Abstract
Objectives. To model the relationship of unstable housing and evictions with physical and sexual violence perpetrated against women sex workers in intimate and workplace settings.
Methods. We used bivariate and multivariable logistic regression with generalized estimating equations to model the association of unstable housing exposure and evictions with intimate partner violence (IPV) and workplace violence among a community-based longitudinal cohort of cisgender and transgender women sex workers in Vancouver, Canada, from 2010 through 2019.
Results. Of 946 women, 85.9% experienced unstable housing, 11.1% eviction, 26.2% IPV, and 31.8% workplace violence. In multivariable generalized estimating equation models, recent exposure to unstable housing (adjusted odds ratio [AOR] = 2.04; 95% confidence interval [CI] = 1.45, 2.87) and evictions (AOR = 2.45; 95% CI = 0.99, 6.07) were associated with IPV, and exposure to unstable housing was associated with workplace violence (AOR = 1.46; 95% CI = 1.06, 2.00).
Conclusions. Women sex workers face a high burden of unstable housing and evictions, which are linked to increased odds of intimate partner and workplace violence. Increased access to safe, women-centered, and nondiscriminatory housing is urgently needed. (Am J Public Health. 2023;113(4):442–452. https://doi.org/10.2105/AJPH.2022.307207)
Housing is a critical social determinant of health and well-being and has been described as a key pathway through which health inequities emerge and are sustained over time.1 In recent years, Vancouver, Canada, like many other places in North America, has experienced an intensifying housing crisis. Community concerns regarding affordable and appropriate housing have continued to intensify, as low-income family housing, single-room occupancy hotels, and social housing continue to be upscaled or replaced by new housing and retail units.
Women sex workers often face intense marginalization stemming from criminalization, stigma and discrimination, and limited labor protections. These are all structural factors that have been strongly associated with an elevated burden of violence among sex workers, including both occupational and intimate partner violence (IPV).2,3 The estimated prevalence of workplace violence (i.e., physical or sexual violence perpetuated against sex workers) in the past 6 months is 50% to 70% for physical violence and 80% for sexual violence.4 In regard to IPV, an estimated 60% of women sex workers have experienced physical violence and 40% have experienced sexual violence over their lifetimes.4
In comparison with the general population, sex workers face disproportionate risks related to housing instability and evictions because of structural factors, including pervasive criminalization and stigma.5,6 Despite growing concerns regarding the negative health and social consequences of unaffordable housing and, in particular, evictions amid the COVID-19 pandemic,7,8 epidemiologic data are lacking on the role of housing instability and evictions as structural determinants of violence across both intimate partner and occupational contexts among sex workers. This information is needed to inform housing-centered interventions for marginalized women. Previous literature suggests that women experiencing IPV may be more likely to face housing precarity.9–13 Research from Baltimore, Maryland,11 and Vancouver12 documents associations of homelessness with younger age, drug use, sexual IPV, client volume, and working in public or outdoor workplaces.
Research on housing and marginalized women’s health has focused mainly on homelessness, which is the most visible and commonly studied form of housing precarity and, more narrowly, refers to experiences of being unhoused or unsheltered. However, less attention has been paid to the needs of marginalized women affected by other forms of housing precarity, such as evictions and housing instability. We define “housing instability” as requiring temporary or marginal overnight sleeping arrangements—such as in a shelter, a hotel, supportive housing, a car, or a recovery house; on the street; with family or friends; or couch surfing—that could lead to loss of housing and eventual homelessness. We operationalize “eviction” as changing one’s place of living or sleeping owing to being evicted or forced to move by one’s landlord or housing operator. Research has found that some types of sleeping arrangements (e.g., couch surfing) among those experiencing homelessness are more common for youths and are associated with deleterious mental and physical health outcomes.14,15
Previous qualitative literature has described housing as a critical determinant shaping marginalized women’s health and safety. Women who use drugs and experience poverty report frequent exposure to rental discrimination, gender-based violence, exploitation, and lack of security and privacy in traditional co-ed low-income housing settings (e.g., single-room occupancy hotels). These inequities may be heightened for transgender women.16,17 Furthermore, the criminalization of aspects of sex work can result in rental discrimination, such as proprietors being less likely to rent to and more likely to evict sex workers.6 Additionally, in the absence of a safe place to take dates, sexual transactions may be relegated to unsafe or isolated settings where sex workers have few protections from occupational violence.6,18 However, few epidemiological studies have analyzed housing instability or evictions in relation to health and safety outcomes, particularly experiences of violence, among women sex workers.19,20 Such research remains critically needed given previous work that has highlighted strong links between IPV and housing precarity among women in the general population and that shows that women experiencing violence are often forced to rely on provisional, overcrowded, or unaffordable housing.21
Building on previous research indicating a high burden of homelessness12 and of violence22,23 in intimate partner and occupational settings among women sex workers, we assessed the independent association of exposure to unstable housing and evictions with IPV and occupational violence among women sex workers in Vancouver, Canada, over a 9-year study period (2010–2019).
METHODS
We collected data from An Evaluation of Sex Workers Health Access (AESHA), an open longitudinal community-based cohort of sex workers in Vancouver, Canada. AESHA is overseen by a community advisory board of 15 or more community agencies. Eligibility criteria were self-identifying as a woman (cisgender and transgender inclusive), being aged 14 years or older, having exchanged sex for money in the past 30 days in Metro Vancouver, and being able to provide informed consent. We recruited participants through time–space sampling, as described previously,24 across various workplaces (e.g., streets, indoor venues) and online. After providing written informed consent, participants completed baseline and biannual study visits composed of an interviewer-administered questionnaire and serological testing. Visits took place at participants’ location of choice (i.e., study office or women’s homes or workplaces) and were conducted by experienced (current or former sex workers) and community-based interviewers and clinical staff. We used Biolytical INSTI (Biolytical Laboratories, Inc., Richmond, BC) rapid tests for HIV screening, with confirmation of reactive tests by blood draw for Western blot. We collected urine samples to test for gonorrhea and chlamydia, and drew blood to test for syphilis, HSV-2 (herpes simplex virus 2) antibody, and HCV (hepatitis C). Project nurses provided pre- and posttest counseling as well as treatment of symptomatic sexually transmitted infections onsite and offered referrals for HIV care to participants living with HIV. Serology and Papanicolaou testing were available to community members regardless of study participation. Participants received an honorarium of 40.00 CAD (Canadian dollars) per study visit.
Measures
We assessed demographics and other lifetime exposures using baseline data only; for other study variables (e.g., housing, violence, substance use), we used time-updated data, examining events with recall periods in the past 6 months at baseline and each biannual study visit. Except for age, income, and place of solicitation, all study variables were binary.
Housing exposure variables
We defined “unstable housing” with participants’ response to the question “In which of the following types of places have you slept overnight in the last six months?” We coded having temporary or marginal housing experiences in the past 6 months (e.g., street, shelter, hostel, hotel, supportive housing, car, family or friends, couch surfing, recovery house) as “yes.” We coded permanent housing (e.g., living in apartment or house on own or with others) as “no.” We evaluated evictions with participants’ response to the question “Have you changed the main place where you live or sleep overnight in the last 6 months? If yes, why?” We coded women who changed their place of living or sleeping because of being evicted or forced to move in the past 6 months as “yes.” Those who did not change their place of living or sleeping or who did so for reasons other than evictions made up the comparison group. We measured unstable housing across the duration of the study (January 2010–February 2019), and we collected data on evictions from September 2014 through February 2019.
Violence outcomes
As with previous research,22 we defined IPV as moderate to severe physical or sexual violence perpetrated by any male intimate partners in the past 6 months, as measured on the World Health Organization Intimate Partner Violence Scale.25 We evaluated IPV based on participants’ response to the question “Have you ever experienced any of the following by your intimate male partners (boyfriends, spouse) in the past six months?” Types of IPV were “moderate physical IPV” (e.g., being slapped, thrown at, pushed, or shoved); “severe physical IPV” (e.g., being hit, kicked, dragged, beaten, choked, or burned); and “sexual IPV” (e.g., being forced to have sex against one’s will). We considered women who responded yes to any of the measures of violence to have experienced IPV.
Workplace violence consisted of physical or sexual violence perpetrated by aggressors posing as clients. We evaluated workplace violence based on participants’ response to the question “Have you experienced any of the following bad dates or experienced violence by clients in the past six months?” We considered women who reported any experiences of physical or sexual violence perpetrated by aggressors posing as clients (e.g., abduction or kidnap, sexual or physical assault or rape, strangulation, or being locked or trapped in a car or room) to have experienced workplace violence.
Potential confounders
Time-fixed (i.e., lifetime) covariates assessed only at baseline were age (in years), educational attainment (whether graduated from high school), Indigenous ancestry (First Nations, Metis, or Inuit ancestry), whether immigrant to Canada (i.e., born outside Canada), sexual orientation (gay, lesbian, bisexual, two spirit, asexual, or queer vs straight), gender identity (transgender woman, transexual woman, or other transfeminine identity vs cisgender woman), whether ever diagnosed with a mental health condition (yes vs no), and whether experienced childhood trauma (i.e., physical or sexual assault before 18 years old).
Time-updated confounders included noninjection (excluding alcohol and cannabis) and injection drug use, inconsistent condom use with clients, client condom refusal, being forced to have sex against will with clients, having experienced sexual assault (by anyone other than intimate partners or clients), and having experienced violence from community members (verbal harassment or physical violence by community residents or business owners). Time-updated structural factors were average monthly income from all sources (in CAD), having access to health care services when needed, always or usually having privacy where currently living, ever feeling in danger where currently sleeping, primary place of solicitation, having experienced police harassment while working (excluding arrest), and incarceration (i.e., in detention, prison, or jail overnight or longer).
We defined primary place of solicitation as 1 of 3 mutually exclusive categories: street or public space, indoor establishment (e.g., crack or drug house; bar, nightclub, or strip club; massage or beauty parlor; microbrothel; single-room occupancy hotel or supportive housing), and independent (e.g., escort agency, newspaper ads, online or telephone or texting, arranged by third party). Police harassment consisted of a range of harmful policing practices (e.g., being threatened with arrest, detainment, or fine; verbal harassment; physical assault; confiscation of harm-reduction materials; coercion into providing sexual favors).
Statistical Analyses
We restricted analyses to observations for which participants reported active engagement in sex work in the past 6 months at each study visit (2010–2019). We restricted models examining IPV to observations for which participants had at least 1 intimate male partner in the past 6 months. We further restricted analyses of evictions to participants interviewed after these questions were added (September 2014–February 2019).
We examined descriptive statistics at baseline, stratified by outcomes in the past 6 months at baseline (i.e., unstable housing and evictions). We used the Pearson χ2 test (or the Fisher exact test for small cell counts) for categorical variables and the Wilcoxon rank–sum test for continuous variables to test for significant differences between participants who had and those who had not experienced the outcome of interest. We conducted all bivariate and multivariable analyses using logistic regression with generalized estimating equations (GEE) and an exchangeable correlation matrix. We first used bivariate analyses to examine associations between hypothesized exposures and outcomes. To examine independent effects of housing exposures (i.e., unstable housing and evictions) on violence outcomes (i.e., IPV and workplace violence), we developed 4 separate multivariable GEE logistic regression confounder models using the procedure described by Maldonado and Greenland.26
We included key confounders that we identified in bivariate analyses and hypothesized to impact gender-based violence outcomes as potential confounders in full models. To determine the most parsimonious models, we removed potential confounders in a stepwise manner. We removed all potential confounding variables that altered the association of interest by less than 5% from models. We performed all analyses using SAS version 9.4 (SAS, Cary, NC). All P values were 2 sided.
RESULTS
Analyses of unstable housing included 4765 observations among 946 participants interviewed from January 2010 through February 2019. Over the study period, the majority (85.9%) of participants experienced unstable housing (3796 events), and 45.7% reported feeling in danger where they slept. Participants reported a high burden of violence over the 9-year study, with 26.2% experiencing IPV (451 events) and 31.8% experiencing workplace violence (552 events) in the past 6 months at any study visit. Analyses of evictions included 1891 observations among 550 participants interviewed from September 2014 through February 2019. Among participants who answered questions about evictions from September 2014 through February 2019, 11.1% experienced eviction (74 events) at least once over a 4.5-year period.
At baseline, the median income was $3000 CAD per month (interquartile range [IQR] = $1790–$5620), and 55.6% had graduated from high school (Table 1); 31.9% reported minority sexual orientation, and 38.1% were of Indigenous ancestry. Two thirds (66.2%) had used noninjection drugs and 40.7% had used injection drugs in the past 6 months. Almost half (48.7%) of participants reported soliciting on the street or in public, whereas 30.2% solicited services in indoor establishments, and 20.4% solicited services independently at baseline.
TABLE 1—
Baseline Characteristics of Women Sex Workers Who Experienced Unstable Housing in the Past 6 Months Compared With Those Who Did Not: Vancouver, Canada, 2010–2019
| Characteristic | Total (n = 946), No. (%) or Median (IQR) | Experienced Unstable Housing (n = 697), No. (%) or Median (IQR) | Did Not Experience Unstable Housing (n = 249), No. (%) or Median (IQR) | P |
| Lifetime variables | ||||
| Age, y | 35 (28–42) | 34 (27–42) | 37 (31–43) | < .001 |
| Minority sexual orientation | < .001 | |||
| Yes | 302 (31.9) | 260 (37.3) | 42 (16.9) | |
| No | 642 (67.9) | 436 (62.6) | 206 (82.7) | |
| Transgender woman | .026 | |||
| Yes | 58 (6.1) | 50 (7.2) | 8 (3.2) | |
| No | 886 (93.7) | 646 (92.7) | 240 (96.4) | |
| Indigenous | < .001 | |||
| Yes | 360 (38.1) | 327 (46.9) | 33 (13.3) | |
| No | 585 (61.8) | 370 (53.1) | 215 (86.4) | |
| Mental health diagnosis ever | < .001 | |||
| Yes | 466 (49.3) | 406 (58.3) | 60 (24.1) | |
| No | 480 (50.7) | 291 (41.8) | 189 (75.9) | |
| Experienced childhood trauma | < .001 | |||
| Yes | 635 (67.1) | 552 (79.2) | 83 (33.3) | |
| No | 38 (4.0) | 30 (4.3) | 8 (3.2) | |
| Graduated from high school | < .001 | |||
| Yes | 526 (55.6) | 333 (47.8) | 193 (77.5) | |
| No | 420 (44.4) | 364 (52.2) | 56 (22.5) | |
| Immigrated or migrated to Canada | < .001 | |||
| Yes | 277 (29.3) | 101 (14.5) | 176 (70.7) | |
| No | 668 (70.6) | 596 (85.5) | 72 (28.9) | |
| Recent variables (occurrences in past 6 mo) | ||||
| Experienced IPV | < .001 | |||
| Yes | 135 (14.3) | 129 (18.5) | 6 (2.4) | |
| No | 777 (3.6) | 535 (76.8) | 242 (97.2) | |
| Experienced workplace violence | < .001 | |||
| Yes | 168 (17.8) | 153 (22.0) | 15 (6.0) | |
| No | 764 (80.8) | 530 (76.0) | 234 (94.0) | |
| Noninjection drug use | < .001 | |||
| Yes | 627 (66.3) | 576 (82.6) | 51 (20.5) | |
| No | 317 (33.5) | 119 (17.1) | 198 (79.5) | |
| Injection drug use | < .001 | |||
| Yes | 384 (40.7) | 362 (52.1) | 22 (8.8) | |
| No | 561 (59.3) | 334 (47.9) | 227 (91.2) | |
| Inconsistent condom use | < .001 | |||
| Yes | 167 (17.7) | 153 (22.0) | 14 (5.6) | |
| No | 764 (80.8) | 532 (76.3) | 232 (93.2) | |
| Client condom refusal | < .001 | |||
| Yes | 196 (20.7) | 171 (24.5) | 25 (10.0) | |
| No | 719 (76.0) | 499 (71.6) | 220 (88.3) | |
| Forced into sex against will | < .001 | |||
| Yes | 67 (7.1) | 61 (8.8) | 6 (2.4) | |
| No | 864 (91.3) | 622 (89.2) | 242 (97.2) | |
| Experienced sexual assault | .003 | |||
| Yes | 23 (2.4) | 23 (3.3) | 0 (0.0) | |
| No | 902 (95.4) | 653 (93.7) | 249 (100.0) | |
| Experienced violence from community | < .001 | |||
| Yes | 153 (16.2) | 144 (20.7) | 9 (3.6) | |
| No | 791 (83.6) | 551 (79.1) | 240 (96.4) | |
| Has access to health services | .14 | |||
| Yes | 804 (85.0) | 600 (86.1) | 204 (81.9) | |
| No | 137 (14.5) | 94 (13.5) | 43 (17.3) | |
| Has privacy where they live | < .001 | |||
| Yes | 651 (68.8) | 448 (64.3) | 203 (81.5) | |
| No | 286 (30.2) | 242 (34.7) | 44 (17.7) | |
| Primary place of solicitation | < .001 | |||
| Street or public space | 461 (48.7) | 430 (61.7) | 31 (12.5) | |
| Indoor establishment | 286 (30.2) | 102 (14.6) | 184 (73.9) | |
| Independent | 193 (20.4) | 159 (22.8) | 34 (13.7) | |
| Experienced police harassment | < .001 | |||
| Yes | 295 (31.2) | 259 (37.2) | 36 (14.5) | |
| No | 649 (68.6) | 436 (62.6) | 213 (85.5) | |
| Monthly income, CAD | 3000 (1790–5620) | 3000 (1730–5910) | 3085 (2000–4800) | .7 |
Note. CAD = Canadian dollars; IPV = intimate partner violence; IQR = interquartile range.
At baseline, younger participants were more likely to have experienced unstable housing (34 years; IQR = 27–42 vs 37 years; IQR = 31–43) and evictions (35 years; IQR = 30–42 vs 40 years; IQR = 31–47) than were older participants (Tables 1 and 2). We also observed higher proportions of unstable housing and evictions among those identifying as gender and sexual minorities. Indigenous participants were more likely to have experienced unstable housing (46.9% vs 13.3%) and evictions (63.0% vs 42.8%) than were non-Indigenous women.
TABLE 2—
Baseline Characteristics of Women Sex Workers Who Experienced Eviction in the Past 6 Months Compared With Those Who Did Not: Vancouver, Canada, 2014–2019
| Characteristic | Total (n = 550), No. (%) or Median (IQR) | Evicted (n = 27), No. (%) or Median (IQR) | Not Evicted (n = 523), No. (%) or Median (IQR) | P |
| Lifetime variables | ||||
| Age, y | 39 (31–46) | 35 (30–42) | 40 (31–47) | .028 |
| Minority sexual orientation | .13 | |||
| Yes | 210 (38.2) | 14 (51.9) | 196 (37.5) | |
| No | 340 (61.8) | 13 (48.2) | 327 (62.5) | |
| Transgender woman | .5 | |||
| Yes | 47 (8.6) | 3 (11.1) | 44 (8.4) | |
| No | 503 (91.5) | 24 (88.9) | 479 (91.6) | |
| Indigenous | .041 | |||
| Yes | 241 (43.8) | 17 (63.0) | 224 (42.8) | |
| No | 308 (56.0) | 10 (37.0) | 298 (57.0) | |
| Mental health diagnosis ever | .07 | |||
| Yes | 315 (57.3) | 20 (74.1) | 295 (56.4) | |
| No | 235 (42.7) | 7 (25.9) | 228 (43.6) | |
| Experienced childhood trauma | .003 | |||
| Yes | 392 (71.3) | 27 (100) | 365 (69.8) | |
| No | 125 (22.7) | 0 (0) | 125 (23.9) | |
| Graduated from high school | .9 | |||
| Yes | 294 (53.5) | 15 (55.6) | 279 (53.4) | |
| No | 256 (46.6) | 12 (44.4) | 244 (46.7) | |
| Immigrated or migrated to Canada | .08 | |||
| Yes | 139 (25.3) | 3 (11.1) | 136 (26.0) | |
| No | 410 (74.6) | 24 (88.9) | 386 (73.8) | |
| Recent variables (occurrences in past 6 mo) | ||||
| Experienced IPV | .017 | |||
| Yes | 46 (8.4) | 6 (22.2) | 40 (7.7) | |
| No | 476 (86.6) | 19 (70.4) | 457 (87.4) | |
| Experienced workplace violence | .5 | |||
| Yes | 41 (7.5) | 3 (11.1) | 38 (7.3) | |
| No | 498 (90.6) | 24 (88.9) | 474 (90.6) | |
| Noninjection drug use | .006 | |||
| Yes | 329 (59.8) | 23 (85.2) | 306 (58.5) | |
| No | 220 (40.0) | 4 (14.8) | 216 (41.3) | |
| Injection drug use | .026 | |||
| Yes | 252 (45.8) | 18 (66.7) | 234 (44.7) | |
| No | 297 (54.0) | 9 (33.3) | 288 (55.1) | |
| Inconsistent condom use | .3 | |||
| Yes | 95 (17.3) | 7 (25.9) | 88 (16.8) | |
| No | 444 (80.7) | 20 (74.1) | 424 (81.1) | |
| Client condom refusal | .3 | |||
| Yes | 81 (14.7) | 6 (22.2) | 75 (14.3) | |
| No | 448 (81.5) | 20 (74.1) | 428 (81.8) | |
| Forced into sex against will | .1 | |||
| Yes | 24 (4.4) | 3 (11.1) | 21 (4.0) | |
| No | 515 (93.6) | 22 (81.5) | 493 (94.3) | |
| Experienced sexual assault | .12 | |||
| Yes | 12 (2.2) | 2 (7.4) | 10 (1.9) | |
| No | 527 (95.8) | 25 (92.6) | 502 (96.0) | |
| Experienced violence from community | .51 | |||
| Yes | 58 (10.6) | 4 (14.8) | 54 (10.3) | |
| No | 491 (89.3) | 23 (85.2) | 468 (89.5) | |
| Has access to health services | .09 | |||
| Yes | 470 (85.5) | 20 (74.1) | 450 (86.0) | |
| No | 80 (14.5) | 7 (25.9) | 73 (14.0) | |
| Has privacy where they live | .002 | |||
| Yes | 405 (73.6) | 13 (48.2) | 392 (75.0) | |
| No | 144 (26.2) | 14 (51.9) | 130 (24.9) | |
| Feels in danger where they sleep | .22 | |||
| Yes | 129 (23.5) | 9 (33.3) | 120 (22.9) | |
| No | 420 (76.4) | 18 (66.7) | 402 (76.9) | |
| Primary place of solicitation | .001 | |||
| Street or public space | 212 (38.6) | 19 (70.4) | 193 (36.9) | |
| Indoor establishment | 139 (25.3) | 1 (3.7) | 138 (26.4) | |
| Independent | 191 (34.7) | 7 (25.9) | 184 (35.2) | |
| Experienced police harassment | .1 | |||
| Yes | 51 (9.3) | 5 (18.5) | 46 (8.8) | |
| No | 496 (90.2) | 22 (81.5) | 474 (90.6) | |
| Ever been incarcerated | .02 | |||
| Yes | 34 (6.2) | 5 (18.5) | 29 (5.5) | |
| No | 516 (93.8) | 22 (81.5) | 494 (94.5) | |
| Monthly income, CAD | 2260 (1280–4000) | 2890 (1010–5840) | 2205 (1280–3890) | .3 |
Note. CAD = Canadian dollars; IPV = intimate partner violence; IQR = interquartile range.
Women who used noninjection (82.5% vs 20.5%) and injection (52.1% vs 8.8%) drugs were significantly more likely to experience unstable housing, which followed a similar pattern for evictions (85.2% vs 58.5% and 66.7% vs 44.7%, respectively). Women diagnosed with mental illness were more likely to experience unstable housing (58.3% vs 24.1%) and evictions (74.1% vs 56.4%). A higher proportion of women who solicited on the street or in public spaces experienced unstable housing (61.7% vs 12.5%) and evictions (70.4% vs 36.9%), a pattern that was reversed for women who solicited in indoor establishments (14.6% vs 73.9% for unstable housing and 3.7% vs 26.4% for evictions).
In separate multivariable GEE confounder models (Table 3) adjusted for confounders (e.g., childhood trauma, Indigenous ancestry, drug use), exposure to recent unstable housing was significantly correlated with elevated odds of both recent IPV (adjusted odds ratio [AOR] = 2.04; 95% CI = 1.45, 2.87) and workplace violence (AOR = 1.46; 95% CI = 1.06, 2.00) over the 9-year period (2014–2019). Additionally, exposure to recent eviction was associated with increased odds of recent IPV (AOR = 2.45; 95% CI = 0.99, 6.07) over a 4.5-year period (2014–2019) after adjustment for key confounders (e.g., age, childhood trauma abuse, income). Evictions were not significantly associated with workplace violence in GEE analyses.
TABLE 3—
Time-Updated Models of the Relationship Between Exposure to Housing Instability and Evictions and Violence Outcomes Among Women Sex Workers: Vancouver, Canada, 2010–2019
| Primary Exposures | Outcomes | OR (95% CI) | AOR (95% CI) |
| Model 1 | |||
| Experienced unstable housinga | Experienced IPVa | 1.74 (1.48, 2.03) | 2.04 (1.45, 2.87)b |
| No unstable housing | No IPV | 1 (Ref) | 1 (Ref) |
| Model 2 | |||
| Experienced unstable housinga | Experienced workplace violencea | 1.57 (1.37, 1.81) | 1.46 (1.06, 2.00)c |
| No unstable housing | No workplace violencea | 1 (Ref) | 1 (Ref) |
| Model 3 | |||
| Experienced evictiona | Experienced IPVa | 2.52 (1.35, 4.69) | 2.45 (0.99, 6.07)d |
| No eviction | No IPVa | 1 (Ref) | 1 (Ref) |
Note. AOR = adjusted odds ratio; CI = confidence interval; IPV = intimate partner violence; OR = odds ratio.
Time-updated measure refers to occurrences in the past 6 months at each biannual study visit.
Model 1 adjusted for childhood trauma; n = 612; 2010–2019.
Model 2 adjusted for age, Indigenous status, injection drug use,a mental health diagnosis, childhood trauma, and place of solicitationa; n = 898; 2010–2019.
Model 3 adjusted for age, childhood trauma, and average monthly incomea; n = 273; 2014–2019.
DISCUSSION
We have documented a high burden of unstable housing and evictions in a diverse cohort of women sex workers in Metro Vancouver, British Columbia, Canada. Housing instability was associated with increased odds of intimate partner and workplace violence, and evictions were further associated with increased odds of IPV. More than three quarters of sex workers faced housing insecurity, almost half reported feeling in danger where they slept, and 1 in 10 experienced eviction.
We also documented a high prevalence of gender-based violence, with more than one quarter of women reporting recent IPV and almost one third reporting recent workplace violence. These human rights violations highlight housing precarity as an important form of institutionalized discrimination faced by sex workers and highlight the need for urgent policy reforms—including decriminalization of sex work—to support access to safe, stable, and nondiscriminatory housing options for women sex workers. This is particularly important for reducing inequities in housing and health faced by marginalized subgroups that our analysis showed experienced a higher burden of unstable housing and evictions. These subgroups include sexual and gender minorities, youths, Indigenous women, women who use drugs, and women with mental health diagnoses.
Our study provides some of the first rigorous prospective cohort data regarding the relationship between housing and violence among women sex workers. Our findings are consistent with previous research demonstrating a high prevalence of violence among precariously housed women.27–29 Our findings provide new insights into the relationship between various indicators of housing precarity—housing instability and evictions—with physical and sexual violence across both intimate partner and occupational settings. These findings are particularly important, as rising gendered violence and housing precarity have become increasing concerns amid the COVID-19 pandemic.30
In Canada—as in other high-income contexts—many sex workers have been unable to access government benefits and supports afforded to other service workers during the pandemic,31 and temporary moratoria on rental evictions ended in August 2020. Many sex workers in British Columbia further reside in housing settings that were not covered by these moratoria (e.g., emergency shelters, transitional or recovery housing, some single-room occupancy hotels, and accommodations shared with property owners), contributing to housing precarity. In addition, many buildings in which sex workers live and work in Vancouver implemented “no-guest” policies because of COVID-19 concerns, which may have forced women to work in more dangerous settings or lose their housing.
Finally, women in our study who used both injection and noninjection drugs faced far higher proportions of unstable housing and evictions than those who did not use drugs. Previous work examining the relationship between housing and violence among women who use drugs has articulated the inadequacy of single-room occupancy hotels, which are among the most common affordable housing options for marginalized women who use substances in Vancouver.32,33 More research is needed on harm-reduction housing strategies as a potential means to reduce IPV. The scale-up of programs that provide women who use substances with affordable, safe, and dignified housing options is urgently needed.
Limitations and Strengths
Our study has several limitations of note. There is potential for information bias resulting from the use of self-reported measures; to mitigate this, trained interviewers with lived experiences and strong personal connections to the community conducted outreach and administered surveys. In addition, our measure of IPV captured experiences of violence perpetuated by only male partners and did not address IPV by other perpetrators. Some effect sizes, particularly for analyses of exposure to evictions, may be a result of limited statistical power.
Our study also has several strengths. Our analysis provides a foundation for beginning to quantitatively understand the relationship between housing instability and gender-based violence among women sex workers. Future mixed-methods and path analyses may be beneficial to understand specific explanatory mechanisms underpinning the observed associations. Future research should also evaluate structural interventions to improve housing stability and safety among sex workers, as well as the impact of the COVID-19 pandemic on housing instability and gender-based violence among sex workers.
Policy and Practice Implications
There is a crucial need to address housing instability, evictions, and high rates of IPV and workplace violence experienced by sex workers, including youths, gender and sexual minorities, and Indigenous women. The high burden of housing instability and evictions documented in this study highlights the urgent need to decriminalize sex work to address barriers related to criminalization and stigma, which undermine sex workers’ human rights related to housing. Decriminalization is a recognized best practice and evidence-based intervention for improving sex workers’ health, safety, and human rights, which includes the right to an adequate standard of living.
Women-specific supportive housing options that meaningfully address sex workers’ health and safety are also needed, for example, affordable, women-centered housing with flexible guest policies, security measures, community-based support staff, and integrated harm-reduction supports.6,18,34 More broadly, scale-up of affordable housing options is needed for marginalized women, as research has shown that rising gentrification poses a high risk of disconnecting women from essential services.35 In addition, the extension of government-implemented eviction moratoria could potentially mitigate the risk of gender-based violence for sex workers facing housing instability.
Conclusions
We have documented a high burden of unstable housing and evictions among women sex workers in Metro Vancouver. We found that exposure to unstable housing was associated with significantly higher odds of experiencing IPV and workplace violence over a 9-year observation period and that exposure to evictions was associated with elevated odds of IPV over a 4.5-year observation period. To support sex workers’ occupational safety and human rights, public health interventions should include structural changes to scale up safe, affordable, and supportive housing for marginalized women as well as decriminalization of sex work to address housing barriers that are produced and reproduced by pervasive criminalization and stigma.
ACKNOWLEDGMENTS
This research was supported by the US National Institutes of Health (NIH; grant R01DA028648) and the Canadian Institutes of Health Research. S. M. Goldenberg was partially supported by the NIH. N. Buglioni was partially supported by Mitacs. S. Moreheart was supported by a Simon Fraser University doctoral scholarship. K. Shannon was partially supported by a Canada Research Chair in Gender Equity, Sexual Health and Global Policy and the NIH.
We are tremendously grateful to all An Evaluation of Sex Workers Health Access (AESHA) participants for their valuable contributions of time and expertise to this research. We thank all those who contributed their time and expertise to this project, including AESHA community advisory board members and partner agencies and the AESHA team, including the following: Emma Ettinger, Christie Gabriel, Jennifer McDermid, Jennifer Morris, Emily Luo, Becca Norris, Alka Murphy, Natasha Feuchuk, Alex Martin, Jennie Pearson, Yinong Zhao, Amber Stefanson, Maddie Hibbs, Arveen Kaur, Peter Vann, and Preet Dhanda.
Note. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the funders.
CONFLICTS OF INTEREST
The authors have no conflicts of interest to report.
HUMAN PARTICIPANT PROTECTION
The study received ethical approval from the Providence Health Care and University of British Columbia and Simon Fraser University research ethics boards. All participants provided written informed consent before participating in the study.
REFERENCES
- 1.Swope CB, Hernández D. Housing as a determinant of health equity: a conceptual model. Soc Sci Med. 2019;243:112571. doi: 10.1016/j.socscimed.2019.112571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Shannon K, Crago AL, Baral SD, et al. The global response and unmet actions for HIV and sex workers. Lancet. 2018;392(10148):698–710. doi: 10.1016/S0140-6736(18)31439-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Decker MR, Crago AL, Chu SKH, et al. Human rights violations against sex workers: burden and effect on HIV. Lancet. 2015;385(9963):186–199. doi: 10.1016/S0140-6736(14)60800-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Deering KN, Amin A, Shoveller J, et al. A systematic review of the correlates of violence against sex workers. Am J Public Health. 2014;104(5):e42–e54. doi: 10.2105/AJPH.2014.301909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lyons T, Krüsi A, Edgar E, Machat S, Kerr T, Shannon K. The impacts of intersecting stigmas on health and housing experiences of queer women sex workers in Vancouver, Canada. J Homosex. 2021;68(6):957–972. doi: 10.1080/00918369.2019.1694337. [DOI] [PubMed] [Google Scholar]
- 6.Lazarus L, Chettiar J, Deering K, Nabess R, Shannon K. Risky health environments: women sex workers’ struggles to find safe, secure and non-exploitative housing in Canada’s poorest postal code. Soc Sci Med. 2011;73(11):1600–1607. doi: 10.1016/j.socscimed.2011.09.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Sandoval-Olascoaga S, Venkataramani AS, Arcaya MC. Eviction moratoria expiration and COVID-19 infection risk across strata of health and socioeconomic status in the United States. JAMA Netw Open. 2021;4(8):e2129041. doi: 10.1001/jamanetworkopen.2021.29041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Leifheit KM, Linton SL, Raifman J, et al. Expiring eviction moratoriums and COVID-19 incidence and mortality. Am J Epidemiol. 2021;190(12):2503–2510. doi: 10.1093/aje/kwab196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Pavao J, Alvarez J, Baumrind N, Induni M, Kimerling R. Intimate partner violence and housing instability. Am J Prev Med. 2007;32(2):143–146. doi: 10.1016/j.amepre.2006.10.008. [DOI] [PubMed] [Google Scholar]
- 10.Gilroy H, McFarlane J, Maddoux J, Sullivan C. Homelessness, housing instability, intimate partner violence, mental health, and functioning: A multi-year cohort study of IPV survivors and their children. J Soc Distress Homeless. 2016;25(2):86–94. doi: 10.1080/10530789.2016.1245258. [DOI] [Google Scholar]
- 11.Footer KHA, White RH, Park JN, Decker MR, Lutnick A, Sherman SG. Entry to sex trade and long-term vulnerabilities of female sex workers who enter the sex trade before the age of eighteen. J Urban Health. 2020;97(3):406–417. doi: 10.1007/s11524-019-00410-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Duff P, Deering K, Gibson K, Tyndall M, Shannon K. Homelessness among a cohort of women in street-based sex work: the need for safer environment interventions. BMC Public Health. 2011;11(1):643. doi: 10.1186/1471-2458-11-643. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Surratt HL, Inciardi JA. HIV risk, seropositivity and predictors of infection among homeless and non-homeless women sex workers in Miami, Florida, USA. AIDS Care. 2004;16(5):594–604. doi: 10.1080/09540120410001716397. [DOI] [PubMed] [Google Scholar]
- 14.Petry L, Hill C, Milburn N, Rice E. Who is couch-surfing and who is on the streets? Disparities among racial and sexual minority youth in experiences of homelessness. J Adolesc Health. 2022;70(5):743–750. doi: 10.1016/j.jadohealth.2021.10.039. [DOI] [PubMed] [Google Scholar]
- 15.Hail-Jares K, Vichta-Ohlsen R, Nash C. Safer inside? Comparing the experiences and risks faced by young people who couch-surf and sleep rough. J Youth Stud. 2020;24(3):305–322. doi: 10.1080/13676261.2020.1727425. [DOI] [Google Scholar]
- 16.Knight KR, Lopez AM, Comfort M, Shumway M, Cohen J, Riley ED. Single room occupancy (SRO) hotels as mental health risk environments among impoverished women: the intersection of policy, drug use, trauma, and urban space. Int J Drug Policy. 2014;25(3):556–561. doi: 10.1016/j.drugpo.2013.10.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lyons T, Krüsi A, Pierre L, Smith A, Small W, Shannon K. Experiences of trans women and two-spirit persons accessing women-specific health and housing services in a downtown neighborhood of Vancouver, Canada. LGBT Health. 2016;3(5):373–378. doi: 10.1089/lgbt.2016.0060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Krüsi A, Chettiar J, Ridgway A, Abbott J, Strathdee SA, Shannon K. Negotiating safety and sexual risk reduction with clients in unsanctioned safer indoor sex work environments: a qualitative study. Am J Public Health. 2012;102(6):1154–1159. doi: 10.2105/AJPH.2011.300638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Reed E, Gupta J, Biradavolu M, Devireddy V, Blankenship KM. The role of housing in determining HIV risk among female sex workers in Andhra Pradesh, India: considering women’s life contexts. Soc Sci Med. 2011;72(5):710–716. doi: 10.1016/j.socscimed.2010.12.009. [DOI] [PubMed] [Google Scholar]
- 20.Katsulis Y, Durfee A, Lopez V, Robillard A. Predictors of workplace violence among female sex workers in Tijuana, Mexico. Violence Against Women. 2015;21(5):571–597. doi: 10.1177/1077801214545283. [DOI] [PubMed] [Google Scholar]
- 21.Yakubovich AR, Bartsch A, Metheny N, Gesink D, O’Campo P. Housing interventions for women experiencing intimate partner violence: a systematic review. Lancet Public Health. 2022;7(1):e23–e35. doi: 10.1016/S2468-2667(21)00234-6. [DOI] [PubMed] [Google Scholar]
- 22.Argento E, Muldoon KA, Duff P, Simo A, Deering KN, Shannon K. High prevalence and partner correlates of physical and sexual violence by intimate partners among street and off-street sex workers. PLoS ONE. 2014;9(7):e102129. doi: 10.1371/journal.pone.0102129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Goldenberg SM, Liyanage R, Braschel M, Shannon K. Structural barriers to condom access in a community-based cohort of sex workers in Vancouver, Canada: influence of policing, violence, and end-demand criminalization. BMJ Sex Reprod Health. 2020;46(4):301–307. doi: 10.1136/bmjsrh-2019-200408. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Stueve A, O’Donnell LN, Duran R, San Doval A, Blome J. Time–space sampling in minority communities: results with young Latino men who have sex with men. Am J Public Health. 2001;91(6):922–926. doi: 10.2105/ajph.91.6.922. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Garcia-Moreno C, Jansen HAFM, Ellsberg M, Heise L, Watts C. WHO Multi-Country Study on Women’s Health and Domestic Violence Against Women: Initial Results on Prevalence, Health Outcomes and Women’s Responses. Geneva, Switzerland: World Health Organization; 2005. [Google Scholar]
- 26.Maldonado G, Greenland S. Simulation study of confounder-selection strategies. Am J Epidemiol. 1993;138(11):923–936. doi: 10.1093/oxfordjournals.aje.a116813. [DOI] [PubMed] [Google Scholar]
- 27.Riley ED, Cohen J, Knight KR, Decker A, Marson K, Shumway M. Recent violence in a community-based sample of homeless and unstably housed women with high levels of psychiatric comorbidity. Am J Public Health. 2014;104(9):1657–1663. doi: 10.2105/AJPH.2014.301958. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Riley ED, Vittinghoff E, Kagawa RMC, et al. Violence and emergency department use among community-recruited women who experience homelessness and housing instability. J Urban Health. 2020;97(1):78–87. doi: 10.1007/s11524-019-00404-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kennedy MC, McNeil R, Milloy MJ, Dong H, Kerr T, Hayashi K. Residential eviction and exposure to violence among people who inject drugs in Vancouver, Canada. Int J Drug Policy. 2017;41:59–64. doi: 10.1016/j.drugpo.2016.12.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Roesch E, Amin A, Gupta J, García-Moreno C. Violence against women during COVID-19 pandemic restrictions. BMJ. 2020;369:m1712. doi: 10.1136/bmj.m1712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Pearson J, Shannon K, Krusi A, et al. 2023. https://www.mdpi.com/2076-0760/11/9/383
- 32.Collins AB, Boyd J, Damon W, et al. Surviving the housing crisis: social violence and the production of evictions among women who use drugs in Vancouver, Canada. Health Place. 2018;51:174–181. doi: 10.1016/j.healthplace.2018.04.001. [DOI] [PubMed] [Google Scholar]
- 33.Vancouver Coastal Health Authority. 2023. http://www.vch.ca/Documents/DTES-Womens-Companion-Paper.pdf
- 34.Collins AB, Boyd J, Hayashi K, Cooper HLF, Goldenberg S, McNeil R. Women’s utilization of housing-based overdose prevention sites in Vancouver, Canada: an ethnographic study. Int J Drug Policy. 2020;76:102641. doi: 10.1016/j.drugpo.2019.102641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Goldenberg SM, Amram O, Braschel M, Moreheart S, Shannon K. Urban gentrification and declining access to HIV/STI, sexual health, and outreach services amongst women sex workers between 2010–2014: results of a community-based longitudinal cohort. Health Place. 2020;62:102288. doi: 10.1016/j.healthplace.2020.102288. [DOI] [PMC free article] [PubMed] [Google Scholar]