Does housing insecurity impact health care? Is the research question. This is the article. Clough,

Does housing insecurity impact health care? Is the research question. This is the article. Clough, J. B., Hanson, G. C., & Bloom, T. L. (2012). Housing instability is as strong a predictor of poor health outcomes Download Housing instability is as strong a predictor of poor health outcomes. Journal of Interpersonal Violence, 27(4), 623-643. Housing Instability Is as Strong a Predictor of Poor Health Outcomes as Level of Danger in an Abusive Relationship: Findings From the SHARE Study Journal of Interpersonal Violence 27(4) 623-643 © The Author(s) 2012 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0886260511423241 http://jiv.sagepub.com Chiquita Rollins, PhD,1 Nancy E. Glass, PhD, MPH, RN,2 Nancy A. Perrin, PhD,3 Kris A. Billhardt, MEd, EdS,4 Amber Clough, MSW,1 Jamie Barnes, MA,1 Ginger C. Hanson, MS,3 and Tina L. Bloom, PhD, MPH, RN5 Abstract Advocates, clinicians, policy makers, and survivors frequently cite intimate partner violence (IPV) as an immediate cause of or precursor to housing problems. Research has indicated an association between homelessness and IPV, yet few studies examine IPV and housing instability. Housing instability differs from homelessness, in that someone experiencing housing instability may currently have a place to live but faces difficulties with maintaining the residence. We present baseline findings from a longitudinal cohort study of 278 female IPV survivors with housing as a primary concern. Our analysis indicates the greater the number of housing instability risk factors (e.g., eviction 1Multnomah Department of County Human Services, Portland, OR, USA 2 Johns Hopkins University School of Nursing, Baltimore, MD, USA 3Kaiser Center for Health Research, Portland, OR, USA 4Volunteers of America-Oregon, Home Free, Portland, OR, USA 5University of Missouri, Columbia, MO, USA Corresponding Author: Tina L. Bloom, University of Missouri, S326 Sinclair School of Nursing, Columbia, MO 65211, USA Journal of Interpersonal Violence 27(4) notice, problems with landlord, moving multiple times), the more likely the abused woman reported symptoms consistent with PTSD (p < .001), depression (p < .001), reduced quality of life (p < .001), increased work/school absence (OR = 1.28, p < .004), and increased hospital/emergency department use (OR = 1.22, p < .001). These outcomes persist even when controlling for the level of danger in the abusive relationship and for survivors' drug and alcohol use. Importantly, both housing instability and danger level had stronger associations with negative health outcomes than other factors such as age, alcohol, and drug use; both make unique contributions to negative health outcomes and could contribute in different ways. Housing instability is an important and understudied social determinant of health for IPV survivors. These findings begin to address the literature gap on the relationship between housing instability, IPV, and survivors' health, employment, and utilization of medical care services. Keywords domestic violence, mental health and violence, assessment Introduction and Background Intimate partner violence (IPV) is a significant public health and human rights issue. Annually, IPV results in an estimated 1,200 deaths and 2 million injuries among women in the United States (Centers for Disease Control and Prevention, 2008). More than 23% of U.S. women report a lifetime history of IPV and 40% to 50% of female homicides are attributable to IPV (Campbell, Webster, & Glass, 2009; Campbell et al., 2003; Mercy & Saltzman, 1989). IPV is more than physical violence—it is a pattern of abusive and coercive behaviors including physical violence, sexual violence, threat of physical or sexual violence, psychological abuse, and stalking by a spouse, ex-spouse, or current or former intimate partner (Saltzman, Fanslow, McMahon, & Shelley, 1999). The health sequelae of IPV are not limited to injuries but also to less obvious or chronic health problems (e.g., chronic pain, depression, posttraumatic stress disorder [PTSD] that often interfere with or limit daily functioning, employment, and quality of life (Bassuk, Dawson, & Huntington, 2006; Coker, Weston, Creson, Justice, & Blakeney, 2005; Eby, 2004; Glass, Perrin, Campbell, & Soeken, 2007; Gorde, Helfrich, & Finlayson, 2004; Stewart et al., 1989; Weaver & Clum, 1995). A woman remains in an abusive relationship for complex reasons, ranging from fear of reprisal to financial barriers that prevent her from supporting Rollins et al. 625 herself and her children (Dutton & Goodman, 2005). A lack of income to sustain safe housing has been consistently reported as one important barrier to women leaving an abusive relationship (Anderson & Sauders, 2003; Hardesty & Campbell, 2004; Sheridan, 2001). The perceived and actual availability of safe, affordable housing and programs in the community to assist survivors of IPV in accessing housing is also linked to women's decision and ability to safely leave an abusive relationship (Hirst, 2003). Clinicians, advocates, policy makers, and survivors themselves frequently cite IPV as an immediate cause of or precursor to housing problems (Rollins, Saris, & Johnston-Robledo, 2001; Tischler, Karim, Rustall, & Vostanis, 2004). Researchers have found that the need for housing and economic resources were the most pressing concerns among abused women who had recently left abusers (Anderson & Sauders, 2003; Goodman, Smyth, Borges, & Singer, 2009). Furthermore, among U.S. cities surveyed in 2005, 50% identified IPV as a primary cause of homelessness in their city (U.S. Conference of Mayors-Sodexho, 2005). Research on the association between homelessness and IPV indicates that survivors and their children are at increased risk for negative health and social outcomes including revictimization, poor physical and mental health, job loss, decline in school performance, and possible fracturing of the family unit (Bassuk et al., 1996; Bures, 2003; Culhane, Webb, Grim, & Culhane, 2003; Gagne & Ferrer, 2006; Rodriguez, 2006; Schanzer, Dominguez, Shrout, & Caton, 2007). Prior research associates homelessness with IPV and poor health and social outcomes. Yet there have been few studies examining IPV and its effect on housing instability (Baker, Cook, & Norris, 2003; Burman & Chantler, 2005). Housing instability is different from homelessness. Someone experiencing housing instability may currently have a place to live but faces multiple ongoing difficulties, both personal and economic, associated with maintaining the residence. Indicators of housing instability include difficulty paying rent, mortgage, or utility bills and being denied affordable housing because of past credit problems, eviction threats or notices, moving frequently, living in overcrowded conditions, or "doubling-up" in a residence with family or friends (Kushel, Gupta, Gee, & Haas, 2006; Pavao, Alvarez, Baumrind, Induni, & Kimerling, 2007). Housing instability is typically defined in the literature as a binary variable that is scored positive if someone has had a period of homelessness, paid more than 50% of income on housing costs, or had difficulty finding safe, adequate, and affordable housing (Kushel et al., 2006; Ma, Gee, & Kushel, 2008; Reid, Vittinghoff, & Kushel, 2008). A binary measure does not 626 Journal of Interpersonal Violence 27(4) capture the complexity of the phenomenon of housing instability among abused women, which is often linked to the abusive behavior of the partner. For example, the abuser's behaviors (e.g., destroying property at the home, threatening the neighbors or landlord with violence if they intervened on her behalf) may have resulted in eviction, therefore making it difficult for her to find a landlord willing to rent to her again (Martin & Stern, 2005). The federal definition of homeless is (a) an individual who lacks a fixed, regular, and adequate nighttime residents or (b) an individual who has a primary nighttime residence that is (i) a supervised, publicly or privately operated shelter designed to provide temporary living accommodations; (ii) an institution that provides temporary residence for individuals intended to be institutionalized; or (iii) a public or private place not designed for, or ordinarily used as, a regular sleeping accommodation for human being (U.S. Department of Housing and Urban Development, n.d.). Women, especially women with children, are hesitant to live on the streets or in shelters because of safety concerns for themselves and their children and the fear of Child Welfare involvement if their children are homeless. There is a well-documented lack of capacity in shelters for women or victims of IPV (National Network to End Domestic Violence, 2010). Thus, abused women, especially women with children, are more likely to "double-up" or to live in highly unstable situations. Research is needed to describe the relationship between housing instability, IPV, and poor health and social outcomes. To begin to address the gap in research, this article presents baseline findings from our quasi-experimental longitudinal study called the SHARE study. The purpose of the SHARE study is to evaluate the effectiveness of a permanent housing program on the health and well-being of abused women and their children. We hypothesized that housing instability is an independent predictor of health outcomes. Specifically, among a sample of female IPV survivors with housing instability, we examined symptoms consistent with PTSD and depression as well as their level of danger, absence from work and/or school, self-reported quality of life and general health, and utilization of hospital and emergency medical care. Method Setting and Participants SHARE was conducted in partnership with four community-based domestic violence and housing programs serving the Portland Metropolitan Area of Rollins et al. 627 Oregon. Participants recruited from our four partner agencies were English- or Spanish-speaking adult women aged 18 to 64 who self-reported all of the following: (a) physical and/or sexual violence by an intimate or ex-intimate partner in the previous 6 months; (b) newly enrolled in services through one our partner agencies; (c) had housing stabilization as a primary need; and (d) planned to stay in the surrounding metro area for the duration of the study. Procedures Direct service staff at partner agencies referred potentially eligible participants to the research team through a signed release of information authorizing staff to provide contact information to research assistants. Participants could also self-refer by responding to flyers with the study contact information posted in the agency or distributed by agency staff. A trained research assistant assessed eligibility, described the purpose of the study, answered any questions, and performed informed consent prior to beginning the study interview. The research assistant then conducted a face-to-face interview at a location and time deemed safe and convenient by the participant. The human subjects review committees at Johns Hopkins University, Oregon Health & Science University, Kaiser Center for Health Research, and the Center for Disease Control and Prevention approved the study. Interview Instruments Demographics. Age was measured in years. Insurance was measured using a single item, "Did you have any type of health insurance for yourself in the last 6 months?" Responses were coded 0 (no insurance) or 1 (yes, private insurance or Medicaid). General health. We assessed general health for the past month using a single item from the SF-8, "In general, would you say your health is: 1 (poor), 2 (fair) 3 (good), 4 (very good), 5 (excellent)?" The reliability and validity of this item have been well established (Ware, Kosinski, Dewey, & Gandek, 2001). Danger assessment (DA). The DA was designed to assess abused women's risk of lethal violence (or serious injury) by their intimate partner or ex- intimate partner. The scale consists of 20 risk-factor items that are scored as 0 (no) or 1 (yes). Scores are computed using a weighted sum of items and have a possible range from 0 to 38, with a higher score indicating a greater risk of lethal violence (e.g., 0-7 = variable danger; 8-13 = increased danger; 14-17 = severe danger; and 18-above = extreme danger). Validation of the 628 Journal of Interpersonal Violence 27(4) weighted scoring of the 20-item DA is published elsewhere (Campbell et al., 2009). Cronbach's alpha for the unweighted items in this sample was .72. Housing instability. Based on a review of published literature on homelessness and housing instability, we developed the Housing Instability Index (Martin & Stern, 2005; Melbin, Sullivan, & Cain, 2003; Menard, 2001; Pavao et al., 2007). The index is a count of 10 possible risk factors for housing instability in the past 6 months. Eight items elicit a dichotomous, yes/no response and two items were recoded to be dichotomous. The question, "In the past 6 months, how many times have you moved?" was counted as a risk factor if participants reported moving more than twice in the past 6 months. The question, "How likely is it that you will be able to pay for your housing this month?" was recoded so that 0 represented a response of "very likely" or "somewhat likely" and 1 represented a response of "unlikely" or "very unlikely." One item, "Do you expect that you will be able to stay in your current housing for the next 6 months?" was reverse-coded so that a response of "no" was counted as a risk factor. Cronbach's alpha for the 10-item measure was .70, and the average item-total correlation was .37. Analysis of variance found the measure differentiates between those who can pay their own rent and those living in domestic violence shelters and motels (p < .001). PTSD. Past-month PTSD symptoms was measured using the PTSD Checklist-Civilian version (PCL-C). The scale contains 17-items rated on a 5-point scale ranging from 1 (not at all) to 5 (extremely). The total score is computed by summing the items and scores. The internal consistency for this sample was .91. A score of 30 or higher indicates high likelihood of PTSD for women in the general population (Walker, Newman, Dobie, Ciechanowski, & Katon, 2002). Depression. Past-week depression symptoms were measured using the Center for Epidemiologic Studies Depression Scale (CES-D). The CES-D is a widely used 20-item scale rating each item on a scale of 0 (rarely) to 4 (most or all of the time). The items represent the major components of depression (depressed mood and feelings of worthlessness or hopelessness). Scores range from 0 to 60 with higher scores indicating greater depressive symptoms, and a score of 16 or greater is suggestive of clinically significant depression. The internal consistency for this sample was .91. Quality of life. The nine-item scale was used to assess women's perceived quality of life over the past 6 months using a 7-point scale ranging from 1 (terrible) to 7 (extremely pleased). A total score is computed by taking the mean of the items. Cronbach's alpha for this sample was .88.Rollins et al. 629 Alcohol and drug abuse. Past-6-month alcohol and drug abuse were measured using the CAGE scale. The scale asks four yes/no questions, each for alcohol use behavior and drug use behavior, with each yes response scored as 1. Based on sensitivity (.78 to .81) and specificity (.76 to .92) analyses a cutoff score of 2 or more is considered the best indicator of substance abuse (Leonardson et al., 2005; Soderstrom et al., 1997). Using a cutoff score of 2 or more, we coded alcohol and drug abuse as 0 (no) or 1 (yes). Work/school absence. Past-6-month absence was measured using two items: "Did you have to take time off from work for which you were getting paid?" and "Did you have to take time off from school?" If a participant answered yes to one or both of these items, they were coded as 1 (yes). If not, they were coded as 0 (no). We also asked how many of these missed days were related to IPV. From these items, we computed one or more IPV-related days off from work or school that was also coded 1 (yes, one or more IPVrelated days off) or 0 (none). Hospital and emergency medical utilization. Past-6-month hospital/emergency medical use and reason for use was measured using three items: "Received ambulance and/or paramedic care"; "Went to the emergency room or some other urgent care facility"; and "Spent the night in a hospital." If a participant indicated that they had received one or more of these services in the past 6 months, they were coded as 1 (yes); otherwise, they were coded as 0 (no). We also asked how many of these incidents were related to IPV. From these items, we computed IPV-related hospital/emergency medical use, which was also coded 1 (yes, at least one incident was IPV-related) or 0 (no IPV-related incidences). Data Analyses The analyses examined the relationship of housing instability with victimization, PTSD, depression, substance use, quality of life, absence from work or school in general and due to IPV, and utilization of hospital/emergency medical care in general and use related to IPV. Multiple linear regressions with all variables entered into the model simultaneously were conducted to estimate the unique contribution of housing instability as reflected by the standardized regression coefficient in predicting PTSD, depression, and quality of life controlling for age, alcohol abuse, illegal drug use, and the level of danger in the abusive relationship. The distribution of the dependent variables were examined for normality prior to conducting the multiple linear regressions. Similarly, logistic regression was used to estimate the unique contribution of housing instability in predicting absence from work/school, 630 Journal of Interpersonal Violence 27(4) use of either hospital/emergency medical care in general and IPV-related use of hospital/emergency medical care. When examining the health care utilization variables, we also controlled for perceived general health and health insurance. Results Participant Characteristics Participants were English- or Spanish-speaking women living in dangerous and often unstable situations due to IPV (N = 278). Table 1 provides the descriptive statistics for the sample. Slightly more than half self-reported their race as White (53.69%) followed by African American (26.62%) and other race (20.86%). Twenty-five percent of participants self-identified as Latina, and the majority then chose to complete the interviews in Spanish. Twenty-six percent indicated that they were married. Only 4.04% reported currently living with an abusive partner. About one quarter reported less than a high school diploma, another 25% had a GED or high school diploma, and approximately another 25% had some college. Twenty-nine percent of participants were employed at the time of the interview. The majority of women (67.61%) made US$1,000 or less per month, and the majority had some type of health insurance (79.5%), primarily Medicaid. At the time of recruitment, 31.74% were receiving some type of rent assistance, 22.63% lived in their own house or apartment where they paid all of the rent, 14.96% were living in the domestic violence shelter, and 10.95% were living in a hotel. The remainder of participants was living in various situations such as transitional housing, their car or the street, or with family or friends. A small percentage (26.62%) reported one or more of the following types of homelessness in the 6 months prior to the survey; 16.55% had lived in a motel/hotel they paid for themselves, 2.88% had stayed at a homeless shelter, and 13.67% had lived on the street, in their car, or camped out. On average, women perceived their health as being fair to good health (M = 2.64, range = 1-5, with score of 5 indicating excellent health). Their mean score on the DA was 21.57, indicating extreme danger in the abusive relationship. Participants experienced symptoms consistent with PTSD (M = 55.65, score of 30 or greater indicates symptoms of PTSD) and depression (M = 31.87, score of 16 or greater indicates symptoms of depression). The CAGE classified less than 16% of the participants as alcohol or illegal drug abusers.Rollins et al. 631 Table 1. Characteristics of the Sample Variables Total N Percentage Racea 278 White 53.69 African American 26.62 American Indian/Alaska Native 7.55 Asian 1.08 Native Hawaiian/Other Pacific Islander 1.08 Other race 20.86 Hispanic/Latina 277 25.00 Currently married 278 26.00 Living with partner 278 5.40 Living with abusive partner 223 4.04 Education 277 Less than high school diploma 25.99 GED/high school diploma 23.10 Some college 27.80 Vocational graduate/associate's degree 16.61 Bachelor's degree or higher 6.50 Currently employed 278 29.00 Income 278 US$0-US$99/month 9.71 US$100-US$500/month 19.07 US$501-US$1,000/month 38.83 US$1,001-US$1,500/month 21.22 US$1,501-US$2,000/month 8.27 US$2,001-US$4,000/month 2.88 Has health insurance 278 79.50 Positive for alcohol abuse on CAGE 277 15.88 Positive for illegal drug abuse on CAGE 275 13.45 Used hospital/emergency medical service in the past 6 months 277 53.79 Had IPV-related hospital/emergency medical use in the past 6 months 277 27.80 Absent 1 or more days from work or school in the past 6 months 193 71.50 Absent 1 or more days from work or school due to IPV 191 65.45 M (SD) Number of housing instability risk factors 278 4.83 (2.24) General health 278 2.64 (1.07) (continued)632 Journal of Interpersonal Violence 27(4) Table 1. (continued) Variables Total N M (SD) Percentage Danger assessment PLC: PTSD CES-D: Depression Quality of life 278 278 277 278 21.57 (7.27) 55.65 (14.85) 31.87 (13.52) 4.10 (1.20) a. Participants could select more than one race category. In the previous 6 months, 54% of participants had accessed hospital or emergency medical services, with half reporting that the hospital or emergency medical services was needed due to IPV. More than 65% had been absent at least 1 day of work or school because of IPV. Out of a possible 10 risk factors for housing instability, on average, women reported 4.83 risk factors, with a higher number of risk factors indicating greater risk for instability. The percentage of women experiencing each housing instability risk factor is presented in Table 2. Housing Instability, Level of Danger, and Health Outcomes The relationship between housing instability and PTSD, depression, and quality of life controlling for the covariates (age, alcohol and drug use, level of danger in relationship) is presented in Table 3. All three variables were normally distributed (skewness = -.391 for PTSD, -.216 for depression and -.142 for quality of life). Greater housing instability was related to more severe PTSD, worse depression, and poorer quality of life. The level of danger as measured by the DA was also significantly associated with symptoms consistent with PTSD and higher levels of depression. Older age and screening positive for alcohol abuse were associated with more severe PTSD, higher depression, and lower quality of life (see Table 3). Housing Instability, Level of Danger, Absence From Work/ School and Health Utilization Housing instability and absence from work/school and hospital/emergency medical care utilization was examined controlling for covariates (see Table 4). With each additional risk factor for housing instability, the odds of being absent from work or school for any reason increased by 28% and absence from work or school due to IPV increased by 32%. Likewise, with each Rollins et al. 633 Table 2. Percentage of Participants Reporting Risk Factors for Housing Instability Items % Yes 1. In the past 6 months, have you had to live somewhere that you did not want to live 278 79.86 2. In the past 6 months, have you had difficulty (or were unable to) pay for your housing 276 79.35 3. Have you had trouble getting housing in the past 6 months 278 65.93 4. Do you expect that you will be able to stay in your current housing for the next 6 months (reversed) 265 56.98 5. In the past 6 months, have you had to borrow money or ask friends/family or others for money to pay your rent/mortgage payment 276 51.09 6. In the past 6 months, how many times have you moved (more than twice) 278 42.09 7. Have you had trouble with a landlord in the past 6 months 278 36.46 8. In the past 6 months, has your landlord threatened to evict you 275 28.10 9. In the past 6 months, have you been served an eviction notice 276 18.18 10. How likely is it that you will be able to pay for your housing (e.g., rent/mortgage) this month (unlikely-very unlikely) 212 16.04 Table 3. Multiple Regressions Examining the Relationship Between Housing Instability and PTSD, Depression, and Quality of Life Controlling for Covariates PLC: PTSD (N = 275) CES-D: Depression (N = 274) Quality of Life (N = 275) R2 = .23, p < .001 R2 = .12, p < .001 R2 = .08, p < .001 β p β p β p Age .18 .001 .12 .040 −.14 .018 Alcohol abuse .19 <.001 .14 .016 −.15 .013 Illegal drug abuse .06 .237 .02 .703 −.10 .086 Level of danger .29 <.001 .15 .009 −.05 .367 Housing instability .22 <.001 .22 <.001 −.15 .015 Note: All variables were entered simultaneously. Alcohol abuse and illegal drug abuse were coded 0 (no) or 1 (yes). The Danger Assessment scores ranged from 0 to 38 with higher scored indicating greater lethality. Housing instability was a count of the number of risk factors ranging from 0 to 10. The PLC ranged from 0 to 85 with higher scores indicating greater PTSD. The CES-D ranged from 0 to 60 with higher scores indicating greater depression. Quality of life ranged from 1 (terrible) to 7 (extremely pleased).634 Journal of Interpersonal Violence 27(4) additional housing instability risk factor, the odds of use of hospital/emergency medical care for any reason increased by 22%. Housing instability was not a significant predictor of the use of hospital/emergency medical care related to IPV. For each additional risk factor on the DA (proxy for severity of IPV), the odds of being absent from work or school for any reason increased by 6% and the odds of being absent from work or school due IPV increased by 6%. Similarly, the odds of hospital/emergency medical use and IPV-related hospital/emergency medical use increased by 6% and 5%, respectively, for each additional risk factor identified by participants on the DA. An increase in DA score from the "increased" danger (score of 11) to "severe" danger (score of 14) was associated with a 24% increase in missed days from school/work and hospital/emergency medical use. Each additional year of age was associated with a 4% increase in the odds of IPV-related hospital/emergency medical use. For each additional unit of improvement in general health, the odds of hospital/emergency medical use and IPV-related hospital/emergency medical use decreased by 36% and 27%, respectively. Table 4. Logistic Regressions Examining the Relationship Between Housing Instability and Medical Use and Days Off Controlling for Covariates Days Off From Work/ School (N = 191) IPV-Related Days Off From Work/School (N = 189) Hospital/ Emergency Medical Use (N = 274) IPV-Related Hospital/ Emergency Medical Use (N = 274) OR p OR p OR p OR p General health 0.64 .001 0.73 .028 Insurance 1.88 .063 1.00 .492 Age 1.01 .693 1.01 .656 1.00 .852 1.04 .021 Alcohol abuse 1.23 .696 1.31 .583 0.81 .564 1.49 .289 Illegal drug abuse 1.41 .556 0.82 .695 2.01 .107 1.36 .453 Level of danger 1.06 .011 1.06 .020 1.06 .002 1.05 .014 Housing instability 1.28 .004 1.32 .001 1.22 .001 1.11 .097 Note: All variables were entered simultaneously. Insurance, alcohol abuse, illegal drug abuse, hospital/emergency medical use, IPV-related hospital/emergency medical use, 1 or more days off from work or school, and one or more IPV-related days off from work or school were coded 0 (no) or 1 (yes). General health was coded 1 (poor), 2 (fair) 3 (good), 4 (very good), and 5 (excellent). The Danger Assessment ranged from 0 to 38 with higher scored indicating greater lethality. Housing instability was a count of the number of risk factors ranging from 0 to 10.Rollins et al. 635 Discussion The findings highlight the complex interrelated challenges many abused women face as they seek safety from an abusive partner. Women reported living with extreme violence (mean DA score = 21.57) in their intimate relationships in the past 6 months. Symptoms of PTSD were common, with a mean score of 55.65 PCL-C, and scores higher than returning war veterans who have reported mean scores of 50 (Walker et al., 2002). They report high levels of depression, with mean score of 31.87 on the C-ESD, where a score of 16 or greater is indicative of clinical depression (American Psychiatric Association, 2000). Only 26% report homelessness in the 6 months prior to accessing services from which they were recruited for this study. However, almost all participants reported several risk factors for housing instability (M = 4.83, range = 0-10). In particular, 80% reported having to live somewhere that they did not want to live in the past 6 months, 80% reported having difficulty (or were unable) to pay for housing in the past 6 months, and 42% reported having to move more than two times in the past 6 months. Our findings indicate that the greater the number of risk factors for housing instability, the more likely participants reported symptoms consistent with PTSD, depression, reduced quality of life, and increased absence from work and/or school as well as hospital and emergency department use. These outcomes associated with housing instability persist even when controlling for the level of danger in the abusive relationship and for drug and alcohol use by the survivor. Importantly, although age and alcohol abuse are significantly related to negative outcomes, housing instability and danger level had stronger associations with negative health outcomes. Both housing instability and danger level make unique contributions to negative health outcomes and could be contributing in different ways. For example, trying to secure housing may lead to absence from work/school separately from absence due to an injury or the need to obtain a restraining order. Study Limitations This is a cross-sectional analysis, and therefore we are not able to examine cause and effect. The population studied represents a particular and limited group of IPV survivors—those who sought services and who needed housing support services; thus the findings are not generalizable to all survivors of IPV. As noted above, general health for the past month was assessed using only a single item from the SF-8, "In general, would you say your health is: 636 Journal of Interpersonal Violence 27(4) 1 (poor), 2 (fair) 3 (good), 4 (very good), 5 (excellent)?" However, previous research has documented the reliability and validity of this item (Ware et al., 2001). The Housing Instability Index was developed for the study; therefore, additional research will require a validation and potential modification of the measure. Implications for Practice The findings suggest that IPV, housing instability, and health outcomes are clearly linked and that for some survivors of IPV the issues must be addressed simultaneously to improve safety and health outcomes. Clinical practitioners should include careful assessment for housing and appropriate linkage to support services, including housing programs and domestic violence agencies. Clinicians should also be trained to understand and recognize housing instability as important in the survivor's risk for PTSD, depression, physical symptoms, and medical care utilization as the severity of IPV. Health care, victim services, and housing programs need to be aware of and provide assistance to survivors in making use of existing federal and state employment laws that provide time off from work for victims of violence without fear of losing position and other safety accommodations and landlord-tenant laws that specifically address the issues of perpetrator-caused landlord problems, such as the need to break a lease to relocate for safety reasons. Implications for Policy To address the linkages between housing instability, IPV, and health, existing housing policies that address IPV survivors need to be adopted and fully implemented and enforced. This must include efforts to boost awareness of existing policies (e.g., those put forward in Violence Against Women and Department of Justice Reauthorization Act of 2005; GovTrack.us, n.d.) as well

******CLICK ORDER NOW BELOW TO GET THE ANSWER TO THIS ASSIGNMENT OR ANY OTHER ASSIGNMENT, DISCUSSION, ESSAY, HOMEWORK OR QUESTION*******."