the volunteer is white 53%, mulato 28.5% and negro 9%.

The marital status information (Table IV) revealed that 83% were single and 12% married or separated. Twelve percent of the study participants had children. Half lived with their parents, 23% lived alone, 20.5% lived with either a friend or companion and 2.0% lived with their wives. Half of the volunteers claimed to have a religion, most frequently "spiritual" and catholic faiths, and a smaller percentage had the "candomblé" and protestants faiths.

The educational level of the participants showed that 31.5% had incomplete primary education, 40% had complete secondary education and 20% had university education. Employment status revealed that 70.5% were working at the time of the interview and 46% earned as from one to six minimum salaries (MS) (1 MS = R$100,00 = US$105,00 at the time of the study period).

DISCUSSION

Brazil was selected by the World Health Organization (now coordinated by UNAIDS) as one of the four developing countries which could be prepared for future Phase III HIV vaccine (efficacy) trials. The country was chosen for its supposedly high incidence in AIDS [fourth in world ranking (WHO 1994), totalling 76,396 reported cases up to the end of 1995 (Ministério da Saúde 1995)] and for the availability of laboratory facilities, combined with technical expertise. Political willingness and continued financial support have made it feasible for Brazil to develop this multicentered preparedness and the capacity to participate in future efficacy trials, on the advent of a future experimental vaccine.

The decision to investigate HIV incidence in Brazil by studying a cohort of homo- and bisexual men has proven to be adequate in Rio de Janeiro due to the high STD prevalence encountered in our study population. Also the social organization of "grass root" NGOs made access to this population group feasible and is effective in mobilizing their peers. Morever, the absolute number of AIDS cases in men who practice sex with other men, notified to the Brazilian Ministry of Health, continues to grow (Bastos et al. 1995).

Recruitment strategies for homosexual cohort studies in Brazil are limited and includes a study performed in Rio de Janeiro (Parker 1994). The summary of recruitment strategies in the present study includes: (1) direct information - Although dissemination among target populations (bars, gay territories, etc) has low recruitment efficacy it ensures awareness of the Project’s existence and availablity; (2) health units - Due to multiple problems encountered in the public system in Rio de Janeiro their contribution was poor; (3) NGOs involvement - Was crucial to the success and became apparent through the continuous and bilateral support between the NGOs and the Project. This was possible after the credibility and seriousness of the Project was established; (4) snowball recruitment - Has proven to be very effective and is now the main source of volunteers. Availability of "on demand" counselling and group activities have contributed to the volunteer’s personal involvement.

The STD disease prevalence in our study population was high. As one third of the volunteers in our cohort stated that they continue to practice unsafe sex, the incidence of the disease in this group can be expected to remain high, during the remainder of the cohort study.

The 23% HIV prevalence in this homosexual population compares to the 22% found at an STD clinic in Rio de Janeiro (Oscar Berro, personal communication). By comparison the prevalence in blood donors was 0.4% in Rio de Janeiro (Vanderborght et al. 1993) and 0.28% in southern Brazil (Vasconcelos et al. 1994). In other homosexual populations similar prevalence rates have been found (Morales et al. 1994) and even higher rates were found in North Carolina (Schoenbach et al. 1993), a state wthin the United States with low overall STD incidence. Our prevalence was higher than those found in San Francisco, within the United States (Lemp et al. 1994), Amsterdam in the Neterlands (Keet et al. 1993) and Tijuana, Mexico (Güerenã-Burgueño et al. 1991) where the rates, in similar populations, were 9, 5 and 12%, respectively.

In our study one third of the volunteers showed evidence of contact with syphilis. This data compares to blood donors studies (0.86%) in Rio de Janeiro (Vanderborght et al. 1993) and in southern Brazil (0.28%) (Vasconcelos et al. 1994). In a study with AIDS patients in São Paulo (Gir et al. 1994) the rate was similar (30%).

The hepatitis B antibody and/or HBsAg carrier rate in our study was high (46%) and rates found at blood donation centers were less than 20% (Clara Yoshida, personal communications). The HBsAg rate (3.7 in HIV negative and 9.1 in HIV positive volunteers) was also higher than that found among blood donors in Rio de Janeiro (Vanderborght et al1993) and southern Brazil (Vasconcelos et al. 1994) where rates were 1.6 and 0.78, respectively. In a study with patients with AIDS from São Paulo (Gir et al. 1994) this rate was 33% and in Europe (Lemp et al.1994) the rate was 20%. Our 80% positivity rate was similar to that found in Italy where anti-HBc in HIV (+) was 77% (Francisci et al. 1995).

Although this study is being carried on in a city where the overall prevalence in the general population of HIV and STD seems to be relatively low, the prevalence among the homosexual population remains high. This indicates that a high incidence can be expected among the subjects of our ongoing cohort.

Lessons learned from the syphilis epidemic should be an important guide in our response to the present HIV epidemic (Cutler & Arnold 1988). Efforts at individual and public training, frequent serological testing, scientific information programs, prevention methods (condom distribution) and goverment subsidised treatment for infected patients made difference even when effective treatments were not available.Another important by-product of this study refers to hepatitis B infection. If the results of our investigation are confirmed by the other centers participating in this multicentered study, this would provide support for public health authorities for enhancing preventive vaccination efforts and control strategies for hepatitis B infection. This has also been suggested by Meheus et al (1995).

The socio-demographic information from the first 200 volunteers in the HIV negative cohort is relevant to determine which type of population might be available for efficacy trials. In this study, the lower income population seems to be the most motivated group and the most willing to participate. The average age of the men recruited for the HIV negative cohort was lower than the male 18-50 year male population of the State of Rio de Janeiro. This is a relevant finding on the grounds that this is the age of expected higher risk due to the involvement in sexual experimentation.

The educational level reported in this study (46% of the volunteers have completed secondary education) is slightly higher than that reported for the general public (FIBGE 1991) where 31% of the urban males, between the ages of 20-49, completed secondary education. In one of our previous reports (Souza 1996) this rate was 49.1% and in a similar study perfored in Rio de Janeiro (Parker et al. 1995) this rate was 86.4%. This suggests that our investigation reflects a recruitment of men whose educational level is more compatible with that of the local population. Whenever evaluating recruitment techniques it is important to be aware of some selection bias that could jeopordise the study, by making it not representative of the general public.

The results obtained in Rio de Janeiro in the Brazilian multi-centered study offered evidence that the recruitment of la arge number of volunteers is feasible, both when multiple recruitment sources are used and there is direct involvement of the target population and the general public. The STD prevalence findings suggest that high incidence rates will be found in this community, making it feasible for Rio de Janeiro to be an effective partner of the Brazilian Ministry of Health’s efforts to prepare for future HIV vaccine efficacy trials.

 

ACKNOWLEDGMENTS

To the volunteers and NGO leadership which have understood the social importance of participating in this scientific effort, withstanding dificulties due to social and cultural pressures. Support provided by the Hospital Evandro Chagas (HEC) Clinical Laboratory and the National Reference Center of Viral Hepatitis made it possible to execute this study. To our staff members: Amilton Xavier Jr, Paulo Starling, Helena Santos, Dionne Peluso, Maria José Maia da Silva and Nelly Medeiros da Cruz. To Dr Carlos Serapião and Ms Therezinha Piemtzaeur for reviewing the manuscript.

REFERENCES

Bastos FI, Szwarcwald CL, Castilho EA 1995. Epidemia de AIDS em São Paulo: Subsídios da análise espacial. Bol Epidemiol AIDS 7: 5.

Cutler JC, Arnold RC 1988. Veneral Disease Control by Health Departments in the Past: Lesson for the Present. Am J Publ Health 78: 372-376.

Francisci D, Baldelli F, Papili R, Stagni R, Stagni G, Pauluzzi S 1995. Prevalence of HBV, HDV and HCV hepatitis markers in HIV positive patients. Eur J Epidemiol 11:123-126.

FIBGE - Fundação Instituto Brasileiro de Geografia e Estatística 1991. Censo Demográfico 1991: resultados do universo relativo às variáveis da população e dos domicílios. Rio de Janeiro, IBGE, 20, 262-281.

Gir E, Duarte G, Martinez R, Moriya TM, Figueredo IF, da Costa JC 1994. Expressão epidemiológica de outras doencas sexualmente transmissíveis entre pacientes de AIDS. Rev Saú Púb, 28: 93-99.

Greco D 1994. Aspectos éticos dos ensaios clínicos com vacinas anti-HIV/AIDS no Brasil, 16-21p. In PNDST/AIDS/ MS Cadernos técnicos 1..

Güerenã-Burgueño F, Benenson AS, Sepúlveda-Amor J 1991. HIV-1 prevalence in selected Tijuana sub-populations. Am J Publ Health 81: 623-625.

Heyward WL, Osmanov S, Saba J, Esparza J, Belsey E, Stoneburner R, Kaldor J, Smith PG 1994. Preparation for Phase III HIV vaccine efficacy trails: methods for the determination of HIV incidence. AIDS 8: 1285-1291.

Keet IP, van der Bergh HS, van Griensven GJ, Coutinho RA,Sandfort TG, van der Hoek JA 1993. HIV infectie en riskant sexueel gedrag onder jonge homoseksuele mannen te Amserdam, 1992. Ned Tijdschr Geneeskd 137: 2709-2712.

Lemp GF, Hirozawa AM, Givertz D, Nieri GN, Anderson L, Lindegren ML, Janssen RS, Katz M 1994. Seroprevalence of HIV and risk factors among young homosexual and bisexual men: The San Francisco / Berkeley Young Men’s Survey. JAMA 272:449-459.

Meheus A 1995. Risk of hepatitis B in adolescence and young adulthood. Vaccine 13 (Suppl. 1): S31-34.

Ministério da Saúde 1995. Boletim Epidemiológico de AIDS 8 :18.

Morales MA, Pineda JA, Leal M, Pino R, Torronteras R, Sanchez Quijano A, Lissen E 1994. Prevalencia de anticuerpos frente al virus de la hepatitis C en um coletivo de varones homosexuales. Med Clin Barc 102: 478-479.

Parker R 1994. Sexo entre homens: consciência de AIDS e comportamento sexual entre homens homossexuais e bissexuais no Brasil, p.129-149 In R Parker, C Bastos, J Galvão, JS Pedrosa (eds). A AIDS no Brasil. Rio de Janeiro, Relume Dumará, ABIA. IMS/UERJ.

Parker R, Mota MPda, Guimarães K, Terto Jr. V 1995. Práticas sexuais e conscientização sobre AIDS: Uma pesquisa sobre o comportamento homossexual e bissexual. Rio de Janeiro, ABIA: IMS, UERJ, mimeo. 160 pp.

Schoenbach VJ, Landis SE, Weber DJ, Mittal M, Koch GG, Levine PH 1993. HIV seroprevalence in sexually transmitted disease of clients in a low prevalence southern state. Evidence of endemic sexual transmission. Ann Epidemiol 3: 281-288.

Souza CTV 1996. Subsídio ao estudo de incidência do Vírus da Imunodeficiência Humana entre homens com práticas homossexuais e bissexuais no Rio de Janeiro. Msc Thesis, Escola Nacional de Saúde Pública / FIOCRUZ, Rio de Janeiro, 107 pp.

Sutmoller F, Morgado M, Bongertz V 1996. Por que o Brasil testa vacina anti-HIV? Ciência Hoje 20: 62-64.

Sutmoller F, Souza CTV, Penna T, Starling P, Martins H, Costa D, Silva MJ, Santos A, Silva G, Monteiro JC 1994 Aspectos críticos na condução de ensaios clínicos de vacinas na fase I / II no Brasil. Programa Nacional de DST / AIDS Ministério da Saúde Cadernos Técnicos 1: 22-26.

Vanderborgt BOM, Reis AMM, Rouzere CD, Salgado da Silva R, Yoshida CFT, Franco LGP, Maertens G, Van Heuverswijn H, Moreira Pereira J 1993. Prevalence of anti-hepatitis C virus in the blood donor population of Rio de Janeiro. Vox Sang 65:122-125.

Vasconcelos HCFF, Yoshida CFT, Vanderborght BOM, Schatzmayr HG 1994. Hepatitis B and C prevalences among blood donors in the South Region of Brazil. Mem Inst Oswaldo Cruz 89: 503-507.

WHO - World Health Organization 1994. Images of the epidemic. Geneva. 122 pp.

TABLE I

Recruitment sources of homo- and bisexual men for a HIV negative cohort per trimester (1994 through April 1996)

_____________________________________________________________

Trimester

Health units

Public information

Community

Volunteer

Total

           

1994

         

Jan. Mar.

1

-

1

1

3

Apr. Jun.

4

-

2

2

8

Jul. Set.

12

-

3

10

25

Oct. Dec.

8

4

5

3

20

Total

25

4

11

16

56

%

(45%)

( 7%)

(20%)

(29%)

 
           

1995

         

Jan. Mar.

13

12

11

4

40

Apr. Jun.

2

6

9

6

23

Jul. Set.

3

0

4

13

20

Oct. Dec.

1

0

28

39

68

Total

19

18

52

62

151

%

(13%)

(12%)

(34%)

(41%)

 
           

1996

         

Jan. Mar.

3

1

23

37

64

Apr.

2

1

12

18

33

Total

5

2

35

55

97

%

( 7%)

( 2%)

(36%)

(57%)

 
           

Total

49

24

98

133

304

Percent

(16%)

( 8%)

(32%)

(44%)

 
           

TABLE II

Seroprevalence of sexually transmitted diseases during the recruitment visit

for a homo- and bisexual HIV negative cohort study (1994-1996)

____________________________________________________________

Trimester HIV (+) Syphilis (+) Hepatitis B

n= 318 n=271 n=234

# (% pos) # (% pos) # (% pos)
             

1994

           

Jan. Mar.

1

33.3

1

100

0

-

Apr. Jun.

3

42.9

0

-

2

100

Jul. Set.

4

15.4

1

8.3

7

63.6

Oct. Dec.

5

26.3

4

25.0

7

46.6

Total

12

22.2

5

23.1

16

57.1

             
             

1995

           

Jan. Mar.

8

20.5

6

222

11

40.7

Apr. Jun.

5

20.0

5

20.8

12

54.5

Jul. Set.

1

4.8

7

41.2

6

42.9

Oct. Dec.

17

23.6

27

39.1

27

44.3

Total

31

19.7

44

32.4

56

45.2

             
             

1996

           

Jan. Mar.

20

32.2

28

46.7

28

49.1

Apr.

11

25.0

12

27.9

12

48.0

Total

31

 

40

 

40

 
             
             

Total

75

23.6

91

33.5

112

47.9

             

TABLE III

Seroprevalence of syphilis and hepatitis B at time of recruitment for the

HIV negative cohort study according to HIV serological status (1994-1996)

_____________________________________________________________

Syphilis Hepatitis B HBsAg

n=291 n=255 n=245

Serological status

#

%

#

%

#

%

             

HIV (+)

26

47.2

47

79.7

5

9.1

             

HIV ( -)

65

27.5

72

36.7

7

3.7

             

p-value

<.001

 

<.001

 

0.1

 

Chi-Square

TABLE IV

Age, income distribution and educational level of the cohort participants and comparison with the State of Rio de Janeiro 991 census

_____________________________________________________________

Cohort volunteers IBGE a

Number Percentage Number Percentage

_____________________________________________________________
 

n=200

 

2929449

 

Age (yrs):

       

18 - 19

14

7.0

216880

7.4

20 - 24

80

40.0

532460

18.2

25 - 29

37

18.5

532048

18.2

30 - 34

33

16.5

498277

17.0

35 - 39

20

10.0

437874

14.9

40 - 44

12

6.0

371687

12.7

45 - 49

4

2.0

286433

9.8

50

0

 

53790

1.8

         

Volunteers’ income

       

No income

19

9.5

59720

3.7

<1 MS b

5

2.5

26948

16.5

1-3

80

40.0

678673

42.0

3-6

38

16.0

263401

16.4

6-10

18

9.0

179958

11.2

10-14

5

2.5

76917

4.8

+14

12

6.0

85943

5.4

Unknown

5

2.5

-

-

         

Educational level

       

Illiterate

1

0.5

166677

16.6

Primary - incomplete

62

31.0

325555

32.4

- complete

40

20.0

204421

20.3

Secondary level

50

25.0

212636

21.2

University level

42

21.0

95820

9.5

Unknown

5

2.5

-

-

_____________________________________________________________

a : FIGBE 1991

b : minimum salary = R$100.00 = US$ 105.00

-

TABLE V

Socio-demographic variables of the first 200 participants of a homo-and bisexual HIV negative cohort (1994-1996)

__________________________________________________________

Number Percentage

Marital status

single

168

84.0

 

separated or married

23

11.5

 

other

9

4.5

       

Children

no

165

82.5

 

yes n=27

29

14.5

 

unknown

6

3.0

       

Living with

parents

100

50.0

 

alone

46

23.0

 

friend/companion

41

20.5

 

wife/ children

4

2.0

 

others

9

4.5

       

Employment

active

141

70.5

 

unemployed

46

23.0

 

student

7

3.5

 

unknown

6

3.0

       

Ethnic group

white

107

53.5

 

mulate

59

28.5

 

negro

18

9.0

 

indian

3

1.5

 

yellow

2

1.0

 

unknown

7

3.5

       

Sexually Transmitted Diseases prevalence among the candidates for the HIV negative cohort.

 

No STD = No sexually transmitted diseases

HIV+ = HIV positive, yet negative for syphylis and hepatitis B

HIV+/S/H = HIV positive and also positive for both or either syphilis and

hepatitis B

Rio, 30/08/96

Memórias do Instituto Oswaldo Cruz

Editorial Office

Dear Editor,

Thank you for your letter of August 19 in which you inform me that the manuscript has been accepted for publications wth the final corrections.

Please find enclosed thhe diskette with the changes and the two new versions of the article. The signed affidavit and the copyright form has also been included.

Thanking you for accepting this paper for publication, we remain,

Sincerely yours,

Frits Sutmoller

 

 

needs modifications and requested me to resubmit it with correction for a final decision. We made the suggested corrections proposed by the editors as well as the modifications proposed by the reviewers.

The title has changed and now reflects better the contents of this paper. A sentence was added clarifying the fact that this investigation is the Rio de Janeiro site of a multi-centered study. We have made a few minor organizational changes as recommended by reviewer 01 (i.e. recruitment instead of initial visit) and the English was revised. Specific comments such as the definition of "snowball" were rewritten and the abbreviations were explained. Statistical analysis was limited in this article accordingg with its descriptive nature. The question as to why so many references were made to the study by Parker is simple: up to the present moment, the other two Brazilian sites have not yet published their data and the only local study with a similar population is that done by ABIA and IMS/UERJ.

The statement "selection of a cohort ... has proven to be correct in Rio de Janeiro" has been rewritten to "seems to have been a proper choice in Rio de Janeiro". We agree that this will only be proven with incidence data and comprehensive socio-behavioral studies. The suggestion (reviewer 02) to include incidence data and social -behavioral aspects was appreciated and is being prepared for future publications.

Figure 1 has been removed. In our opinion the tables should not be removed or condensed as suggested. The detailed data will be revelvant whenever the Sao Paulo and Minas Gerais sites compare their studies to ours. If the editors wish to diminish the number of tables this can be discussed directly with you. The headings of tables have been changed and clarification included. The ethnics groups terms being used are those used in the questionnaire. References have been rechecked.

We feel that the above changes have improved the manuscript and we would like to thank you for your help, hoping now that you find it acceptable for publication.

Sincerely yours,

Frits Sutmoller

 

 

4.2 - Percepção de Vulnerabilidade à Infecção pelo HIV numa coorte de Homossexuais/Bissexuais Masculinos no Rio de Janeiro, Brasil. (Perception of Vulnerability to HIV infection in a cohort of homosexual/bisexual men in Rio de Janeiro, Brazil).

(AIDS Care - 11(5):567-579, 1999)

Financial support was provided by the National STD/AIDS Program of the Brazilian Ministry of Health; the Global Programme on AIDS/WHO; UNAIDS; Brazilian National Research Council (CNPq) and the Oswaldo Cruz Foundation (FIOCRUZ). Dr. Bastos was supported by a short-stay visiting researcher scholarship, sponsored by MRC (Medical Research Council of Canada) and CNPq.

Socio-demographic and Behavioural Characteristics, and Perception of Vulnerability to HIV Infection, in Homo/Bisexual Men Enrolled in a Cohort Study on HIV Incidence in Rio de Janeiro, Brazil

Introduction

The rapid spread of HIV-1 infection throughout the world, particularly in developing countries, motivated the World Health Organisation (WHO), and subsequently UNAIDS, to propose a comprehensive action plan for the evaluation of anti-HIV vaccine candidates in different geographic areas. An evaluation previously carried out by WHO’s Vaccine Development Committee (1991) concluded that Brazil, Thailand, Uganda and Rwanda fulfilled the minimum requirements to become trial sites. The HIV Vaccines Program in Brazil was launched in June 1992, with the appointment of a National Committee on HIV Vaccines. Three cities, Rio de Janeiro, São Paulo and Belo Horizonte, were selected to become potential vaccine test sites. These cities are located in the industrialised south-east of Brazil, the region with the highest cumulative AIDS incidence (Bastos et al., 1995). As of November 1997, more than120,000 AIDS cases had been registered in Brazil, with 39,667 (33%) of these occurring in homo- and bisexual men (Brazilian Ministry of Health, 1997). As of December 1997, Rio de Janeiro state registered 18,610 and8,091 (43.5%) cases, respectively (Rio de Janeiro State Health Secretariat, 1997).

In 1994, a multi-centre cohort study of HIV-negative men who have sex with other men (MSM) was initiated in order to determine HIV incidence and follow-up attrition rates, as well as assessingsocio-demographic and behavioural characteristics. The decision by the three regional centres that the cohorts would be composed of MSM was based on: the relevance of this population segment to the HIV/AIDS epidemic in Brazil; the fact that it is a population traditionally engaged in attempts to introduce safer sex practices; and its involvement in self-organising initiatives and actions concerned with human rights advocacy. Sexual transmission has been the main route of HIV spread in Brazil. Until 1995, the yearly number of AIDS cases registered among MSM increased continuously, although since the beginning of the present decade the proportion of cases in this exposure category has decreased as a result of the increase in numbers of cases due to heterosexual transmission and to sharing of injecting equipment (Bastos et al., 1995).

Epidemiological studies in a number of different countries indicate that several factors are consistently associated with new HIV infections in MSM, such as unprotected anal intercourse, having a known HIV-positive partner, having a sexually transmitted disease (STD), and injection drug use (Buchbinder at al., 1996; DiFranceisco et al., 1996; Williams et al., 1996). Some recent reports suggest that HIV transmissionmay also occur as a result of unprotected oral sex (Page-Schafer et al., 1997; Bratt et al., 1997). With respect to psychosocial factors, poor knowledge about AIDS (Kippax et al., 1995; Kelly et al., 1990), lack of perceived severity of AIDS as an illness, negative attitudes and low self-efficacy with respect to condom use (Aspinwall et al., 1991; Godin et al., 1993), as well as normative factors in the social environment (McKusick et al., 1990), have all been found to be associated with risky sexual behaviour.

The relationship between perception of personal vulnerability to HIV infection and the enactment of risky/safe sexual behaviour is however less clear. Some investigations, such as that carried out by Moore and Rosenthal in adolescents (1991), indicate a contradiction between high levels of reported risk behaviours and perception of invulnerability to HIV infection. Other studies suggest, however, that MSM who practise risky sexual behaviour are also more aware of their risk of becoming infected (Kelly et al., 1995; Bosga et al., 1995). Differing results with respect to this issue may be due to a number of different issues such as cultural context, underlying conceptual and theoretical models, study design and the nature of the measures employed (Gerrard et al., 1996; Poppen & Reisen, 1997). A better understanding of this issue is important, since perception of personal vulnerability may be seen as a necessary first step towards behaviour change.

In addition, characterisation of patterns of risk behaviours, as well as the perception of volunteers of their vulnerability to HIV infection, are core issues in vaccine preparedness studies. Koblin et al. (1998) and Bartholow et al. (1997) have shown that participants reporting recent high-risk behaviours were more likely to be willing to participate in vaccine preparatory studies than those not reporting these behaviours. These findings are essential to the feasibility of future vaccine trials, since they demonstrate the capacity to recruit volunteers at risk. There is also a need to gather data relevant to these questions in different cultural settings, in order to cross-compare perceived and actual risks.

In this paper, we present baseline socio-demographic and behavioural data on MSM enrolled in the cohort in Rio de Janeiro from January 1994 through March 1997. We also analyse the relationship between perception of vulnerability to HIV infection and socio-behavioural characteristics, specifically the enactment of risky sexual behaviour, as well as with laboratory data indicating exposure to STDs. Cohort members engaged in regular commercial sex activities (known locally as "michês"), as well as transvestites and transsexuals (who are usually also engaged in commercial sex), were excluded from the present analysis, since exploratory analyses of our data, as well as previously published studies of comparable groups (Parker et al., 1992), indicate that these groups have distinct socio-demographic characteristics and behaviours, composing a true sub-sample of the whole cohort.

Methodology

The study was reviewed and approved by the FIOCRUZ Ethics Committee. Study procedures, recruitment sources and sexually transmitted disease (STD) seroprevalence rates among potential volunteers have been reported previously (Sutmoller et al., 1997). We briefly describe here the procedures used in the cohort study, focusing on issues relevant to this manuscript.

Study population: Entry criteria for this study were being male, HIV-seronegative, having sexual activity with other men (male homosexual intercourse), being aged between 18 and 50 years of age, not being engaged in regular commercial sex nor being a transvestite/transsexual, and having agreed with the informed consent procedures. Volunteers were recruited through a combined strategy involving mainly outreach activities, e.g. recruitment by study volunteers themselves ("snow-ball") (45% of the volunteers here analysed) and by non-governmental organisations (NGOs) (34.8%), as well as by media contacts (7.7%) with advertising, targeted interviews, etc., and referrals from health care facilities (12.5%). The most important factors determining the enrollment of these volunteers were: the desire to know their HIV status and their wish to engage in a project that provides volunteers with preventive resources, e.g. condoms, education and the support of a multidisciplinary team.

Study procedures: At the first (recruitment) interview basic socio-demographic data, opinions, HIV knowledge and attitudes on health issues were obtained, and an explanation of the Programa Rio study given. The volunteers were then given pre-test counselling with subsequent blood specimen collection for HIV, syphilis (VDRL, with confirmation using TPHA) and hepatitis B (anti-HBc, anti-HBs, HBsAg) testing. On their return, one week later, the volunteers received the HIV and other test results during a session of post-test counselling. The HIV-positive individuals were referred to public health care facilities for follow-up care and those who were HIV-negative were asked to enroll in the study after signing the informed consent form. A more detailed questionnaire was administered after enrollment.

Standardised questionnaires were developed and applied by specifically trained study staff at the three Centres. The socio-behavioural questionnaire was composed of 100 (mainly closed) questions on: socio-demographic characteristics, knowledge about HIV transmission routes and prevention strategies, beliefs and attitudes towards AIDS and sexual life, sexual practices, and STD/AIDS preventive initiatives.

The questions on knowledge of transmission routes, perception of vulnerability and protection from infection were formulated as multiple-option questions, permitting the interviewees to give at their discretion as many answers as they wished. Perception of vulnerability to HIV infection was addressed through a direct question: "Do you think that you are at risk for HIV infection?", answers being yes/no, with further detailing of the reasons for feeling "vulnerable" or not.

Statistical analysis: Data were entered on a Foxpro (2.0) spreadsheet and analysed (frequency, correlation and statistics) with SPSS for Windows 95 package for PCs (Norusis, 1997). Significance was defined by p-values less than 0.05. In the univariate analyses, associations with vulnerability were assessed through chi-square tests, Yates corrected, and 95% confidence intervals for the odds-ratios (OR). Fisher’s exact test was used when necessary.

To further assess the role of the different variables in a multivariate form, a stepwise forward logistic regression was carried out, maximizing at each step the likelihood ratio, with perception of vulnerability to HIV infection as the dependent variable. The significance level for variable inclusion was 5% and that for exclusion, 10%.

Results

During the period of the study, 818 men were interviewed as potential study participants; however, 24% of these tested positive for HIV. By March 1997, 470 persons had actually enrolled in the study.The sample analysed here was composed of 295 volunteers, of whom 177 (60%) were men reporting strictly homosexual practices, and 118 (40%) reporting sexual intercourse with both men and women.

Mean age was 27.6 years, with a standard deviation of 7.26 and a mode of 23 years old (Table 1). With respect to marital status, 251 volunteers (85.1%) were single, 33(11.2%) married or divorced and 11(3.7%) classified in a miscellaneous category which includes stable relationships with boyfriends (not currently defined by Brazilian law as "legal unions"). Most of the interviewees - 227 (77%) - lived with relatives and friends, while 68 (23%) lived alone. The majority of interviewees (70.5%) were in paid employment; 187 (65%) interviewees received a monthly wage of between 1 and 6 minimum Brazilian salaries (approx. US$ 122.00 - 732.00). Half (54.7%) of the sample had completed full high school education.

TABLE 1

The vast majority of interviewees (75.6%) reported having changed their sexual practices since the onset of the AIDS epidemic (Table 2); around 58% of those reporting behavioural change mentioned the systematic use of condoms. Other reported behavioural changes included: a redefinition of partnerships (less partners, exclusive relationship with a principal partner, "sex only with chosen partners", sex only with former - supposedly better known - partners), and changes in sexual practices themselves (such as lower frequencies of penetrative intercourse).

Two hundred and two subjects (68.5%), in an open question with multiple answers permitted, stated that HIV infection could be transmitted through unprotected sexual intercourse (non-specified); 208 (71%) volunteers mentioned blood transfusion as a risk factor and 164 (55.6%) spontaneously mentioned the risk of needle sharing among injecting drug users.

Nevertheless, the practice of unprotected anal intercourse, both insertive and receptive, continued to be reported with relatively high frequencies among those with both principal and casual partners in the six months prior to the interview.Among those with a principal partner (168/295 or 57% of the cohort), 40% (67/168) engaged in unprotected insertive, and 33% (60/168) in unprotected receptive anal intercourse. Among those with casual partners (180/295 or 61% of the cohort), 36% (65/180) had engaged in unprotected insertive, and 27% (49/180) in unprotected receptive anal sex.

For the whole sample under analysis, the frequency of unprotected insertive and receptive anal sex was therefore 23% (67/295) and 20% (60/295) respectively for sex with permanent partners; and 22% (65/295) and 17% (49/295) respectively for occasional partners.

Despite the fact 40% of the men interviewed described themselves as bisexuals, actual rates of sexual intercourse of these men with their female partners were very low, precluding any kind of valid analysis.

The prevalence of STDs in the cohort was discussed in a previous paper (Sutmoller et al., 1997), and STD risk data presented here refer to sexual partnerships: 11.8% reported having had sexual intercourse with partners supposedly infected with a STD other than HIV in the six months prior to interview. Twenty-four (8%) interviewees reported having had sexual intercourse with known HIV-positive partners during the same period, 19 (79%) reporting systematic condom use with the latter.

Illicit drug use was rarely reported, with 83.3% of the subjects reporting never having used any kind of illicit drug. The consumption of alcoholic beverages was much more frequently reported, and almost all interviewees had used it on at least a few occasions during the six months prior to interview; 51.5% reported having drunk alcohol before and during dating and ensuing sexual intercourse, but only 9.8% subjects affirmed having engaged in unsafe sexual practices "due to the fact they were under the influence of alcohol".

TABLE 2

With respect to perception of vulnerability to HIV infection, 185 (72% of the valid answers) of the volunteers interviewed perceived themselves as being "vulnerable" (at risk). The main reasons (in a open question with multiple answers permitted) given for perceiving themselves as "vulnerable" are presented in Table 3. Variables related both to individual/dyadic behaviours (e.g. prediction of the difficulty of being engaged only in safe sexual relationships in the foreseeable future), as well as variables related to the broader social context (e.g. fear of receiving a transfusion of infected blood or the rather fatalistic statement that "everyone is at risk of being infected with HIV"), were mentioned by the interviewees. No interviewees perceived themselves as vulnerable based only on such "broader social context" variables: all mentioned at least one variable associated with risky sexual behaviour.

Seventy two (28% of the valid answers) interviewees did not perceive themselves as being at risk (i.e. perceived themselves as "non-vulnerable") for HIV infection. Among these respondents (free to list multiple answers) the vast majority "trust in condoms as a preventive measure", referring use of them in all sexual relationships in which they engaged. The 38 men with no answer to this question, or with inconclusive answers, were excluded from the final analysis.

TABLE 3

We present in Table 4 all statistically significant associations between perception of vulnerability and socio-demographic/behavioural variables. As shown in Table 4, perception of vulnerability was associated with higher school attainment (OR=2.4; p=0.002). Other socio-demographic variables assessed but showing no significant association were age, housing, employment, income, religion, and attending or not health care facilities (data not shown).

With regard to sexual behaviour during the six months prior to interview, we found that those reporting any unprotected anal intercourse (OR=2.3; p=0.004), were significantly more likely to perceive themselves as vulnerable to HIV infection. Other associations were observed with respect to those reporting unprotected sexual intercourse with casual partners (OR=2.6, p=0.001). No other significant associations were found among other variables tested, including age of first sexual intercourse with other men and/or women, different "scenes" from where partners were recruited, behavioural changes after the AIDS epidemic and sex with HIV-positive partners (the lack of association with this latter variable may be due to the relatively low numbers of MSM reporting sex with known HIV-positive partners).

TABLE 4

The final multiple logistic regression model (model chi-square = 24.688; p=0.0000) identified as factors predictive of perception of vulnerability, following the inclusion order as defined by the stepwise procedure: "educational level" (defined as an ordinal variable); "anal intercourse without condoms, with any male partners in the last 6 months" and"any unprotected intercourse with casual partners in the last 6 months" (Table 5).

TABLE 5

Data on syphilis and hepatitis B seroprevalence (HBsAg, anti-HBs and anti-HBc) were available for all interviewees. As shown in Table 6, no statistically significant associations were found between perception of vulnerability/non-vulnerability and seropositivity for syphilis (p=0.06); hepatitis (p=0.8), or the presence of HBsAg (p=0.5). Syphilis seroprevalence was actually higher among volunteers perceiving themselves as "non-vulnerable", although this difference was not statistically significant.

TABLE 6

Discussion

In this study we present socio-demographic and behavioural profiles, as well as data relating to knowledge of / attitudes towards HIV/AIDS, from 295 HIV-negative volunteers taking part in the first cohort study on incidence of HIV infection in MSM in Brazil, at the Rio de Janeiro HIV Vaccine Centre. We also analyse the relationship between perception of vulnerability to HIV infection and other socio-demographic, behavioural and laboratory variables in the study population.

The socio-demographic characteristics of our sample are comparable to those from another study on MSM in Rio de Janeiro (Parker et al., 1998). Educational attainment and salary levels were somewhat higher in our sample than those of the general population. A significant proportion of the sample here analysed were involved in NGOs and/or recruited through cohort participants, which in a country with profound social inequalities could be interpreted as belonging to more "conscious" social strata, with an associated tendency towards higher educational levels / higher professional status (Sutmoller et al., 1997; FIBGE, 1991). The proportion of MSM defining themselves as bisexual was higher in our sample than studies carried out by Parker et al. (1998), where, in 1995, 73% of MSM perceived themselves as homosexual/gay. However, in our study, we found that most MSM defining themselves as bisexuals had actually engaged very infrequently in sexual intercourse with women. The latter issue, however, merits further consideration, given the recent dramatic increase in the number of women infected with HIV through heterosexual intercourse in Brazil, many of whom report having had bisexual male partners (Bastos et al., 1995).

Our results indicate that knowledge regarding HIV transmission and prevention was generally adequate in our sample, as was also found in the study by Parker et al. (1998). Most (76%) of the volunteers reported having changed their behaviour as a consequence of the AIDS epidemic; these changes included: reduction in the number of sexual partners, more careful choice of sexual partners, more frequent use of condoms, and lower frequencies of penetrative anal intercourse. Despite this, unprotected anal sex with both principal and occasional partners remained a common practice (almost one quarter of the sample under analysis reported unprotected anal intercourse in the 6 months prior to interview), suggesting that preventive attitudes are probably not being consistently acted out as safer behaviours. MSM from other countries also report relatively high levels of risky behaviour: over 50% of a sample of 250 MSM in San Francisco reported having unprotected anal intercourse in the last 6 months (Peterson et al., 1992); these men were more likely to be poor and to have a higher perceived risk vis-à-vis HIV infection. In New York, 48.7% of a sample of 698 MSM reported receptive anal sex in the last 3 months, with 34% of these acts unprotected or with condom breakage or slippage (Koblin et al., 1997).

Of particular note in our study is the high frequency of reported unprotected anal sex with casual partners (22% of the sample), and the fact that the proportion of the cohort reporting this behaviour is very similar to that reporting unprotected anal sex with principal partners (23%). This differs from results from studies in other countries, which tend to show that the usual intimacy of stable relationships, with its atmosphere (accurate or not) of mutual trust, is one of the privileged contexts for unprotected anal intercourse. Data from two cohort studies of MSM in Canada (Vancouver and Montreal) show that approximately 26% of cohort members engage in unprotected sex with principal partners, in contrast to only 9-14% who engage in unprotected anal sex with casual partners (Dufour et al., 1997: Strathdee et al., 1997).

Such findingsmay be a reflection of differences in the definition of what constitutes a safe as opposed to a risky sexual relationship for MSM in different contexts. It would seem a very common practice for MSM to be aware of their serostatus in some countries: for instance, 87% of MSM enrolling in a cohort study in Montreal had already been tested for HIV at least once (Turmel et al., 1997). Similar figures were found by Koblin et al. (1997), with 84% of volunteers of Project ACHIEVE, in New York City, reporting having been previously tested for HIV. The figure for the members of the Rio cohort previously tested was much lower, at 45.4%.

HIV testing in Rio de Janeiro,although available at no cost in public facilities, is performed with significant delay. This probably explains the fact that the most frequent reason given by volunteers for participation in the Projeto Rio cohort was the desire to have a reliable and rapid HIV test result. It is also noteworthy that the prevalence of HIV in initial volunteers (i.e. before exclusion of seropositive individuals from the cohort study) was 24% in Rio, despite the fact that the cohort recruitment material stated that only seronegative MSM or MSM with unknown serostatus should consider participation. This contrasts with baseline HIV prevalences of less than 2% in the two cohorts cited above (Dufour et al., 1997; Strathdee et al., 1997), suggesting that large numbers of MSM in Rio are unaware of their serostatus, although many of them perceive themselves as being at risk of infection.

Among the 72 subjects perceiving themselves as being "not vulnerable to HIV infection", the main reason given was the regular use of condoms. Participants who perceived themselves as being at risk tended more frequently to have practised unprotected anal sex both with their principal and casual partners in the 6 months prior to interview, and to have had casual sex partners more frequently.

The observed association between the reporting of recent risky sexual practices and perception of vulnerability is encouraging, since it may represent a first step towards behaviour change. Indeed, some of the more influential models of behaviour change, such as the Health Belief Model (Becker, 1974) and the ARRM (AIDS Risk Reduction Model - Catania et al., 1990), highlight "perceived personal susceptibility" as one of their core components. Relatively high frequencies of risky behaviour were nevertheless reported by cohort members in this study, suggesting, as found in many other studies that knowledge does not necessarily translate into safer practices (Valdisserri et al., 1988; Perkins et al., 1993). Barriers to behavioural change identified in such studies include difficulties in using condoms in challenging situations, for example where a high degree of affective bonding and/or sexual attraction exists, as well as factors such as lack of social support (Dilley et al., 1998).

The lack of observed association between serological markers of recent and past STD infection and perception of vulnerability may also reflect that fact that cohort members who perceive themselves as non-vulnerable do not in fact consistently practise safe sex. Alternatively, this may be a function of the cross-sectional design of the present study (Gerrard, et al., 1996). While acute Hepatitis B infection may be a relatively good surrogate marker of recent exposure to infectious agents through unprotected sexual intercourse (particularly since the use of illicit drugs is exceptional in our cohort), the small number of cohort members positive for HbsAg (only five), and the resultant lack of statistical power, may account for the lack of association observed between this marker and perception of vulnerability to HIV infection.

We found that interviewees with higher levels of formal education more frequently perceived themselves as being vulnerable to HIV infection. Although people with lower educational attainment seem to be, in fact, less empowered to face the HIV/AIDS challenge (Mann et al., 1992), the general level of information in the sample is good, with no meaningful differences among volunteers with different socio-economic backgrounds. Other factors such as denial of risk, or the relationship between specific information on HIV/AIDS and broader attitudes and beliefs, could explain the differences observed.

In a community survey carried out in São Paulo, in 1993-5 (De Franco et al., 1998), a subgroup of MSM, composed of younger and less educated men, belonging to the lower social strata, reported higher levels of risk behaviours. The same men perceived themselves as being at less risk of HIV infection, reported more difficulties in negotiating and enacting safer sex, and were less integrated in the gay community, as well as being out-of-reach of most preventionprogrammes.

Finally, we would like to register here that ten subjects from the sample under analysis seroconverted between February 1996 and October 1998, among these being 6 who considered themselves as vulnerable to HIV infection, 3 as non-vulnerable, and with one no response.

In conclusion, although the data presented in this paper demonstrate that the HIV-negative participants of the Rio de Janeiro MSM cohort had good levels of knowledge regarding HIV transmission, in addition to a high rate of reported behavioural change as a result of the AIDS epidemic, it would appear that a gap continues to exist between knowledge and action in the sense that relatively high proportions of interviewees continued to engage in risky sexual behaviours. The association between perception of vulnerability/invulnerability to HIV infection and risky/safe sexual behaviours is encouraging, but whether this perception has already or could in the future translate into actual behavioural change and effective risk reduction is as yet unclear.

Acknowledgments

To the study volunteers and NGO leaders who have understood the social importance of participating in this scientific effort, notwithstanding difficulties due to social and cultural pressures. Support provided by the staff of the Hospital Evandro Chagas (HEC), the Clinical Laboratory/National Reference Centre of Viral Hepatitis (IOC/FIOCRUZ) made it possible to execute this study.

References:

ASPINWALL, L.G., KEMERY, M.E., TAYLOR, S.E., SCHNEIDER, S.G., DUDLEY. J.P. (1991). Psychosocial predictors of gay men’s AIDS risk-reduction behavior. Health Psychology, 10, 432-444.

BARTHOLOW, B.N., MacQUEEN, K.M., DOUGLAS Jr., J.M., BUCHBINDER, S.,

McKIRNAN, D. & JUDSON, F.N. (1997). Assessment of the changing willingness to participate in Phase III Vaccine trials among Men who have sex with men. Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology, 16, 108 - 115.

BASTOS, F.I., TELLES, P.R., CASTILHO, E.A., BARCELLOS C. (1995). A epidemia de AIDS no Brasil [The AIDS epidemic in Brazil]. In: M.C.S. Minayo (Ed.), Os Muitos Brasis: Saúde e População na década de 80 (pp. 245-268). Rio de Janeiro: Hucitec/ABRASCO.

BECKER, M.H. (1974). The health belief model and personal health behavior. Health Education Monographs, 2, 324-508.

BOSGA, M.B., de WIT, J.B., de VROOME, E.M., HOUWELING, H., SCHOP, W., SANDFORT, T.G. (1995). Differences in perception of risk for HIV infection with steady and non-steady partners among homosexual men. AIDS Education and Prevention, 7, 103-115.

BRATT, G.A., BERGLUND, T., GLANTZBERG, B.L., ALBERT, J., SANSTROM, E. (1997). Two cases of oral-to-genital HIV-1 transmission. International Journal of STD & AIDS, 8, 522-5.

BRAZILIAN MINISTRY OF HEALTH (1997). Boletim Epidemiológico de AIDS, X, 1-41.[Epidemiological Bulletin].

BUCHBINDER, S.P., DOUGLAS JR, J.M., MCKIRNAN. D.J., JUDSON, F.N., KATZ. M.H., MacQUEEN. K.M. (1996) Feasibility of human immunodeficiency virus vaccine trials in homosexual men in the United States: Risk behavior, seroincidence, and willingness to participate. Journal of Infectious Diseases, 174, 954-61.

CATANIA, J.A., KEGELES, S. & COATES, T. (1990). Towards an understanding of risk behavior: An AIDS risk reduction model (ARRM). Health Education Quarterly, 17, 53-72.

DE FRANCO, E., D’ANGELO, FERNANDES, M.E.L. & PARKER, R.G. (1998). Práticas sexuais e conscientização sobre AIDS: Uma pesquisa sobre o comportamento homossexual e bissexual em São Paulo [Sexual practices and knowledge on AIDS: A survey on homosexual and bisexual in São Paulo]. In: R. Parker & V. Terto Jr. (Eds.), Entre Homens: Homossexualidade e AIDS no Brasil (pp. 49-62). Rio de Janeiro: ABIA.

DIFRANCEISCO, W., OSTROW. D.G. & CHMIEL, J.S. (1996). Sexual adventurism, high-risk behavior, and human immunodeficiency virus-1 seroconversion among Chicago MACS-CCS cohort, 1984-92: A case-control study. Sexually Transmitted Diseases, 23, 453-60.

DILLEY, J.W., McFARLAND, W., SULLIVAN, P. & DISCEPOLA, M. (1998). Psychosocial correlates of unprotected anal sex in a cohort of gay men attending an HIV-negative support group. AIDS Education and Prevention, 10, 317-326.

DUFOUR, A., ALARY, M., MÂSSE, B., REMIS, R., OTIS, J., PARENT, R., TURMEL, B.,

LAVOIE, R., LE CLERC, R., VINCELETTE, J. & The Omega Study Group (1997). Risk behaviours and incidence among Omega Cohort participants: Preliminary data. Canadian Journal of Infectious Diseases, 8(suppl. A), 23A.

FUNDAÇÃO INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATÍSTICA (FIBGE) (1991). Censo Demográfico 1991[Brazilian National Census - 1991]. Rio de Janeiro: FIBGE.

GERRARD, M., GIBBONS, F.X. & BUSHMAN, B.J. (1996). Relation between perceived vulnerability to HIV and precautionary sexual behavior. Psychological Bulletin, 119, 390-409.

GODIN, G., MYERS, T., LAMBERT, J., CALZAVARA, L.M., LOCKER, D., and the Canadian AIDS Society (1993). Understanding the intention of gay and bisexual men to use condoms for insertive or receptive anal sex. Canadian Journal of Infectious Diseases, 4(suppl.), 43B.

KELLY, J.A., St-LAWRENCE, J.S., BRASFIELD, T.L., LEMKE, A., AMIDEI, T., ROFFMAN, R.E., HOOD, H.V., SMITH, J.E., KILGORE, H. McNEILL, C. JR. (1990). Psychological factors that predict AIDS high-risk versus AIDS precautionary behavior. Journal of Consultant Clinical Psychology, 58, 117-120.

KELLY, J.A., SIKKEMA, K.J., WINETT, R.A., SOLOMON, L.J., ROFFMAN, R.A., HECKMAN, T.G., STEVENSON, L.Y., PERRY, M.J., NORMAN, A.D., DESIDERATO, L.J. (1995). Factors predicting continued high-risk behavior among gay men in small cities: psychological, behavioral, and demographic characteristics related to unsafe sex. Journal of Consultant Clinical Psychology, 63, 101-107.

KIPPAX, S., CRAWFORD, J., RODDEN, P. & NOBLE, J. (1995). Predictors of unprotected male-to-male anal intercourse with casual partners in a national sample. Australian Journal of Public Health, 19, 132-138.

KOBLIN, B.A., AVRETT, S., TAYLOR, P.E. & STEVENS, C.E. (1997). Willingness to participate in HIV-1 vaccine efficacy trials and the effect of media events among gay and bisexual men in New York City: Project ACHIEVE. Journal of Acquired Immune Deficiency and Human Retrovirology, 15, 165-171.

KOBLIN, B. A., HEAGERTY, P., SHEON, A., BUCHBINDER, S., CELUM, C., DOUGLAS, J. M., GROSS, M., MARMOR, M., MAYER, K., METZGER, D.& SEAGE, G. (1998). Readiness of high-risk populations in the HIV Network for Prevention Trials to participate in HIV vaccine efficacy trials in the United States. AIDS, 12, 785-793.

MANN, J., TARANTOLA, D.J.M. & NETTER, T.W. (Eds.) (1992). AIDS in the World. Cambridge/London: Harvard University Press.

McKUSICK, L., COATES, T.J., MORIN, S.T., POLLACK, L. & HOFF, C. (1990). Longitudinal predictors of reduction in unprotected anal intercourse among gay men in San Francisco: The AIDS behavioral project. American Journal of Public Health, 1990, 80, 978-983.

NORUSIS, M.J. (1997). SPSS 7.5. Guide to data analysis. Upper Saddle River, NJ: Prentice Hall.

PAGE-SHAFER, K., VEUGELERS, P.J., MOSS. A.R., STRATHDEE, S.A., KALDOR. J.M., van GRIENSVEN, G.J (1997) Sexual risk behavior and risk factors for HIV-1 seroconversion in homosexual men participating in the Tricontinental Seroconverter Study, 1982-1994. American Journal of Epidemiology, 146, 531-42.

PARKER, R.G., LARVIE, P., CARDOSO JR., R. (1992).Programa Pegação: An outreach program for male commercial sex workers in Rio de Janeiro. In: L. S. Bond (Ed.), A Portfolio of AIDS/DST Behavioral Interventions and Research (pp. 171-178). Washington: PAHO.

PARKER, R.G., MOTA, M.P., ALMEIDA, V, TERTO JR., V. & RAXACH, J.L.C. (1998). Práticas sexuais e mudança de comportamento entre homens que fazem sexo com homens no Rio de Janeiro, 1990 - 1995 [Sexual practices and behavioural change among MSM in Rio de Janeiro, 1990 - 1995]. In: R. Parker & V. Terto Jr. (Eds.), Entre Homens: Homossexualidade e AIDS no Brasil (pp. 15-48). Rio de Janeiro: ABIA.

PERKINS, D.O., LESERMAN, J., MURPHY, C.& EVANS, D. L. (1993). Psychosocial predictors of high-risk sexual behavior among HIV-negative homosexual men. AIDS Education and Prevention, 5, 141-152.

PETERSON, J.L.., COATES, T.J., CATANIA, J.A.., MIDDLETON, L., HILLIARD, B., & HEARST, N. (1992). High-risk sexual behavior and condom use among gay and bisexual African-american men. American Journal Public Health, 82, 1490-1494

POPPEN, P.J. & REISEN, C.A. (1997). Perception of risk and sexual self-protective behavior: A methodological critique. AIDS Education and Prevention, 9, 373-390.

RIO DE JANEIRO STATE HEALTH SECRETARIAT (1997). Boletim DST/AIDS, August-December[Epidemiological Bulletin].

STRATHDEE, S.A., MARTINDALE, S.L., HOGG, R.S., CORNELISSE, P.G., EGAN, J., COOK, D., REKART, M.L., MONTANER, J.S.G., O’SHAUGHNESSY, M.V., SCHECHTER, M.T. for the Vanguard Project (1997). HIV Prevalence, incidence and risk behaviours among a cohort of young gay/bisexual men. Canadian Journal of Infectious Diseases , 8(suppl. A), 24A.

SUTMOLLER, F., SOUZA, C.T.V., MONTEIRO, J.C. & PENNA, T. (1997). The Rio de Janeiro HIV Vaccine Site - I Recruitment strategies and socio-demographic data of a HIV negative homosexual and bisexual male cohort in Rio de Janeiro, Brazil. Memórias do Instituto Oswaldo Cruz, 92, 39-46.

TURMEL, B., LAVOIE, R., DUFOUR, A. , LE CLERC, R., ALARY, M., REMIS, R., OTIS, J., MÂSSE, B., VINCELETTE, J., PARENT, R. et Le Groupe de Recherche Oméga (1997). Expériences antérieures de déspistage VIH et degré de satisfaction chez les participants à la Cohorte Oméga. Canadian Journal of Infectious Diseases, (suppl. A), 33A.

VALDISSERRI, R.O., LYTER, D., LEVITON, L.C., CALLAHAN, C. M., KINGSLEY, L.A., RINALDO, C.R. (1988). Variables influencing condom use in a cohort of gay and bisexual Men. American Journal Public Health, 78, 801-805.

WILLIAMS, D.I., STEPHENSON, J.M., HART, G.J., COPAS, A., JOHNSON. A.M. & WILLIAMS, I.G. (1996). A case control study of HIV seroconversion in gay men, 1988-1993: What are the current risk factors? Genitourinary Medicine, 72, 193-6.

WHO (World Health Organization) (1990). Prevenção contra a transmissão sexual do vírus da imunodeficiência humana. São Paulo: Ed. Santos [Portuguese edition].

Table 1

Socio-demographic characteristics of the 295 participants in the HIV-negative cohort of MSM in Rio de Janeiro

Variables

 

Number

Percentage

Age (yrs.):

18 - 19

29

9.8

 

20 - 24

96

32.5

 

25 - 29

70

23.7

 

30 - 34

50

17.0

 

35 - 39

26

8.8

 

40 - 44

17

5.8

 

45 - 49

7

2.4

       

Marital status

Single

251

85.1

 

currently or previously married

33

11.2

 

Other

11

3.7

       

Living with

Parents

150

51.0

 

Alone

68

23.0

 

friend/companion

60

20.3

 

wife/ children

9

3.0

 

Others

8

2.7

       

Employment

Active

208

70.5

 

Unemployed

81

27.5

 

Unknown

6

2.0

       

Volunteers’ income

No income

27

9.1

 

<1 MS a

12

4.1

 

1-3

131

44.4

 

3-6

56

19.0

 

6-10

32

10.8

 

10-14

7

2.4

 

+14

12

4.1

 

Unknown

18

6.1

       

Educational level

Illiterate

-

-

 

Primary - incomplete

46

15.6

 

- complete

86

29.1

 

Secondary level

104

35.2

 

University level

56

19.0

 

Unknown

3

1.1

a Minimum salary currently about R$120.00/US$ 125.00

Table 2

Behavioural change as a result of the AIDS epidemic as reported by the 295 participants of the HIV-negative cohort of MSM, in Rio de Janeiro.

Behavioural change

n

%

No behavioural change

52

17.6

Behavioural change #

223

75.6

Condom use in most relationships

129

57.8*

Reduction in the number of sexual partners

72

32.3*

Sex only with carefully chosen sexual partners

76

34.0*

No new partners, due to fear of AIDS

27

12.0*

Only mutually faithful partnerships

30

13.5*

Safer sexual practices (e.g. lower frequency of penetrative anal sex )

150

67.2*

Other answers

22

9.8*

No answer/incoherent answer

20

6.8

# Behavioural change items allow for multiple answers.

* Of those reporting behavioural change.

Table 3

Grounds (multiple answers allowed) for self-perception of (in)vulnerability, given by volunteers in the HIV-negative cohort of MSM in Rio de Janeiro.

 

Responses

n

%

 

"Afraid to receive a transfusion in the near future"

68

37.0

 

"Everyone is at risk of being infected with HIV"

82

45.0

 

"Afraid condoms could break.."

73

40.0

Vulnerable

"At risk due to very fact one is a male homo/bisexual"

30

16.0

(n=185)

"Has seropositive sexual partners"

09

4.9

 

"Scared by own lack of control over his sexual desires and behaviour"

32

17.0

 

"Speculate about having unprotected sexual intercourse in the foreseeable future "

80

44.0

 

"Trust in condoms as a preventive measure, using them in all sexual relations"

57

79.0

Non- Vulnerable

"Has sex only with principal partners"

21

29.0

(n=72)

"Now, has a smaller number of sexual partners"

21

29.0

 

"Chooses partnerships carefully"

19

26.0

Table 4

Associations between main socio-demographic and risky behavioural variables and perception of (in)vulnerability to HIV infection among the 257 HIV-negative cohort participants with valid answers

Variables

"Vulnerability"*

"Non-vulnerability"**

OR

95%

CI

p-value***

 

N

%

N

%

     

Social-demographic

             

Higher educational level ****

76

41

45

64

2.4

1.4-4.2

0.002

Behaviours / Practices

             

Anal intercourse without condoms, with (male) partners, in the last 6 months

109

59

28

39

2.3

1.3-3.9

0.004

Any unprotected intercourse with occasional partners, in the last 6 months

107

58

25

35

2.6

1.5-4.5

0.001

* 185 valid answers, excluding 38 subjects who did not answer/gave incoherent answers; ** 72 valid answers; *** Chi-Square - Yates corrected; **** Cut-off being (over/under) high school level

Table 5

Variables associated with perception of vulnerability to HIV infection in the final multiple logistic regression model

Selected variables

Coefficient (β)

p-value

Exp(B)

95% CI

Educational level

(ordinal)

 

0.1392

0.0032

1.1494

1.0478-1.2608

Anal intercourse without condoms, with (male) partners, in the last 6 months (any partner)

(1.yes/0.no)

0.8477

0.01

2,3343

1.2244-4.4501

Any unprotected intercourse with occasional partners, in the last 6 months

(1.yes/0.no)

0.6130

0.0581

1.8460

0.9790-3.4808

 

Seroprevalence of Syphilis and (ever/acute) Hepatitis B in the 295 HIV-negative cohort participants vis-à-vis perception of vulnerability to HIV-infection

Self-perception of vulnerability

Positive for Syphilis

Positive for Hepatitis B (ever)

Positive for HBsAg

(acute Hepatitis B)
 

#

%

#

%

#

%

"Vulnerability"

37

18.8

54

27.4

4

2.0

"Non-vulnerability"

23

27.0

23

27.0

1

1.2

p-value*

0.06

0.8

0.9

* Chi-square, Yates corrected / Fisher’s exact test

 

 

4.3 - Desejo de participar em testes de vacinas anti-HIV entre homens que fazem sexo com homens, com e sem história de sexo comercial, participantes de um estudo de coorte no Rio de Janeiro, Brasil. (Willingness to participate in HIV vaccine trials among men who have sex with men, with and without a history of commercial sex, enrolled in a cohort study in Rio de Janeiro, Brazil)

(AIDS Care - submetido)

Willingness to participate in HIV vaccine trials among men who have sex with men, with and without a history of commercial sex, enrolled in a cohort study in Rio de Janeiro, Brazil

Claudia Teresa Vieira de Souza, MPH1; Frits Sutmöller MD, PhD2 ; Catherine Mary Lowndes, PhD3, Célia Landman Szwarcwald 4 & Francisco Inácio Bastos, MD, PhD4

1. Evandro Chagas Hospital Research Centre, FIOCRUZ, RJ, Brazil; 2. National School of Public Health, FIOCRUZ, RJ, Brazil; 3. Dept. of Social Science and Medicine, Imperial College of Science, Technology and Medicine, University of London, UK; 4. Department of Health Information, Centre for Information on Science & Technology, FIOCRUZ, RJ, Brazil.

FIOCRUZ is a UNAIDS Collaborative Centre.

Corresponding author: Claudia Teresa Vieira de Souza, Coordenadoria de Epidemiologia e Avaliação, Centro de Pesquisa Hospital Evandro Chagas, FIOCRUZ. Av. Brasil, 4365 - Manguinhos - 21045-900 - Rio de Janeiro - RJ - Brasil. Fax: +55-21-260.9749. E-mail: clau@cpqhec.fiocruz.br

 

Willingness to participate in HIV vaccine trials among men who have sex with men, with and without a history of commercial sex, enrolled in a cohort study in Rio de Janeiro, Brazil

Abstract

The study objective was to assess willingness of homo and bisexual men enrolled in a vaccine preparedness study ("Projeto Rio") to participate in phase III anti-HIV/AIDS vaccine trials. Roughly 57% of Projeto Rio participants stated they would participate in a putative vaccine trial, because of "humanitarian concerns/solidarity" and "to be protected against HIV-infection". In multivariate analyses, variables associated with WTP were positive serology for syphilis, and "engagement, under the influence of alcohol, in risky sexual practices that would normally be avoided". The potential enrolment in vaccine trials of volunteers with different socio-demographic and behavioural backgrounds, including those at particular risk due to engagement in unprotected commercial sex, seems thus to be feasible and associated with actual and perceived risk. We must keep in mind the need to tailor preventive interventions to specific behavioural and socio-demographic profiles and needs, including interventions targeting alcohol and drug consumption, and their influence upon risky behaviours.

Willingness to participate in HIV vaccine trials among men who have sex with men, with and without a history of commercial sex, enrolled in a cohort study in Rio de Janeiro, Brazil

Introduction

The rapid diffusion of HIV/AIDS worldwide, especially in developing countries, prompted WHO and UNAIDS to establish a comprehensive programme co-ordinating vaccine preparedness studies/vaccine trials in developing countries in different geographic areas (Esparza, Osmanov, Kallings, Wigzell, 1991; Heyward et al., 1994). Brazil was one of the countries selected to implement such studies, after consultations evaluating the nature and extent of the epidemic in the country, as well as its scientific infrastructure and local co-ordinating capacity (Heyward, Osmanov, Esparza, 1996).

In the design and implementation of anti-HIV/AIDS vaccine phase III studies, it is of utmost importance to carry out a thorough assessment of HIV incidence rates, as well as of behavioural and socio-demographic characteristics, in populations where the vaccine protocols are to be evaluated (Esparza et al., 1991; Heyward et al., 1994). Besides factual data, ethical and operational aspects are of particular concern (Esparza et al., 1991; Heyward et al., 1994; Lurie et al., 1994; Guenter, Esparza, Macklin, 2000), including capacity to enrol volunteers and to maintain attrition rates as low as possible (Esparza et al., 1991; Heyward et al., 1994; Boily, Masse, Desai, Alary, Anderson, 1999).

In the Brazilian context, comprehensive evaluations of socio-demographic and behavioural aspects of vaccine preparedness studies are as yet scarce and preliminary (Sutmöller, Souza, Monteiro, Penna, 1997; Souza et al. 1999; Harrison et al., 1999; Carneiro et al., 2000, Périssé et al. 2000; Hofer et al., 2000), but new initiatives under the co-ordination of VPTN (NIH, USA) are soon to be launched. The Brazilian effort has so far comprised the establishment of four cohorts of men who have sex with men (MSM) three of them sponsored by the Brazilian Ministry of Health and WHO/UNAIDS, in the cities of Sao Paulo, Belo Horizonte and Rio de Janeiro, as well as one in Rio de Janeiro sponsored by the National Institutes of Health (NIH).

Recently, Souza et al. (1999) published a paper addressing socio-demographic and behavioural characteristics, as well as perception of vulnerability to HIV infection, among 295 volunteers enrolled in a cohort study of MSM in Rio de Janeiro ("Projeto Rio") between January 1994 and March 1997. Exploratory analyses indicated striking differences between volunteers who reported engagement in commercial sex (male prostitutes - locally known as "michês" -, and male transvestites engaged in commercial sex), and those who did not. The present paper presents data from baseline interviews of a larger sample of 675 MSM enrolled in the "Projeto Rio" study up to 1999, comparing characteristics of MSM engaged or not in commercial sex, and analysing factors associated with willingness to participate in HIV vaccine trials among study participants.

The "Projeto Rio" is, to the best of our knowledge, the only cohort worldwide to enrol a significant number of male prostitutes. This understudied population seems to be particularly vulnerable to HIV infection, both in the Brazilian context (Parker, Larvie, Cardoso, 1992) and in other settings in both developed (Elifson, Boles, Sweat, 1993a; Elifson, Boles, Posey, Sweat, Darrow, Elsea, 1993b; Boles and Elifson, 1994; de Graaf, Vanwesenbeek, van Zenssen, Straver, Visser, 1995; Miller, Katz, Eckholdt, 1998; Elifson, Boles, Darrow, Sterk, 1999) and developing countries (Lubis et al., 1997; Baqi, Shah, Baig, Mujeeb, Memon, 1999). Previous reports (Parker et al., 1992), as well as data presented in the current paper, indicate that Brazilian male commercial sex workers (CSW) quite frequently engage in unprotected sex with both male and female partners, indicating that this population may act as a core group with respect to further spread of HIV both within the gay community as well as to women, this latter a population particularly affected by the epidemic, especially in its intersection with poverty, in Brazil in recent years (Lowndes, Bastos, Giffin, Vaz dos Reis, d’Orsi, Alary, 2000; Szwarcwald, Castilho, Lauria, Durovni, Bastos, 2000).

Materials and Methods

Study procedures and preliminary findings have been reported elsewhere (Sutmoller et al., 1997; Souza et al., 1999), and are briefly described here.

Study population: Entry criteria for this study were being male, HIV-seronegative, having sexual activity with other men (male homosexual intercourse), being aged between 18-50 years of age and having agreed with the informed consent procedures (approved by FIOCRUZ IRB). Volunteers were recruited through a combined strategy mainly involving outreach activities, e.g. recruitment by study volunteers themselves ("snow-ball") and by non-governmental organisations (NGOs), as well as through media advertising and referrals from health care facilities (Sutmoller et al., 1997; Souza et al., 1999).

Study procedures: At the first (recruitment) interview basic socio-demographic and behavioural data were obtained, and the "Projeto Rio" study was explained. The volunteers were then given pre-test counselling with subsequent blood sample collection for HIV, syphilis (VDRL, confirmed by TPHA) and hepatitis B (anti-HBc, anti-HBs, HBsAg) testing. On their return one week later, the volunteers received the HIV and other test results during a session of post-test counselling. The HIV-positive individuals were referred to public health care facilities for follow-up care and those who were HIV-negative were asked to enrol in the study after signing the informed consent form. A detailed questionnaire was administered on enrolment.

Study instruments: A standardised questionnaire was developed and applied by trained study staff. This consisted of about 100 (mainly closed) questions on: socio-demographic characteristics, knowledge of HIV transmission routes and prevention strategies, beliefs and attitudes towards AIDS and sexual life, sexual practices, and STD/AIDS preventive initiatives.

Willingness to participate in a putative HIV vaccine trial was evaluated through a simple question "Would you be willing to participate in HIV vaccine trials?", answers being "yes", "no", "it depends", and "do not know". Willingness to participate was defined for our purposes as the equivalent of a "yes" answer and the remaining alternatives ("no", "it depends", "do not know") were defined, for our purposes, as unwillingness. Previous engagement in commercial sex was defined by the question: "Have you accepted money/goods in exchange for sex during the last six months?" Volunteers answering "often", "sometimes" or "rarely" were considered to have a history of commercial sex. Exploratory analyses were carried out stratifying those engaged in commercial sex on a regular ("always" or "often") or irregular ("sometimes" or "rarely") basis during the six months prior to interview; since both subcategories differed markedly from people not reporting any engagement in commercial sex (data not shown), they were considered together for the purpose of the present analyses.

Statistical analysis: In the bivariate analyses, differences in characteristics between MSM with and without a history of commercial sex, as well as variables associated with willingness to participate in HIV vaccine trials, were assessed through contingency table statistics. Criteria of statistical significance were established by p-values less than 0.05.

To further assess the role of the different variables associated with willingness to participate in HIV vaccine trials, a stepwise logistic regression was carried out, maximizing the likelihood ratio at each step, with "willingness to participate in HIV vaccine trials" as the dependent variable. All variables significantly associated with the response variable (p<0.05) as well as borderline associations formerly mentioned in the literature were entered in the model. The stepwise procedures were performed in two different ways: i) all covariates were tested for inclusion at once, without controlling for reported engagement in commercial sex; ii) the covariate "reported engagement in commercial sex" was included in the first step of the procedure and then the other covariates were tested for inclusion.

Results

Forty-eight (7.11%) volunteers were excluded from the analyses due to missing data for core variables. Analyses here described refer to a total sample of 675 subjects, of which 294 reported engagement in commercial sex in the six months prior to the interview.

As shown in Table 1, the majority of volunteers were single (82.9%), young adults (mean age 27.4, s.d. 7.11 years), and currently employed (66.3%). Roughly 63% earned less than US$ 250.00 a month (corresponding to the second tercile for income and taken here as a "poverty index") and close to half of the interviewees (49.1%) had undergone less than 8 years of regular education. Ethnic background, defined as a dichotomous variable (white and non-white) was represented in similar proportions.

Noticeable differences distinguish those reporting and not reporting involvement in commercial sex in the six months prior to interview. The interviewees in the subgroup reporting commercial sex (CS) were younger (26.7 versus 28.1 years old for those not reporting such behaviour - NCS), more frequently married (21.4% vs. 12.9%), unemployed (40.1 vs. 24.6), poorer (71.1% CS earned less than US$ 250.00 vs. 55.3% for NCS) and non-white (58.2% of CS and 41.1% of NCS), as well as being less well-educated (74.1% of CS had less than 8 years of education, compared with 36.9% of NCS) (Table 1).

A high proportion of the interviewees reported risky sexual practices with both regular and casual male partners (Table 1). Unprotected oral sex was reported by 59.9% and 47.4% of those reporting sex with regular and casual partners, respectively.

The corresponding figures for both unprotected receptive and insertive anal sex were of particular concern: 38.9% and 22.5% of those reporting sex with regular and casual partners, reported having practised unprotected anal receptive sex, respectively. For unprotected insertive anal sex the figures were, respectively, 40.4% and 34.0%. These risky practices were significantly more frequent (for the whole sample) for unprotected sex with steady than with casual partners, both with respect to oral sex (p=0.03) and receptive anal sex (p=0.00).

A considerable number of volunteers also reported having had sex with female partners in the last six months. Among those reporting sexual intercourse with regular (157 or 25.0%) and casual (171 or 27.3%) female partners , the majority had had unprotected sex. Unprotected vaginal sex was reported by 77.1% and 60.2% of those with regular and casual female sexual partners, respectively. The corresponding figures for unprotected anal sex were 42.0% and 44.4%. Again, unprotected vaginal sex was significantly more frequently reported for those engaged in steady partnerships than for those in casual partnerships (p=0.003).

Unprotected oral sex with regular male partners was more frequently reported by NCS than by CS (65.6% vs. 49.6%), while proportions of NCS and CS practising unprotected receptive and insertive anal sex with regular male partners were not significantly different. Regarding casual male partners, unprotected oral and receptive anal sex were more frequently reported by NCS (53.0% vs. 27.7% respectively) than CS (41.2% vs. 18.5% respectively). On the other hand, unprotected insertive anal sex with casual male partners was more frequently reported by CS (38.4%) than NCS (28.2%). Higher frequencies of unprotected (insertive) anal sex with casual female partners were also reported by CS than NCS (48.9% vs. 27.8%). No other statistically significant differences in sexual behaviour were found between the two subgroups.

Less than half (44.0%) of the men in the cohort reported having changed their behaviour as a result of the AIDS epidemic, whereas a slightly greater proportion (56.1%) of volunteers perceived themselves as vulnerable to HIV infection. Roughly one-quarter (26.5%) of interviewees reported a history of sexual abuse. NCS more frequently reported behavioural change as a result of the AIDS epidemic than those with a history of commercial sex (48.9% vs. 38.4%).

The engagement, under the influence of alcohol, in risky sexual practices that would normally be avoided was reported by 12.8% of the volunteers. No significant difference was found in this respect between CS and NCS. On the other hand, engagement in risky sexual practices that would normally be avoided, under the influence of (illicit) drugs, was more frequently reported by CS than by NCS (7.5% vs. 2.1%, respectively).

Laboratory data presented in Table 1 show that, overall, 28.7% and 33.5% of the volunteers were seropositive for syphilis and hepatitis B, respectively. Both prevalences were higher for CS, when compared with NCS (32.7% vs. 25.2% and 38.4% vs29.1%, respectively).

Roughly 57% of the volunteers of Projeto Rio stated they would like to participate in a putative vaccine trial, with a higher percentage of CS than NCS stating that they would participate (62.6% vs. 51.4%).

ENTER TABLE 1

Table 2 depicts the main reasons stated by volunteers for their willingness or not to participate in HIV vaccine trials. Among those willing to participate, "humanitarian concerns/solidarity" was the chief motivation, and was mentioned significantly more frequently by NCS than CS (73.7% vs. 49.0%). The second reason most frequently given ("to protect [myself] against HIV-infection") was mentioned less often by CS than NCS (50.0% vs. 38.0%, respectively).

For those MSM who did not manifest willingness to participate in an HIV vaccine trial, roughly one-third of the interviewees alleged "not having enough information about vaccines", with no significant differences for the two subgroups. "Concern about possible adverse effects of the vaccine" was mentioned by 33.8% of interviewees in the total sample, with a significantly higher proportion for those who did not report commercial sex, when compared with those who did (40.0% vs. 24.5%). Other reasons mentioned by a significant proportion of volunteers were: "afraid to get AIDS after vaccination" (25.3%) and "afraid to be used as a guinea pig" (14.2%), the former motive particularly by NCS.

ENTER TABLE 2

The final multiple logistic regression model, controlling for reported engagement in commercial sex (model chi-square=13.71; p=0.003), identified as factors independently associated with "willingness to participate in HIV vaccine trials", following the inclusion order as defined by the stepwise procedure: "positive serology for syphilis" (Adjusted OR = 1.52; 95% CI 1.04-2.22) and "engagement, under the influence of alcohol, in risky sexual practices that would normally be avoided" (Adj OR 1.83; 95% CI= 1.05-3.20).

The final multiple logistic regression model not controlling for previous history of commercial sex (model chi-square = 15.28; p=0.002) identified as factors predictive of the willingness to participate in HIV vaccine trials: "lower educational level" (defined as an ordinal variable, with a cut-off point of 8 years of regular school attendance), (Adjusted OR 1.47; 95% CI 1.03-2.11); and a "positive serology for syphilis" (Adj OR = 1.48; 9%% CI 1.01-2.16) and "engagement, under the influence of alcohol, in risky sexual practices that would normally be avoided" (Adj OR = 1.85; 95% CI 1.06-3.23) (Table 3).

ENTER TABLE 3

Discussion

The "Projeto Rio" recruited, from 1994 to 1999, 723 men who have sex with men (MSM), most of them belonging to the low middle class and lower social strata, including a sizeable number of men reporting commercial sex. Volunteers with and without a history of commercial sex differ substantially in many socio-demographic and behavioural aspects.

Practices, behaviours and attitudes related to engagement in commercial sex by MSM have been studied in different social and cultural contexts in the framework of the AIDS epidemic. A previous study carried out in Rio de Janeiro showed that paid sex is far more commonly reported by MSM belonging to the dispossessed social strata (Parker et al., 1992).

Some findings are of serious concern, especially among those reporting engagement in commercial sex. Reported levels of unprotected sex were found to be unacceptably high - significantly higher for CS, for most risky behaviours -, despite the fact that subjects were voluntarily taking part in a scientific study which includes various preventive initiatives. We can hypothesize that other MSM from the community, who did not present themselves for vaccine preparedness studies, may have higher levels of risky behaviours.

Men with a history of commercial sex enrolled in "Projeto Rio" frequently reported sex with both men and women, corroborating previous findings by Parker et al. in a study targeting male prostitutes recruited from gay meeting places in the streets of Rio de Janeiro (Parker et al., 1992). Given the high levels of unprotected sex reported by those men, irrespective of the nature of their partnerships, they have probably functioned as a bridging population for HIV transmission between the gay community and women.

Findings from a sample of male prostitutes from New Orleans, Louisiana, USA, recruited in 1988-9, indicated high levels of risk behaviours and a disquieting seroprevalence for HIV infection of 17.5%. A sizeable proportion of such men (42.3%) had stable female partners. Although well informed about risks and ways to protect themselves, their use of condoms was inconsistent.

In Atlanta, Georgia, another sample of MSM engaged in commercial sex, assessed in 1990-1, was also found to have high levels of infection for different infectious agents: HIV, T. pallidum and Hepatitis B virus (HBV), with prevalences of 29.0%, 25.0% and 58.0%, respectively (Elifson et al., 1993a). Male transvestites engaged in paid sex in the same area were also evaluated (Elifson et al. 1993b), and very high levels of infection, with prevalences of 68%, 81% e 80%, for HIV, T. pallidum and HBV, respectively, were found. Data from a study by the same researchers on clients of male and female sex workers showed higher levels of HIV, T. pallidum and HBV infection among clients of male sex workers than those of female sex workers (Elifson et al., 1999).

In our study, over half (57%) interviewees stated they would be willing to participate in vaccine trials, a proportion slightly higher than the 50% reported for a similar study carried out in Belo Horizonte, Minas Gerais, Brazil (Carneiro et al., 2000), but substantially lower than the 69.8% reported in the other vaccine preparedness study implemented in Rio de Janeiro (Périssé et al., 2000). We must observe, however, that the other studies (Projects "Horizonte", in Belo Horizonte, and "Praça XV", in Rio de Janeiro) enrolled few MSM with a history of commercial sex, and in this sense differ substantially from our study in terms of the socio-demographic and behavioural characteristics of their volunteers.

Most studies on MSM carried out in different contexts have pointed to a strong willingness of such populations to participate in vaccine trials, with a moderate variance. Koblin et al. (1998) found that 76.2% of the volunteers enrolled in cohorts from eight American states stated they would like to participate in vaccine trials. Similar results (70-74% having stated they would like to participate) were found by Scheer et al. (1999) in the CDC Collaborative HIV Seroincidence Study in Chicago, IL, Denver, CO and San Francisco, CA. A slightly lower percentage (68.0%) was seen in the Project ACHIEVE and in cohorts from Boston (65.0%), as reported by Gross, Seage, Mayer, Goldstein, Losina, Wold (1996).

The main reasons given by the volunteers in our study as to why they would putatively enrol in vaccine trials - "humanitarian concerns/solidarity" and "to protect [myself] against HIV-infection" - are reported by other national (Carneiro et al., 2000; Périssé et al., 2000) and international studies (Harrison et al., 1999; Koblin et al., 1998; Gross et al., 1996; Hays and Kegeles, 1999; Koblin, Avrett, Taylor, Stevens, 1997). Interviewees with a history of commercial sex alleged more frequently stated "to protect [myself] against HIV-infection" as their motivation than those without such a history, a perception that could reflect the increased risks such populations are exposed.

The main motives for unwillingness to participate in vaccine trials given by study participants were similar to those in other national and international studies. For instance, in Project "Horizonte" 30% of the interviewees said they "were insecure and needed more information on the subject before a final decision" (Carneiro et al., 2000), and in Project "Praça XI" the main alleged reasons given were: "fear of becoming HIV infected from the vaccine itself"; "fear of vaccine-induced positive HIV serological test result"; and "do not want to be a human guinea pig" (Périssé et al., 2000).

To the best of our knowledge, no other vaccine preparedness study has recruited such a large proportion of volunteers reporting commercial sex (Harrison et al., 1999; Koblin et al., 1998; Scheer et al., 1999; Gross et al. 1996; Koblin et al., 1997; Koblin, Taylor, Avrett, Stevens, 1996; Bartholow, MacQueen, Douglas, Buchbinder, McKirnan, Judson, 1997). In two vaccine preparedness studies carried out in the USA, one in San Francisco, CA (Gross et al., 1996), and other in New York, NY (Koblin et al., 1997), a history of commercial sex was shown to be associated with willingness to participate in vaccine trials. The small number of volunteers reporting commercial sex (less than 8% in the Project ACHIEVE, in New York, and a negligible percent of the volunteers of the S. Francisco) precluded, in both studies, the inclusion of these men in multivariate analyses.

Financial incentives, although modest (~U$10.00), could influence willingness to participate in vaccine trials, especially for those reporting commercial sex, most of whom are poor and underserved. This hypothesis was discussed, among others, as a factor contributing to willingness to participate in vaccine trials by Koblin et al. (1998). This issue was not specifically addressed in the present study.

Various studies (Périssé et al., 2000; Scheer et al., 1999; Gross et al., 1996; Buchbinder, Douglas, McKirnan, Judson, Katz, Mac Queen, 1996) have shown that volunteers at particular risk of infection (for instance, those reporting unprotected sex with HIV-infected partners) state more frequently they would like to participate in vaccine trials. A low educational level has been shown to be associated with willingness to participate in vaccine trials (Périssé et al., 2000; Koblin et al., 1998; Gross et al., 1996; Koblin et al., 1997; Koblin et al., 1996; Bartholow et al., 1997). In our study, when the whole sample was considered, lower educational level was consistently associated with willingness to participate in vaccine trials. This association did not remain when multivariate models were controlled for a history of commercial sex, probably due to the striking differences in SES (socio-economic status) between NCS and CS. This is important since dispossessed and underserved population are currently being increasingly infected by HIV in Brazil (Fonseca, Bastos, Derrico, Andrade, Veras, Szwarcwald , 2000).

Hays and Kegeles (1999), analysing data from a cohort of 390 young gay/bisexual men recruited from three communities on the West Coast of the US (Eugene, OR; Santa Cruz, CA; Santa Barbara, CA), found that the willingness to participate in vaccine trials was associated with alcohol and (illicit) drugs consumption, and with the perception of interviewees that under the influence of such substances they tended to adopt risky sexual practices. Alcohol consumption patterns were assessed only indirectly in the present study, which limits itself to data relative to the relationship between alcohol (and drug) consumption patterns and sexual intercourse. Nevertheless,one could hypothesize that MSM reporting difficulty in initiating/maintaining safer behaviours under the influence of alcohol perceive themselves to be at higher risk of HIV infection, and thus as most likely to benefit from taking part in vaccine trials.

Subjects in our study received, before they signed the informed consent and answered the baseline interview, results for syphilis, hepatitis B and HIV infection. Whereas the latter constituted one of the exclusion criteria for enrolment in the cohort, volunteers found to be positive for syphilis and hepatitis were informed immediately before they were interviewed that they had a medical condition, explicitly linked, in the counselling sessions, to risky sexual behaviour. In this sense, it seems plausible that such volunteers will more frequently assent to be engaged in a future vaccine trial, since these studies (as they were informed before they sign the Projeto Rio’s consent form) offer permanent psycho-social support, distribute condoms, and provide easy and confidential referral for the treatment of any medical condition. Engagement in a vaccine trial could thus also signify an attempt to protect against other infections acquired through sexual transmission .

Summarizing, our results indicate the feasibility of enrolling volunteers engaged in commercial sex, keeping in mind the necessity to tailor preventive interventions to their specific profile and needs. The fact that those interviewees, and/or volunteers with lower educational levels and higher prevalence for syphilis, wish to enrol in vaccine trials seems encouraging. The same reasoning applies to men reporting difficulties in maintaining safer behaviours under the influence of alcohol and drugs. The finding that those perceiving themselves as at risk under the effects of alcohol would be willing to participate in vaccine trials are encouraging. Beyond the specific scope of vaccine preparedness studies, MSM engaged in commercial sex, as a population highly vulnerable to HIV infection, should be the target of comprehensive preventive efforts..

Acknowledgments

To the study volunteers and NGO leaders who have understood the social importance of participating in this scientific effort, notwithstanding difficulties. The support provided by the staff of "Centro de Pesquisa Hospital Evandro Chagas" (CPqHEC/FIOCRUZ), the "National Reference Centre of Viral Hepatitis" (IOC/FIOCRUZ) made it possible to execute this study.

Financial support was provided by the National STD/AIDS Program of the Brazilian Ministry of Health; the Global Programme on AIDS/WHO; UNAIDS; Brazilian National Research Council (CNPq) and the Oswaldo Cruz Foundation (FIOCRUZ).

Participants of "Projeto Rio" include: Frits Sutmöller (coordinator), Claudia Teresa Vieira de Souza, Amilton Xavier Junior, Helena Santos Martins, Dionne Peluso de Oliveira Costa, Paulo Starling Brandão Junior & Therezinha Lucy Penna.

References

Baqi, S., Shah, A. S., Baig, M. A., Mujeeb, S.A., Memon, A. (1999). Seroprevalence of HIV, HBV, and syphilis and associated risk behaviours in male transvestites (Hijras) in Karachi Pakistan. International Journal of STD & AIDS, 19, 300-304.

Bartholow, B.N., MacQueen, K.M., Douglas, Jr.J.M., Buchbinder, S., McKirnan, D., and Judson, F.N. (1997). Assessment of the changing willingness to participate in phase III HIV vaccine trials among men who have sex with men. Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology, 16, 108-115.

Boily, M.C., Masse, B.R., Desai, K., Alary, M., Anderson, R. M. (1999). Some important issues in the planning of phase III HIV vaccine efficacy trials. Vaccine, 17, 989-1004.

Boles, J., Elifson, K.W. (1994). The social organization of transvestite prostituition and AIDS. Social Science Medicine, 39, 85-93.

Buchbinder, S.P., Douglas, Jr. J.M, McKirnan, D.J., Judson, F.N., Katz, M.H., and MacQueen, K.M. (1996). Feasibility of human immunodeficiency virus vaccine trials in homosexual men in the United states: Risk behavior, seroincidence, and willingness to participate. Journal of Infectious Diseases, 174, 954-961.

Carneiro, M., de Figueredo Antunes, C. M., Greco, M., Oliveira, E., Andrade, J., Lignani, Jr.L., Greco, D. B. for the Project Horizonte. (2000). Design, implementation, and evaluation at entry of a prospective cohort study of homosexual and bisexual HIV-negative men in Belo Horizonte, Brazil : Project Horizonte. Journal of Acquired Immune Deficiency Sindromes and Human Retrovirology, 25, 182-187.

de Graaf, R., Vanwesenbeeck, I., van Zessen, G., Straver, C.J., Visser, J.H. (1995). Alcohol and drug use in heterosexual and homosexual prostituition, and its relation to protection behaviour. AIDS care, 7, 35-47.

Elifson, K.W., Boles, J., and Sweat, M. (1993a). Risk factors associated with HIV infection among male prostitutes. American Journal of Public Health, 83, 79-83.

Elifson, K.W., Boles, J., Posey, E., Sweat, M., Darrow, W., and Elsea, W. (1993b). Male transvestite prostitutes and HIV risk. American Journal of Public Health, 83, 260-262.

Elifson, K.W., Boles, J., Darrow, W.W., and Sterk, C.E. (1999). HIV Seroprevalence and risk factors among clients of female and male prostitutes. Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology, 20, 195-200.

Esparza, J., Osmanov, S., Kallings, L.O., Wigzell, H. (1991). Planning for HIV vaccine trials: The World Health Organization perspective. AIDS, 5, S159-S163.

Fonseca, M.G., Szwarcwald, C.L., Derrico, M., Andrade, C.L.T., Veras, C.T., Bastos, F.I. (2000). AIDS and level of education in Brazil: temporal evolution from 1986 to 1996. Reports in Public Health,16, S77-S87.

Gross, M., Seage, G.R., Mayer, K.H., Goldstein, R.S., Losina, E, and Wold, C. (1996). Interest among gay/bisexual men in greater Boston in participating in clinical trials of preventive HIV vaccines Journal of Acquired Immune Deficiency Sindromes and Human Retrovirology, 12, 406-412.

Guenter, D., Esparza, J., Macklin, R. (2000). Ethical considerations in international HIV vaccine trials: summary of a consultative process conducted by the Joint united Nations Programme on HIV/AIDS (UNAIDS). Journal of Medical Ethics, 26, 37-43.

Harrison, L.H., Lago, R.F., Friedman, R.K, Rodrigues, J., Santos, E.M., Melo, M.F., Moulton, L.H., Schechter, M., and the Projeto Praça Onze Study Group. (1999). Incident HIV Infection in a high-risk, homosexual, male cohort in Rio de Janeiro. Journal of Acquired Immune Deficiency Sindromes and Human Retrovirology, 21, 408-412.

Hays, R.B, and Kegeles, M.K. (1999). Factors related to the willingness of young gay men to participate in preventive HIV vaccine trials. Journal of Acquired Immune Deficiency Sindromes and Human Retrovirology, 20, 164-171.

Heyward, W.L., Osmanov, S., Esparza, J. (1996). Establishment of WHO-sponsored field sites for HIV vaccine evaluation in developing countries. Antibiotics and Chemotherapy, 48, 139-144.

Heyward, W.L., Osmanov, S., Saba, J., Esparza, J., Belsey, E., Stoneburner, R., Kaldor, J., and Smith, P.G. (1994). Preparation for phase III HIV vaccine efficacy trials: methods for the determination of HIV incidence. AIDS, 8, 1285-1291.

Hofer, C.B., Harrison, L.H., Struchiner, C.J., Moreira, R.I., do Lago, R.F., de Melo, M.F., Schecheter, M. (2000). Acute retrovirus syndrome among prospectively identified homosexual incident HIV infection in Brazil. Journal of Acquired Immune Deficiency Sindromes and Human Retrovirology, 25, 188-191.

Koblin, B.A., Avrett, S., Taylor, P.E., and Stevens, C.E. (1997). Willingness to participate in HIV-1 vaccine efficacy trials and the effect of media events among gay and bisexual men in New York City: Project ACHIEVE. Journal of Acquired Immune Deficiency Sindromes and Human Retrovirology, 15, 165-171.

Koblin, B.A., Heagerty, P., Sheon, A.., Buchbinder, S., Celum, C., Douglas, J.M., Gross, M., Marmor, M., Mayer, K., Metzger, and Seage, G. (1998). Readiness of high-risk populations in the HIV network for prevention trials to participate in HIV vaccine efficacy trials in the United States. AIDS, 12, 785-793.

Koblin, B.A., Taylor, P.E., Avrett, S., and Stevens, C.E. (1996). The feasibility of HIV-1 vaccine efficacy trials among gay/bisexual men in New York City: Project ACHIEVE. AIDS, 10, 1555-1561.

Lowndes, C.M., Bastos, F.I., Giffin, K.M., Vaz dos Reis, A.C., d’Orsi, E., Alary, M. (2000). Differential trends in mortality from AIDS in men and women in Brazil (1984-1995). AIDS, 14, 1269-1273.

Lubis, I., Master, J., Munif, A., Iskandar, N., Bambang, M., Papilaya, A., Roesin, R., Manurung, S, and Graham, R. (1997). Second report of AIDS related attitudes and sexual practic of the Jakarta Waria (male transvestites) in 1995. The Southeast Asian Journal Tropical Medicine and Public Health, 28, 525-529.

Lurie, P., Bishaw, M., Chesney, M. A., Cooke, M., Fernandes, M.E., Hearst, N., Katongole-Mbidde, E., Koetsawang S, Lindan, C.P., Mandel, J., Mhloyi, M., Coates, T.J.(1994). Ethical, behavioral and social aspects of HIV vaccine trials in developing countries. The Journal of the American Medical Association, 271, 295-301.

Miller, R.L., Klotz, D., Eckholdt, H.M. (1998). HIV prevention with male prostituites and patrons of hustle bars: replication of na HIV preventive intervention. American Journal of Community Psychology, 26, 97-131.

Parker, R.G., Larvie, P., Cardoso, Jr. R. (1992). Programa Pegação: An outreach program for male commercial sex workers in Rio de Janeiro. In: A Portfolio of AIDS/DST Behavioral Interventions and Research. Edited by Bond LS. Washington: PAHO, 171-178.

Périssé, A.R.S., Schechter, M., Moreira, R.I., Lago, R.F., Santoro-Lopes, G., Harrison, L.H., and Projeto Praça Onze Study Group. (2000). Willingness to Participate in HIV Vaccine Trials Among Men Who Have Sex With Men in Rio de Janeiro. Journal of Acquired Immune Deficiency Sindromes and Human Retrovirology, 25, 459-463.

Scheer, E.L., Douglas Jr., J.M., Vittinghoff, E., Bartholow, B.N., McKirnan, D., Judson, F.N., MacQueen, K.M., and Buchbinder, S. (1999). Feasibility and suitability of targeting young gay men for HIV vaccine efficacy trials. Journal of Acquired Immune Deficiency Sindromes and Human Retrovirology, 20, 172-178.

Souza, C.T.V., Bastos, F.I., Lowndes, C.M., Szwarcwald, C.L., Santos, E.M., Castilho, E.A., Sutmöller, F., on behalf of the Oswaldo Cruz Foundation STD/HIV Prevention Group. (1999). Perception of vulnerability to HIV infection in a cohort of homosexual/bisexual men in Rio de Janeiro, Brazil. AIDS CARE, 11, 567-579.

Sutmöller, F., Souza, C.T.V., Monteiro, J.C., Penna, T. (1997). The Rio de Janeiro HIV Vaccine Site-I. Recruitment strategies and socio-demographic data of a HIV negative homosexual and bisexual male cohort in Rio de Janeiro, Brazil. Memórias do Instituto Oswaldo Cruz, 92:39-46.

Szwarcwald, C.L., Castilho, E.A., Lauria, L. , Durovni, B. & Bastos, F.I. (2000). The spatial spread of AIDS in the city of Rio de Janeiro: the role of district poverty levels among women. In: Flahault, A., Toubiana, L. & Valleron, A. J. (eds). Geography na Medicine Geomed99, Paris: Elsevier.

 

Table 1

Socio-demographic, behavioral characteristics of participants (All Sample, Non-Commercial Sex, and reporting Commercial Sex) of "Projeto Rio", according to baseline interview. Rio de Janeiro, 1994-1999
         

Variables

 

All Sample

NCS

CS

   

N=627

N=333

N=294

Age (mean in years; S.D.)

 

27.4; 7.11

28.1; 7.21

26.7; 6.92

         

Marital status

Single

520 (82.9%) a

289 (86.8%)

231 (78.6%)

 

Married/ divorced b

106 (16.9%)

43 (12.9%)

63 (21.4%)

         

Employment

Formally employed

416 (66.3%) a

242 (72.7%)

174 (59.2%)

 

Unemployed/alike c

200 (31.9%)

82 (24.6%)

118 (40.1%)

         

Income

< U$250.00

393 (62.7%) a

184 (55.3%)

209 (71.1%)

 

³ U$250.00

202 (32.2%)

134 (40.2%)

68 (23.1%)

         

Race

White

303 (48.3%) a

186 (55.9%)

117 (39.8%)

 

Non-white

308 (49.1%)

137 (41.1%)

171 (58.2%)

         

Educational level

Fundamental

341 (54.4%) a

123 (36.9%)

218 (74.1%)

 

High School/college

279 (44.5%)

204 (61.3%)

75 (25.5%)

         

Unprotected Sexual Practices d

With Males

     
 

Regular partners

334 (53.3%)

215 (64.6%)

119 (40.5%)

 

Oral

200 (59.9%)

141 (65.6%)

59 (49.6%)

 

Anal receptive

130 (38.9%)

88 (40.9%)

42 (35.3%)

 

Anal insertive

135 (40.4%)

86 (40.0%)

49 (41.2%)

         
 

Casual partners

409 (65.2%)

177 (53.2%)

232 (78.9%)

 

Oral

194 (47.4%)

94 (53.0%)

100 (41.2%)

 

Anal receptive

92 (22.5%)

49 (27.7%)

43 (18.5%)

 

Anal insertive

139 (34.0%)

50 (28.2%)

89 (38.4%)

         
 

With Females

     
 

Regular partners

157 (25.0%)

46 (13.8%)

111 (37.8%)

 

Vaginal

121 (77.1%)

34 (74.0%)

87 (78.4%)

 

Anal

66 (42.0%)

15 (32.6%)

51 (45.9%)

         
 

Casual partners

171 (27.3%)

36 (10.8%)

135 (45.9%)

 

Vaginal

103 (60.2%)

16 (44.4%)

87 (64.4%)

 

Anal

76 (44.4%)

10 (27.8%)

66 (48.9%)

         

Other behavioral

and attitudinal

Behavioral change after AIDS

276 (44.0%)

163 (48.9%)

113 (38.4%)

variables

Perception of vulnerability

352 (56.1%)

177 (53.2%)

175 (59.5%)

 

History of sexual abuse

166 (26.5%)

85 (25.5%)

81 (27.6%)

 

Engaged in risky sexual practices that normally would be avoided, under the influence of alcohol

80 (12.8%)

34 (10.2%)

46 (15.6%)

 

Engaged in risky sexual practices that normally would be avoided, under the influence of drugs

29 (4.6%)

7 (2.1%)

22 (7.5%)

 

Willingness to participating in Phase III HIV vaccines

355 (56.6%)

171 (51.4%)

184 (62.6%)

Laboratory data

       
 

Serology (+) for syphilis

180 (28.7%)

84 (25.2%)

96 (32.7%)

 

Serology (+) for hep. B

210 (33.5%)

97 (29.1%)

113 (38.4%)

a Total different from 100% due to invalid answers;

b Refers to the civil status of men not reporting engagement in commercial sex, married with women. Brazilian law does not recognize the civil union of gay couples;

c Includes living from social security, informal sources of income, etc.

d Data refer to those engaged in each one of these partnerships in the last 6 months (Having 358, 433, 161, 173 volunteers reported sexual relationship with, respectively, male regular partners, male casual partners, female regular partners, and female casual partners.

** In bold = Statistically significant differences (p<0.05)

Table 2

Main responses (multiple answers allowed) for "willingness" and "Unwillingness" to participating in HIV vaccine trials among the cohort participants
 

Responses

ALL

NCS

CS

Willingness

 

N=355 (56.6%)

N= 171 (51.4%)

N=184 (62.6%)

         
 

Human concerns/solidarity

216 (60.8%)

126 (73.7%)

90 (49.0%)

 

To protect [myself] against HIV-infection

157 (44.2%)

65 (38.0%)

92 (50.0%)

 

To enjoy sex without being concerned with AIDS

51 (14.4%)

29 (17.0%)

22 (12.0%)

 

I trust scientific achievements

67 (18.9%)

39 (22.8%)

28 (15.2%)

         

Unwillingness

 

N=272 (43.4%)

N=162 (48.6%)

N=110 (37.4%)

         
 

Afraid to be used as a guinea pig

36 (13.2%)

23 (14.2%)

13 (11.8%)

 

Concern about possible adverse effects of the vaccine

92 (33.8%.)

65 (40.0%)

27 (24.5%)

 

Afraid to get AIDS after vaccination

57 (21.0%)

41 (25.3%)

16 (14.5%)

 

Not having enough information about vaccines

94 (34.6%)

59 (34.4%)

35 (31.8%)

* In bold = Statistically significant associations (p<0.05)

Table 3

Variables associated with willingness to participate of HIV vaccine trial as made evident byin the final multiple logistic regression model

Selected variables

Exp(B)

95% CI

Controlling for CS

   

Positive serology for syphilis

(1.yes/0.no)

1.52

1.04-2.22

Engaged in risky sexual practices that normally would be avoided, under the influence of alcohol

1.83

1.05-3.20

Not controlling for CS

   

Lower educational level

(ordinal)

1.47

1.03-2.11

Positive serology for syphilis

(1.yes/0.no)

1.48

1.01-2.16

Engaged in risky sexual practices that normally would be avoided, under the influence of alcohol

1.85

1.06-3.23

 

4.4 - A associação do status socioeconômico e uso de crack/cocaína com sexo anal desprotegido numa coorte de homens que fazem sexo com homens no Rio de Janeiro, Brasil. (The association of socioeconomic status and use of crack/cocaine with unprotected anal sex in a cohort of men who have sex with men in Rio de Janeiro, Brazil).

(J Acquir Immune Defic Syndr - submetido)

The association of socioeconomic status and use of crack/cocaine with unprotected anal sex in a cohort of men who have sex with men in Rio de Janeiro, Brazil

Claudia Teresa Vieira de Souza, MPH1; Theresa Diaz MD, MPH1,2; Frits Sutmoller MD, PhD3, & Francisco Inácio Bastos, MD, PhD4

1 - Research Center Evandro Chagas Hospital, FIOCRUZ, RJ, Brazil.

2 - Pan American Health Organization, Brazil / Centers for Disease Control and

Prevention, Global AIDS Program, Atlanta, GA

3 - National School of Public Health, FIOCRUZ, RJ, Brazil.

4 - Department of Health Information, Centre for Information on Science & Technology, FIOCRUZ, RJ, Brazil.

FIOCRUZ is a UNAIDS Collaborative Centre.

Corresponding author: Claudia Teresa Vieira de Souza, Coordenadoria de Epidemiologia & Avaliação, Centro de Pesquisa Hospital Evandro Chagas, FIOCRUZ. Av. Brasil, 4365 - Manguinhos - 21045-900 - Rio de Janeiro - RJ - Brasil. Fax: +55-21-260.9749. E-mail: clau@cpqhec.fiocruz.br

Abstract

To evaluate the relationship of illicit drug use, sexual attitudes and practices, and socioeconomic status (SES) we analyzed data from the baseline interview of a cohort of men who have sex with men (MSM) conducted from 1994 to 1999, in Rio de Janeiro, Brazil. We found that crack/cocaine users were significantly (p<0.05) more likely than non drug users to be unemployed to have an income of <$250 per month, to be non-white, and to have <8 years of education. Crack/cocaine users were significantly more likely than non drug users to drink alcohol (76.8% versus 45.7%) , report bisexual activity (81.7% versus 41.7%) and engage in commercial sex (72.0% versus 37.9%). In logistic regression the following variables were associated with unprotected anal sex with a casual partner: income <$250 USD a month (Adjusted odds ratio [AOR] 1.73, CI 1.04-2.87), less than 8 years of education (AOR 2.21, CI 1.38-3.53), self perception of vulnerability (AOR 2.58, CI 1.54-4.33), willingness to participate in vaccine trials (AOR 1.91, CI 1.20-3.05) and use of crack/cocaine (AOR 1.91, CI 1.05-3.46). Our findings suggest that HIV prevention programs for these men need to address drug use and how drug use may influence sexual behaviors.

 

Key words: HIV/AIDS, Men who Have sex with Men, cocaine, Brazil, socio-economic status, vaccine preparedness studies.

Introduction

Unprotected sexual intercourse has a pivotal role in AIDS spread worldwide, and in many different regions of the world has been intertwined with illicit drug use as a source of unabated epidemics (1). Recent papers have highlighted two core aspects of the relationship between drug consumption and risky behavior: a) the relationship is complex and bidirectional in nature (i.e. psychoactive substances modulate behavior, but it is also the case that people with certain life styles seek specific mind-altering substances as a way to enhance/reinforce pre-existing behaviors) (2), and b) the extent unprotected sexual intercourse impacts the spread of HIV in populations double-exposed to sexual and parenteral risks (3).

As of December 2000, sexual transmission has been the mode of transmission of HIV infection in 107,004 of the 203,348 AIDS cases registered in Brazil (52.6%). Among those cases attributed to unprotected sexual intercourse, 54,777 (51.2%) cases have been reported among men having sex with other men (MSM) (4).

Although Brazilian literature is quite informative respecting risk behaviors and infection rates for different blood-borne and sexually transmitted infections among injection drug users (IDUs) (5-9), very few papers have explored the intersection between non-injecting drug use and sexual risk behaviors in Brazil (10,11).

To the best of our knowledge, the inter-relationships between drug use and sexual risk behaviors were never studied among MSM in Brazil. However, available evidence is of concern; 6,059 AIDS cases registered in Brazil so far are among men who have men sex with other men and who also inject drugs (4).

Men constitute the vast majority (81%) of AIDS cases among IDUs registered in Brazil (4) and predominate in all studies addressing IDUs and non-injecting drug users carried our so far, whether or not they were recruited in street settings (5,7), or drug treatment centers (11) or both (6,9). Consistent condom use has been seldom reported by both IDUs (5-7) and non-injecting drug users (10-11) in different Brazilian settings; some of these unprotected sexual acts involve gay and bisexual men (6,11).

In the current paper we analyze data assessing alcohol and illicit drug using behaviors that were included in the baseline interview of a MSM cohort conducted between 1994-1999 in Rio de Janeiro, Brazil. Our purpose here is to evaluate the relationship of illicit drug use and sexual attitudes and practices.

Materials and Methods

Study procedures have been reported elsewhere (12,13), and are briefly described here.

Study population: Entry criteria for this study were being male, HIV-seronegative, ever having sexual activity with other men (male homosexual intercourse), being aged between 18-50 years of age. Volunteers were recruited through a combined strategy involving mainly outreach activities, e.g. recruitment by study volunteers themselves ("snow-ball"), and by non-governmental organisations (NGOs), as well as by media advertising and referrals from health care facilities. The study was approved by the Oswaldo Cruz Foundation Ethics Committee.

Study procedures: At the first (recruitment) interview basic data were obtained, and an explanation of the "Projeto Rio" study given. The volunteers were then given pre-test counselling with subsequent blood specimen collection for HIV, syphilis (VDRL, confirmed by TPHA) and hepatitis B (anti-HBc, anti-HBs, HBsAg) testing. On their return, one week later, the volunteers received the HIV and other test results during a session of post-test counselling. The HIV-positive individuals were referred to public health care facilities for follow-up care and those who were HIV-negative were asked to enroll in the study after signing the informed consent form. A detailed questionnaire was administered on the enrollment.

Study instruments: A standardized questionnaire was developed and administred by trained study staff. The form contains around 100 (mainly closed) questions on: socio-demographic characteristics, knowledge about HIV transmission routes and prevention strategies, beliefs and attitudes towards AIDS and sexual life, sexual practices, and STD/AIDS preventive initiatives.

The use of illicit drugs was initially assessed through the, following question: "In the last 6 months have you used drugs in situations associated with sexual activity?". The possible responses were: "always", "frequently", "casually", "rarely" and "never".

An exploratory analysis revealed no significant differences between men who reported doing this activity "always" or "frequently", and those who reported using drugs with sex "casually" or "rarely" (data not shown). We therefore considered for the purpose of current analysis men who reported any drug use with sex under the designation "drug user". We further classified respondents as to whether or not they used cocaine/crack. We found that 82 men reported using cocaine/crack; 46 of these men also use other drugs (e.g. marijuana, amphetamines, inhalants). Because there was a small number (n=8) of drug using respondents that did not use cocaine/crack we removed this group from the analysis.

Eight interviewees reported to have injected cocaine. These injectors were included in the "any powder cocaine/crack cocaine category".

Statistical analysis: In the bivariate analyses, we compared proportions for "drug using" and "not using" groups with chi-square test, and T tests for means were used for continuous variables. Statistical significance was defined by p-values less than 0.05.

We then specifically examined two sexual behavior outcomes: "unprotected anal sex with steady male partner" and "unprotected anal sex with casual male partner" in relation to drug consume. Although we did ask about unprotected sex with a women (bisexual behavior) these data were not presented in this paper.

Because "unprotected anal sex with casual male partner" was found to be associated with "powder cocaine/crack cocaine use" we further examined the relationship of several variables with this outcome analysis. We compared proportions between groups with odds ratios (OR) and 95% confidence intervals. Statistical significance was defined by OR that did not overlap 1.

To further assess the role of the different variables, we created logistic regression models, with "unprotected anal sex with casual partners males in the last 6 months" as the dependent variable. All variables significantly associated with the response variable (p<0.05) and/or borderline associations formerly mentioned in the literature were entered in the multivariate analysis which was performed using SPSS version 7.5 (Upper Saddle River, NJ: Prentice Hall).

We found in earlier analyses that there were substantial differences (e.g. socio-economic status, levels of unprotected sex) between participants that did are did not engage in commercial sex (13,14). Therefore all multivariate analyses were carried out, controlling and not controlling for reported engagement in commercial sex, but the adjusted OR and p-value were similar irrespectively of controlling, so only the models adjusting for commercial sex are shown. All variables that remained in the model were checked for colinearity (i.e. Pearson > 0.5), but no colinearity was found. Additionally we did not find any interaction between all significant variables in the model and use of "cocaine powder/crack cocaine".

Results

Of 675 participants, 37 (5.5%) could not be classified in relation to their drug use because information was missing. We found 501 (74.2%) participants did not use drugs and 137 (20.3%) used drugs. Among drug users 82 (59.9%) used powder cocaine or crack; 46 (33.6%) used other types of drugs (e.g., marijuana, inhalants) and 9 (6.5%) did not report the type of drug used. For our analysis we excluded the 55 drug users that did not use cocaine or crack.

In relation to sociodemographic characteristics we found there was not a significant difference between crack/cocaine users and men who did not use drugs by marital status nor age (Table 1). However, we did find that crack/cocaine users were significantly (p<0.05) more likely than non drug users to be unemployed (42.7% vs. 29.1%) to have an income of < $250 USD per month (considered below poverty) (70.7% vs. 60.9%), to be non-white (65.8% vs. 47%), and have <8 years of education (69.5% vs. 50.9%). We did not find differences in use of health care services.

When we examined the type of sexual partner with whom they had sex in the past 6 months we found that non drug users were significantly more likely to have a steady male partner (57.9%) than crack/cocaine users (31.7%) (Table 1). On the contrary, crack/cocaine users were significantly more likely to have casual male partners (75.6%) than non drug users (62.9%).

Additionally we found that crack/cocaine users were significantly more likely than non drug users to drink alcohol (76.8% vs. 45.7%), report bisexual activity (81.7% vs.41.7%) and engage in commercial sex (72.0% vs. 37.9%). We did not find differences between the two groups in relation to self-perception of vulnerability to HIV infection, willingness to participate in a HIV vaccine trial and serologic markers for syphilis and hepatitis B.

We further examined whether other variables were associated with unprotected anal sex with a casual partner (Table 2). We found that income <$250.00 USD a month (OR 1.80), being non white (OR 2.31) , < 8 years of education (OR 2.31), self-perception of vulnerability (OR 2.06), commercial sex (OR 1.98), use of alcohol (OR 1.47), bisexuality (OR 1.74), and willingness to participate in HIV vaccine trials (OR 1.56) were also associated with unprotected anal sex with a casual partner. No other sociodemographic or behavioral variables were associated with this unprotected anal sex.

In multiple logistic regression we found 5 factors were associated with unprotected anal sex with a casual partner (Table 2). These were income <$250 USD a month (Adjusted odds ratio [AOR] 1.73, CI 1.04-2.87), less than 8 years of education (AOR 2.21, CI 1.38-3.53), self perception of vulnerability (AOR 2.58 (1.54-4.33), willingness to participate in vaccine trials (AOR 1.91, CI 1.20-3.05) and use of crack/cocaine (AOR 1.91, CI 1.05-3.46).

Discussion

We found that among a cohort of HIV negative MSM in Rio de Janeiro 20.3% reported using drugs in the last six months. We also found that drug use was associated with a lower socioeconomic status. When we specifically examined which factors were associated with unsafe sexual practices we found that both drug use and lower socioeconomic status were independently associated with unprotected anal sex. Our study is one of the few studies in Brazil among a group of persons at high risk for HIV infection who are known to be HIV negative. In fact, it is the only study we could find in Brazil that analyses the intersection of drug use and sexual behavior among HIV negative MSM; behaviors that are important to understand in order to design appropriate HIV prevention programs for these men.

Approximately 60% of the men in our cohort reported using crack/cocaine. Crack/cocaine use in our cohort did not influence sexual behavior with steady sex partners in that both drugs users and non drug users were less likely to report using condoms with their steady partner. This has been found in many studies (15-17). It is most likely that decisions to not use condoms with one’s steady sex partner are associated with mutual trust (13).

On the other hand, participants who reported crack/cocaine use were more likely to engage in unsafe sexual practices with casual partners; a behavior that can place them at high risk for HIV and STD infections. Many studies have shown that drug use influences sexual behavior often resulting in sexual practices that are riskier than those practiced when not under the influence of drugs (15,18-27). HIV prevention programs for MSM should address drug use associated with sexual behavior.

These crack/cocaine users were of low socioeconomic status, frequently unemployed, with incomes of less that $250 a month, and with low levels of education. This finding is similar to that of studies of current and former crack/cocaine users in the city of São Paulo (28-30) and commercial sex workers who used cocaine in Santos, São Paulo (10). Poverty in itself can contribute to HIV transmission in that poverty is often associated with instability, inequality, discrimination, and disrespect of human rights; factors that make it difficult to protect oneself against HIV infection (31-34).

Besides being of lower socioeconomic status crack/cocaine users may be more likely to engage in multiple risky behaviors than persons who do not use drugs. Crack/cocaine users, both in our study and other studies, were more likely than men who did not use drugs to report using alcohol, engaging in commercial sex (35,36), practicing bisexual activity (37) and perceiving themselves to be vulnerable to HIV infection (38). This last finding has been found in several studies to be associated with greater willingness, than others, to participate in vaccine trials (39,40). We also found, in an earlier analysis of our cohort, men who perceived themselves vulnerable to HIV infection to be more willing to participate in vaccine trials than other men (13, 14). This suggests that recruitment of drug using MSMs into vaccine trials should be feasible.

There are several limitations to our study. The question specifically asked only about drug use associated with sex, we may actually have more drug users in the cohort than we detected. The small number of IDUs do not allow any separate analysis of this group; however, evidence suggests the majority of drug users in Brazil do not inject (30).

Not only are drug using MSMs at higher risk for HIV infection but once they become HIV infected they can facilitate the spread of HIV into other segments of society through their commercial sex and bisexual activity. Our findings suggest that HIV prevention programs for these men need to address drug use, not only to inform MSM of how drug use may influence their sexual practices but also to assist MSM who use crack/cocaine to stop using these drugs. Incorporation of our findings into HIV prevention programs in Brazil may contribute to a reduction in the spread of HIV.

Acknowledgments

We would like to thank the study volunteers and NGO leaders who have understood the social importance of participating in this scientific effort, notwithstanding difficulties. We also acknowledge the support provided by the staff of "Centro de Pesquisa Hospital Evandro Chagas" (CPqHEC/FIOCRUZ), the "National Reference Centre of Viral Hepatitis" (IOC/FIOCRUZ) made it possible to execute this study.

Financial support was provided by the National STD/AIDS Program of the Brazilian Ministry of Health; the Global Programme on AIDS/WHO; UNAIDS; Brazilian National Research Council (CNPq) and the Oswaldo Cruz Foundation (FIOCRUZ).

The staff of "Projeto Rio" include: Frits Sutmöller (coordinator), Claudia Teresa Vieira de Souza, Amilton Xavier Junior, Helena Santos Martins, Dionne Peluso de Oliveira Costa, Paulo Starling Brandão Junior & Therezinha Lucy Penna.

References:

1. UNAIDS (The Joint United Nations Programme on HIV/AIDS), 2000. Report on the global HIV/AIDS epidemic. Genebra: UNAIDS.

2. Fortenberry JD, Orr DP, Katz BP et al. Sex under the influence - A diary self-report study of substance use and sexual behavior among adolescent women. Sex Transm Dis 1997; 24(6):313-319.

3. Strathdee S, Galai N, Safaiean M et al. Gender Differences in Risk Factors for HIV Seroconversion among Injection Drug Users: A Ten Year Perspective. Arch of Internal Medine (in press).

4. Brazilian Ministry of Health. Boletim Epidemiológico de AIDS 2000a; XIII: 3-56.[Epidemiological Bulletin].

5. Carvalho HB, Mesquita F, Massad E et al. HIV and infections of similar transmission patterns in a drug injectors community of Santos, Brazil. J Acquir Immune Defic Syndr Hum Retrovirol 1996; 12: 84-92.

6. Brazilian Ministry of Health. Bela Vista e Horizonte: Estudos Comportamentais e Epidemiológicos entre Homens que fazem Sexo com Homens. Série Avaliação: 216 pp., 2000b.

7. Dourado I, Andrade T, Galvão-Castro B. HTLV-I in Northeast Brazil: differences for male and female injecting drug users. J Acquir Immune Defi Syndr Hum Retrovirol 1998; 19(4):426-9.

8. Oliveira MLA, Bastos FI, Telles PR et al. Prevalence and risk factors of HBV, HCV and HDV infections among injecting drug users from Rio de Janeiro, Brazil. Braz J Med Biol Res 1999; 32:1107-14.

9. Guimarães ML, Bastos FI, Telles PR et al. Retroviruses infections in a sample of injecting drug users in Rio de Janeiro City, Brazil: prevalence of HIV-1 subtypes, and co-infection with HTLV. Journal of Clinical Virology 2001; 14 (in press).

10. Szwarcwald CL, Bastos FI, Gravato N, Lacerda R, Chequer PN & Castilho EA. The relationship of illicit drug consume to HIV-infection among commercial sex workers (CSWs) in the city of Santos, São Paulo, Brazil. Int J Drug Polic 1998; 9:427-36.

11. Bastos FI, Lowndes CM, Derrico M, et al. Sexual behaviour and infection rates for HIV, blood-borne and sexually transmitted infections among patients attending drug treatment centres in Rio de Janeiro, Brazil. International Journal of STD & AIDS 2000; 11:383-392.

12. Sutmoller F, Souza CTV, Monteiro JC, et al: The Rio de Janeiro HIV Vaccine Site-I. Recruitment strategies and socio-demographic data of a HIV negative homosexual and bisexual male cohort in Rio de Janeiro, Brazil. Mem Inst Oswaldo Cruz 1997; 92:39-46.

13. Souza CTV, Bastos FI, Lowndes CM et al. Perception of vulnerability to HIV infection in a cohort of homosexual/bisexual men in Rio de Janeiro, Brazil. AIDS CARE 1999; 11:567-579.

14. Souza CTV, Sutmöller F, Lowndes CM et al. Willingness to participate in HIV vaccine trials among men who have sex with men, with and without a history of commercial sex, enrolled in a cohort study in Rio de Janeiro, Brazil (submitted).

15. Hospers H J & Kok G. Determinants of Safe and Risk-Taking Sexual Behavior among Gay Men: A Review. AIDS Education and Prevention 1995; 7(1):74-94.

16. Anderson JE, Wilson R, Doll L et al. Condom use and HIV risk behaviors among U.S. adults: data from a national survey. Fam Plann Prespect 1999; 31(1):24-8.

17. Lau JT, Thomas J. Risk behaviours of Hong Kong male residents travelling to mainland China: a potencial bridge population for HIV infection. AIDS Care 2001; 13(1):71-81.

18. Kalichman SC, Heckman T, Kelly JA. Sensation seeking as an explanation for the association between substance use and HIV-related risky sexual behavior. Arch Sex Behav 1996; 25(2):141-54.

19. Hays RB, Paul J, Ekstrand M et al. Actual versus perceived HIV status, sexual behaviors and predictors of unprotected sex among young gay and bisexual men who identify as HIV-negative, HIV-positive and untested. AIDS 1997; 11:1495-1502.

20. Waldo CR, McFarland W, Katz MH et al. Very young gay and bisexual men are at risk for HIV infection: the San Francisco Bay Area Young Men’s Survey II. J Acquir Immune Defic Syndr 2000; 24(2):168-74.

21. Ruiz J, Facer M, Sun RK. Risk factors for human immunodeficiency virus infection and unprotected anal intercourse among young men who have sex with men. Sex Transm Dis 1998; 25(2):100-7.

22. Strathdee AS, Hogg RS, Martindale SL, et al. Determinants of sexual risk-taking among young HIV-negative gay and bisexual men. J Acquir Immune Defic Syndr Hum Retrovirol 1998; 19(1):61-6.

23. Chesney MA, Barrett DC, Stall R. Histories of substance use and risk behavior: precursors to HIV seroconversion in homosexual men. Am J Public Health 1998; 88(1):113-6.

24. Benotsch EG, Kalichman SC, Kelly JA. Sexual compulsivity and substance use in HIV-seropositive men who have sex with men: prevalence and predictors of high-risk behaviors. Addict Behav 1999; 24(6):857-68.

25. MacNall M, Remafedi G. Relationship of amphetamine and other substance use to unprotected intercourse among young men who have sex with men. Arch Pediatr Adolesc Med 1999; 153(11):1130-5.

26. Crosby GM, Stall RD, Paul JP, Barrett DC. Substance use and HIV risk profile of gay/bisexual males who drop out of substance abuse treatment. AIDS Educ Prev 2000; 12(1):38-48.

27. Craib KJ, Weber AC, Cornelisse PG, et al. Comparason of sexual behaviors, unprotected sex, and substance use between two independent cohorts of gay and bisexual men. AIDS 2000; 14(3):303-11.

28. Dunn J & Laranjeira RR. Transitions in the route of cocaine administration-characteristics, direction associated variables. Addiction 1999; 94(6):813-24.

29. Ferri CP & Gossop M. Route of cocaine administration: Patterns of use and problems among a brazilian sample. Addict Behav 1999; 24(6):815-821.

30. Dunn J & Laranjeira RR. HIV-risk behaviour among non-heroin using cocaine injectors and non-injectors in São Paulo, Brazil. AIDS Care 2000; 12(4):471-81.

31. Farmer P, Connors M, & Simons J. Women, Poverty, and AIDS: Sex, Drugs and Structural Violence. Monroe: Comon Courage Press, 1996.

32. Bastos FI & Szwalrcwald CL. 2000. AIDS and pauperization: principal concepts and empirical evidence. Reports in Public Health 2000; 16 (Supp.1): 65-76.

33. Parker R & Camargo Jr KR. Poverty and HIV/AIDS: anthropological and sociological aspects. Reports in Public Health 2000; 16 (Supp.1): 89-102.

34. Parker RG, Easton D, Klein CH. Structural barriers and facilitators in HIV prevention: a review of international research. AIDS 2000; 14(Suppl 1):S22-S32.

35. de Graaf R, Vanwesenbeeck I, van Zessen G, et al. Alcohol and drug use in heterosexual and homosexual prostitution, and its relation to protection behaviour. AIDS Care 1995; 7(1):35-47.

36. Word CO, Bowser B. Background to crack cocaine addiction and HIV high-risk behavior: the next epidemic. Am J Drug Alcohol Abuse 1997; 23(1):67-77.

37. Diaz T, Chu SY, Frederick M et al. Sociodemographics and HIV risk behaviors of bisexual men with AIDS: results from a multistate interview project. AIDS 1993; 7(9):1227-32.

38. Robles RR, Cancel LI, Colon HM, et al. Prospective effects of perceived risk of developing HIV/AIDS on risk behaviors among injection drug users in Puerto Rico. Addiction 1995; 90(8): 1105-11.

39. Hays & Kegeles. Factors related to the Willingness of Young Gay Men to Participate in Preventive HIV Vaccine Trials. J Acquir Immune Defic Syndr Hum Retrovirol 1999; 20:164-171.

40. Koblin BA, Holte S, Lenderking et al. Readiness for HIV Vaccine Trails: Changes in Willingness and Knowledge among High-Risk Populations in the HIV Network for Prevention Trials. J Acquir Immune Defic Syndr Hum Retrovirol 2000; 24:451-457.

 

Table 1

Socio-demographic, behavioral characteristics of HIV negative MSM, according to drug use. Rio de Janeiro, Brazil, 1994-1999.

Variables

 

Not use drugs

Any Cocaine/Crack use a

   

N=501)

N=82

Age (mean in years; S.D.)

 

27.6; 7.41

27.6; 7.07

       

Marital status

Single

413 (82.4%)

67 (81.7%)

 

Married/ Divorced b

85 (17.0%)

15 (18.3%)

       

Employment

Currently Employed

347 (69.3%)

45 (56.3%)

 

Unemployed/alike c

146 (29.1%)

35 (42.7%)

       

Income (per month)

< U$250.00

305 (60.9%)

58 (70.7%)

 

³ U$250.00

166 (33.1%)

16 (19.5%)

       

Ethnic

White

265 (53.0%)

28 (34.2%)

 

Non-White

236 (47.0%)

54 (65.8%)

       

Educational level

Fundamental

255 (50.9%)

57 (69.5%)

 

High School/College

237 (47.3%)

23 (28.0%)

       

Corrently attending any health facility d

Yes

351 (70.0%)

61(74.4%)

       

Sexual partners type e

Male steady

290 (57.9%)

26 (31.7%)

 

Male casual

315 (62.9%)

62 (75.6%)

       

Other behavioral

and attitudinal

Use of alcohol

229 (45.7%)

63 (76.8%)

Variables

Sexual practices

   
 

Strictly homosexual

practices

289 (57.7%)

14 (17.0%)

 

Bisexual practice

209 (41.7%)

67 (81.7%)

 

Any commercial sex

190 (37.9%)

59 (72.0%)

 

Self-perception of vulnerability

282 (56.3%)

49 (59.8%)

 

Willingness to participate

279 (55.7%)

52 (63.4%)

       

Laboratory data

Serology (+) for syphilis

147 (29.3%)

25 (30.5%)

 

Serology (+) for hep. B

174 (34.7%)

28 (34.1%)

a Reported to have used powder cocaine (snorted or injected) and/or crack cocaine;

b Refers to civil status of men engaged with women. Brazilian law does not recognize the civil union of gay couples;

c Includes living from social security, informal sources of income, etc.

d Other than the cohort itself

e Data refer to those engaged in each one of these partnerships in the last 6 months

** In bold = Statistically significant (p<0.05)

Table 2

Factors associated with unprotected sex* with casual male sex partners among HIV negative MSM, according to baseline interview. Rio de Janeiro, Brazil, 1994-1999.

   

Univariate analysis

Multivariate analysis

Variables

Unprotected Anal Sex

OR

p- value

Ajusted OR

p-value

Income

         

< U$250.00

35.4% (146/413)

1.80 (1.23-2.62)

0.002

1.73 (1.04-2.87)

0.036

³ U$250.00

23.3% (49/210)

       
           

Ethnic

         

Non-white

36.9%(120/325)

1.82 (1.30-2.62)

0.001

1.16 (0.74-1.81)

0.526

White

24.0% (78/321)

       
           

Educational level

         

< 8 years

38.6% (139/360)

2.31 (1.63-3.27)

0.000

2.21 (1.38-3.53)

0.000

³ 8 years

21.4% (64/299)

       
           

Self-perception of vulnerability

         

Yes

35.0% (129/368)

2.06 (1.33-3.20)

0.001

2.58 (1.54-4.33)

0.000

No

20.6% (33/159)

       
           

Commercial sex**

         

Yes

39.0% (115/294)

1.98 (1.41-2.78)

0.000

1.42 (0.86-2.33)

0.167

No

24.5% (85/347)

       
           

Use of alcohol

         

Yes

34.2% (116/339)

1.47 (1.05-2.08)

0.027

1.39 (0.89-2.19)

0.150

No

35.8% (77/215)

       
           

Sexual behavior

         

Bisexual

36.9% (121/328)

1.74 (1.25-2.43)

0.001

0.99 (0.61-1.62)

0.977

Homosexual

25.1% (85/338)

       
           

Willingness to participate in anti-HIV vaccine trials

         

Yes

35.1% (129/368)

1.56 (1.12-2.18)

0.009

1.91 (1.20-3.05)

0.010

No

25.7% (79/307)

       
           

Use cocaine/crack

         

Yes

48.8% (40/82)

2.40 (1.51-3.91)

0.000

1.91 (1.05-3.46)

0.034

           

No

28.1% (144/512)

       

* Receptive or insertive anal sex without a condom in the last 6 months.

** Received money for sex

In bold = Significant differences at the level of p<0.05

 
 
  Início