FEMALE SEX WORKERS (FSWs) ARE A high-risk group of HIV/STI.1 Their male clients may also play an important role in the spread of infections to the general population. These men are considered a “bridge” of transmission between FSWs and women in the general population, through unprotected sex.2 Despite their high risk behaviors, clients of FSWs have rarely been studied and few interventions have targeted this group. The belief that this population is hard to reach may partially explain this. Some studies indicate that the prevalence of HIV/STI is higher among clients of FSWs than among men in general.3–7
Although HIV has reached epidemic proportions in Haiti, it remains concentrated within specific groups. HIV prevalence was estimated at 2.5% in 2005–2006.8 Sentinel surveillance among pregnant women has indicated a prevalence of 3.7% for syphilis.9 For herpes simplex virus 2 (HSV-2), a seroprevalence of 54% was observed in women attending a health care centre.10 Currently, heterosexual contact seems to be the primary route for HIV transmission.11,12 Among FSWs, a recent Haitian study indicates a prevalence of 10% for HIV and 18% for syphilis.13 This high prevalence could lead to high HIV/STI risk for their sexual partners. To date, no studies on clients of FSWs in Haiti have been conducted. Therefore, it is essential to learn more about the characteristics and risk behaviors of these men, to devise effective interventions targeting this population.
The objectives of this study were to: 1) examine characteristics and risk behaviors of clients of FSWs; and 2) estimate the prevalence of HIV, syphilis, HSV-2 infections, and their associated risk factors.
Materials and Methods
A cross-sectional survey was conducted in commercial sex sites in St-Marc (population 62,200) and Gonaives (population 104,825), two port cities located in the Artibonite region in Haiti. The study is part of the Projet D’Appui à la Lutte contre les ITS/VIH/SIDA en Haiti (PALIH), which supports local efforts to promote STI/HIV transmission control; reinforce a network of STI services; prevent mother-to-child HIV transmission; and organize education and behavior change campaigns.
Study Population and Sampling
The study population consisted of clients of FSWs, aged ≥18 years, recruited from various sites in St-Marc and Gonaives. Commercial sex sites (brothels, dance clubs, bars, restaurants, and street locations) were identified by local FSWs and field workers collaborating with this study. The choice of sites was based on their size (≥5 clients per night), security profile, accessibility, diversity, and geographical location. A client was defined as a man who was present at the commercial sex site during fieldwork and who had had a sexual encounter in the last 3 months with an FSW for which he had paid in money or goods.
Data Collection and Study Procedure
Data were collected in December and January 2006–2007, in the evening (6–10 pm). During fieldwork, it was difficult to register and count each person who refused to participate, but participation was high; interviewers estimated nonresponse at less than 5%. Three field teams collected data in each city, consisting of 1 supervisor, 4 interviewers, a collaborating local FSW, and a nurse. Interviewers received 4 days of training. The sites were mapped through visits to estimate the number of clients present. Owners were informed about the project and their permission to conduct the study was obtained. Collaborating local FSWs approached clients, explained the purpose of the study and invited them to participate. After giving their verbal informed consent, clients were interviewed for 20 to 30 minutes in a quiet place. The data collection instrument was a structured questionnaire, translated into Creole, on sociodemographic characteristics, and behaviors regarding STI/HIV/AIDS. A finger prick was performed by the nurse and capillary blood deposited on a filter paper (Whatman no. 3; Whatman International, Maidstone, UK). Dried blood spots (DBS) were stored at room temperature for a maximum of 2 weeks and then transported to Montreal, Canada, where they were stored at 4°C until testing, a maximum of 1 month later. Of the 378 clients interviewed, 351 (92%) provided a blood sample. Each questionnaire and blood sample received its own identification number. Clients wishing to know their STI/HIV status were referred to a health clinic for free testing, counseling, and treatment, if necessary. Free condoms, information on STI/HIV/AIDS, and T-shirts were offered to clients for participation. This study was completely anonymous and approved by the Ethics Committee of the Universite de Montreal (Quebec, Canada) and the Departement Sanitaire de l’Artibonite (Haiti).
HIV serology was performed using Detect-HIV (Adaltis Inc., Montreal, Canada) with 100 μL of eluted sample obtained from a 6-mm DBS eluted in 250 μL of sample diluent for 16 hours at 4°C. Nonreactive samples were considered seronegative whereas reactive samples were tested with Genie II HIV-1/HIV-2 (Bio Rad Laboratories, Marnes La Coquette, France). Genie II dually reactive samples (to HIV-1 and HIV-2), as well as discordant samples (Detect-HIV reactive/Genie II nonreactive) were further tested by INNO-LIA HIV I/II Score (Innogenetics, Ghent, Belgium), using 1 mL of eluted sample (resulting from two 6-mm DBS in 1.2 mL of INNO-LIA sample diluent).
Syphilis antibodies were screened with Trep-check Treponema antibody EIA (Phoenix Bio-Tech Corp., Mississauga, ON) by using100 μL of eluted sample (6-mm DBS eluted in 200 μL of sample diluent). Nonreactive samples were considered seronegative whereas reactive and equivocal samples were tested with SeroDia TP-PA (Fujirebio, Malvern, PA) as described previously.14
The HerpesSelect HSV type specific ELISA (Focus Technologies, Cypress, CA) was used to determine the presence of HSV-2 antibodies in 25 μL of eluted sample (6-mm DBS eluted in 150 μL of sample diluent). Nonreactive samples were considered seronegative. Reactive and equivocal samples were tested with Kalon HSV-2 IgG (Kalon Biologic, Aldershot, UK), using 180 μL of eluted sample obtained from a 6-mm DBS in 250 μL of sample diluent, and Captia HSV-2 type specific IgG (Trinity Biotech, Jamestown, NY), using 100 μL of eluted sample obtained from a 6-mm DBS in 175 μL of sample diluent. Samples reactive with at least 2 of the 3 commercial kits were considered HSV-2 positive and those nonreactive or equivocal with both Kalon and Captia were considered seronegative.
Outcome variables were HIV, Treponema pallidum and HSV-2 serostatus. A summary STI variable distinguished those “infected” (with at least 1 STI/HIV) from those “uninfected.”
Age was categorized as: 18 to 20; 21 to 25; 26 to 30; and 30 or older; religion as Catholic, Protestant, Voodoo, and other (mostly atheist); education as none or primary, and secondary or higher; occupation as student, drivers, and others (mason, mechanic, security agent, etc.). Prostitution sites were classified as brothels, streets, or other; city of residence as St-Marc and Gonaives. Other background variables were living with a partner (yes/no), and length of time since subject began visiting FSWs (1–3/4–6/7 or more years).
Frequency of condom use was examined by type of sex partner: FSWs, regular, and occasional partners. A respondent was classified as “consistent” if he always used a condom with a type of partner. The number of sex partners and visits to sex FSWs within the previous 3 months was categorized as: 1 to 4, 5 to 9, and 10 or more years. An indicator was used to distinguish those with only FSW contacts from those with relationships with both FSW and non-FSW. Marijuana use was investigated via a single question: Have you ever smoked marijuana?
Associations between each outcome variable and potential predictors were examined using simple logistic regression. For “HIV infection,” multivariate logistic models were fitted entering variables in staggered fashion, using the stepwise backward procedure and retaining variables with P <0.10 within each of the following blocks: 1) background; 2) behaviors; 3) STI variables. The last block was omitted for outcomes “syphilis” and “HSV-2.” Variables were entered in the models according to level of significance in univariate analysis (P <0.25).15 Estimations of odds ratio (OR), 95% confidence interval (CI), and P values are presented in the final model. Only significant variables (P <0.05) were retained in the final models. EPI-INFO 3.3.2 and SPSS version 12 were used.
Of the 378 clients, 191 were from St-Marc and 187 from Gonaives. Men were young (mean age 24), 70.1% had at least secondary education and over half (60.7%) were living with a partner (married or common-law). Twenty-six percent were students, 14.3% drivers (taxi, truck, bus), and 59.7% worked at other occupations. Most were Catholics (55.0%), with Protestants (22.9%) and Voodoo practitioners (15.1%) ranking next. Most respondents were recruited in brothels (70.0%), others on the streets (20.2%), or at commercial sex sites (mostly bars and disco). The majority had been frequenting FSWs for 4 years or more (64.0%; mean 6.3 years). St-Marc’s population contained more drivers (22.5% vs. 5.9%; P <0.001). Most clients in Gonaives were recruited in brothels (87.3% vs. 53.2% in St-Marc) and very few frequented FSWs working in the street (4.4% vs. 35.5%).
Clients reported a high number of sex partners; 4 of 10 had 10 or more different partners and nearly half of them had visited FSWs at least 10 times in the previous 3 months (Table 1). Very few had exclusively FSWs as partners (5.3%). Clients in Gonaives had visited FSWs more often (P = 0.05). Respondents used condoms more frequently with FSWs (59.5%) than with occasional (44.9%) or regular partners (32.8%). Consistent condom use with all partners was higher among clients from St-Marc. One of 5 men reported having tried marijuana. More men in Gonaives had tried marijuana than those in St-Marc (P <0.001).
Prevalence of HIV, Syphilis, and HSV-2
Among all respondents, 16.8% reported having had a previous STI and only 17.2% had been tested (Table 2). Seven percent (7.2%) were infected with HIV-1. Antibodies against Treponema pallidum were detected in 13.4% of clients, reflecting past or active syphilis infection. Finally, 22.0% were infected with HSV-2. HSV-2 and syphilis infections were higher among respondents in Gonaives than St-Marc (P <0.001).
Behaviors of STI-Infected and Uninfected Clients
Clients infected with any STI had the same number of sexual partners and visits to FSWs compared with those uninfected (data no shown). There was no difference concerning condom use with FSWs. Consistent condom use with regular partners tended to be less frequent among infected than uninfected men (25.5% vs. 36.9%; P = 0.05). The same trend was observed with occasional partners, but nonsignificant. STI-infected clients were also more susceptible to have smoked marijuana (30.5% vs. 16.8%; P <0.01).
Risk Factors for STI/HIV Among Clients
The population of St-Marc and Gonaives combined was used in univariate analyses (Table 3). Clients in Gonaives were more likely to be infected with HSV-2 or syphilis than those in St-Marc (OR = 2.84; 95% CI = 1.68–4.78 and OR = 4.50; 95% CI = 2.16–9.36, respectively). HIV prevalence was similar in both cities (6.4% in St-Marc and 8.1% in Gonaives, Table 2). The older the client and the longer they had been visiting FSWs, the greater the prevalence of all STI. Clients with little education were more likely to test positive for HSV-2. A similar, but not significant, trend was observed with HIV and syphilis. Students were less likely to be infected by STI compared with other clients. Protestants seemed to be protected from STI and those practicing Voodoo were more likely to be infected with all 3 diseases. Living with a partner and type of commercial sex site were not associated with any of these STI.
Clients who had tried marijuana at least once were more frequently infected with syphilis and HIV (OR = 3.5; 95% CI = 1.82–6.72 and OR = 3.09; 95% CI = 1.35–7.04) (Table 3). A similar, but nonsignificant association was observed for HSV-2. The greater the number of partners and number of visits to FSWs, the more the prevalence of syphilis and HIV increased. However, no significant associations were observed for HSV-2.
Reporting a history of STI was significantly associated with HIV (OR = 2.99; 95% CI = 1.26–7.07) (Table 3). Prevalence of HSV-2, syphilis, and HIV infections were highly correlated. The association between HIV and HSV-2 was stronger (OR = 8.9; 95% CI = 3.72–21.5).
Table 4 shows the multivariate results for HIV infection. Of all the background variables, older age, and practicing voodoo were associated with HIV. When behavioral variables were entered, associations with religion and age were attenuated. Having tried marijuana was the only behavioral variable whose association with HIV reached statistical significance, adjusting for age. Including history of STI and HSV-2 explained most of the association with HIV, indicating that STI and HIV share the same determinants and/or STI serve as an entry point for HIV infection.
City of residence, religion, and occupation were independently associated with syphilis (Table 5). Among the behavioral variables, having tried marijuana was the only one associated with syphilis infection, after adjusting for sociodemographic variables.
Living in Gonaives, practicing Voodoo and older age were the only sociodemographic variables associated with HSV-2 infection (Table 6). None of the behavioral variables was associated with HSV-2, adjusting for sociodemographic variables.
This is the first study examining the characteristics of clients of FSWs in Haiti, as well as STI prevalence and related risk factors. Most respondents were young, Catholic, living with a partner and high-school educated. Nearly one-fifth had tried marijuana. Their young age might be related to the fact that a commercial transaction with an FSW is cheap and visiting FSWs is socially accepted.
Clients reported having had multiple sex partners over the previous 3 months and almost all of them alternate FSWs and non-FSW partners. Frequency of visits to FSWs was also high. Condom use with FSWs was relatively consistent. However, it was less consistent with casual partners and lesser still with regular partners. Furthermore, clients infected with any STI were less likely to use condom consistently with regular partners. Therefore, clients of FSWs are likely to act as a bridge for HIV transmission.
STI prevalence was high. Approximately 7.2% of clients of FSWs were infected with HIV-1; more than 3 times higher than the figure reported in a recent population-based study.8 This high prevalence is cause for concern, considering that most of these men are young, untested for HIV, and probably unaware of their serological status. The prevalence of HIV as well as HSV-2 and syphilis increased with age. The latter 2 STI probably cause ulcers facilitating HIV transmission. HSV-2 and syphilis prevalence was higher in Gonaives as expected because clients in this city were using condom less consistently, had more sexual partners and visits to FSWs. The safer sexual behaviors of clients in St-Marc could be the result of well-established STI/HIV prevention programs in this city. The fact that these 2 STI are considered cofactors for HIV transmission12,16,17 indicates a high risk for clients in Gonaives.
Overall, factors associated with the 3 STI are similar, with few exceptions possibly attributable to correlations between the factors considered. Age was associated with HIV and HSV-2, whereas it did not enter the equation for syphilis. Occupation entered the equation for syphilis instead of age because students are younger than other occupational groups. Men who had had many sex partners ran a higher risk of being infected with HIV. However, this association was attenuated when STI history and HSV-2 was included in the model. This finding reinforces the importance of controlling STI, especially HSV-2, to reduce HIV transmission within this population.
Clients who had tried marijuana at least once were more likely to be infected with syphilis and HIV. Individuals using cannabis have been shown to engage in risky behaviors and run a higher risk of developing STI/HIV.18–21 Cannabis use has been associated with a greater number of partners22,23 and early sexual initiation.24 Cannabis consumption may impair judgment and decision- making, diminish risk perception, reduce behavioral control, and increase the risk of unprotected intercourse. Alternatively, an underlying disposition to risk-taking may induce both cannabis use and sexually risky behavior.25,26
Interestingly, Protestant clients were less likely to be infected with STI. Clients practicing Voodoo were at higher risk of HSV-2, syphilis, and HIV infection. Those in the “other” religious category (mostly atheist) were more likely to be infected with syphilis. Religious affiliations may generate positive effects on STI risk by promoting delayed sexual initiation and abstinence.27 However, certain religious groups discourage condom use, increasing the risk of infection.28,29 The literature has shown that religiosity, the strength of religious beliefs, may be more important than religious affiliation.30–33 This aspect was not examined in our study and could explain the observed difference of STI/HIV infections across religious groups.
Our study has limitations that should be considered when interpreting the results. This study mostly sampled clients frequenting busy commercial sex sites that may lead to underrepresentation of clients of high class and occasional FSWs. Moreover, clients who rarely visit these places were less likely to be included. STI prevalence may be lower among them. However, the fact that this study was conducted during the early evening (until 10 pm), and that more dangerous sites were not visited might have led to a possible underestimation of STI prevalence. Self-selection bias could be present despite high participation because it was not possible to determine the exact number of clients invited to participate who declined the invitation.
Data on sexual behaviors based on self-report may be subject to recall problems and desirability bias. Variables not taken into consideration in this study, such as alcohol consumption, may affect our results. Finally, a cross-sectional study can only provide statistical associations and no causality can be ascertained. Although the analysis of factors associated with HIV has limited power because of low prevalence within the population, this design was appropriate for the purpose of this study, which was to determine the profile of the clients of FSWs and to estimate STI prevalence. Finally, results may not be representative of the entire Haitian client population, because the study was conducted in two cities of the Artibonite region. Moreover, it is unclear whether these findings can be safely generalized to other settings. Although Haitian culture may differ in several ways, we believe that the characteristics and behaviors of our population may be similar to the clients of FSWs living in other developing countries having similar social acceptability and frequenting of commercial sex venues.
Poverty, political instability, violence, traditional gender norms, and patriarchal beliefs, limited access to health care, lack of an efficient surveillance system, and risky sexual behaviors are some key factors that may have contributed to the spread of HIV in Haiti. Efforts targeting clients of FSWs in Haiti are nearly nil. Our findings highlight the importance of STI/HIV prevention interventions addressing this vulnerable population. Interventions should promote consistent condom use, because this is suboptimal among this clientele, especially with occasional and regular partners. The importance of having fewer sexual partners should also be addressed. Activities should be included in already existing programs targeting FSWs. Given the high prevalence of STI and their association with HIV, STI testing, counseling, and treatment should be incorporated into HIV prevention programs.
1. Plummer FA, Nagelkerke NJ, Moses S, et al. The importance of core groups in the epidemiology and control of HIV-1 infection. AIDS 1991;5 Suppl 1:S169–S176.
2. Lowndes CM, Alary M, Meda H, et al. Role of core and bridging groups in the transmission dynamics of HIV and STIs in Cotonou, Benin, West Africa. Sex Transm Infect, 2002;78 Suppl 1:i69–i77.
3. Lowndes CM, Alary M, Gnintoungbe CA, et al. Management of sexually transmitted diseases and HIV prevention in men at high risk: targeting clients and non-paying sexual partners of female sex workers in Benin. AIDS 2000; 14:2523–2534.
4. Hor LB, Detels R, Heng S, et al. The role of sex worker clients in transmission of HIV in Cambodia. Int J STD AIDS 2005; 16:170–174.
5. Vuylsteke BL, Ghys PD, Traore M, et al. HIV prevalence and risk behavior among clients of female sex workers in Abidjan, Cote d’Ivoire. AIDS 2003; 17:1691–1694.
6. Espirito Santo MF, Etheredge GD. How to reach clients of female sex workers: a survey by surprise in brothels in Dakar, Senegal. Bull World Health Organ 2002; 80:709–713.
7. Cote AM, Sobela F, Dzokoto A, et al. Transactional sex is the driving force in the dynamics of HIV in Accra, Ghana. AIDS 2004; 18:917–925.
8. EMMUSIV, Enquête mortalité, morbidité et utilisation des services: Haïti. Institut haïtien de l’enfance (IHE)-Institut haïtien de statistique et d’informatique (IHSI), 2006. Rapport préliminaire.
9. Ministère de la Santé Publique et de la Population, IHD and CG l’Enfance, Centers for Disease Control and Prevention, Etude de séro surveillance par méthode sentinelle de la prévalence du VIH, de la syphilis, de l’hépatite B et de l’hépatite C et chez les femmes enceintes en Haïti 2003/2004. 2004:Haïti.
10. Boulos R, Ruff AJ, Nahmias A, et al. Herpes simplex virus type 2 infection, syphilis, and hepatitis B virus infection in Haitian women with human immunodeficiency virus type 1 and human T lymphotropic virus type I infections. The Johns Hopkins University (JHU)/Centre pour le Developpement et la Sante (CDS) HIV Study Group. J Infect Dis 1992; 166:418–420.
11. Pape JW, Stanback ME, Pamphile M, et al. Prevalence of HIV infection and high-risk activities in Haiti. J Acquir Immune Defic Syndr 1990; 3:995–1001.
12. Deschamps MM, Pape JW, Hafner A, et al. Heterosexual transmission of HIV in Haiti. Ann Intern Med 1996; 125:324–330.
13. CCISD-CECI, Résultats de la première enquête de surveillance de seconde génération chez les travailleuses du sexe de St-Marc, Artibonite-Haïti. 2005a, Centre de coopération internationale en santé et développement-Centre canadien d’études et de coopération internationale.
14. Dada Y, Milord F, Frost E, et al. The Indian Ocean paradox revisited: HIV and sexually transmitted infections in the Comoros. Int J STD AIDS 2007; 18:596–600.
15. Hosmer DW, Lemeshow S. Applied Logistic Regression.
2nd ed. Wiley-Interscience; 2001.
16. Freeman EE, Weiss HA, Glynn JR, et al. Herpes simplex virus 2 infection increases HIV acquisition in men and women: systematic review and meta-analysis of longitudinal studies. AIDS 2006; 20: 73–83.
17. Reynolds SJ, Risbud AR, Shepherd ME, et al. High rates of syphilis among STI patients are contributing to the spread of HIV-1 in India. Sex Transm Infect 2006; 82:121–126.
18. Poulin C, Graham L. The association between substance use, unplanned sexual intercourse and other sexual behaviours among adolescent students. Addiction 2001; 96:607–621.
19. Bautista CT, Sanchez JL, Montano SM, et al. Seroprevalence of and risk factors for HIV-1 infection among female commercial sex workers in South America. Sex Transm Infect 2006; 82:311–316.
20. Cook RL, Comer DM, Wiesenfeld HC, et al. Alcohol and drug use and related disorders: an underrecognized health issue among adolescents and young adults attending sexually transmitted disease clinics. Sex Transm Dis 2006; 33:565–570.
21. Simbayi LC, Kalichman SC, Jooste S, et al. Risk factors for HIV-AIDS among youth in Cape Town, South Africa. AIDS Behav 2005; 9: 53–61.
22. Castilla J, Barrio G, Belza MJ, et al. Drug and alcohol consumption and sexual risk behaviour among young adults: results from a national survey. Drug Alcohol Depend 1999; 56:47–53.
23. Wingood GM, DiClemente RJ. The influence of psychosocial factors, alcohol, drug use on African-Am women’s high-risk sexual behavior. Am J Prev Med 1998; 15:54–59.
24. Stanton M, Leukefeld C, Logan TK, et al. Gender differences in substance use and initiation of sexual activity. Pop Res Policy Rev 1999; 18:89–100.
25. Brodbeck J, Matter M, Moggi F. Association between cannabis use and sexual risk behavior among young heterosexual adults. AIDS Behav 2006; 10:599–605.
26. Stall R, McKusick L, Wiley J, et al. Alcohol and drug use during sexual activity and compliance with safe sex guidelines for AIDS: the AIDS Behavioral Research Project. Health Educ Q 1986; 13:359–371.
27. Agha S, Hutchinson P, Kusanthan T. The effects of religious affiliation on sexual initiation and condom use in Zambia. J Adolesc Health 2006; 38:550–555.
28. Arie S. Crusading for change. BMJ 2005; 330:926.
29. Condoms and the Vatican. Lancet 2006; 367:1550.
30. Hardy SA, Raffaelli M. Adolescent religiosity and sexuality: an investigation of reciprocal influences. J Adolesc 2003; 26:731–739.
31. Lefkowitz ES, Gillen MM, Shearer CL, et al. Religiosity, sexual behaviors, and sexual attitudes during emerging adulthood. J Sex Res 2004; 41:150–159.
32. Rostosky SS, Wilcox BL, Comer ML, et al. The impact of religiosity on adolescent sexual behavior: a review of the evidence. J Adolesc Res 2004; 19:677–697.
33. Zaleski EH, Schiaffino KM. Religiosity and sexual risk-taking behavior during the transition to college. J Adolesc 2000; 23:223–227.