No Safe Place: Sexual Assault in the Lives of Homeless Women
Lisa A. Goodman, Ph.D., Katya Fels, & Catherine Glenn, M.A. With contributions from Judy Benitez
Despite over two decades of media and public policy attention, homelessness remains an enormous social problem in the United States, due in large part to the continual closings of institutions for people with mental illness, persistent poverty, a shortage of affordable housing, changes in welfare and mental health policy, and economic trends that favor the wealthy (Burt, Aron, Lee, & Valente, 2001; Evans & Forsyth, 2004; Haber & Toro, 2004; Lee & Schreck; Wolch & Li, 1997). Although women without custodial children and mothers taking care of young children represent two of the most rapidly growing subgroups of this population (Burt, Aron, Douglas, Lee & Valente, 2001; National Coalition for the Homeless, 2001; Urban Institute, 2000; U.S. Conference of Mayors, 1990, 2000), their needs remain relatively unexplored and largely unmet. Furthermore, these women are particularly vulnerable to multiple forms of interpersonal victimization, including sexual and physical assault at the hands of strangers, acquaintances, pimps, sex traffickers, and intimate partners on the street, in shelters, or in precarious housing situations.
All forms of victimization endured by homeless women and children deserve our immediate attention and action. In this paper, we summarize available research on sexual violence in particular -- that is, unwanted sexual activity that is forced, coerced or manipulated -- and we suggest ways of understanding and responding to the varied, critical needs of homeless survivors of such violence. We focus on adult women only, leaving for another paper the enormous problem of violence against runaway children and teenagers. Our goals are to consolidate knowledge about the damaging interplay between homelessness and sexual violence, and clarify what steps researchers, policy-makers, and service providers might take to intervene with victims and prevent future sexual assaults from occurring.
Although prevalence rates of homelessness have been estimated at up to seven to eight percent among adults (Haber & Toro, 2004), these statistics are based on studies of the ìliterally homelessî (Toro, 1998); that is, people who have spent nights in homeless shelters, on the street, or in other settings not intended for human habitation. This narrow definition leaves out the much larger population of the hidden homeless : women and children who may stay on friends', neighbors', and family members' couches night after night (ìcouch surfingî) or return to their abusers when emergency shelters are full, women in rural areas where no shelters are available, and women who trade sex for a place to sleep (Evans & Forsyth, 2004). As cuts to welfare and social services have deepened over the last decade, the hidden homeless population has grown steadily, with African-American women and female heads of households at greatest risk (Wolch & Li, 1997). Although most research on sexual assault among homeless women focuses on the narrower definition of homelessness, we will include information on the hidden homeless where possible in this review.
General prevalence of sexual assault and its consequences among homeless women
There are many limitations to the available research on homeless women who have experienced sexual assault (Nyamathi, Wenzel, Lesser, Flaskerud, & Leake, 2001; Wenzel, Leake, & Gelberg, 2000), We focus on recent studies that are methodologically sound and that approach the issue with an awareness of the complex and inter-related factors so often present in homeless women's lives.
The most comprehensive and rigorous studies on homeless women conducted to date continue to note the extraordinarily high levels of abuse and victimization that homeless women endure before, during, and after episodes of homelessness. In fact, although rates of victimization in this country have decreased overall, rates of victimization among homeless women remain relatively unchanged (Lee & Schreck, 2005). Research also highlights the grim finding that homeless women often report multiple episodes of violent victimization at the hands of multiple perpetrators, beginning in childhood and extending into adulthood (Browne & Bassuk, 1997; Goodman, 1991; Goodman, Dutton & Harris, 1995; Felix, 2004; Lee & Schreck, 2005; Stermac & Paradis, 2001; Wenzel et al., 2004). Indeed, homeless women have been described as enduring a ìtraumatic lifestyleî (Goodman, et al., 1995)ñ one in which traumatic incidents such as sexual assaults are layered upon ongoing traumatic conditions such as struggling to meet basic survival needs and living with ongoing dangers and threats.
Many large-scale studies report findings that repeatedly emphasize the violent and traumatic lives of homeless women. One of largest and most in-depth studies on this topic found that 92% of a racially diverse sample of homeless mothers had experienced severe physical and/or sexual violence at some point in their lives, with 43% reporting sexual abuse in childhood and 63% reporting intimate partner violence in adulthood. Over half (57.6%) reported experiencing violence in at least two out of four age periods (0-5; 6-12; 13-18, 18+) (Browne & Bassuk, 1997). In another study, 13% of homeless women reported having been raped in the past 12 months and half of these were raped at least twice (Wenzel, et al., 2000). In yet another study, 9% of homeless women reported at least one experience of sexual victimization in the last month (Wenzel, Koegel & Gelberg, 2000). Women who do not have children with them are at particularly high risk for sexual violence after the age of 18 (Zugazaga, 2004), in part because they are more likely to sleep outside than are women with children, who might fear for their children's well being or worry about the intervention of child protective services. Also disturbing is the finding that compared to their low-income housed counterparts, the sexual assault experiences of homeless women are more likely to be violent, and to include multiple sexual acts (Stermac & Paradis, 2001).
Homeless women who experience sexual assault may suffer from a range of emotional and physical challenges (D'Ercole & Streuning, 1990; Goodman, Saxe & Harvey, 1991; Ingram et al, 1996; Padgett & Streuning, 1992; Rayburn, Wenzel, Elliot, Hambrasoomians, Marshall, & Tucker, 2005; Salomon, Bassuk, & Huntington, 2002). In one study of homeless women who had been victimized (Browne & Bassuk, 1997; see also Bassuk, Buckner, Weinreb, Browne, Bassuk, Dawson, & Perloff, 1997) most participants reported mental health problems ranging from suicide attempts (45%) and depression (47%) to alcohol or drug dependence (45%) and posttraumatic stress disorder (39%). Other studies report similar types of psychological difficulties (e.g., Nyamathi et al., 2001; North & Smith, 1992; North, Smith & Spitznagel, 1994; Wenzel, Leake, and Gelberg, 2000).
Sexual assault also effects homeless women's physical health. For example, in one study of homeless women, those who reported a rape in the last year were significantly more likely than nonvictims to suffer from two or more gynecological conditions and two or more serious physical health conditions in the past year (Wenzel et al., 2000). They were also significantly more likely to report that although they needed to see a physician during the past year, they could not manage to do so, and that although they desired treatment for substance abuse they were unable to obtain appropriate services.
Homeless victims of sexual assault must contend with these psychological and physical difficulties within the context of poor access to legal, mental health and medical resources, social alienation and isolation, unsafe living environments, constant exposure to reminders of the experience, and lack of transportation and information about available services (Goodman, Saxe, and Harvey, 1991). Homeless women of color, lesbians and bisexuals, and women with physical, emotional, and developmental disabilities face even greater barriers. Those who are mothers must take care of their children in chaotic situations while under the scrutiny of a range of social service providers in shelters, food pantries, and other settings. It is no wonder, then, that homeless women are at high risk for state involvement in their parenting, and the potential removal of their children (Cowal, Shinn, Weitzman, Stojanovic, & Labay, 2002; Zlotnick et al., 2003).
Question of Causality
A range of factors increase homeless women's risk of adult sexual victimization, including childhood abuse, substance dependence, length of time homeless, engaging in economic survival strategies (such as panhandling or involvement in sex trade), location while homeless (i.e. sleeping on the street versus sleeping in a shelter) and presence of mental illness (Kushel, Evans, Perry, Robertson, & Moss, 2003; Nyamathi, Wenzel, Lesser, Flaskerud, & Leake, 2001; Wenzel, Koegel, & Gelberg, 2000; Wenzel, Leake, & Gelberg, 2001). Many of these factors, discussed in more detail below, coexist, interact with, and exacerbate each other over time, creating a complex and distinctive context for each woman.
It is important to note that all of these factors would have a much more tenuous connection with sexual assault if social institutions were in place to prevent homelessness, to protect vulnerable women, and to help them recover and become safe following an initial assault while addressing the myriad other challenges they face. And yet to date, no research has been conducted on the impact of institutional failures on the prevalence or correlates of sexual assault among homeless women. For example, research has not yet examined the unsuitability of traditional sexual assault crisis services, such as hotlines and in-office counseling, for individuals who lack access to a telephone, transportation, literacy skills, and safe housing.
Sexual Assault Prior to Homelessness
The relationship between sexual assault and homelessness is complex, with either experience potentially laying the groundwork for the other. Indeed, given the traumatic lifestyles of so many homeless women, sexual abuse may precede and follow from homelessness in a vicious cycle downward. In the next two sections, we take a closer look at existing research on two different types of sexual assault (child sexual abuse and sexual violence at the hands of a partner) as precursors to adult homelessness and subsequent victimization.
Childhood Sexual Abuse
A number of studies have emphasized the correlation between childhood sexual abuse and homelessness among adult women (Bassuk and Rosenberg, 1988; Davies-Netzley & Hurlburt, & Hough, 1996; Simons & Whitbeck, 1991; Stermac & Paradis, 2001; Wenzel et al., 2004; Zugazaga, 2004). For example, one study of women seeking help from a rape/sexual assault crisis center found that childhood sexual abuse was reported by 43% of the homeless participants, compared to 24.6% of the housed participants (Stermac et al., 2004). Another study that took a qualitative approach found that homeless women identified child sexual victimization as a cause of their homelessness (Evans & Forsyth, 2004).
Childhood sexual abuse is also correlated with adult victimization among homeless women (Nyamathi et al., 2001; Terrell, 1997; Tyler, Hoyt, & Whitbeck, 2000; Whitbeck, Hoyt, & Ackley 1997). One study found that homeless women with histories of childhood sexual abuse were twice as likely to experience adult violent victimization as those without such histories (Nyamathi et al., 2001). For homeless women with serious mental illness (SMI), the connection between child sexual abuse and adult victimization is even stronger. In one study of women with serious mental illness and histories of homelesness, the chance of revictimization for women who had experienced child physical or sexual abuse was close to 100% ñ difficult odds to beat (Goodman, Dutton & Harris, 1995; Goodman, Johnson, Dutton, & Harris, 1997).
A number of explanations have been offered for the relationship between child sexual abuse and subsequent homelessness and sexual assault, respectively. It is possible, for example, that child sexual abuse survivors may find it difficult to trust others, so they develop fewer of the sustaining and supportive relationships necessary to avoid homelessness (Bassuk, 1993). Also, the posttraumatic stress disorder that often results from child sexual abuse can cause women to miss danger cues in their environments due to hypervigilance (attending to everything as a threat) or dissociation (shutting down when faced with threatening situations), resulting in risk for further victimization (Salomon, Bassuk, & Huntington, 2002; Tyler, Hoyt, & Whitbeck, 2000; Whitbeck, Hoyt, & Ackley, 1997). Finally, women who experience childhood sexual abuse have been shown to be at increased risk for developing substance abuse disorders, which put women at increased risk for both assault and homelessness (Burnam, Stein, Golding, Siegal, Sorenson, & Telles, 1988; Salomon, Bassuk, & Huntington, 2002; Simmons & Whitbeck, 1991; Tyler, Hoyt, & Whitbeck, 2000).
However, these explanations alone do not tell the whole story. A much fuller explanation for these devastating correlations emerges from an exploration of the complex array of historical and current contextual factors many women face, including multiple oppressions, lack of appropriate, culturally relevant, and timely resources, and growing up in unsafe settings without sufficient material and emotional support. Rather than one causing the other, we suggest that the contextual factors that often precede child sexual abuse (and repeated victimization) also precede homelessness. For example, poor families; people of color; and immigrants, refugees, and victims of sex trafficking may experience systems such as law enforcement, social services, foster care, or welfare not as sources of care and assistance, but of neglect or punishment. Childhood sexual abuse survivors in particular may have experienced caregivers acting appropriately in public and inappropriately in private, and therefore may be reluctant to trust people whose job it is to help them. As children and as adults, they may be reluctant to seek help from people in ìthe systemî and therefore remain particularly vulnerable to ongoing victimization and homelessness, in addition to self-medication through substances and isolation.
Abuse by Partners
Not surprisingly, a number of studies point to abuse--including rape--at the hands of a current or former partner, as a risk factor for homelessness among women (Toro, Bellavia, Daeschler, Owens, Wall, Passero, & Thomas, 1995). This is particularly evident for women who experience partner violence at the more severe end of the continuum, and who have been isolated by their abusers from family and friends who might have offered to help them (Baker, Cook, & Norris, 2003). Indeed, it is estimated that half of all homeless women and children have become homeless while trying to escape abusive situations (Browne & Bassuk, 1997, as cited in Evans & Forsyth, 2004). Experiences of partner violence have also been shown to predict risk of repeat homelessness and shelter use (Metraux & Culhane, 1999). Yet, there are few studies documenting the impact of partner violence on women who are currently homeless, how the threat of such violence might shape women's decision-making while homeless, or the nature of the complicated tradeoffs many partner violence victims make to survive on the streets. For example, a homeless woman may stay in a relationship with a person who abuses her physically or sexually because the risks associated with leavingóhomelessness, hunger, poverty, violence on the streets, lack of resources for children, risk of further abuse by additional perpetrators óseem worse than the abuse. Furthermore, the abusive partner may also provide protection and companionship some of the time.
Homelessness as Risk Factor for Sexual Assault
Although childhood sexual abuse and intimate partner violence often precede, and may contribute to women's homelessness and risk for revictimization, the condition of homelessness itself dramatically increases women's risk of being sexually assaulted. Women on the streets do not enjoy the same degree of safety as women who have four walls and a roof to protect them. Despite being in very close quarters with many others, women staying in shelters often lack robust and nurturing social connections, as people in crisis have fewer resources to dedicate to developing mutual trust than those who feel safer and more grounded (Goodman, 1991). The need to serve a maximum number of people with limited dollars, combined with some communities' unwillingness to host shelters in their neighborhoods, often leads shelters to locate within or close to high-crime areas (Burt, et al., 2001; Wenzel, Koegel & Gelberg, 2000). Moreover, as discussed in subsequent sections, many homeless women have little choice but to participate in activities that place them at further risk for sexual assault, such as panhandling or trading sex for needed resources (Kushel, et al., 2003; Lee & Schreck, 2005).
Individual vulnerabilities also play a role. Homeless women are more likely than non-homeless women to suffer from substance abuse (Toro et al., 1995; Wenzel et al, 2004), a mental illness that may include psychosis (Toro et al., 1995; Wenzel et al, 2004), domestic violence (Toro et al., 1995), or severe physical health limitations (Wenzel, Leake & Gelberg, 2000) that make self-defense in a dangerous situation harder. In one of the most rigorous studies of antecedents of sexual assault while homeless, Wenzel, Koegel, and Gelberg (2000) found that women who were dependent on drugs or alcohol; who received income from survival strategies such as panhandling, selling items on the street, or trading sex for drugs or other items; who lived outdoors; who experienced mania or schizophrenia; or who had physical limitations were especially likely to have endured a recent (at most, 30 days prior) sexual assault. The next sections review our knowledge of some of these factors in more detail.
Survival Sex and Prostitution
Survival for some homeless women is contingent on trading sex for money, goods (food, shelter, clothes, medicine, drugs), services, transportation, and protection on the street (Wenzel et al, 2001). It is debatable whether sex under these circumstances is ever really a choice; certainly, it is often a requirement last resort strategy for survival. Further, outright sexual violence is a common occurrence for women who engage in sex trade (Dalla, Xia, & Kennedy, 2003; Nyamathi, et al., 2001). Wenzel, Koegel and Gelberg (2000) found that over the course of a year, homeless women who panhandled or traded sexual favors for drugs or money were three times more likely to experience sexual assault and other forms of violence relative to their homeless peers who did not engage in sex trade. Indeed, 84% of women who use prostitution as an income strategy report current or past homelessness ñ which can mean living with abusive pimps or ìcustomersî in the absence of a more stable option (Farley & Barkan, 1998); and homeless prostituted women are at much greater risk for sexual assault than their non-homeless counterparts (El Bassel, Witte, Wada, Gilbert, & Wallace, 2001). When substance use (often ìpaid forî by sex) is a factor, the risk of sexual assault increases further, as described in the next section. Because these assaults often occur in the context of an illegal act (prostitution) and among drug users, victims may be seen by perpetrators as attractive targets, as they are less likely to report the crime or to be believed or seen as worthy of services and protection by authorities.
Homeless women are more likely to have substance abuse problems and to engage in substance use than low-income housed women (Wenzel et al., 2004). Although substance use and abuse among homeless women may represent their best method of coping with the chaos, unpredictability, and isolation of homelessness, as well as previous victimizations, it is also strongly associated with risk for further sexual assaults. One study found that homeless women who had experienced either physical or sexual victimization in the past month were three times more likely to report both drug and alcohol abuse or dependence than homeless women who were not victimized (24.3% vs. 7.9%) (Wenzel, Leake, & Gelberg, 2000). As with so many aspects of homeless women's lives, the causal relationships between substance abuse and victimization are far less clear than the correlation itself. Nevertheless, substance abuse and dependence may put women at risk for victimization in a number of ways, such as by altering women's perceptions of what is dangerous; leading them to engage in risky survival strategies; causing disorientation that may make it difficult to ward off an attacker; making them a target for assault because authorities will be less likely to believe them; or putting them in an environment that involves interactions with criminals. Indeed, offenders often rely on drugs and alcohol to incapacitate their victims (Lisak & Miller, 2002). Furthermore, drug and alcohol services and rape crisis services largely remain fragmented, which can make it difficult for individuals to receive the services they need to recover.
Severe Mental Illness (SMI)
Homeless women with serious mental illnesses such as major depression, schizophrenia, and bipolar disorder are highly vulnerable to victimization. Indeed, in one in-depth study 97% of the participants, all of whom were homeless and had a mental illness, reported experiences of violent victimization at some point in their lives (Goodman, Dutton & Harris, 1995; Goodman, Johnson, Dutton, & Harris, 1997), with an astonishing 28% reporting at least one physical or sexual assault in the month preceding the interview. Another large-scale study of 1,839 ethnically diverse, homeless women and men with mental illnesses from 15 cities across the US found that 15.3% of the women participants reported being raped in the past 2 months (Lam & Rosenheck, 1998), compared to 1.3% of the men. For homeless women with mental illnesses, rape appears to be a shockingly normative experience. This is deeply troubling, as no one should ever become ìusedî to being raped or assaulted. To the contrary, there is evidence that the cumulative effects of multiple victimizations may be far deeper than single rape events (Goodman & Dutton, 1996). Moreover, these women's ability to get help are greatly compromised by social attitudes that people with mental illnesses do not experience violation as searingly as others; that their accounts of the abuse and assault are ìmade upî (Goodman & Dutton, 1996; Goodman, et al., 1999); or that women with mental illnesses cannot clearly communicate a lack of consent. Homeless women with mental illnesses who are also victims of sexual violence shoulder the burden of three forms of social stigmaóagainst poor or homeless people, people with mental illnesses, and victims of rape.
Barriers to Accessing Institutional Support
Although more research is needed to understand the relationship between sexual assault and homelessness, especially research that explores the social and institutional contributions to this enormous social problem, action is also needed. In this section, we provide an overview of situational, contextual and systemic barriers homeless women face in finding the support they need to heal in the wake of sexual assault.
Homelessness often involves spiraling crises, which means that homeless women might not deal with, attend to, or process sexual assault in the same way as housed women do. For example, a rape may be followed only weeks later by a notice of loss of social security disability benefits because the victim failed to appear at a hearing scheduled the day after she was raped. This new crisis may shift the woman's attention temporarily, but the impact of the previous crisisóthe rapeóbecomes interwoven with the impact of other crises. It is important, therefore, that the sexual assault be addressed in culturally sensitive ways as part of a complex context of trauma and crises. Unfortunately, few services available to homeless victims of sexual assault are set up to deal with these compounding crises. This complexity presents a range of challenges both to staff at programs responding to the homeless, who are rarely trained to detect and respond appropriately and sensitively to trauma or sexual violence, and to rape crisis counselors, who are often unequipped to deal with the multiple challenges brought on by homelessness.
By their very nature, homeless shelters can worsen women's psychological distress and compromise their ability to do what is necessary to regain residential stability and increased quality of life. Homelessness is inherently chaotic, internally and externally, with others controlling access to such basic resources as food, clothing, and shelter. Indeed the very process of accessing the variety of programs necessary to rise out of homelessness may itself create a chaotic situation. There is little privacy, and entering many programs requires subjecting the private details of one's life to regulation and/or scrutiny. This lack of privacy and power differential can mirror and exacerbate the impact of the violence many homeless women have survived. This combination of chaos, power dynamics and feeling watched can trigger traumatic memories or symptoms that, in turn, make it more difficult to abide by shelter rules or stay ìin controlî as shelters require. Many shelters are neither culturally sensitive nor ìtrauma-informed,î and have not provided staff adequate training to, for example, deal with women's angry outbursts therapeutically rather than punitively, or recognize the differences between flashbacks and psychosis. Overburdened staff must balance the needs of the individual with the needs of many. A woman whose trauma-related nightmares wake up an entire dorm, for example, may be told to leave.
At the same time, many of the options for self-care and self-soothing following a sexual assault are not available to homeless women. As noted earlier, homeless women lack telephones, making hotlines irrelevant. A woman may become alienated from a traditional sexual assault support group when she cannot make weekly meeting times or finds that unlike her peers, her history includes so many assaults when others report significantly fewer. To make matters worse, general shelters are often full to capacity and may have to turn women away, while battered women's shelters rarely offer beds to women who fear violence from people who are not traditional partners, leaving them no choice but to return to dangerous and out-of-the-way places to sleep (Amster, 1999, as cited in Evans & Forsyth, 2004).
There is a widespread, although increasingly disputed, belief that trauma must not, indeed cannot, be addressed before a woman is in a stable situation with regard to food, shelter, physical safety, and housing (Herman, 1992) Yet, few rape crisis centers are equipped to help provide the stability that they prescribe, making services fragmented at best, and possibly even irrelevant. Furthermore, stability may be elusive until the trauma is named and at least partially explored. Fragmented services that force an individual to separate problems that are inextricable can exacerbate existing trauma.
The relationship between homeless sexual assault victims and law enforcement is equally complex. Sexual assault and rape reporting rates are very low in general (Rennison, 2002). Homeless women may lack someone, whether peer, volunteer or advocate, to support them through the often-intimidating process of reporting an assault. Homeless women, already turning to bureaucracies for even their most basic needs (e.g. food stamps, housing vouchers and the like), may be reluctant to engage with yet one more system that they expect will be unresponsive.
Homeless women may not see the police in particular as providing protection and safety. They may be afraid to report a rape because they are involved in illegal activities (e.g. drug related, prostitution) or have outstanding warrants from other activities. They may distrust police officers because their only contact with them is when they are kicked off park benches and forced to sleep under bushes that are far from the public eye and therefore more dangerous. For women who engage in street-based sex trade, harassment and abuse by police is so commonplace that many women no longer perceived police as sources of help. Homeless women of color, immigrants, refugees, and victims of sex trafficking may be even more skeptical about law enforcement and less likely to turn to them for help or protection. Further, law enforcement personnel are not immune from general social attitudes about stigmatized groups such as homeless, mentally ill, prostituting, or substance abusing women, resulting in discriminatory behavior. Last, because homeless women are highly transient, they generally make poor witnesses in victimization cases; and the very public nature of life on the streets means that few women have a place to hide if an abuser or rapist learns she has ìrattedî on him. These obstacles result in shared feelings of helplessness between even the most sympathetic criminal justice personnel and homeless women.
Suggestions for System Improvements
Given that homeless women are raped more than housed women, addressing the grave shortage of affordable housing in the United States would not only reduce the rates of homelessnessóit would reduce the incidence of sexual assault. Yet, there are severe shortages of affordable housing, supported housing, and housing vouchers across the country (Clampet-Lundquist, 2003). Beyond housing, substance abuse and mental health treatment programs are severely under-funded; rape crisis programs are struggling to meet the needs of those who already come to them for help and often are not funded to provide shelter to victims and shelters for homeless women have long waitlists. Further, all of these services have developed independently and are now invested in maintaining their ìsiloedî services in order to hold onto standard funding streams. These gaps are creating a growing number of ìspecial needsî victims whose needs are grossly neglected.
Clearly, the systems that impact homeless women who are sexual assault survivors require new funds and new forms of collaboration to be able to respond to the particular needs and challenges that face them. What follows is a list ñ albeit not an exhaustive one -- of recommendations, which will provide a baseline upon which communities can build. The recommendations made here require the combined energies and resources of funders, policy makers, service providers, and communities.
Much has already been written about the need for trauma-informed homeless services (Elliott, Bjelajac, Fallot, Markoff, & Reed, 2005; Markoff, Reed, Fallot, Elliott, & Bjelajac, 2005), and these efforts must be expanded and supported. Collaboration between homeless providers and rape crisis advocates is critical to meet the needs of homeless victims of sexual violence. Homeless service providers must be given training and support in trauma and its consequences and staff must be given the authority to respond flexibly and appropriately to sexual assault survivors who come to them. Indeed, all homeless women would benefit from a trauma-informed approach since, as illustrated throughout this article, homelessness is itself a form of trauma and a significant risk factor for assault; and survivors of child sexual abuse, regardless of economic status, may not remember or label their experience as such, but continue to suffer its damaging effects. However, being trauma-informed goes far beyond staff training. Organizations must examine and reframe their practices and protocols based on an understanding that most homeless women are survivors of trauma, and are likely to be revictimized if not given emotional support, the ability to have some control over their daily lives, and a safe and calm place to stay 24 hours a day.
Rape crisis centers must be given the funding and the technical support to provide new services responsive to the particular challenges of homeless women, without further marginalizing them. Given that many housed women who seek the services of rape crisis centers also contend with a range of other difficulties, providing staff and volunteers with tools to effectively help survivors facing multiple challenges would clearly strengthen their offerings. Indeed, whether situated within a rape crisis center or elsewhere, counseling and therapy around trauma cannot be divorced, temporally or practically, from assistance addressing other crises threatening a woman's health and wellness, whether physical, psychological, economic or situational (Fels, Goodman, & Glenn, 2006).
Survivors of sexual assaultóhomeless or housed, poor or wealthyólive with shame and fear. If they are homeless, they are further shamed by society for being poor and requiring help. If they are also women of color; immigrants; refugees; victims of sex trafficking; prostitutes; lesbian, bisexual, and transgender individuals; or women with disabilities, they face even greater stigma and discrimination. At every turn, homeless women are dehumanized by systems that collect information on intake, assess them, process them and attempt to move them forward. The expectation that these women will reach out to strangers for help around issues as deeply personal as sexual assault, whether volunteers at a hotline or an assigned case manger, without time to develop a relationship and a foundation of trust is often unrealistic.
Creating initiatives, programs, systems and communities that respond collaboratively, respectfully, and holistically to homeless women with sexual assault histories in their distant or immediate past requires an understanding of the constellation of issues raised above. But it also requires organizational readiness and capacity. Systems do not prevent homeless survivors of assault from falling through the cracks ñ the people who work in these systems and their programs do. Although a significant discussion of the harmful effects of high staff turnover in many social services is beyond the scope of this review, we recognize its deleterious effects on program effectiveness and, ultimately, on survivors' lives. Further, it is the authors' experience that poor and homeless women are acutely tuned-in to the possibility of classicism, racism and other explicit and subtle forms of oppression imposed by the very people offering them services. Cultural competency training that attends to issues of class, as well as race and position, is critical.
The production and dissemination of this publication was supported by Cooperative Agreement Number U1V/CCU324010-02 from the Centers for Disease Control and Prevention. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the CDC, VAWnet, or the Pennsylvania Coalition Against Domestic Violence.