Betrayal Trauma Among Homeless Adults

Abstract

Betrayal trauma theory postulates that traumas perpetrated by a caregiver or close other are more detrimental to mental health functioning than are traumatic experiences in which the victim is not affiliated closely with the perpetrator. This study is the first to examine the concept of betrayal among a sample of individuals with a history of homelessness. A total of 95 homeless or formerly homeless adults completed the Brief Betrayal Trauma Survey, the Posttraumatic Stress Disorder Checklist–Civilian Version, the Center for Epidemiologic Studies Depression Scale the Perceived Stress Scale, and a demographics questionnaire assessing participants’ histories of homelessness, health, and relationships with their families. Regression analyses were conducted to explore the associations between high betrayal (HB) and low betrayal (LB) trauma exposure, relationship with family, and physical and mental health symptoms. Exposure to HB traumas in childhood and poor family relationships predicted earlier age at first episode of homelessness, and participants who had been exposed to a greater number of traumas during childhood were more likely to be revictimized during adulthood. Trauma exposure as an adult and earlier age of first homeless episode predicted symptoms of posttraumatic stress disorder, while trauma exposure alone predicted symptoms of depression and perceived stress. Number of medical diagnoses was associated with trauma exposure and becoming homeless at an older age. These findings emphasize that even among the most marginalized and multiply victimized individuals in our society, traumas that are characterized by a higher degree of betrayal are associated with more adverse outcomes.

Keywords

violence exposure PTSD mental health and violence homelessness betrayal trauma theory

Homelessness is a serious public health problem that occurs at the intersection of personal, situational, demographic, economic, and familial factors (Koegel, Melamid, & Burnam, 1995). A recent point-in-time estimate reported that 636,017 homeless persons were identified in communities across the United States (National Alliance to End Homelessness, 2012), and as many as 3.5 million Americans are believed to experience homelessness in any given year (National Alliance to End Homelessness, 2009).

The lives of homeless individuals are laden with adversity, and the chaotic and often traumatic circumstances that accompany homelessness may precipitate or exacerbate mental health and physical problems. Research documents high rates of childhood and adulthood trauma among the homeless (e.g., Ryan, Kilmer, Cauce, Watanabe, & Hoyt, 2000; Stein, Leslie, & Nyamathi, 2002; Tyler, Melander, & Noel, 2009). In a sample of 372 homeless adolescents, Tyler and Cauce (2002) found that 29% and 47% of participants reported being sexually and physically abused, respectively. Childhood abuse among homeless young people has been associated with subsequent intimate partner violence (Slesnick, Erdem, Collins, Patton, & Buettner, 2010; Tyler et al., 2009); a greater likelihood of trading sex for food, shelter, money, or drugs (Tyler, Hoyt, & Whitbeck, 2000); and elevated risk of self-directed violence (Torchalla, Strehlau, Li, Schuetz, & Krausz, 2012). North, Smith, and Spitznagel (1994) discovered that more than 40% of homeless men and women confirmed at least one violent trauma in their adult lives, most commonly assault for men and rape for women. Conservative estimates report that 20% of homeless men and more than 30% of homeless women meet lifetime criteria for posttraumatic stress disorder (PTSD; North & Smith, 1992), compared with a rate of 6.8% in the general population (Kessler et al., 2005). A recent Australian study found that lifetime PTSD diagnosis among homeless adults may even exceed 50% (Taylor & Sharpe, 2008).

Lack of insurance limits access to health care, and immediate survival demands (e.g., obtaining food and shelter) compete with seeking preventive health care (Barrett, Fogel, Garrett, & Young, 2011). A large survey of patients at federally qualified health centers confirmed disparities among homeless and housed patients in the prevalence of chronic illnesses, mental health conditions, and substance use disorders (Lebrun-Harris et al., 2012). A growing body of literature also links trauma exposure to adverse health outcomes through behavioral, social, cognitive, and emotional pathways (Kendall-Tackett, 2002), all of which may be especially influential in the lives of individuals who are homeless.

Homeless individuals often describe histories of chaotic early family lives marked by a dearth of social support and abuse (Lalonde & Nadeau, 2012). Social support has been shown to exert a buffering effect against the development of PTSD (Hyman, Gold, & Cott, 2003) and may also be protective against revictimization. People who suffer abuse or neglect may have difficulties trusting other people or institutions (e.g., law enforcement), even those that may have benevolent intentions (Gobin & Freyd, 2009). This may preclude abuse survivors from cultivating healthy social support networks later in life, which may make them more vulnerable to adverse outcomes, including homelessness and revictimization. Indeed, homeless individuals who perceive themselves as having good social support are less likely to be revictimized (Hwang et al., 2009).

Betrayal Trauma Theory

Betrayal trauma theory posits that traumatic experiences vary in degree of interpersonal violation based on the relationship between the victim and the perpetrator, and that a victim will suffer greater consequences when the perpetrator is a close other (Freyd, 2012; Freyd, DePrince, & Gleaves, 2007). Traumas higher in betrayal occur “when the people or institutions on which a person depends for survival significantly violate that person’s trust or well-being” (Freyd, 2008, p. 76). For instance, sexual or physical abuse by a caretaker or intimate partner would be considered a high betrayal (HB) trauma, while traumas that are not interpersonal in nature (e.g., natural disasters) or are perpetrated by aggressors with whom there is little or no relationship (e.g., a stranger) are regarded as low betrayal (LB) traumas.

According to betrayal trauma theory, when trauma is perpetrated by a close other, especially a caretaker, a child is placed in an unsolvable bind. To deem a caretaker culpable yields great risk, as the child is dependent on the adult perpetrator to have his or her basic survival needs met (Goldsmith, Barlow, & Freyd, 2004). Moreover, leaving the abusive situation may result in loss of access to life-sustaining resources, ultimately resulting in homelessness. To avoid these situations, a child may experience a form of unknowing (e.g., dissociation), an adaptive blindness to the betrayal, which enables him or her to continue to trust and to rely on the caretaker for survival needs (Freyd, 1997). While this lack of awareness offers a short-term solution, it carries with it psychological risks. Children abused by an immediate or an extended family member tend to suffer greater internalizing problems than children who were victimized by a distant other (Tremblay, Hébert, & Piché, 1999). Likewise, HB traumas are associated with greater symptoms of PTSD, depression, and anxiety than are LB traumas (Goldsmith, Freyd, & DePrince, 2012; Tang & Freyd, 2012). As trauma survivors grow older and become increasingly self-reliant, they “may become conscious of strong feelings of betrayal” (DePrince & Freyd, 2002, p. 75), leading to a distrust of other people, and even institutions, on which they rely (Gobin & Freyd, 2009). Indeed, institutional betrayal adds yet another layer of victimization. For example, college-aged women who reported coercive or unwanted sexual experiences that were followed by failure of an institution, such as a university, to take action to respond to the event, suffered increased posttraumatic symptoms (Smith & Freyd, 2013). Disabling psychological symptoms, coupled with lack of trust, may be another path to homelessness.

Current Study

Although betrayal and its relationship with trauma sequelae have been researched in a variety of populations, including undergraduates and community samples, this concept has not been investigated in homeless populations. Moreover, scant research has explored the relationship of homeless individuals to perpetrators who victimized them, and the emotional and physical health ramifications of the proximity of those relationships have not been elucidated. Tyler and Cauce (2002) found that among homeless adolescents with a history of physical abuse, the majority had been harmed for the first time by a family member, most commonly a biological parent. Of those who had been sexually abused during childhood, 35% were victimized first by either a relative, an older sibling, or a biological parent (Tyler & Cauce, 2002). To date, a limited number of studies have reported that traumas with a higher degree of betrayal are more strongly associated with mental and physical health difficulties (Edwards, Freyd, Dube, Anda, & Felitti, 2012; Freyd, Klest, & Allard, 2005; Goldsmith et al., 2012), but this has not been examined among homeless individuals. The present study sought to investigate the association between self-reported trauma exposure and various aspects of physical and mental health in a sample of homeless adults. Three hypotheses based on betrayal trauma theory were proposed:

Hypothesis 1: Childhood exposure to HB trauma and poorer relationships with family will be associated with younger age of first homelessness episode.

Hypothesis 2: Exposure to LB and HB childhood traumas and age at first homeless episode will predict exposure to trauma in adulthood.

Hypothesis 3: Age of first homeless episode and exposure to trauma in adulthood will predict symptoms of PTSD, depression, and perceived stress, as well as number of medical diagnoses.

Method

Participants

Participants were recruited between April 2009 and February 2011 using fliers posted at an outpatient clinic that provides services to homeless and impoverished persons in Miami, Florida, and through word-of-mouth in the homeless community. Participants were English-speaking adults (N = 94) ages 18 to 65 years who were homeless or who reported a history of homelessness (see Table 1 for demographic information). The measures were part of a larger neuropsychological test battery. Prospective participants were excluded if they reported using drugs or alcohol in the 8 hours prior to participation or if they were suspected to be under the influence. Four participants who initiated the consent process withdrew participation midway through the study, citing conflicting obligations (e.g., need to retrieve personal belongings from a day program before it closed for the day). No participants who began the consent process were excluded on the basis of substance use concerns.

Table 1.

Descriptive Statistics for Study Participants (N = 94).

Variable	n	%
Age
18-24	3	3.2
25-34	14	14.8
35-44	12	12.8
45-54	46	48.9
55-65	19	20.2
Gender
Female	23	24.5
Male	71	75.5
Ethnicity
Black, African American, or Afro-Caribbean	52	55.3
Hispanic or Latino	14	14.9
Caucasian	21	22.3
Other	7	7.4
Sexual orientation
Heterosexual	86	91.5
Gay or lesbian	2	2.1
Bisexual	6	6.4
Marital status
Single, never married	51	54.3
Married or partnered	5	5.3
In a relationship	8	8.5
Separated or divorced	27	28.7
Widowed	3	3.2
Employment status
Working part-time	7	7.4
Looking for work	28	29.8
Able to work but not looking for work	7	7.4
Disabled	38	40.4
Retired	3	3.2
Other	11	11.7
Education
8th grade or less	4	4.3
9th-11th grade	36	38.3
High school or GED	27	28.7
Some college	17	18.1
College graduate	10	10.6

Note: GED = General Equivalency Diploma

Measures

Assessments were administered by the principal investigator (first author) and a team of research assistants, all of whom were doctoral-level graduate students in a clinical psychology program who had received prior training administering the assessment battery.

Brief Betrayal Trauma Survey (BBTS)

The BBTS (Goldberg & Freyd, 2006) inquires about 12 traumatic events an individual may have endured, each rated on a 3-point scale (i.e., never, 1 or 2 times, more than that). Two time intervals are queried—before and after age 18. Good psychometric properties have been reported (Goldberg & Freyd, 2006). Responses were recoded into dichotomous categories of “never” and “1 or more times” and summed (Freyd et al., 2005). Items describing traumas not perpetrated by another person or involving a perpetrator with whom the victim was not close are considered LB, whereas interpersonal traumas perpetrated by a close other are considered HB. Total scores were derived for traumas experienced during both time intervals, as well as for subscale scores for HB and LB traumas in each age range.

PTSD Checklist–Civilian Version (PCL-C)

The PCL-C (Weathers, Litz, Herman, Huska, & Keane, 1993) is a 17-item inventory that assesses for PTSD symptoms (Diagnostic and Statistical Manual of Mental Disorders [4th ed., text rev.]; DSM-IV-TR; American Psychiatric Association, 2000, criteria) and quantifies the severity of symptoms during the preceding month. Symptoms are rated on a 5-point Likert-type scale (i.e., not at all, a little bit, moderately, quite a bit, extremely), with total scores ranging from 17 to 85. The PCL-C is used as a screening measure but has demonstrated 90% diagnostic efficiency when compared with the Clinician Administered PTSD Scale (Blanchard, Jones-Alexander, Buckley, & Forneris, 1996). It has good test–retest reliability and internal consistency (Ruggiero, Del Ben, Scotti, & Rabalais, 2003; Weathers et al., 1993).

Center for Epidemiologic Studies Depression Scale (CES-D)

The CES-D (Radloff, 1977) is a 20-item measure designed to assess depressive symptomatology in the prior week. Items are rated on a 4-point Likert-type scale (i.e., rarely or none of the time, a little or some of the time, occasionally or a moderate amount of the time, most or all of the time). Scores range from 0 to 60, with higher scores indicating more severe symptoms. The CES-D has demonstrated good psychometric properties in community and psychiatric settings (Radloff, 1977), as well as in homeless samples (LaGory, Ritchey, & Mullis, 1990; Wong, 2000).

Perceived Stress Scale (PSS)

The PSS (Cohen, Kamarck, & Mermelstein, 1983; Cohen & Williamson, 1988) is a 10-item measure designed to gauge the degree to which a person has experienced life as uncontrollable and/or unpredictable in the last month. Items are rated on a 5-point Likert-type scale (i.e., never, almost never, sometimes, fairly often, very often). Scores range from 0 to 40. Higher scores indicate a greater degree of stress. Among a nationally representative sample of 2000 participants, the average score was 15.52 (SD = 7.44; Cohen & Janicki-Deverts, 2012). Good psychometric properties have been reported (Roberti, Harrington, & Storch, 2006).

Demographic and personal information

A questionnaire was utilized to gather demographic and background information and participants’ histories of homelessness. Items used in the current analyses included age at first homeless, quality of relationship with family (i.e., great, okay, poor, I do not have a relationship), and medical conditions. Chronic conditions included arthritis, diabetes, hypertension, high cholesterol, HIV/AIDS, physical disability, sexually transmitted disease (not including HIV), seizures/epilepsy, tuberculosis, and “other.” A variable was created by summing the number of number of reported medical diagnoses.

Research Ethics

Research ethics of relevance to this population were considered (Runnels, Hay, Sevigny, & O’Hara, 2009). Materials were written at no higher than a sixth-grade reading level. A synopsis of key points in the informed consent form was provided verbally to each participant, and the complete form was read aloud, unless a participant indicated a preference for reading it independently. To thank them for contributing to the study, participants were provided a gift card to a nearby sandwich shop and an up-to-date list of nearby shelters, mental health resources, and services. The Institutional Review Board at Nova Southeastern University approved this study.

Data Analysis

Fewer than 1% of data points were missing, and missing data were handled differently for each measure. Missing data on the BBTS were coded as 0 (no exposure; see Goldsmith et al., 2012). One participant’s age at first homeless episode was missing, and that participant was excluded from analyses where that variable was considered. On the PCL-C and the CES-D, missing values were replaced using two-way imputation, in which a participant’s other scores on the same measure, as well as other participants’ scores on the missing item, are used to estimate scores for missing items (Sijtsma & van der Ark, 2003; van Ginkel & van der Ark, 2005). For regression analyses, all variables were entered into the models simultaneously as no particular variable or set of variables was considered to be more or less relevant in testing our hypotheses. Indices of collinearity were well within acceptable limits, except as noted.

Results

Participants’ mean age at the time of their first homeless episode was 35.9 years (SD = 12.4; range = 14-63 years). Twenty percent of participants became homeless for the first time during adolescence or young adulthood (i.e., by the age of 25 years). Participants had been homeless for a total of 6.1 years (SD = 6.8), and most (64.9%) had experienced multiple episodes of homelessness. All but two participants (97.9%) reported at least one of the traumatic events queried by the BBTS. Because events were assessed for two age categories (before and after age 18), scores ranged from 0 to 24. The mean number of traumas reported was 8.7 (SD = 5.7). Most participants reported one or more LB traumas during childhood (68.1%) and adulthood (85.1%), with 90.4% reporting at least one lifetime LB trauma (see Table 2). Similarly, at least one lifetime HB trauma was reported by 87.9% of the sample, with 70.2% reporting one or more childhood HB traumas and 72.3% reporting at least one adulthood HB trauma. Participants were multiply exposed, with 47.9% and 66.0% reporting exposure to four or more types of traumas during childhood and adulthood, respectively.

Table 2.

Exposure to LB and HB Traumas During Childhood and Adulthood.

	<18 years old		≥18 years old
Traumas endorsed on the BBTS	n	%	n	%
Lower betrayal
Been in a natural disaster (e.g., fire, flood, hurricane)	27	28.7	39	41.5
Been in a major accident (e.g., automobile, boat, bus)	33	35.1	40	42.6
Witnessed someone with whom you were not very close harming themselves or being harmed severely	34	36.2	57	60.6
Deliberately attacked very severely by someone with whom you were not very close	33	35.1	51	54.3
Made to have sexual contact (e.g., touching, penetration) by someone with whom you were not very close	15	16.0	22	23.4
High betrayal
Witnessed someone with whom you were very close harming themselves or being harmed very severely	36	38.3	52	55.3
Witnessed someone with whom you were very close deliberately attack another family member severely	40	42.6	33	35.1
Deliberately attacked very severely by someone with whom you were very close	38	40.4	44	46.8
Made to have sexual contact (e.g., touching, penetration) by someone with whom you were very close	24	25.5	21	22.3
Emotionally/psychologically mistreated over a period of time by someone with whom you were very close	38	40.4	46	48.9
Items not included in subscales
Experienced the death of your own child	13	13.8	14	14.9
Other trauma not covered	28	29.8	37	39.4

Note. LB = low betrayal; HB = high betrayal; BBTS = Brief Betrayal Trauma Survey.

Mean scores on symptom measures were high and fell in the clinically significant range (see Table 3). A majority of participants (77%) had scores greater than 16 on the CES-D, and 40% scored above 50 on the PCL-C. Scores on the PSS indicated high levels of perceived stress (M = 21.0, SD = 7.5), with 79.8% of the sample scoring higher than average (Cohen & Janicki-Deverts, 2012). Most participants (75.5%) indicated they had been diagnosed with one or more of the major medical conditions surveyed, and more than half (55.3%) endorsed at least two diagnoses.

Table 3.

Descriptive Statistics (N = 94).

Variable	M	SD
Adulthood trauma total	4.8	3.1
Adulthood LB trauma	2.2	1.5
Adulthood HB trauma	2.1	1.7
Childhood LB trauma	1.5	1.4
Childhood HB trauma	1.9	1.7
PCL-C	46.5	17.4
CES-D	27.5	13.1
PSS	21.0	7.51
Number of medical diagnoses	2.00	1.6

Note. LB = low betrayal; HB = high betrayal; PCL-C = PTSD Checklist–Civilian Version; CES-D = Center for Epidemiologic Studies Depression Scale; PSS = Perceived Stress Scale; PTSD = posttraumatic stress disorder.

Hypothesis 1

Childhood exposure to HB trauma and poorer relationship with family were hypothesized to be associated with younger age at first homeless episode. As hypothesized, exposure to more childhood HB trauma (β = −.33, t = −2.53, p < .05) and poorer self-reported relationship with family (β = .32, t = 3.03, p < .01) each predicted younger age of first homeless episode. Exposure to childhood LB trauma was not a significant predictor (β = .07, t = .52, p > .10).

Hypothesis 2

It was hypothesized that exposure to childhood trauma and age of first homeless episode would predict exposure to trauma in adulthood. Exposure to a greater number of childhood HB (β = .31, t = 2.89, p < .01) and LB traumas (β = .37, t = 3.43, p < .01) and younger age of first homeless episode (β = −.19, t = −2.22, p < .05) each predicted more exposure to LB trauma as an adult. In partial support of this hypothesis, exposure to more childhood HB trauma (β = .61, t = 6.23, p < .01), but not LB trauma (β = .14, t = 1.45, p > .10) or age at first homelessness (β = −.01, t = −.18, p > .10), predicted more exposure to HB trauma during adulthood.

Hypothesis 3

Becoming homeless at a younger age and being exposed to more traumas during adulthood were hypothesized to predict worse mental health symptoms and greater numbers of medical diagnoses. First, attempts were made to explore the independent effects of HB trauma and LB trauma experienced during adulthood on current symptoms by entering BBTS subscales as separate variables in regression models predicting current symptoms. By entering those subscales of the BBTS as separate variables in regression models predicting current symptoms. Although the overall models were significant, individual betas were not, a situation that may occur when multicollinearity exists (Friedman & Wall, 2005). Although collinearity diagnostic tests were well within acceptable limits (variance inflation factor = 2.03-2.04), a correlation matrix revealed that adult HB trauma and adult LB trauma had very similar correlations with symptom measures (see Table 4). Thus, there was concern that measuring those subscales separately was redundant in this sample. To remedy this problem, a total score was calculated by summing the number of traumas during adulthood, and this variable was used in place of subscale scores in regression models predicting current symptoms.

Table 4.

Pearson r Correlations Among Trauma Measures and Current Symptoms (N = 94).

Variable	Adulthood LB trauma	Adulthood HB trauma	Adulthood total trauma
PCL-C	.35**	.30**	.32**
CES-D	.35**	.33**	.39**
PSS	.25*	.30**	.27**
Number of medical diagnoses	.26*	.28**	.28**

Note. LB = low betrayal; HB = high betrayal; PCL-C = PTSD Checklist–Civilian Version; CES-D = Center for Epidemiologic Studies Depression Scale; PSS = Perceived Stress Scale; PTSD = posttraumatic stress disorder.

p < .05. **p < .01.

Separate regression analyses were conducted with the PCL-C, CES-D, PSS, and number of medical diagnoses (see Table 5). For PTSD symptoms, the hypothesis was supported. Exposure to more trauma as an adult and younger age of first homeless episode each predicted more severe PTSD symptoms. Trauma exposure, but not age of first homeless episode, predicted worse symptoms of depression and stress. Exposure to a greater number of traumas was associated with endorsing more medical diagnoses, but, contrary to our hypothesis, so was older age of first homeless episode.

Table 5.

Associations Among Trauma Exposure, Age of First Homeless Episode, and Current Symptoms (N = 94, df = 2, 92).

			Adulthood trauma	Age of first homeless episode
Variable	F	R	β	β
PCL-C	10.36**	.43	.39**	−.20*
CES-D	6.41**	.35	.32**	−.15
PSS	4.18*	.29	.28**	−.09
Number of medical diagnoses	6.61**	.36	.26*	.24*

Note. PCL-C = PTSD Checklist–Civilian Version; CES-D = Center for Epidemiologic Studies Depression Scale; PSS = Perceived Stress Scale; PTSD = posttraumatic stress disorder.

p < .05. **p < .01.

A set of post hoc regression analyses were conducted to determine whether duration of homelessness rather than age at first episode of homelessness predicted symptoms in the hypothesized direction. Trauma exposure but not duration of homelessness significantly predicted worse symptoms of PTSD, depression, and stress, and more medical diagnoses (see Table 6).

Table 6.

Associations Among Trauma Exposure, Duration of Homelessness, and Current Symptoms (N = 94, df = 2, 92).

			Adulthood trauma	Duration of homelessness
Variable	F	R	β	β
PCL-C	8.30**	.39	.39**	.08
CES-D	5.32**	.32	.32**	.02
PSS	4.11*	.29	.27**	.09
Number of medical diagnoses	4.70*	.31	.28**	.13

Note. PCL-C = PTSD Checklist–Civilian Version; CES-D = Center for Epidemiologic Studies Depression Scale; PSS = Perceived Stress Scale; PTSD = posttraumatic stress disorder.

p < .05. **p < .01.

Discussion

The purpose of this study was to examine the relationship among traumas with varying degrees of betrayal and psychological symptoms and medical diagnoses in a sample of adults who were at the time of the study or had been homeless. As expected, participants endorsed very high rates of trauma exposure during childhood and adulthood. Being exposed to multiple types of traumas was the rule rather than the exception in this sample, and participants endorsed high scores on the PCL-C and the CES-D. Likewise, scores on the PSS were substantially higher than those found in the general population, which aligns with prior research that demonstrated a relationship between lower socioeconomic status and elevated levels of stress (Cohen & Janicki-Deverts, 2012).

Exposure to a greater number of HB traumas during childhood and a poorer relationship with family predicted younger age at first episode of homelessness. Given that HB traumas are perpetrated by someone close to the victim, it is likely that family members were responsible for at least a portion of the violence to which participants were exposed early in their lives. As predicted, LB traumas were not associated with earlier onset of homelessness. LB traumas may or may not be interpersonal in nature and can occur in chaotic families, as well as in families where a greater degree of stability exists.

Exposure to a greater number of LB traumas during adulthood was predicted by HB and LB traumas during childhood, a finding that is consistent with research on revictimization (McIntyre & Widom, 2011; Widom, Czaja, & Dutton, 2008). Younger age at first homeless episode also predicted LB traumas in adulthood and was associated with a greater total number of years being homeless (r = −.30, p = .003). It is not surprising that participants who were homeless longer were exposed to more LB traumas, given that homeless individuals are typically in closer physical proximity to potential offenders, and the likelihood of repeat victimization increases with duration of homelessness (Tyler, Gervais, & Davidson, 2012). Consistent with the work of Gobin and Freyd (2009), experiencing HB but not LB traumas during childhood predicted exposure to HB traumas during adulthood, indicating, tragically, that circumstances on the streets yield further betrayal, victimization, and violence rather than a reprieve from the abusive familial relationships that were endured at home (Whitbeck, Hoyt, & Ackley, 1997). Zurbriggen and Freyd (2004) asserted that exposure to traumas high in betrayal may be detrimental to cognitive mechanisms that enable an individual to make self-protective decisions later in life. For example, a survivor of HB traumas may be less astute at recognizing an unhealthy relationship and, consequently, less likely to make the choice to remove himself or herself from a situation that may risk further victimization. This mechanism may help to explain the high rates of recurrent HB trauma in this sample. The sequelae of betrayal trauma may also make it more difficult for those who have been victimized to establish a healthy social network, which would help to insulate an individual against adverse outcomes such as homelessness. These findings corroborate those of prior studies, lending further support to the tenets of betrayal trauma theory.

The hypothesis that trauma exposure and earlier onset of homelessness would predict more severe PTSD symptoms was supported. This is consistent with recent research that indicates greater exposure to trauma and poorer socioeconomic situations are related to PTSD (Klest, Freyd, & Foynes, 2012). Perceived stress and depression were predicted by trauma exposure only and were not significantly associated with age of first homeless episode. On examining these findings, we hypothesized that duration of time homeless may be a more influential factor than age of first homeless episode, as the effects of homelessness (e.g., victimization, social and economic deprivation) are likely cumulative over time. However, this alternative hypothesis was also not supported. These results were not consistent with predictions, and several explanations are offered. Earlier onset of homelessness may be associated with symptoms of stress and depression, but the effect size may be too small to be detected in this sample. Alternatively, and perhaps more likely, homelessness itself may be a risk factor for stress and depression, regardless of when in life it occurs or for how long it lasts. Variation in symptom severity may be better explained by the additive impact of stressors and traumas above and beyond the shared stressor of homelessness.

It was predicted that a greater number of traumas would be associated with poorer health, which was, indeed, the case. A growing literature links mood disorders with hyperarousal of the sympathetic nervous system. Psychosocial stressors, including trauma, may have an impact on the development of mood disorders but may also be pathways to inflammatory health conditions (e.g., cardiovascular disease and diabetes; Currier & Nemeroff, 2010; Miller, Maletic, & Raison, 2009). It was also expected that individuals who became homeless younger would report a greater number of medical conditions, because the harsh conditions on the street may have a deleterious effect on health status; however, this was not the case. Older age at the time of first homeless episode was associated with more medical diagnoses. Of course, older adults are more likely to suffer certain health conditions; thus, more health problems might be expected among an older cohort. The supposition is supported by Schanzer, Dominguez, Shrout, and Caton (2007) who found that many newly homeless adults had medical and psychiatric conditions prior to becoming homeless. It is possible that for at least some individuals in our sample, poor health was more a contributor to, rather than an effect of, homelessness. Individuals who become homeless later in life may have been burdened and/or made destitute by poor health through interactive effects of comorbid mental health conditions, limited financial resources, and poor social support. Moreover, it is possible that individuals who become homeless later have remained for longer periods of time with his or her abuser(s). This supposition is consistent with the tenets of betrayal trauma theory, which suggest that cognitive coping mechanisms, such as dissociation, may serve a survival function for the victim by allowing him or her to remain attached to an individual on whom he or she is dependent (Zurbriggen & Freyd, 2004).

This study has a number of limitations that warrant discussion. Participants were a convenience sample and were recruited at a primary care clinic, which may have led to a biased recruiting of individuals who had greater health needs or who presumably had better access to medical and behavioral health services. Likewise, excluding individuals who had used alcohol or drugs in the 8 hours prior to study participation may have led some prospective participants to choose not to participate. Thus, this sample may not be representative of the larger homeless population, and findings may not generalize to other, more socially isolated individuals or to those who have more severe substance use difficulties. Because it was cross-sectional in design and retrospective in nature, this study may have involved over- or under-reporting of traumas and family difficulties, particularly given that participants were asked to recall events from their childhoods. In addition, although participants were reassured that their responses were confidential, some participants may have been reluctant to disclose fully the extent of their past victimization, given the sensitive and potentially stigmatizing nature of the topic. Our sample size was also small, which precluded more complex analyses and limited its power to examine differences between subgroups (e.g., males vs. females). Despite these limitations, this is the first study of which we are aware to explore betrayal in relation to psychological and physical health functioning in a homeless population.

Conclusion

These results support betrayal trauma theory and substantiate findings of recent research that notes the relationship between HB trauma exposure and adverse long-term physical and mental health outcomes (Edwards et al., 2012; Gobin & Freyd, 2009). This study adds uniquely to this burgeoning area of research by introducing betrayal as a potentially important factor in understanding symptoms among homeless adults and by demonstrating that among individuals who are homeless—the most marginalized and vulnerable individuals in society—those persons whose lives were marred by victimization perpetrated at the hands of caregivers tend to fare worse psychologically and to suffer more victimization later in their lives. These findings underscore the importance of early intervention after trauma to prevent long-term adverse outcomes, including homelessness. In effect, what would be considered a tertiary intervention for trauma exposure becomes a primary or even secondary prevention for homelessness. In addition, these findings add to the body of literature on trauma among individuals who are homeless and have implications for frontline service providers. Physicians, mental health providers, social workers, and other service providers (e.g., shelter staff) may benefit from education regarding the emerging findings about betrayal trauma and its relationship with mental and physical health.

Footnotes

Acknowledgements

First and foremost, we wish to thank the individuals who kindly offered their time to participate in this study. We also thank Jessica Karle, Robyn Hardie, Cynthia King, Tracy Chisholm, Patrick Aragon, Rachel Orleck, and Alexandra Bodden for their assistance with data collection, as well as Dr. Frederick P. Rivara for his editorial commentary on an earlier draft of this manuscript.

Authors’ Note

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interests with respect to the authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Dr. Mackelprang received fellowship support from the National Institute of Child Health and Human Development (T32HD057822) during the preparation of this article.

Author Biographies

Jessica L. Mackelprang, PhD, is a postdoctoral research fellow in the Department of Pediatrics at the Harborview Injury Prevention and Research Center at the University of Washington (UW). She earned a doctorate in clinical psychology from Nova Southeastern University (NSU) and completed a predoctoral internship in behavioral medicine and neuropsychology in the Department of Psychiatry and Behavioral Sciences at the UW. Prior to her present fellowship, she completed a clinical postdoctoral fellowship in rehabilitation psychology at Harborview Medical Center. Her current research explores the prevention and sequelae of self-directed violence and unintentional injury among medically vulnerable and underserved populations.

Bridget Klest, PhD, is assistant professor of psychology at University of Regina in Saskatchewan, Canada. Her current research explores the interrelations among traumatic stress, health, and social context variables such as socioeconomic status.

Shadae J. Najmabadi, MS, is a doctoral student in clinical psychology at NSU. Her primary research interest is in school violence prevention. After completing graduate school, she intends to pursue a career working with children and adolescents.

Sarah Valley-Gray, PsyD, is a licensed psychologist, certified school psychologist, associate professor, and director of continuing education and special projects at the Center for Psychological Studies at NSU. She led the development of the specialist program (2001) and doctoral program (2011) in school psychology at NSU. Her current research is in the area of clinical supervision, the assessment of competency within professional psychology, and the use of standardized patients to evaluate clinical skills.

Efrain A. Gonzalez, PsyD, ABPP, MSClinPharm, is the director of the Medical Psychology/Consultation Liaison Service at the University of Miami/Jackson Memorial Medical Center. He is also an attending psychologist at Camillus Health Concern and an adjunct professor at NSU. His clinical and research interests are in the integration of psychoanalytic theory and behavioral medicine, diagnosis and treatment of adult patients who are homeless, neuropsychological functioning among homeless persons, and psychopharmacology.

Ralph E. (Gene) Cash, PhD, NCSP, is professor of psychology at NSU’s Center for Psychological Studies, director of NSU’s school psychology clinic, former president of the Florida Association of School Psychologists (FASP), and former president of the National Association of School Psychologists. Gene is one of the few recipients of FASP’s Lifetime Achievement Award for his service to the profession. His current research involves suicide prevention, training of psychology students using simulated patients, fostering critical thinking in elementary school students, and assessing effects of public policy advocacy training on advocacy behaviors.

References

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

Barrett

Fogel

S. J.

Garrett

Young

(2011). Assessing health care needs among street homeless and transitionally housed adults. Journal of Social Service Research, 37, 338-350. doi:10.1080/01488376.2011.564073

Blanchard

E. B.

Jones-Alexander

Buckley

T. C.

Forneris

C. A.

(1996). Psychometric properties of the PTSD Checklist (PCL). Behaviour Research and Therapy, 34, 669-673. doi:10.1016/0005-7967(96)00033-2

Cohen

Janicki-Deverts

(2012). Who’s stressed? Distributions of psychological stress in the United States in probability samples from 1983, 2006, and 2009. Journal of Applied Social Psychology, 42, 1320-1334. doi:10.1111/j.1559-1816.2012.00900.x

Cohen

Kamarck

Mermelstein

(1983). A global measure of psychological stress. Journal of Health and Social Behavior, 24, 385-396. doi:10.2307/2136404

Cohen

Williamson

(1988). Psychological stress in a probability sample of the United States. In Spacapan

Oskamp

(Eds.), The social psychology of health: Claremont symposium on applied social psychology (pp. 31-67). Newbury Park, CA: Sage.

Currier

M. B.

Nemeroff

C. B.

(2010). Inflammation and mood disorders: Proinflammatory cytokines and the pathogenesis of depression. Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 9, 212-220.

DePrince

A. P.

Freyd

J. J.

(2002). The harm of trauma: Pathological fear, shattered assumptions, or betrayal? In Kauffman

(Ed.), Loss of the assumptive world: A theory of traumatic loss (pp. 71-82). New York, NY: Brunner-Routledge.

Edwards

V. J.

Freyd

J. J.

Dube

S. R.

Anda

R. F.

Felitti

V. J.

(2012). Health outcomes by closeness of sexual abuse perpetrator: A test of betrayal trauma theory. Journal of Aggression, Maltreatment & Trauma, 21, 133-148. doi:10.1080/10926771.2012.648100

10.

Freyd

J. J.

(1997). Violations of power, adaptive blindness and betrayal trauma theory. Feminism & Psychology, 7, 22-32.

11.

Freyd

J. J.

(2008). Betrayal trauma. In Reyes

Elhai

J. D.

Ford

J. D.

(Eds.), Encyclopedia of psychological trauma (p. 76). New York, NY: John Wiley.

12.

Freyd

J. J.

(2012). What is a betrayal trauma? What is betrayal trauma theory? Retrieved from http://dynamic.uoregon.edu/jjf/defineBT.html

13.

Freyd

J. J.

DePrince

A. P.

Gleaves

D. H.

(2007). The state of betrayal trauma theory: Reply to McNally—Conceptual issues and future directions. Memory, 15, 295-311.

14.

Freyd

J. J.

Klest

Allard

C. B.

(2005). Betrayal trauma: Relationship to physical health, psychological distress, and a written disclosure intervention. Journal of Trauma & Dissociation, 6, 83-104. doi:10.1300/J229v06n03_04

15.

Friedman

Wall

(2005). Graphical views of suppression and multicollinearity in multiple linear regression. The American Statistician, 59(2), 127-136.

16.

Gobin

R. L.

Freyd

J. J.

(2009). Betrayal and revictimization: Preliminary findings. Psychological Trauma: Theory, Research, Practice, and Policy, 1, 242-257. doi:10.1037/a0017469

17.

Goldberg

L. R.

Freyd

J. J.

(2006). Self-reports of potentially traumatic experiences in an adult community sample: Gender differences and test–retest stabilities of the items in a Brief Betrayal-Trauma Survey. Journal of Trauma & Dissociation, 7, 39-63. doi:10.1300/J229v07n03_04

18.

Goldsmith

R. E.

Barlow

M. R.

Freyd

J. J.

(2004). Knowing and not knowing about trauma: Implications for therapy. Psychotherapy: Theory, Research, Practice, Training, 41, 448-463. doi:10.1037/0033-3204.41.4.448

19.

Goldsmith

R. E.

Freyd

J. J.

DePrince

A. P.

(2012). Betrayal trauma: Associations with psychological and physical symptoms in young adults. Journal of Interpersonal Violence, 27, 547-567.

20.

Hwang

S. W.

Kirst

M. J.

Chiu

Tolomiczenko

Kiss

Cowan

Levinson

(2009). Multidimensional social support and the health of homeless individuals. Journal of Urban Health, 86, 791-803. doi:10.1007/s11524-009-9388-x

21.

Hyman

S. M.

Gold

S. N.

Cott

M. A.

(2003). Forms of social support that moderate PTSD in childhood sexual abuse survivors. Journal of Family Violence, 18, 295-300. doi:10.1023/A:1025117311660

22.

Kendall-Tackett

(2002). The health effects of childhood abuse: Four pathways by which abuse can influence health. Child Abuse & Neglect, 26, 715-729.

23.

Kessler

R. C.

Berglund

Demler

Jin

Merikangas

K. R.

Walters

E. E.

(2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62, 593-602. doi:10.1001/archpsyc.62.6.593

24.

Klest

Freyd

J. J.

Foynes

M. M.

(2012). Trauma exposure and posttraumatic symptoms in Hawaii: Gender, ethnicity, and social context. Psychological Trauma: Theory, Research, Practice, and Policy. Advance online publication. doi:10.1037/a0029336

25.

Koegel

Melamid

Burnam

M. A.

(1995). Childhood risk factors for homelessness among homeless adults. American Journal of Public Health, 85, 1642-1649.

26.

LaGory

Ritchey

F. J.

Mullis

(1990). Depression among the homeless. Journal of Health and Social Behavior, 31, 87-102. doi:10.2307/2137047

27.

Lalonde

Nadeau

(2012). Risk and protective factors for comorbid posttraumatic stress disorder among homeless individuals in treatment for substance-related problems. Journal of Aggression, Maltreatment & Trauma, 21, 626-645. doi:10.1080/10926771.2012.694401

28.

Lebrun-Harris

L. A.

Baggett

T. P.

Jenkins

D. M.

Sripipatana

Sharma

Hayashi

A. S.

Ngo-Metzger

(2012). Health status and health care experiences among homeless patients in federally supported health centers: Findings from the 2009 patient survey. Health Services Research. Advance online publication. doi:10.1111/1475-6773.12009

29.

McIntyre

J. K.

Widom

C. S.

(2011). Childhood victimization and crime victimization. Journal of Interpersonal Violence, 26, 640-663. doi:10.1177/0886260510365868

30.

Miller

A. H.

Maletic

Raison

C. L.

(2009). Inflammation and its discontents: The role of cytokines in the pathophysiology of major depression. Biological Psychiatry, 65, 732-741. doi:10.1016/j.biopsych.2008.11.029

31.

National Alliance to End Homelessness. (2009). Homelessness looms as potential outcome of recession. Retrieved from http://www.endhomelessness.org/content/general/detail/2161

32.

National Alliance to End Homelessness. (2012). The state of homelessness in America 2012: A research report on homelessness. Retrieved from http://www.endhomelessness.org/library/entry/the-state-of-homelessness-in-america-2012

33.

North

C. S.

Smith

E. M.

(1992). Posttraumatic stress disorder among homeless men and women. Hospital and Community Psychiatry, 43, 1010-1016.

34.

North

C. S.

Smith

E. M.

Spitznagel

E. L.

(1994). Violence and the homeless: An epidemiologic study of victimization and aggression. Journal of Traumatic Stress, 7, 95-110. doi:10.1002/jts.2490070110

35.

Radloff

L. S.

(1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401. doi:10.1177/014662167700100306

36.

Roberti

J. W.

Harrington

L. N.

Storch

E. A.

(2006). Further psychometric support for the 10-item version of the Perceived Stress Scale. Journal of College Counseling, 9, 135-147. doi:10.1002/j.2161-1882.2006.tb00100.x

37.

Ruggiero

K. J.

Del Ben

Scotti

J. R.

Rabalais

A. E.

(2003). Psychometric properties of the PTSD Checklist–Civilian Version. Journal of Traumatic Stress, 16, 495-502. doi:10.1023/A:1025714729117

38.

Runnels

Hay

Sevigny

O’Hara

(2009). The ethics of conducting community-engaged homelessness research. Journal of Academic Ethics, 7, 57-68. doi:10.1007/s10805-009-9083-2

39.

Ryan

K. D.

Kilmer

R. P.

Cauce

A. M.

Watanabe

Hoyt

D. R.

(2000). Psychological consequences of child maltreatment in homeless adolescents: Untangling the unique effects of maltreatment and family environment. Child Abuse & Neglect, 24, 333-352.

40.

Schanzer

Dominguez

Shrout

P. E.

Caton

C. L.

(2007). Homelessness, health status, and health care use. American Journal of Public Health, 97, 464-469. doi:10.2105/AJPH.2005.076190

41.

Sijtsma

van der Ark

L. A.

(2003). Investigation and treatment of missing item scores in test and questionnaire data. Multivariate Behavioral Research, 38, 505-528. doi:10.1207/s15327906mbr3804_4

42.

Slesnick

Erdem

Collins

Patton

Buettner

(2010). Prevalence of intimate partner violence reported by homeless youth in Columbus, Ohio. Journal of Interpersonal Violence, 25, 1579-1593.

43.

Smith

C. P.

Freyd

J. J.

(2013). Dangerous safe havens: Institutional betrayal exacerbates sexual trauma. Journal of Traumatic Stress, 26, 119-124. doi:10.1002/jts.21778

44.

Stein

J. A.

Leslie

M. B.

Nyamathi

(2002). Relative contributions of parent substance use and childhood maltreatment to chronic homelessness, depression, and substance abuse problems among homeless women: Mediating roles of self-esteem and abuse in adulthood. Child Abuse & Neglect, 26, 1011-1027.

45.

Tang

S. S. S.

Freyd

J. J.

(2012). Betrayal trauma and gender differences in posttraumatic stress. Psychological Trauma: Theory, Research, Practice, and Policy, 4, 469-478. doi:10.1037/a0025765

46.

Taylor

K. M.

Sharpe

(2008). Trauma and post-traumatic stress disorder among homeless adults in Sydney. Australian & New Zealand Journal of Psychiatry, 42, 206-213. doi:10.1080/00048670701827218

47.

Torchalla

Strehlau

Schuetz

Krausz

(2012). The association between childhood maltreatment subtypes and current suicide risk among homeless men and women. Child Maltreatment, 17, 132-143.

48.

Tremblay

Hébert

Piché

(1999). Coping strategies and social support as mediators of consequences in child sexual abuse victims. Child Abuse & Neglect, 23, 929-945.

49.

Tyler

K. A.

Cauce

A. M.

(2002). Perpetrators of early physical and sexual abuse among homeless and runaway adolescents. Child Abuse & Neglect, 26, 1261-1274. doi:10.1016/S0145-2134(02)00413-1

50.

Tyler

K. A.

Gervais

S. J.

Davidson

M. M.

(2012). The relationship between victimization and substance use among homeless and runaway female adolescents. Journal of Interpersonal Violence. Advance online publication. doi:10.1177/0886260512455517

51.

Tyler

K. A.

Hoyt

D. R.

Whitbeck

L. B.

(2000). The effects of early sexual abuse on later sexual victimization among female homeless and runaway adolescents. Journal of Interpersonal Violence, 15, 235-250.

52.

Tyler

K. A.

Melander

L. A.

Noel

(2009). Bidirectional partner violence among homeless young adults: Risk factors and outcomes. Journal of Interpersonal Violence, 24, 1014-1035. doi:10.1177/0886260508319364

53.

van Ginkel

J. R.

van der Ark

L. A.

(2005). SPSS syntax for missing value imputation in test and questionnaire data. Applied Psychological Measurement, 29, 152-153. Doi:10.1177/0146621603260688

54.

Weathers

Litz

Herman

Huska

Keane

(1993, October). The PTSD Checklist (PCL): Reliability, validity, and diagnostic utility. Paper presented at the Annual Convention of the International Society for Traumatic Stress Studies, San Antonio, TX.

55.

Whitbeck

L. B.

Hoyt

D. R.

Ackley

K. A.

(1997). Abusive family backgrounds and later victimization among runaway and homeless adolescents. Journal of Research on Adolescence, 7, 375-392. doi:10.1207/s15327795jra0704_2

56.

Widom

C. S.

Czaja

S. J.

Dutton

M. A.

(2008). Childhood victimization and lifetime revictimization. Child Abuse & Neglect, 32, 785-796. doi:10.1016/j.chiabu.2007.12.006

57.

Wong

Y.-L. Y. I.

(2000). Measurement properties of the Center for Epidemiologic Studies–Depression Scale in a homeless population. Psychological Assessment, 12, 69-76. doi:10.1037//1040-3590.12.1.69

58.

Zurbriggen

E. L.

Freyd

J. J.

(2004). The link between child sexual abuse and risky sexual behavior: The role of dissociative tendencies, information-processing effects, and consensual decision mechanisms. In Koenig

L. J.

Doll

L. S.

O’Leary

Pequenat

(Eds.), From childhood abuse to adult sexual risk: Trauma, revictimization, and intervention (pp. 135-158). Washington, DC: American Psychological Association.