You are here

Article Text

Article menu
Epidemiology
Original article
Risk-taking behaviours and HIV infection among sex workers in Portugal: results from a cross-sectional survey
  1. Sónia Dias1,
  2. Ana Gama1,
  3. Ricardo Fuertes2,
  4. Luís Mendão2,
  5. Henrique Barros3
  1. 1Instituto de Higiene e Medicina Tropical & CMDT, Universidade Nova de Lisboa, Lisboa, Portugal
  2. 2GAT—Grupo Português de Activistas sobre o Tratamento de VIH/Sida Pedro Santos, Lisboa, Portugal
  3. 3Institute of Public Health of University of Porto, University of Porto Medical School, Porto, Portugal
  1. Correspondence to Professor Sónia Dias, Instituto de Higiene e Medicina Tropical & CMDT, Universidade Nova de Lisboa, Rua da Junqueira, 100, Lisboa 1349-008, Portugal; sfdias{at}ihmt.unl.pt, smfdias{at}yahoo.com

Abstract

Objectives Sex workers (SW) are key populations at an increased risk of HIV infection. This study aimed to characterise risk-taking behaviours and assess HIV prevalence among SW in Portugal.

Methods A cross-sectional survey was conducted with 1040 SW using a participatory research approach. SW were recruited in sex-work locations and community-based organisation offices. Data were collected through a questionnaire with trained interviewers. An HIV rapid test was performed in 213 respondents.

Results Reported HIV prevalence was 8%: 17.6% of man-to-woman transgenders, 7.4% of women and 5% of men. Of SW reportedly living with HIV, 52.2% reported ever injecting drug use. Inconsistent condom use with clients in the last month was higher among male SW (26.5%) and with non-paying partners in the last year was higher among women (71.3%). Among reported HIV-positive SW, the proportions of inconsistent condom use were high. In multivariate regression analysis, reported HIV infection remained significantly higher among transgenders (OR 6.4; 95% CI 1.7 to 24.3), those older (OR 5.1; 95% CI 1.3 to 21.1), working outdoors (OR 5.4; 95% CI 1.9 to 15.6), having ever used psychoactive substances (OR 4.1; 95% CI 2.2 to 7.7) and earning ≤€1000 per month (OR 2.6; 95% CI 1.2 to 5.9). Of those who had an HIV rapid test, 8.9% were reactive; 73.7% were unaware of their seropositivity.

Conclusions The HIV infection burden in SW is high. Efforts to promote HIV testing must be sustained in order to reduce undiagnosed infection. The diverse risk profiles of SW must be addressed in targeted HIV interventions. Prevention interventions should be systematically implemented within most-at-risk subgroups of SW.

  • HIV
  • SEXUAL BEHAVIOUR
  • COMMERCIAL SEX
  • SEXUAL HEALTH
  • DRUG ADDICTION

https://doi.org/10.1136/sextrans-2014-051697

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    HIV continues to be a major concern in public health.1 Sex workers (SW) have been considered to be at an increased risk of HIV infection. Limited data about HIV prevalence among SW in Europe show wide variation between countries,2 ,3 being estimated to be >1% in 22 countries and >5% in six of these countries (6.1%–22.2%).2

    Sex work is viewed as extremely widespread; yet, the actual number of people involved is difficult to determine.4 Sex work varies in terms of formality or organisation and settings in which it occurs, ranging from dedicated establishments and private homes to hotels, bars and streets or parks.5 ,6

    Little scientific attention has been given to the heightened behavioural, social and structural vulnerabilities of SW to HIV. Increased risk of HIV infection can result from multiple exposures: large number and concurrency of sexual partners, inconsistent condom use,7 ,8 intersection with injecting drug use9 ,10 and presence of other sexually transmitted infections (STIs).7 ,11 In addition, among SW, infection risk and the adoption of unsafe behaviours vary across economic strata, sociocultural context and work environments.12 Outdoor sex work is associated with higher vulnerability to HIV and violence.13 The health and safety of these SW are frequently compromised by the inability to assert control over their working environments and negotiate safer sex.7 Access to health services is often hindered by discriminatory attitudes and social exclusion, particularly among men and transgenders.6 ,14 Additionally, criminalisation and enforcement-based policies can impact directly and indirectly on HIV vulnerability.3 ,15

    Despite the widely recognised importance of characterising the burden of HIV among SW, epidemiological and socio-behavioural data remain scarce in Europe. In particular, there is no information on HIV prevalence among SW in Portugal, one of the countries presenting the largest burden of infection in Europe.16 It probably reflects the difficulty of addressing hidden populations, which are stigmatised and not represented in traditional national HIV surveillance systems.4 Injecting drug users have driven the Portuguese HIV epidemic, but sexually transmitted cases are arising. Routine surveillance data showed that in 2011 about 62% of the newly diagnosed infections were related to heterosexual and 26.2% to homosexual/bisexual transmission and 9.6% to injecting drug use.17 The contribution of sex work to transmission of HIV in the country remains unknown. Thus, we aimed to characterise risk-taking behaviours and assess HIV prevalence among SW in Portugal.

    Methods

    Behavioural study

    A cross-sectional survey of SW was conducted using a participatory research approach. A Community Advisory Board (comprising SW, representatives of non-governmental and governmental organisations, and academics) actively participated in the study design, implementation and interpretation of results. SW were defined as ‘female, male and transgender adults and young people who received money or goods in exchange for sexual services, either regularly or occasionally’ whatever the settings.6 The study inclusion criteria were being ≥18 years old and reporting sex work in the last 12 months. To estimate the sample size that would be large enough for detailed analyses, we assumed a hypothetical worst case prevalence of 50%, at 95% confidence level with precision of 3.5%; it was estimated that 784 SW would be required. Given that there is no adequate sampling frame of this population enabling to draw a probability sample of SW in Portugal, this study used a convenience sampling approach to generate a sufficiently robust sample that could capture different elements of the population. In a first stage, a geographic and network mapping was conducted based on a formative research with community partners in which the locations and networks where SW congregate were systematically listed to identify data collection sites. We sampled SW from the three large urban centres in Portugal: Great Lisbon (n=470), Great Porto (n=424) and Coimbra (n=146). Data were collected in venues frequented by large and diverse numbers of SW, including sex-work locations (streets, bars, private residences) and community-based organisations targeted at SW. Interviewers approached eligible respondents inviting them to participate. After data collection, respondents were asked to advertise the study among persons involved in sex work from their social networks. In total, 1040 SW were enrolled in the study.

    Data were collected between January and September 2011 using a questionnaire administered by trained interviewers recruited from community organisation partners. Interviewer training included a presentation of study's purpose, ethical considerations, instrument, data collection procedures and practice of interviewing techniques through role plays. The questionnaire comprised information on socio-demographic characteristics, sex-work setting, sexual behaviour, drug use, HIV testing, prevention initiatives, HIV infection and other STIs.

    The study was approved by the Ethics Committee for Health of North Regional Health Administration (No.18.09CES). Before responding to the questionnaire, participants were informed of the purpose and procedures of the study; confidentiality was guaranteed and informed consent was obtained. All participants were given information leaflets on HIV prevention and the testing services available.

    HIV prevalence study (rapid test)

    Throughout 1 month, we offered an HIV rapid test to every respondent in sex-work venues in the centre of Lisbon using a mobile unit that provided adequate conditions of privacy and room for pretest/post-test counselling. Logistic constraints limited the provision of HIV rapid tests in other areas of the country and for the whole study period. Of the 272 eligible participants who were approached, 213 accepted to be tested onsite using a finger-stick whole-blood rapid test (see online supplementary figure 1). Those with a reactive HIV test were referred to HIV care services for confirmatory testing and linkage to care. All the participants referred had their positive result confirmed. A numerical code was used to link behavioural and biological information, maintaining participant anonymity.

    Statistical analysis

    χ2 Or exact Fisher test was used for comparisons. The magnitude of univariate and multivariate associations among reported HIV infection and socio-demographic, behavioural and environmental factors was estimated by means of OR and 95% CIs calculated using logistic regression. A p<0.05 was set to retain variables in the model. For statistical analysis, SPSS V.19.0 was used.

    Results

    Characteristics of participants

    Overall, the refusal rate was 35%; no significant differences were found between refusals and respondents on gender, age and education. As shown in table 1, 853 participants (82%) were women, 106 (10.2%) men and 81 (7.8%) transgenders (man-to-woman). Men were significantly younger and more frequently non-nationals than transgenders and women. Transgenders had a higher level of education. More female SW reported lower incomes and worked outdoor (in streets/cars). Most outdoor SW were Portuguese (71.7% vs 34.9% indoor, p<0.001; among non-nationals, Africans: 53.5% outdoor vs 4.1% indoor, Europeans: 19.4% vs 3.3%, Latin Americans: 27.1% vs 92.6%, p<0.001) (data not shown in table). Also, the majority of outdoor SW had a monthly income ≤€1000 (75.1% vs 40.0% indoor, p<0.001) and had ≤9 schooling years (84.5% vs 52.5% indoor, p<0.001).

    View this table:
    Table 1

    Characteristics and risk behaviours of the participants by gender

    Of those who had an HIV rapid test (n=213), 176 were women (82.6%), 28 men (13.2%) and nine transgenders (4.2%) (data not shown). No significant gender differences were found regarding age (overall, 59.6% were ≤35 years old) and nationality (50.2% were Portuguese). More women reported low income (≤€1000: 73.5% vs 50.0% of transgenders and 42.9% of men, p=0.009) and worked outdoors (85.8% vs 55.6% of transgenders and 47.6% of men, p<0.001).

    HIV risk behaviour

    A higher number of clients was more frequently reported by transgenders than men and women. Men declared a higher number of non-paying partners. Inconsistent condom use with clients was higher among men (26.5%) and with non-paying partners was higher among women (71.3%). Condom failure and forced sexual relations were reported, with no significant gender differences (table 1). Forced sexual relations were more likely among outdoor SW (14.3% vs 5.3% indoor, p<0.001) (data not shown in table). Those who had forced sexual relations reported more frequently inconsistent condom use with clients (19.8% vs 10.1%, p=0.003).

    Ever use of psychoactive substances was reported by 68.9% of men, 46.9% of transgenders and 31.1% of women. Ever injecting drug use was more prevalent among women (18.1%) than men (13.7%) or transgenders (7.9%) (table 1). Injecting drug use was also higher among Portuguese SW (22.4% vs 3.3% non-nationals, p<0.001) and outdoor SW (25.4% vs 1.7% indoor, p<0.001). Those who ever injected drugs also reported more frequently having had forced sexual relations (23.3% vs 14.4%, p=0.085) (data not shown).

    HIV testing and prevention

    Most participants reported having been tested for HIV (table 2). Further analysis showed a higher proportion of those never tested among non-nationals (24.1% vs 18.9% of nationals; p=0.05), but no significant differences were found across education, sex-work setting, number of partners and drug use. Those SW who ever tested compared with those never tested reported more frequently consistent condom use with clients and non-paying partners (90.2% vs 81.6%, p=0.001; 36.5% vs 25.9%, p=0.010) and having been reached by HIV prevention programmes (44.2% vs 24.4%, p<0.001).

    View this table:
    Table 2

    HIV prevention initiatives, HIV testing, self-reported HIV infection and STI by gender

    HIV prevalence

    Self-reported HIV infection

    Of those ever tested for HIV, 66 (8%, 95% CI 6.0 to 10.0) reported living with HIV: 17.6% of transgenders (95% CI 8.0 to 27.0), 7.4% of women (95% CI 5.0 to 9.0) and 5% of men (95% CI 1.0 to 10.0) (table 2). In additional analysis, approximately 10% of participants admitted not knowing their current serostatus. Recalling at least one STI episode in the last 12 months was more frequent among HIV-positive than HIV-negative participants (24.4%, 10/41 vs 6.8%, 39/577, p=0.001).

    Among SW reportedly living with HIV, transgenders more often reported inconsistent condom use with clients (33.3%, 4/12), than men (25.0%, 1/4) and women (8.9%, 4/45), (p=0.034). The opposite was observed with non-paying partners: 43.6% (17/39) of women, 33.3% (1/3) of men and 18.2% (2/11) of transgenders (p=0.086). Condom failure was referred by 54.4% and forced sexual relations by 15.4% of reported HIV-positive SW (no significant differences across gender). Ever injecting drug use was reported by 52.2% of HIV positives (vs 10.6% among HIV negatives): 66.7% (2/3) of men, 57.1% (20/35) of women and 25.0% (2/8) of transgenders (p=0.099). Among these injecting drug users, 55.0% referred that the probable mode of HIV transmission was sharing injection material.

    HIV infection (rapid test)

    Of the 213 SW who accepted to be tested, 8.9% (n=19) were reactive: 22.2% (n=2; 95% CI 0.0 to 56.0) of transgenders, 10.7% (n=3; 95% CI 0.0 to 23.0) of men and 8% (n=14; 95% CI 4.0 to 12.0) of women. Of the 19 who tested HIV positive, 14 (73.7%, 95% CI 52.0 to 95.0) had reported being HIV negative or not knowing their serostatus: two transgender cases, three men and nine (64.3%) of positive women. Around 64% (9/14) of unknown HIV-positives were non-nationals: 55.6% (five) women, 22.2% (two) men and 22.2% (two) transgenders.

    Of the 14 HIV-positive participants unaware of their status, 42.9% inconsistently used condoms with clients: 50% (1/2) of transgenders, 44.4% (4/9) of women and 33.3% (1/3) of men. About 75.0% (6/8) inconsistently used condoms with non-paying partners: all women (3/3), 66.7% (2/3) of men and 50.0% (1/2) of transgenders. Condom failure occurred in 33.3%. Forced sexual relations were reported by 33.3%. Around 43% reported ever consuming psychoactive substances: all men (3/3), 50% (1/2) of transgenders and 22.2% (2/9) of women. No unknown HIV-positive participant reported ever injecting drug use.

    Factors associated with self-reported HIV infection

    In the multivariate model, reported HIV infection remained significantly higher among transgenders (OR 6.4; 95% CI 1.7 to 24.3), those older (OR 5.1; 95% CI 1.3 to 21.1), earning ≤€1000 per month (OR 2.6; 95% CI 1.2 to 5.9), having used psychoactive substances (OR 4.1; 95% CI 2.2 to 7.7) and doing outdoor sex work (OR 5.4; 95% CI 1.9 to 15.6) compared with indoor. HIV infection was lower among non-nationals (OR 0.4; 95% CI 0.1 to 0.9) compared with Portuguese (table 3).

    View this table:
    Table 3

    Factors associated with self-reported HIV infection*

    Discussion

    This is the first Portuguese survey conducted in a large sample of SW to characterise risk behaviours and assess HIV prevalence. The frequency of high-risk sexual behaviours remains concerning. Even with gender differences, SW reported high client frequency, multiple partners and engagement in unprotected sex with clients and non-paying partners. A high reported prevalence of HIV was found, particularly among transgenders. Other European studies show highly disparate levels of infection in different SW populations.8 ,18–20 It is worth noting that in general risk behaviours were frequent among reported HIV-positive SW but also among those screened who tested positive and did not know their seropositivity. The risk of sexual behaviours undertaken combined with HIV prevalence increases the likelihood of spreading of HIV to general population. This indicates that both SW and their clients/non-paying partners are key populations to be targeted as they may act as bridging populations.

    Self-reported HIV prevalence was significantly higher among drug users compared with non-drug users (15.3% vs 3.8%). Among HIV-positive SW, 52.2% ever injected drugs. This is consistent with research demonstrating that HIV infection is driven largely by the overlap of injecting drug use and sex work.10 ,11 ,21 Our study also found a higher proportion of injecting drug users working outdoors which heightens vulnerability to HIV infection. As in earlier surveys, SW working in streets presented higher levels of infection than those working in brothels, clubs and apartments.3 ,22 Outdoor SW had lower income and educational levels. In a context of socioeconomic disadvantage, these SW are more likely to engage in higher risk behaviours.7 ,13 We observed that among outdoor SW, a higher proportion engaged in inconsistent condom use with clients (12.9% vs 8.1% of indoor SW, p=0.063). Additionally, an important proportion of SW experienced forced sexual relations (14.3% vs 5.3% of indoor, p<0.001), which increases vulnerability to HIV. As noted in the literature, policies supporting safer environments and integrated interventions including harm-reduction programmes should be promoted.23 ,24

    The results from the rapid tests show that 74% of the SW who tested HIV positive (reactive test) were unaware of their serostatus. Among these, 64.3% were migrants from highly endemic areas. Although the sample screened was small and CIs were wide, this can be an important finding. Those unaware of their HIV status are at risk of transmitting HIV to others through high-risk behaviours. Early diagnosis is crucial to limit further transmission of HIV.25 This finding may partly explain the higher self-reported HIV prevalence among Portuguese SW compared with non-nationals. Other possible explanations for this disparity are that a larger proportion of Portuguese SW worked outdoor, reported lower income and higher injecting drug use than non-nationals.

    Our findings indicate a considerable HIV testing rate and proportion of participants tested during the previous year. In Europe, data on HIV testing among SW are scarce and the reported rates vary considerably between countries, ranging from 33% in Czech Republic and 35% in France to 93.8% in Belgium and 100% in Greece.26 The rate of HIV testing found in our study may be a positive outcome of prevention programmes among SW implemented in the last few years as community-based HIV testing initiatives. Nevertheless, the proportion of SW who never tested for HIV is worth highlighting. Indeed, these reported more frequently inconsistent condom use with both clients and non-paying partners. In Portugal, HIV testing is non-mandatory and can be done anonymously, confidentially and for free. Potential reasons for not getting tested are mistrust of health services, confidentiality, fear of discrimination or exclusion, lack of awareness of services available and low self-perception of HIV risk.27 Unfortunately, we did not collect data on duration of sex work and so we cannot determine its effect on HIV-testing behaviour. There remains a need to understand factors that may underlie unwillingness to be tested. Yet, our findings reinforce that efforts should be made to increase HIV testing, particularly among individuals who do not perceive themselves to be at risk or who do not contact healthcare services.28 Particular emphasis should be given to increase health service use by migrant SW.6 ,13 Furthermore, as supported by other studies,29 ,30 integrating outreach strategies for HIV prevention is valuable: having been reached by HIV prevention programmes in the previous year appeared positively associated with having been tested for HIV.

    Almost a fourth of reported HIV-positive SW reported an STI. STIs increase the risk of HIV infection by increasing HIV infectiousness and susceptibility.25 In a context of frequent high-risk sexual exposure, STIs may have a stronger role on HIV infection and transmission. Preventing and controlling STIs are key strategies in controlling the spread of HIV/AIDS, and HIV/STI programmes and services should be integrated for SW.

    Limitations of this study must be acknowledged. First, due to sampling procedure, results may not reflect the situation of the SW population in general. Nevertheless, interviewing SW at different times and working places allowed reaching a heterogeneous sample of 1040 SW comprised of outdoor and indoor, nationals and non-nationals, with reported HIV and other STIs. The socioeconomic characteristics of our sample are in line with recent data on sex work in Europe showing high rates of low education and low income among SW.5 Also, the socio-demographic profile of individuals who refused to participate in the study was similar to those who participated. However, as we did not collect information about reasons for refusal to behavioural study or to rapid test, it is not possible to determine to what extent HIV prevalence and related risk behaviours might have differed between participants and refusals. Second, given the relatively low numbers of HIV positives, the wide CIs of the OR in the multivariate analysis and the missing data in some variables, results must be treated with caution. Finally, to include those of low-level literacy and migrants, we opted for an interview-administered questionnaire. Interviewers might have interpreted the responses in a biased way, although they were trained thoroughly prior to fieldwork. Social desirability and recall bias also potentially led to under-reported risk behaviours and HIV infection from respondents. Nevertheless, the high response rate and the obtained data on reported HIV prevalence make us confident of the validity of the responses. Overall, the HIV burden in SW population could be under-represented.

    Research among SW remains relatively scarce, which reflects the ongoing challenge in studying this population. Further research is required for an indepth understanding of the effect of sex work-related variables and factors that increase vulnerability.

    In conclusion, to prevent further transmission of HIV it is essential to acknowledge the diverse risk profiles of SW and address the links between these key features and HIV/STIs. The findings of this study suggest most-at-risk subgroups that need to be additionally focused for future prevention interventions and research such as transgenders, drug users and those who are unaware of HIV status. Safe sex promotion should also pay attention to HIV-positive SW. Overall, targeted interventions aimed at improving uptake of testing and reducing risky behaviours in this population should be sustained.

    Key messages

    • The HIV burden in sex workers (SW) is high and the overlap between drug use and sex work is a major driver of HIV infection.

    • The prevalence of risky sexual behaviour among SW, such as high numbers of partners and inconsistent condom use with clients and non-paying partners, is concerning.

    • The findings suggest a potential high level of undiagnosed infection among SW.

    • Prevention interventions should be systematically implemented particularly within most-at-risk subgroups of SW including transgenders, drug users and those not tested for HIV.

    Acknowledgments

    The authors would like to thank the project team, namely, Daniel Simões, Gabriela Cohen, Inês Rego, Ricardo Fernandes, Ricardo Rosa and Sara Trindade. Special thanks to all the participants of this study. The authors are also very grateful to all the community partners of the project.

    References

    1. ↵
      UNAIDS. The gap report 2014. Geneva: UNAIDS, 2014.
    2. ↵
      ECDC 2013. Thematic report: Sex workers. Monitoring implementation of the Dublin Declaration on Partnership to Fight HIV/AIDS in Europe and Central Asia: 2012 progress report. Stockholm: ECDC, 2013.
    3. ↵
      1. Platt L,
      2. Jolley E,
      3. Rhodes T, et al
      . Factors mediating HIV risk among female sex workers in Europe: a systematic review and ecological analysis. BMJ Open 2013;3:e002836.
      OpenUrlAbstract/FREE Full Text
    4. ↵
      1. Vanderpitte J,
      2. Lyerla R,
      3. Dallabetta G, et al
      . Estimates of the number of female sex workers in different regions of the world. Sex Transm Infect 2006;82(Suppl 3):iii18–25.
      OpenUrlAbstract/FREE Full Text
    5. ↵
      TAMPEP. Sex work in Europe: a mapping of the prostitution scene in 25 European countries. Amsterdam: TAMPEP, 2009.
    6. ↵
      UNAIDS. UNAIDS guidance note on HIV and sex work. Geneva: UNAIDS, 2009.
    7. ↵
      1. Li Y,
      2. Detels R,
      3. Lin P, et al
      . Difference in risk behaviors and STD prevalence between street-based and establishment-based FSWs in Guangdong Province, China. AIDS Behav 2012;16:943–51.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. van Veen MG,
      2. Götz HM,
      3. van Leeuwen PA, et al
      . HIV and sexual risk behavior among commercial sex workers in the Netherlands. Arch Sex Behav 2010;39:714–23.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Day S,
      2. Ward H
      . Approaching health through the prism of stigma: research in seven European countries. In: Day S, Ward H, eds. Sex work, mobilityand health in Europe. London: Kegan Paul Limited, 2004:139–61.
    10. ↵
      1. Stockman JK,
      2. Strathdee SA
      . HIV among people who use drugs: a global perspective of populations at risk. J Acquir Immune Defic Syndr 2010:55(Suppl 1):S17–22.
      OpenUrlCrossRefPubMed
    11. ↵
      1. Strathdee SA,
      2. Lozada R,
      3. Martinez G, et al
      . Social and structural factors associated with HIV infection among female sex workers who inject drugs in the Mexico-US border region. PLoS ONE 2011;6:e19048.
      OpenUrlCrossRefPubMed
    12. ↵
      1. Harcourt C,
      2. Donovan B
      . The many faces of sex work. Sex Transm Infect 2005;81:201–6.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Yi H,
      2. Mantell JE,
      3. Wu R, et al
      . A profile of HIV risk factors in the context of sex work environments among migrant female sex workers in Beijing, China. Psychol Health Med 2010;15:172–87.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Scorgie F,
      2. Nakato D,
      3. Harper E, et al
      . ‘We are despised in the hospitals’: sex workers’ experiences of accessing health care in four African countries. Cult Health Sex 2013;15:450–65.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Shannon K,
      2. Goldenberg SM,
      3. Deering KN, et al
      . HIV infection among female sex workers in concentrated and high prevalence epidemics: why a structural determinants framework is needed. Curr Opin HIV AIDS 2014;9:174–82.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      ECDC/WHO Regional Office for Europe. HIV/AIDS surveillance in Europe 2012. Stockholm: ECDC, 2013.
    17. ↵
      INSA. Infecção VIH/SIDA: a situação em Portugal a 31 de Dezembro de 2011. Lisboa: INSA, 2012.
    18. ↵
      1. Verster A,
      2. Davoli M,
      3. Camposeragna A, et al
      . Prevalence of HIV infection and risk behavior among street prostitutes in Rome 1997–1998. AIDS Care 2001;13:367–2.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Belza MJ
      . Risk of HIV infection among male sex workers in Spain. Sex Transm Infect 2005;81:85–8.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Grath-Lone L,
      2. Marsh K,
      3. Hughes G, et al
      . The sexual health of male sex workers in England: analysis of cross-sectional data from genitourinary medicine clinics. Sex Transm Infect 2014;90:38–40.
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. El-Bassel N,
      2. Terlikbaeva A,
      3. Pinkham S
      . HIV and women who use drugs: double neglect, double risk. Lancet 2010;376:312–4.
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Uuskula A,
      2. Fischer K,
      3. Raudne R, et al
      . A study on HIV and hepatitis C virus among commercial sex workers in Tallinn. Sex Transm Infect 2008;84:189–1.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Deering KN,
      2. Amin A,
      3. Shoveller J, et al
      . A systematic review of the correlates of violence against sex workers. Am J Public Health 2014;104:e42–54.
      OpenUrl
    24. ↵
      1. Shannon K,
      2. Csete J
      . Violence, condom negotiation, and HIV/STI risk among sex workers. JAMA 2010;304:573–4.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Del Romero J,
      2. Castilla J,
      3. Hernando V, et al
      . Combined antiretroviral treatment and heterosexual transmission of HIV-1: cross sectional and prospective cohort study. BMJ 2010;340:c2205.
      OpenUrlAbstract/FREE Full Text
    26. ↵
      ECDC. Implementing the Dublin Declaration on Partnership to Fight HIV/AIDS in Europe and Central Asia: 2010 progress report. Stockholm: ECDC, 2010.
    27. ↵
      1. Deblonde J,
      2. De Koker P,
      3. Hamers FF, et al
      . Barriers to HIV testing in Europe: a systematic review. Eur J Public Health 2010;20:422–32.
      OpenUrlAbstract/FREE Full Text
    28. ↵
      1. Dias S,
      2. Severo M,
      3. Gama A, et al
      . Factors associated with HIV testing among migrants in Portugal. Int J Public Health 2011;56:559–66.
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Platt L,
      2. Grenfell P,
      3. Bonell C, et al
      . Risk of sexually transmitted infections and violence among indoor-working female sex workers in London: the effect of migration from Eastern Europe. Sex Transm Infect 2011;87:377–84.
      OpenUrlAbstract/FREE Full Text
    30. ↵
      1. Shahmanesh M,
      2. Patel V,
      3. Mabey D, et al
      . Effectiveness of interventions for the prevention of HIV and other sexually transmitted infections in female sex workers in resource poor setting: a systematic review. Trop Med Int Health 2008;13:659–79.
      OpenUrlCrossRefPubMedWeb of Science
    View Abstract

    Supplementary materials

    • Abstract in Portuguese

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • Handling editor Jackie A Cassell

    • Contributors SD and LM conceived the study; SD initiated and coordinated all aspects of the study; AG performed the statistical analysis and participated in the interpretation of the results; RF and LM were involved in conducting data collection and collaborated in the interpretation of the results; HB was involved in the study conception. All authors contributed to writing and revising the manuscript.

    • Funding This study was cofinanced by the ADIS/SIDA Program from the National Programme for the Prevention and Control of HIV/AIDS Infection.

    • Competing interests None.

    • Ethics approval Ethics Committee for Health of North Regional Health Administration.

    • Provenance and peer review Not commissioned; externally peer reviewed.