Health Education Research Advance Access originally published online on March 14, 2007
Health Education Research 2008 23(1):137-145; doi:10.1093/her/cym011
Community development and HIV/STI-related vulnerability among female sex workers in Rio de Janeiro, Brazil
1 Department of International Health, Bloomberg School of Public Health, Johns Hopkins University, 615 North Wolfe Street, Room E5523A, Baltimore, MD 21205, USA
2 State University of Rio de Janeiro, Rua Fonseca Teles 121, 4 Andar, Sao Cristovao, Rio de Janeiro RJ 20940, Brazil
3 Network of Sex Work Projects (NSWP), The Stableyard, Broomgrove Road, SW9 9TZ London, UK
4 United Nations Educational, Scientific, and Cultural Organization (UNESCO), 7, Place de Fontenoy, 75352, Paris 07 SP, France
* Correspondence to: D. Kerrigan. E-mail: dkerriga{at}jhsph.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
The Sonagachi Project of Kolkata, India has been recognized as a model community development and human immunodeficiency virus/sexually transmitted infection (HIV/STI) prevention intervention among female sex workers. Limited research has been conducted regarding its applicability outside the South Asian context. This study sought to document the process and effectiveness of integrating community development activities based on the Sonagachi model into an ongoing HIV/STI peer education program with female sex workers in Rio de Janeiro, Brazil. Structured cross-sectional surveys examining HIV/STI-related behaviors and community development measures were conducted among
500 sex workers at pre- and post-intervention. We found that several community development components including social cohesion and mutual aid were significantly associated with consistent condom use among sex workers and their paying clients at pre-intervention. However, only a minority of women actively engaged in community-building activities over the 18-month study period. In turn, limited changes in community development components and no significant increases in the HIV/STI-related protective behaviors assessed were documented. Findings indicate that internalized stigma and socioeconomic pressures may have constrained the scope and pace of community mobilization in this setting during the study observation period. |
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
The Sonagachi Project of Kolkata, India brought to the forefront the importance of community mobilization and development among female sex workers as a means to improving their health and well-being and reducing human immunodeficiency virus/sexually transmitted disease (HIV/STI)-related vulnerability [1, 2]. When the Sonagachi Project began in the early 1990s, it utilized a three-pronged approach to HIV/STI-risk reduction including behavior change communication, condom distribution and STI management. Over time, the project became sensitized to the importance of sex workers' social vulnerability to HIV infection. In particular, project leaders and participants came to conceptualize sex workers' heightened HIV/STI-related vulnerability in terms of community powerlessness resulting from the historic exclusion of sex workers from access to material resources, policy considerations and decision making and citizenship [3, 4].
The social vulnerability perspective adopted by Sonagachi over time led them to develop and implement a broader set of program activities to complement the aforementioned three-pronged approach. Five intervention areas were identified through a comprehensive case study as critical components of the Sonagachi operational model, including (i) facilitating a sense of community among sex workers, (ii) decreasing perceived powerlessness and insecurity, (iii) increasing access and control over material resources, (iv) increasing social participation and (v) facilitating the social acceptance of sex workers [3, 4]. It is this framework and the increases in condom use and the decreases in STIs that were achieved through its implementation which has made Sonagachi a UNAIDS Best Practice Model for HIV/STI prevention among sex workers [5, 6].
In an effort to extend the lessons learned from Sonagachi to other sociocultural contexts, an intervention research project was initiated among a group of low-income, predominantly street-based female sex workers in Rio de Janeiro, Brazil. In 2000, a series of participatory workshops were held to discuss the Sonagachi experience and potential applications of their community development approach among this group of women. These meetings made clear the potential importance of community cohesion among sex workers in order to both increase opportunities for mutual support regarding protective health behaviors and for organizing collective action to address structural constraints to improving the their health, including violence and discrimination. In turn, a diverse set of community development activities were defined and implemented by local sex workers. The primary objective of this paper is to describe the process, the effectiveness and the challenges of integrating community development activities into an ongoing HIV/STI peer education program among female sex workers in Rio de Janeiro, Brazil.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
Research setting
Approximately 6% of female sex workers are estimated to be infected with HIV in Brazil [7]. While this is considerably higher than the 0.7% overall national prevalence rate among adults [8], curbing the expansion of the HIV epidemic among female sex workers represents a considerable amount of effort and innovation on the part of the Brazilian Ministry of Health. The Ministry has made concerted efforts to reach sex workers with HIV prevention messages, materials and resources in partnership with local non-governmental organizations (NGOs) across the country, particularly in large urban areas such as Rio de Janeiro and Sao Paulo. One of the more recent media campaigns oriented toward female sex workers in Brazil entitled, ‘Maria without Shame’ relays the Ministry's positive, non-judgmental approach to the sex work community. It is also important to note that sex work is not illegal in Brazil and is recognized as an official occupation by the Brazilian government [9]. These factors made Brazil a logical and feasible setting for adapting the social vulnerability and HIV/STI prevention framework developed and utilized by the Sonagachi Project.
Three study areas were purposively selected for inclusion in this study. The sites were chosen from among those areas where our local implementing partner Programa Integrado de Marginalidade (PIM) had been conducting HIV/acquired immunodeficiency syndrome (AIDS) peer education for over a decade. Low-income areas with a high prevalence of street-based sex work were prioritized for the purposes of the study. The three specific sites included (i) a downtown site near the city's main train station, (ii) a large park and stadium area within the city limits and (iii) a main marketplace and truck stop area right outside the city. These sites were neither intended to be representative of sex work in Rio de Janeiro nor thought to be homogenous in terms of the demographics of the sex workers or clients associated with each. All three have in common, however, several key traits in that they are all areas where drug use and sales as well as both petty and violent crime are quite prevalent. Because of the strong history of HIV/AIDS peer education at each site, they also shared a knowledge base and shared experience which severed as a starting point for the community development activities developed during the course of the intervention.
Intervention activities
As part of the transition to a community development approach, the role of existing peer educators was jointly reconceptualized by NGO workers and sex worker peer leaders. Whereas their previous role focused predominantly on one-on-one and small group discussions regarding the importance of HIV/STI prevention and the provision of educational materials and condoms, the new role of project peer educators emphasized the idea of being an agent of social change. In turn, they were trained and encouraged to meet regularly with other sex workers in their sites to elicit and discuss issues of common concern regarding their health and well-being, not limited to their sexual health, and in particular to encourage their colleagues to participate in the newly formed sex worker organization named, Fio da Alma or The Soul's String.
One of the first community development activities initiated by Fio da Alma, upon the suggestion of many of the sex workers involved in the planning process, was the opening of a Drop-in-Center in Downtown Rio de Janeiro. The center sought to create a ‘safe space’ for sex workers to come together to discuss issues of common concern, hold project workshops and activities and serve as the administrative offices of Fio da Alma. Additionally, sex workers from each of the three study sites were asked to jointly identify priority action areas. They then solicited small funds and technical assistance from NGO project staff to assist them in implementing activities that they perceived as appropriate responses to the needs and concerns documented in open group meetings with sex workers from their area. Intervention activities were implemented during an 18-month period. Examples of intervention activities from each of the five conceptual areas of the project, based on the Sonagachi model described above, are described in Table I.
|
Data collection procedures
The study utilized a pre/post-test cross-sectional design. A pre-intervention survey was conducted with 499 female sex workers during the period May–July 2001 and a post-intervention survey was conducted with 537 sex workers from December 2002–February 2003. The survey component of the study utilized convenience sampling and included as many women >18 years of age who reported exchanging sex for money at the study sites as possible.
In order to estimate the population of women working at these sites, study staff visited each area on several different occasions at differing times of day in order to approximate the number of women working as sex workers at each location based on the knowledge of peer educators associated with each site. Discussion groups were also held with women from each site to discuss and refine these estimates. It was determined that roughly 700 women were working in the three study areas prior to the initiation of the pre-intervention survey. The study also sought to recruit sub-samples from each site proportionate to the estimated population totals found per site at pre-intervention. The research team was able to include 70% of the estimated total population in the survey components of the study with the assistance of peer educators who acted as recruitment assistants and who had longstanding relationships with many of the women working at the sites.
Ethical issues
The study protocol and informed consents were approved by the Institutional Review Board of the State University of Rio de Janeiro. All survey participants were informed of the study objectives and potential risks and benefits of participation. All participants gave written informed consent. No monetary compensation was provided to study participants. Condoms, health education and health and social services referrals were made available to all study participants. The survey interviews were conducted by study staff trained in survey methodology and took place in private settings at each project site. In order to address both ethical issues associated with community participation as well as improve the utility of the data collected, sex workers were actively engaged in a wide range of research components associated with this study. In addition to assisting with recruitment, sex workers participated in developing the research protocol, data collection instruments and informed consent forms and in monitoring survey interviews in the field. They also reviewed data generated from the study and used it to inform and refine the intervention. Several capacity building sessions were conducted with groups of sex workers from each of the three study sites to facilitate their ability to meaningfully participate in these research and program planning activities [10].
Measures
Community development measures
Elements of all five components of the community development model employed by Sonagachi and adapted in the intervention herein were measured at pre- and post-intervention and are described in detail in Table I. The community development measures assessed included perceived social cohesion, level of comfort with sex work profession, ability to manage client risk behavior, access to and management of social and material resources, social participation and social acceptance. All aggregate measures related to community development were left in their original continuous form. Additionally, we sought to assess exposure to and participation in community-building activities by inquiring into whether pre- and post-test survey participants were familiar with Fio da Alma or considered themselves members of the group, knew where the group's drop-in-center was located or had visited the center and had participated in any group activities with other sex workers in the last 4 months.
HIV/STI-related outcomes
Study outcomes included protective sexual behaviors such as consistent condom use with both paying and non-paying partners. Participants were asked to report any instances of unprotected sex with paying clients over the last 4 months as well as the number of protected sex acts with all partners over the last week. They were also asked to report the number of times condoms were used during the last 10 sexual acts with their new clients, regular clients and steady, non-paying partners. Condom use variables were dichotomized into consistent and non-consistent use.
Sociodemographic variables
The study also assessed the following sociodemographic variables: age in years, educational level, number of years exchanging sex for money, average number of clients per week, average monthly income over last 4 months and the number of people the participant helped to financially support with income earned. All sociodemographic variables were measured as continuous variables and later dichotomized at their median.
Data analysis
Chi-square tests of association were conducted to identify significant differences in the sociodemographic characteristics of study participants from pre- to post-intervention. The relationship between community development components and consistent condom use with paying clients at pre-test was examined by conducting bivariate and multivariate logistic regression. Only sociodemographic variables and community development indicators found to be significantly associated with condom use in bivariate analyses at the P < 0.10 level of significance were included in multivariate analyses. Pre- to post-intervention changes in community development and HIV/STI-related behaviors were also assessed using bivariate and multivariate logistic and linear regression, controlling for sociodemographic variables for which significant pre- to post-changes were detected. Bivariate and multivariate logistic regression was also conducted to determine the psychosocial and material factors associated with higher levels of participation in community-building activities at post-test.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
Characteristics of the samples at pre- and post-intervention
As shown in Table II, the median age of pre-intervention survey participants was 28.0. The majority of participants reported low levels of formal education with only a small percentage of women at pre-test, 18.1%, having any secondary school education. The median number of years in sex work at pre-test was 3.0 and the median number of dates with clients per week was 10.0. The median average monthly income from sex work at pre-intervention was $US 228 with almost 40% of participants reporting that they supported more than two individuals in addition to themselves with this income. Significant pre- to post-intervention changes were detected for the following variables: number of years in sex work, number of clients per week, average monthly income and number of survey participants per study site. Changes in clients and income may reflect issues of seasonality and demand given the differing points of the year in which the pre- and post-test surveys were conducted. Our ability to reach our sample size targets at one of our sites was impacted by a series of violent crimes committed against sex workers around the time of the baseline survey. In turn, we had to adjust our site-specific sampling goals pre- to post-intervention contributing to the significant difference in the number of participants recruited per site at the two data collection points.
|
Community development measures and consistent condom use at pre-intervention
Table III demonstrates that many of the community development components examined at pre-intervention were significantly associated with consistent condom use among female sex workers and their paying partners in the last 4 months. Social cohesion and mutual aid, the ability to manage client risk behaviors, ownership and management of material resources, possession of government documents, participation in community organizations and experiences with violence were all significantly associated with condom use in bivariate analysis.
|
With the exception of ownership and management of material resources, all of these variables remained significant in multivariate analyses. We documented a 1.30 increase in the odds of consistent condom use with clients in the last 4 months per each level of social cohesion and mutual aid among sex workers. We found a 1.39 increase in the odds of condom use per each level of sex workers' ability to manage client risk behaviors. The possession of key government documents also proved important with a 1.36 increase in the odds of condom use per each additional document possessed. The level of sex workers' participation in community groups outside of the sex industry was also significantly associated with consistent condom use in multivariate analyses with a 1.56 increase in the odds of condom use per each additional group of affiliation. Violence related to sex work became significant at the P < 0.05 level in multivariate analysis with a decreased odds (OR 0.37, 95% CI 0.17–0.81) of consistent condom use with clients in the last 4 months among those who reported experiencing violence in the last 4 months. Significant sociodemographic factors included age and the number of persons supported by the participants' income from sex work. Whereas older sex workers were less likely (OR 0.48, 95% CI 0.24–0.95) to use condoms than younger sex workers and participants supporting more persons with their income from sex work were less likely to use condoms with clients in the last 4 months (OR 0.29, 95% CI 0.15–0.56).
Pre–post changes in community development measures and HIV/STI-related behaviors
Of the nine community development measures assessed at pre- and post-intervention, significant positive changes were detected among four of them, after controlling for sociodemographic characteristics of the samples. Three of the four positive changes were documented within the domain of perceived powerlessness and insecurity and one within the sphere of social participation. Level of comfort being a sex worker (OR 1.80, 95% CI 1.36–2.38; P < 0.0001) and level of comfort disclosing sex work to friends (OR 1.41, 95% CI 1.07–1.85; P < 0.0001) both increased significantly from pre- to post-intervention. Participants' ability to manage client risk behaviors related to alcohol, drugs and violence increased significantly pre–post intervention, with the mean score on the four-point measure rising from 2.07/4.0 to 2.49/4.0 (P < 0.0001). Social participation also increased significantly with the mean number of community organizations that sex workers participated in during the prior 4 months rising from 0.74 at pre-test to 1.23 at post-test (P < 0.0001).
We did not detect any significant positive changes in the condom use behaviors assessed among participating female sex workers from pre- to post-intervention. The majority of the behaviors assessed remained virtually the same from pre- to post-intervention including consistent condom use with all clients in the last 4 months (87.2–88.6%; P = 0.287) and consistent condom use with all partners, including both paying and non-paying partners, over the last week (80.4–79.0%; P = 0.808).
Exposure to and participation in community development activities
The intervention sought to increase community cohesion and social participation among sex workers and encourage them to come together as a group in a safe collective space. Several significant pre–post changes were detected regarding this process. For example, the percent of participants who could name Fio da Alma as the sex work organization represented by program peer educators, knew about the drop-in-center and its location and considered themselves Fio da Alma members all increased significantly from pre to post. However, the percent of women who had actually visited the center did not increase significantly (8.5–11.8%; P = 0.10); nor did the percent of women who had participated in a group activity with other sex workers during the last 4 months (13.8–16.0%; P = 0.314). Given these low levels of participation, we examined the predictors of participation in group activities with other sex workers in multivariate analyses among post-test survey participants. These analyses revealed that both psychological and material factors were associated with participation in community-building activities. For example, sex workers who were less comfortable with their profession (OR 0.61, 95% CI= 0.38–0.97) were significantly less likely to have participated in a group activity with other sex workers in the last 4 months as were sex workers who supported more than two people other than themselves (OR 0.38, 95% CI = 0.33–0.87) with their income from sex work.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
This study is one of the few to have adapted the community mobilization and development approach to HIV/STI prevention utilized by the Sonagachi Project of Kolkata, India to a sociocultural context outside of South Asia. Through this research, we developed five reliable aggregate measures to represent key components of the Sonagachi model which may be useful to public health programs working from a community development perspective within a variety of settings. Additionally, we found that four of these five community development measures including social cohesion and mutual aid, social participation and access to material and social resources were significantly associated with HIV/STI preventive behavior such as consistent condom use among participating sex workers and their clients, indicating the importance of community-building and community-bridging interventions for this population group.
Intervention activities were implemented in relation to each of the five key components of the Sonagachi operational model during the study period including the development of a local sex worker organization, the establishment of a sex-worker-run drop-in-center and a variety of solidarity-building activities and collective advocacy initiatives. While the activities implemented were well received among those who attended them, we did not achieve widespread participation in the intervention over the study period as hoped. Additionally, we did not detect significant pre–post gains in the condom use behaviors assessed among participating sex workers during the 18-month observation period. In interpreting these findings, we propose that it is critical to note that the Sonagachi Project evolved into a comprehensive community development initiative over the course of almost a decade, rather than over a period of 18 months [6]. Additionally, when Sonagachi began in 1992, consistent condom use with clients was quite rare or <10% [6]. However, consistent condom use with clients was already >80% among participating sex workers from this study when the community development activities implemented began.
While our study period may have been too short to document significant changes in the HIV/STI-related outcomes assessed, we did find significant increases in several community development components associated with protective sexual behavior over the study period. These include participants' level of comfort with their work, ability to ensure their safety by managing their client's risk behavior and social participation or involvement in community organizations outside the sex industry. Findings among post-test participants suggest that both psychological and material factors may have constrained a broader community mobilization and development process and in turn reductions in HIV/STI-related vulnerability over the study period. Hence, factors such as internalized stigma surrounding identity as a sex worker and socioeconomic status should be addressed in future programmatic efforts with female sex workers in order to facilitate greater participation in and benefits from the community-building process.
The primary limitation of this study is the use of a pre/post-test design limiting our ability to establish temporal patterns between the cross-sectional associations which we observed. Our sole reliance on reported outcomes measures related to condom use is an additional study limitation, as is our relatively short intervention and observation periods. Given the complexities of developing and implementing community mobilization and development activities, future research in this area should consider the use of longer term longitudinal study designs and a mix of biomedical and behavioral data collection methodologies. While no further operations research has been conducted by our study team with this project to date, the Brazilian government has been supportive of the processes initiated herein, lending both material and moral support to the project and facilitating the participation of project leaders in the ongoing dialogue between government and civil society regarding the prevention of HIV/STI and the care of those living with HIV in Brazil.
Despite its limitations, this study indicates that the community mobilization and development components and processes documented by the Sonagachi experience are relevant and feasible to implement outside of that context. However, the pace by which program planners can adapt elements of the Sonagachi operational model may be influenced by a variety of contextual factors which should be assessed and addressed throughout program implementation in order to increase the possibilities for both project effectiveness and sustainability.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
None declared.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
This research is dedicated to the memory of Paulo Longo, this study's principal investigator and long-time HIV/AIDS and human rights activist. Without his passion and dedication, this study would not have been realized. We would like to thank all of the sex workers who participated in the intervention and research, particularly the leadership of Fio da Alma, the local sex work organization that made this initiative possible. In particular, we would like to extend our gratitude to the work of Fio da Alma's President, Ivanilda Santos de Lima. We would also like to thank Celia Szterenfeld, Rose Soares and Lourdes de Alecrim from the Programa Integrado de Marginalidade, the local NGO that helped to coordinate the implementation of the intervention as well as Zelia Caldeira and Zairine Freire from the Sociedade de Estudos E Pesquisas em Drogadiccao which oversaw the research conducted in Rio de Janeiro. Finally, we would like to thank Naomi Rutenberg, Ellen Weiss, Katie Schenk and Malea Hoepf. The time dedicated to this paper by its first author was partially supported by the National Institute of Mental Health (Grant K01 MH3491).
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
1. Jana S, Basu I, Rotheram-Borus MJ, et al. The Sonagachi Project: a sustainable community intervention program. AIDS Educ Prev (2004) 16:405–14.[CrossRef][Web of Science][Medline]
2. Basu I, Jana S, Rotheram-Borus MJ, et al. HIV prevention among sex workers in India. J Acquir Immune Defic Syndr (2004) 36:845–52.[CrossRef][Web of Science][Medline]
3. Bandyopadhyay N, Ray K, Banerjee A. Operationalizing an Effective Community Development Intervention for Reducing HIV Vulnerability in Female Sex Work: Lessons Learned from the Sonagachi Project in Kolkata, India. Barcelona, Spain: XIV International AIDS Conference. 6–12. July 2002. Oral presentation number [ThOrF1478].
4. Bandyopadhyay N, Mahendra V, Kerrigan D. The Role of Community Development Approaches in Ensuring the Effectiveness and Sustainability of Interventions to Reduce HIV Transmission through Commercial Sex: Case Study of the Sonagachi Project, Kolkata, India (2003) Washington, DC: Prepared for the Horizons Program of the Population Council/USAID. Draft Final Report.
5. UNAIDS. Technical Update: Sex Work and HIV/AIDS (2002) Geneva, Switzerland: UNAIDS Best Practices Collection.
6. UNAIDS. Female Sex Worker HIV Prevention Projects: Lessons Learned from Papua New Guinea, India and Bangladesh. UNAIDS Case Study (2000) Geneva, Switzerland: Best Practices Collection.
7. Okie S. Fighting HIV—lessons from Brazil. N Engl J Med (2006) 354:1977–81.
8. UNAIDS. Brazil: Epidemiological Fact Sheets on HIV/AIDS and Sexually Transmitted Infections (2004) Geneva, Switzerland: UNAIDS Publication.
9. AVERT. HIV & AIDS in Brazil. Accessed: 1 November 2006. Available at: http://www.avert.org/aids-brazil.htm.
10. Longo P, Overs C, Lima IS, et al. From research "subjects" to "partners": getting the ethics right. Int Conf AIDS (2002) 7–12:14. Abstract No. TuPpE2070].
Received on July 31, 2006; accepted on January 14, 2007
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Shannon, S. A. Strathdee, J. Shoveller, M. Rusch, T. Kerr, and M. W. Tyndall Structural and Environmental Barriers to Condom Use Negotiation With Clients Among Female Sex Workers: Implications for HIV-Prevention Strategies and Policy Am J Public Health, April 1, 2009; 99(4): 659 - 665. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S Hawkes, M Collumbien, L Platt, N Lalji, N Rizvi, A Andreasen, J Chow, R Muzaffar, H ur-Rehman, N Siddiqui, et al. HIV and other sexually transmitted infections among men, transgenders and women selling sex in two cities in Pakistan: a cross-sectional prevalence survey Sex Transm Inf, April 1, 2009; 85(Suppl_2): ii8 - ii16. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||