Article Text

Download PDFPDF

Structural barriers to condom access in a community-based cohort of sex workers in Vancouver, Canada: influence of policing, violence and end-demand criminalisation
  1. Shira Goldenberg1,2,
  2. Ruchi Liyanage2,
  3. Melissa Braschel2,
  4. Kate Shannon2,3
  1. 1 Faculty of Health Sciences, Simon Fraser University, Burnaby, British Columbia, Canada
  2. 2 Centre for Gender and Sexual Health Equity, Vancouver, British Columbia, Canada
  3. 3 Department of Medicine, The University of British Columbia, Vancouver, British Columbia, Canada
  1. Correspondence to Dr Shira Goldenberg, Faculty of Health Sciences, Simon Fraser University, Burnaby, BC V5A 1S6, Canada; dr.goldenberg{at}


Objectives Sex workers (SWs) face a disproportionate burden of HIV/sexually transmitted infections (STIs), violence and other human rights violations. While recent HIV prevention research has largely focused on the HIV cascade, condoms remain a cornerstone of HIV prevention, requiring further research attention. Given serious concerns regarding barriers to condom use, including policing, violence and ‘end-demand’ sex work criminalisation, we evaluated structural correlates of difficulty accessing condoms among SWs in Vancouver over an 8-year period.

Methods Baseline and prospective data were drawn from a community-based cohort of women SWs (2010–2018). SWs completed semi-annual questionnaires administered by a team that included lived experience (SWs). Multivariable logistic regression using generalised estimating equations (GEE) modelled correlates of difficulty accessing condoms over time.

Results Among 884 participants, 19.1% reported difficulty accessing condoms during the study. In multivariable GEE analysis, exposure to end-demand legislation was not associated with improved condom access; identifying as a sexual/gender minority (adjusted odds ratio (aOR) 1.62, 95% CI 1.16 to 2.27), servicing outdoors (aOR 1.52, 95% CI 1.17 to 1.97), physical/sexual workplace violence (aOR 1.98, 95% CI 1.44 to 2.72), community violence (aOR 1.79, 95% CI 1.27 to 2.52) and police harassment (aOR 1.66, 95% CI 1.24 to 2.24) were associated with enhanced difficulty accessing condoms.

Conclusions One-fifth of SWs faced challenges accessing condoms, suggesting the need to scale-up SW-tailored HIV/STI prevention. Despite the purported goal of ‘protecting communities’, end-demand criminalisation did not mitigate barriers to condom access, while sexual/gender minorities and those facing workplace violence, harassment or those who worked outdoors experienced poorest condom accessibility. Decriminalisation of sex work is needed to support SWs’ labour rights, including access to HIV/STI prevention supplies.

  • condom
  • sexually transmitted infections

Statistics from

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.

Key messages

  • In this longitudinal study, one-fifth of women sex workers reported difficulty accessing condoms over an 8-year period.

  • Identifying as LGBTQ2S, working in outdoor/public spaces, workplace and community violence, and police harassment were associated with difficulty accessing condoms.

  • Exposure to ‘end-demand’ sex work legislation was not associated with improvements in condom access.

  • Sex worker-friendly HIV/STI programmes, including uninterrupted access to low/no-cost condoms, remain needed to address current gaps in HIV/STI prevention faced by sex workers.

  • Structural interventions including decriminalisation and safer work environment models are recommended to improve condom access.


Women sex workers (SWs) face enhanced health and social inequities, including a disproportionate burden of HIV1–3 and sexually transmitted infections (STIs).3 4 These inequities are driven by high rates of criminalisation, violence and other human rights violations3 5 and vary across work environments and epidemic settings.1 3 Whereas research has increasingly focused on the HIV cascade6 and emerging biomedical interventions (eg, pre-exposure prophylaxis), condoms remain a critical, cost-effective cornerstone of HIV/STI prevention and an ongoing human rights and HIV/STI prevention priority for SWs.7

While numerous studies have reported on patterns of condom use among SWs and their clients,8–10 few have focused on condom accessibility, despite its importance within a rights-based response to HIV and evidence of suboptimal access to health and social services for SWs.11 12 Research has linked barriers to consistent condom use to individual-level factors including younger age13 and drug use,14 as well as structural factors such as migration/mobility,8 sexual/gender minority status,15 lower educational attainment,13 and socioeconomic status.3 16 A 2015 global review highlighted the central role of structural determinants in HIV/STI prevention in sex work,3 with criminalisation and policing,9 17 unsafe work environments,18 and availability and accessibility of healthcare8 10 all strongly linked to sexual risk negotiation and the burden of HIV/STIs.

In recent years, a number of countries have implemented or considered ‘end-demand’ criminalisation, which criminalises purchase of sexual services rather than their sale.19 In Canada, until 2013, sex work was criminalised through provisions against keeping a bawdy house (brothel), living off the avails of prostitution, and public communication for the purposes of prostitution. Based on substantial evidence that these laws violated SWs’ rights, they were struck down by the Supreme Court in 2013. However, these laws were replaced with ‘end-demand’ legislation (ie, the Protection of Communities and Exploited Persons Act, 2014),2 which left the sale of sexual services legal while criminalising their purchase, as well as many third-party activities (eg, advertising). This new legislation conflates sex work with trafficking19 and was enacted despite concerns that it could recreate or worsen harms faced under previous laws.18 20 Research has shown that laws and enforcement targeting the purchase of sex may encourage rushed sexual transactions to avoid police; increase police surveillance; cause displacement to new work environments; increase pressure to see more dangerous clients; impede access to third-party supports; and reduce access to healthcare and community-based supports.19 21–23

In Vancouver, Canada, high rates of policing and workplace violence have been linked to barriers to sexual negotiation9 14 and a disproportionate HIV/STI burden among SWs;3 24 however, robust quantitative evidence on condom access among SWs across diverse work environments remains limited, particularly post-implementation of ‘end-demand’ law reform. As such, this study aimed to evaluate structural correlates of difficulty accessing male condoms among SWs in Vancouver over an 8-year period, including working conditions, policing, and impacts of ‘end-demand’ sex work criminalisation.


This study was based on An Evaluation of Sex Workers’ Health Access (AESHA), an ongoing open community-based prospective cohort of women SWs in Metropolitan Vancouver.20 Reporting conforms to the Strengthening the Reporting of Observational Studies guidelines.25

Patient and public involvement

AESHA is based on deep community collaborations since inception, and is guided by a Community Advisory Board of >15 sex work, HIV and women’s organisations. Recruitment and baseline and semi-annual questionnaires are conducted by a community-based team which prioritises staff with lived experience (current/former SWs). All interviewers are highly trained in maintaining rapport and non-judgmental interactions to ensure a safe, trauma-informed and non-stigmatising approach.

Study design

As previously described,20 participants included cisgender and transgender women aged ≥14 years engaged in sex work within the past month. Participants were recruited through time-location sampling during outreach at various times during the day and evening to diverse indoor and outdoor venues (eg, street, parks, massage parlours, bars, hotels, housing) and online. Venues were identified by mapping with current/former SWs;20 online recruitment was via postings on advertising sites and other SW online venues. Following informed consent, participants completed study visits at the study office or confidential location of their choosing.

Cohort procedures included regular follow-up to work environments by a community-based team, in close collaboration with partner organisations. Participants provided updated contact information at each visit to support follow-up. Between 10% and 15% of individuals screened were deemed ineligible (eg, not engaged in sex work at baseline; live outside Metropoliton Vancouver; unable to give informed consent). Annual retention of participants under active follow-up was >90%; primary reasons for attrition included mortality and migration. Extensive efforts were made to continue to follow participants who move, including phone interviews. Mortality was tracked using linkages to provincial Vital Statistics registries. Our open cohort design allowed us to maintain a sample size of approximately 800 participants. Power calculations for the larger study suggested that this is sufficient to detect associations between structural exposures and HIV/STI incidence outcomes.

Participants received CAD $40 at each visit for their time, expertise and travel. Procedures were approved by the Providence Health Care/University of British Columbia and Simon Fraser University Research Ethics Boards.


The questionnaire included detailed questions regarding demographics, sex work patterns, condom use, drug use, work environment, criminalisation, and access to health supplies and services. At each visit, participants received pre/post-test counselling and voluntary serological testing for HIV, hepatitis C and STIs by the project nurse. Women with symptomatic STIs are provided with treatment on site. Free serology and Papanicolaou testing is offered regardless of enrolment.

Our dependent variable was a time-updated measure of experiencing difficulty accessing condoms in the last 6 months. At baseline and each follow-up visit, participants were asked: “In the last 6 months, have you had any difficulty accessing condoms while working?” Responses were coded as yes or no.

Independent variables were based on known and hypothesised individual and structural variables associated with condom use, condom access, or healthcare access among SWs. Time-fixed sociodemographics included identifying as a sexual/gender minority (ie, lesbian, gay, bisexual, asexual, trans, queer, or two-spirit (a term used by some indigenous people to describe their sexual, gender and/or spiritual identity)), indigenous ancestry, and being Canadian-born (vs born outside of Canada); age was time-updated at each study visit. All other time-updated variables were based on a 6-month recall period and included: recent homelessness; place of service (primarily street/public vs formal/informal indoor establishment); and physical/sexual workplace violence by aggressors posing as clients (‘yes’ to abducted/kidnapped, attempted sexual assault/raped, strangled, physically assaulted/beaten, locked/trapped in car, thrown out of moving car, assaulted with weapon). Harassment by community residents or businesses while working included verbal harassment/threats, physical assault, or NIMBYs (ie, not in my back yard’, referring to local residents who oppose the presence/operation of SWs). Exposure to ‘end-demand’ criminalisation was evaluated based on whether an interview took place before (2010–2013) or after (April 2015-onwards) implementation of the new legislation. Because this legislation was introduced in 2014, we excluded this year from analysis due to variation in how the laws may have been enforced, as well as the first 3 months of 2015 to account for measures referring to the previous 6 months. This measure was selected since the new laws primarily target clients and third parties rather than SWs, alongside evidence that SWs’ health outcomes and access are often greatly shaped by fear/anxiety related to SW laws (eg, avoidance of accessing services or working in certain venues/areas due to fear of criminalisation).23 26 27 Questions on criminalisation and policing included experiences of police harassment without arrest (eg, threatened with arrest/detainment/fine, verbally harassed, physically assaulted, propositioned to exchange sex) and police arrest. Finally, health and social services access questions included access to health services when needed (yes vs no), utilisation of STI and HIV testing, and utilisation of SW outreach programs (eg, SW-tailored outreach/drop-in services, outreach by public health nurses).

Statistical analysis

Analysis was restricted to participants enrolled between January 2010 and February 2018, and observations where participants reported SW within the last 6 months. Interviews completed during the legislation transition period (January 2014 to March 2015) were excluded. A complete case approach was used; visits with missing data were excluded. Descriptive statistics were derived at baseline to generate frequencies and proportions for categorical data, and medians and interquartile ranges (IQRs) for continuous data. Differences between groups were assessed using the Wilcoxon rank sum test for continuous variables and Pearson’s Chi-square test (or Fisher’s exact test) for categorical variables. Logistic regression with generalised estimating equations (GEE) and an exchangeable correlation structure measured associations between independent variables of interest and difficulty accessing condoms in the last 6 months. Bivariate and multivariable GEE analyses included baseline and all follow-up data during the study, accounting for repeated measures among participants.28 Variables were considered for multivariable analysis if they were significant at p<0.10 in bivariate analysis. The multivariable model with the best overall fit, indicated by the lowest quasi-likelihood under the independence model criterion, was determined using a manual backward selection process.29 SAS v9.4 (SAS, Cary, NC, USA) was used. All p values are two-sided.


Of 884 participants, at baseline 9.6% experienced difficulty accessing condoms in the last 6 months, and 19.1% experienced difficulty accessing condoms during the 8-year study. Of 3626 observations, 264 events of difficulty accessing condoms were reported. Participants completed a median of three study visits (IQR 1–6) for a total of 2307 person-years of follow-up. Among women facing difficulty accessing condoms at baseline, 83.5% sometimes/occasionally faced difficulty accessing condoms, whereas 10.6% usually/always faced difficulty. Some 58.9% accessed condoms from mobile outreach, who received a median of 10 (IQR 6–40) condoms on each outreach contact. Participants reported carrying a median of 6 (IQR 4–10) condoms per shift. The median age was 35 (IQR 28–42) years, 33.0% identified as LGBTQ2S (lesbian, gay, bisexual, transgender, transsexual, queer, questioning, and two-spirit), 38.6% were of indigenous ancestry, and 71.3% were Canadian-born (table 1).

Table 1

Demographic and structural variables stratified by difficulty accessing condoms among sex workers (n=884) in Metropolitan Vancouver, BC, Canada at baseline, 2010–2018

In bivariate GEE analysis, younger age, identifying as LGBTQ2S, and homelessness were correlated with greater difficulty accessing condoms (table 2). Women servicing clients in outdoor/public spaces had higher odds of reporting difficulty accessing condoms than those in informal indoor or in-call spaces (eg, hotels, bars, massage parlours, micro-brothels). Although exposure to sex work law reform was negatively associated with difficulty accessing condoms (OR 0.52, 95% CI 0.37 to 0.73), other structural measures of workplace violence, criminalisation, and policing were strongly associated with enhanced difficulty accessing condoms, including recent physical/sexual workplace violence (OR 2.87, 95% CI 2.14 to 3.85), harassment by community residents/businesses (OR 2.57, 95% CI 1.84 to 3.58), police harassment without arrest (OR 2.27, 95% CI 1.75 to 2.94) and police arrest (OR 2.86, 95% CI 1.77 to 4.64).

Table 2

Results from bivariate and multivariable generalised estimating equations (GEE) analysis of factors associated with difficulty accessing condoms among sex workers (n=884) in Metropolitan Vancouver, BC, Canada, 2010–2018

In multivariable GEE analysis (table 2), exposure to sex work law reform was not associated with changes in condom access and was not retained in the best fitting model; identifying as LGBTQ2S (adjusted odds ratio (aOR) 1.62, 95% CI 1.16 to 2.27), servicing clients in outdoor/public spaces (aOR 1.52, 95% CI 1.17 to 1.97), physical/sexual workplace violence (aOR: 1.98, 95% CI 1.44 to 2.72), harassment by community residents/businesses (aOR 1.79, 95% CI 1.27 to 2.52) and police harassment without arrest (aOR 1.66, 95% CI 1.24 to 2.24) were significantly correlated with difficulty accessing condoms over time.


This 8-year study identified significant unmet need for condoms among SWs in Metropolitan Vancouver, with one-fifth reporting persistent challenges accessing condoms. End-demand criminalisation did not mitigate barriers to condom access, and sexual/gender minorities, SWs facing workplace violence and harassment by police and community, and those working outdoors experienced poorest condom accessibility. These findings indicate the need to scale-up SW-friendly and rights-based HIV/STI prevention efforts, including structural interventions (eg, decriminalisation, mobile outreach).

Despite the purported goal of ‘protecting communities’, we found no significant difference in condom access following implementation of ‘end-demand’ criminalisation after adjustment for other variables, suggesting that such legislation may perpetuate barriers faced under previous legislation. Whereas ‘end-demand’ legislation typically aims to combat exploitation and support linkage to support services, evidence suggests that such legislation may exacerbate harms for SWs – including barriers to health services and HIV/STI prevention, violence, and poor working conditions.18 19 21 22

Research has linked barriers to condoms to health service delivery features, including geography, cost, stigma, and limited convenient, low-barrier distribution outlets.16 Our findings move beyond this work to highlight the roles of criminalisation and violence in shaping condom access for marginalised women. In addition to the lack of improvements in condom access post-law reform, the associations between workplace violence and harassment, policing, and barriers to condoms indicate the urgent need to address criminalisation. The relationship between impeded access to HIV/STI prevention, violence, and criminalisation has been documented previously.16 30 Criminalisation can undermine health and human rights by encouraging rushed transactions and pushing SWs to hidden areas where outreach or peer supports may be limited.9 SWs and third parties (eg, managers) often avoid carrying or storing sufficient condoms due to fear that this will be used as evidence of criminalised activities;27 31 given the continued criminalisation of third-party activities under ‘end-demand’, such concerns are likely to persist.22 Additionally, SWs may avoid interacting with health or outreach services due to stigma and fear of legal ramifications.20 23 27 31 Decriminalisation would enable SWs and workplaces to carry, store and use HIV/STI prevention supplies without fear of criminalisation. Scale-up of safer workplace interventions (eg, indoor settings with onsite condoms, security) and peer-led mobile outreach are also recommended to support immediate occupational health.3 32

Where sex work is uniquely stigmatised and singled out from other occupations, SWs’ use of public space becomes restricted by community, which can exacerbate stigma, perpetuate barriers to health and safety, and increase adversarial police interactions due to community complaints.12 15 22 The barriers faced by sexual/gender minorities may relate to intersectional stigmas related to sex work, sexual orientation, and gender identity. There remain unacceptable gaps in gender-sensitive HIV prevention and related health services for sexual/gender minorities, which may be linked to the high levels of stigma, violence, displacement, and unmet health needs documented among this population.12 15 33 Gender-affirming and inclusive policies in HIV/STI prevention services and scale-up of targeted interventions (eg, outreach, drop-in services) that address the needs of sexual/gender minorities remain needed.

Strengths and weaknesses

Strengths of this study include its longitudinal nature, strong community collaborations, and a large, diverse sample. As with most research involving stigmatised populations, there is potential for under-reporting of self-reported risks; our community-based and experiential team, training in non-stigmatising interview techniques, and community collaborations are designed to mitigate this. This study was designed to understand structural determinants of HIV/STI prevention and care; future research is needed to design and test the most effective interventions in this context.


In this 8-year study, one-fifth of women SWs reported difficulty accessing condoms. In multivariable GEE analysis, ‘end-demand’ criminalisation was not associated with improvements in condom access, whereas identifying as LGBTQ2S, working in outdoor/public spaces, workplace violence, community harassment, and police harassment were associated with higher odds of difficulty accessing condoms. SW-friendly HIV/STI programmes, including uninterrupted access to low/no-cost condoms, remain needed to address current gaps in HIV/STI prevention. Structural interventions including decriminalisation and safer work environment models are recommended.


The authors thank all those who contributed their time and expertise to this project, particularly participants, AESHA community advisory board members and partner agencies, and the AESHA team, including: Sarah Moreheart, Shannon Bundock, Brittney Udall, Jennifer Morris, Jennifer McDermid, Alka Murphy, Sylvia Machat, Minshu Mo, Sherry Wu, Emily Leake, Bridget Simpson, Gail Madanayake, Zoe Hassall, Kate Noyes, Emma Kuntz, Akanée Yamaki, Anna Mathen, Peter Vann, Megan Bobetsis and Colette Ryan. They also thank Sarah Watt for her research assistance. This research is supported by the US National Institutes of Health (R01DA028648), a Canadian Institutes of Health Research Foundation Grant, a Canadian Institutes of Health Research Bridge Grant (PTJ-153423) and MacAIDS. SG is partially supported by NIH and a CIHR New Investigator Award. KS is partially supported by a Canada Research Chair in Global Sexual Health, NIH, and HIV/AIDS and Michael Smith Foundation for Health Research.



  • Contributors KS and SG had full access to all the study data and take full responsibility for the integrity of the data and the accuracy of the data analysis. SG and KS made substantial contributions to the study conception and design. KS and SG made substantial contributions to the data acquisition. MB conducted the statistical analysis. SG and RL drafted the manuscript. KS and SG made substantial contributions to the interpretation of the data and revised the article critically for important intellectual content. All authors have approved the final version to be published.

  • Funding This research was supported by operating grants from the US National Institutes of Health (R01DA028648) and Canadian Institutes of Health Research (HHP-98835, PTJ-153423) and MacAIDS. KS is partially supported by a Canada Research Chair in Global Sexual Health and HIV/AIDS and Michael Smith Foundation for Health Research.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Patient consent for publication Not required.

  • Ethics approval Approval provided by the Providence Health Care/University of British Columbia and Simon Fraser University Research Ethics Boards. REB number H09-02803.

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

  • Data availability statement Data are available upon reasonable request. Due to our ethical and legal requirements related to protecting participant privacy and current ethical institutional approvals, all relevant data are available upon request pending ethical approval. Please submit all requests to initiate the data access process to the corresponding author.