The Impact of Specific and Complex Trauma on the Mental Health of Homeless Youth

Abstract

This study investigates the relative impact of trauma experiences that occurred prior to and since becoming homeless on depressive symptoms, posttraumatic stress disorder (PTSD) symptoms, and self-injurious behaviors among a sample of homeless youth (N = 389). Youth (aged 13 to 25) who had been homeless or precariously housed in the past year completed a survey about housing history, experiences of violence and victimization, mental health, and service utilization. In addition to examining the impact associated with specific trauma types, we also considered the effect of “early-on” poly-victimization (i.e., cumulative number of reported traumas prior to homelessness) and the influence of a compound sexual trauma variable created to represent earlier complex trauma. This created-variable has values ranging from no reported trauma, single trauma, multiple non-sexual traumas, and multiple traumas that co-occurred with sexual abuse. Multivariate analyses revealed that specific traumatic experiences prior to homelessness, including sexual abuse, emotional abuse/neglect, and adverse home environment, predicted greater mental health symptoms. Poly-victimization did not add to the prediction of mental health symptoms after the inclusion of specific traumas. Results with early compound sexual trauma revealed significant differences between lower-order trauma exposures and multiple-trauma exposures. Specifically, experience of multiple traumas that co-occurred with sexual trauma was significantly more detrimental in predicting PTSD symptoms than multiple traumas of non-sexual nature. Findings support the utility of an alternate/novel conceptualization of complex trauma, and support the need to carefully evaluate complex traumatic experiences that occurred prior to homelessness, which can impact the design and implementation of mental health care and services for homeless youth.

Keywords

child abuse mental health and violence PTSD sexual assault

Introduction

Research has estimated that about 1 million to 1.6 million young adults experience homelessness each year in the United States (National Alliance to End Homelessness, 2006). Homeless youth experience disproportionately high amounts of trauma, abuse, and neglect prior to becoming homeless and often face assault and violence daily while they are homeless (Bender, Ferguson, Thompson, Komlo, & Pollio, 2010). Many homeless youth report being abused by caregivers (Rabinovitz, Desai, Schneir, & Clark, 2010; Tyler, Hoyt, Whitbeck, & Cauce, 2001; Whitbeck & Simons, 1990) and once on the street, youth who experienced maltreatment may be exposed to other forms of victimization (Lauritsen, Sampson, & Laub, 1991; Whitbeck & Simons, 1990). There is substantial empirical evidence showing a high correlation between earlier and later life trauma and that vulnerability from earlier abuse can intensify subsequent trauma exposure (Classen, Palesh, & Aggarwal, 2005; Follette, Polusny, Bechtle, & Naugle, 1996; Tansill, Edwards, Kearns, Gidycz, & Calhoun, 2012; Walsh et al., 2012). Often isolated from reliable sources of support from family and friends, homeless youth may be unlikely to access services that can help them recover and exit homelessness.

There is an urgent need to better understand the effects of trauma among homeless youth, particularly the potentially complex nature of their traumatic life experiences (Bender, Thompson, Ferguson, Yoder, & Kern, 2014). Complex trauma has been referred to as the experience of one or multiple forms of trauma repeatedly for a sustained period of time (Cloitre et al., 2009; Resick et al., 2012). Complex trauma often begins early on in life, is typically interpersonal in nature, occurs under circumstances where victims are often powerless to escape and is associated with a variety of mental health problems (Briere, Hodges, & Godbout, 2010; D’Andrea, Ford, Stolbach, Spinazzola, & van der Kolk, 2012; Ford, Elhai, Connor, & Frueh, 2010). Developmental consequences such as difficulties in self-regulation and establishment of relationships, non-suicidal self-injury, and substance abuse increase as the number of trauma increases and the types of trauma varies (Briere & Jordan, 2009).

Complex Trauma and Poly-Victimization

Traumatic experiences often co-occur (e.g., physical and psychological abuse) and trauma of the same type can recur within individuals over time (e.g., being verbally abused daily; Finkelhor, Ormrod, Turner, & Hamby, 2005; S. T. Scott, 2007; Turner, Finkelhor, & Ormrod, 2010). The concept of “poly-victimization” captures an individual’s exposure to multiple forms or types of trauma and victimization. Not surprisingly, research indicated that those who have been poly-victimized (often operationalized as the sum of distinct trauma types) exhibited significantly greater number and complexity of psychological symptoms (Briere, Kaltman, & Green, 2008; Cyr, Clement, & Chamberland, 2013; Ford et al., 2010). While researchers have demonstrated the cumulative effects of poly-victimization on mental health, studies have yet to consider the mental health consequences of an alternate way to conceptualize complex traumas that not only considers the co-occurrence of multiple types of traumas but also the nature of specific trauma types that co-occur (e.g., sexual abuse and physical abuse) among homeless youth. There is research to suggest that, among both general and clinical adolescents, child sexual abuse and the co-occurrence of other trauma types carry a markedly stronger association with mental health problems than sexual abuse alone (Briere & Gil, 1998; Gustafsson, Nilsson, & Svedin, 2009).

Trauma Exposure Prior to Homelessness and Mental Health Symptoms

The devastating consequences of childhood maltreatment and victimization have been clearly demonstrated in the literature. Childhood maltreatment can have a lasting impact on a wide range of mental and physical health outcomes (Gilbert et al., 2009; Streeck-Fischer & van der Kolk, 2000). Physical abuse and psychological/emotional abuse are leading risk factors for individuals’ social and emotional development and subsequent poor mental health outcomes, including depression and posttraumatic stress disorder (PTSD; Chapman et al., 2004; Gibb & Abela, 2008; Nooner et al., 2012). In addition, recent work on chaotic or adverse home environments (e.g., witness domestic violence and parental substance abuse) found living in such an environment to be negatively associated with adolescent mental health and development, perhaps due to the unstable and chaotic nature of these types of households (Hunt, Martens, & Belcher, 2011; Stevens, Ruggiero, Kilpatrick, Resnick, & Saunders, 2005). It has been shown that childhood experiences of domestic violence, parental separation/divorce, and exposure to substance abuse in a household place an individual at greater risk for depression later in life (Anda et al., 2002). Nevertheless, among different forms of child maltreatment, sexual abuse has been shown to be the leading risk factor for depression, PTSD, and disturbances in affective and interpersonal self-regulatory capacities (Briere et al., 2010; Tansill et al., 2012; Weierich & Nock, 2008). Research examining the impact of multiple traumatic experiences (including childhood abuse and domestic violence in adulthood) indicated that those who were exposed to childhood sexual abuse were at higher risk for development of PTSD (J. Scott, Chant, Andrews, Martin, & McGrath, 2007). Another study found that not only is child sexual abuse strongly associated with depression, but the more severe the abuse, the greater the likelihood the survivor would develop depression later in life (Chen et al., 2014).

In addition to PTSD and depression, another mental health symptom that would likely be impacted by the experience of complex trauma is self-injury (D’Andrea et al., 2012; Ford et al., 2013). Self-injury (known as self-mutilation, deliberate self-harm, or non-suicidal self-injury) represents a wide range of behaviors that an individual engages in to inflict harm on himself or herself that results in injuries, but the act itself is not intended as a suicide attempt (Favazza, 1998). The act of self-injury often reflects disturbances in affective and interpersonal self-regulatory capacities (Cloitre et al., 2009), which are among the symptoms of those exposed to childhood trauma, particularly sexual abuse (van der Kolk, Perry, & Herman, 1991; Yates, Carlson, & Egeland, 2008). Not surprisingly, self-injury has been found to be related to both depression and PTSD (Asarnow et al., 2012; Dyer et al., 2009). The high prevalence of self-injury among homeless youth (from 25% to 69%) compared with non-clinical populations (about 4%) makes self-injury another important mental health outcome to consider in the study of the impact of trauma on mental health in this population (Moskowitz, Stein, & Lightfoot, 2013).

Current Study

The current study systematically investigates the way homeless youth’s trauma experiences (traumas that occur prior to and since becoming homeless) impact their mental health, with a particular focus on “earlier” traumas or traumas that have occurred prior to homelessness. We first determine the relative importance of different trauma types in predicting depression, PTSD, and self-injury. We expect that earlier traumas, particularly sexual abuse, to be significantly associated with poorer mental health symptoms among homeless youth (Hypothesis 1). Next, we investigate whether “early-on” poly-victimization (i.e., the sum of all assessed traumas that occurred prior to homelessness) might uniquely contribute to the explanation of these mental health symptoms in addition to the specific traumas. Lastly, we propose a novel way to conceptualize complex trauma that incorporates the idea that traumas often co-occur (as in poly-victimization) but also consider the relative impact of the experience of multiple traumas that did not include sexual abuse and multiple traumas that co-occurred with sexual abuse. We expect that the experience of multiple trauma types (regardless of whether it includes sexual abuse or not) that occurred prior to homelessness would be associated with poorer mental health outcomes compared with those with zero reported traumas or those who reported single trauma experiences (Hypothesis 2a). We further speculate that those who reported multiple traumas that co-occurred with sexual abuse would have poorer mental health outcomes than to those who reported multiple traumas that did not include sexual abuse (Hypothesis 2b).

The current study will be among the first to unpack the mental health consequences associated with the breadth of different traumas that homeless youth experienced prior to homelessness and since becoming homeless. This investigation also takes into consideration homeless youth’s exposure to other stressful life events prior to homelessness, such as foster care or incarceration (Milburn, Ayala, Rice, Batterham, & Rotheram-Borus, 2006; Nesmith, 2006), and other risk factors for mental health symptoms related to being homeless, such as amount of time homeless/precariously housed and whether homeless youth had spent any nights in places not fit for human habitation in the last 30 days (Bender et al., 2010). In addition, we considered socio-demographic predictors of mental health problems among homeless youth including age, gender, sexual orientation, and ethnicity (Auerswald & Puddefoot, 2012; Milburn et al., 2006).

Method

Participants

From February to July 2007, research staff from Children’s Hospital Los Angeles, California, partnered with agencies from the Hollywood Homeless Youth Partnership to recruit youth who are homeless for an audio-computer-assisted self-interview (ACASI). Of the 642 prospective participants screened, 532 were eligible and 413 completed the survey, yielding a response rate of 78%. A total of 24 surveys were excluded due to data issues, bringing the total sample to 389. Youth were eligible if they were 13 to 25 years old and had been homeless or precariously housed in the past year. Youth completed a 60- to 90-min survey from drop-in centers, transitional living programs, street locations where homeless youth congregate, and clinics that serve them. Participants were compensated with a $20 gift card. This study received approval from the Institutional Review Board.

Measures

The 120-item ACASI included questions about housing history, experiences of violence, mental health, substance use/abuse, and service utilization. Socio-demographic variables including age, race, gender, and sexual orientation are included as covariates in the analysis. Additional experiences assessed as dichotomous variables included having ever been removed from the home by Child Protective Services (CPS), incarceration prior to homelessness, spending a night on the street in the last 30 days at a place that is not fit for human habitation, and total length of time being homeless (in years, but not necessarily continuously). Questions and standardized measures used were adapted from a previous study with homeless youth (Kipke, Simon, Montgomery, Unger, & Iversen, 1997) or were generated by a group of local experts.

Trauma prior to homelessness

The survey assessed a variety of traumatic and adverse life experiences prior to participants’ homelessness, including sexual abuse by family members or other trusted adults, physical abuse, emotional abuse/neglect, and adverse home life, totaling 10 individual questions. Aggregate categories were created to represent different earlier trauma types as dichotomous variables (i.e., experience of any physical abuse, sexual abuse, etc.). History of any physical abuse was based on the item, “when you were growing up, did either of your parents, guardians, siblings, or relatives ever hit you so hard that it left bruises or other injuries?” History of any sexual abuse was based on whether participants responded affirmative to one or both of the following questions: “When you were growing up, did ____ ever sexually abuse you? By sexual abuse, I mean things like masturbating in front of you, touching you in a sexual way, forcing you to touch their body in a sexual way, or having sex with you.” The first question asked about familial perpetrators (i.e., parents, guardians, siblings, or relatives) and the subsequent question asked about others (i.e., neighbor, teacher, or friend). History of any emotional abuse/neglect was indicated by either or both of the following: a parent/caregiver or relative having repeatedly put the respondent down, said hurtful things, or demeaned them; and if they had been unsupervised as a child without food. Based on previous work, a history of having an “adverse home life” was based on affirmation to any of the following: if participant’s parent/caregiver used alcohol in excess or used other drugs/substances in excess, if the participant’s parent/caregiver involved the participant in selling drugs, and if the participant witnessed physical violence and/or verbal abuse between parents and family members. We also created a continuous “early-on” poly-victimization variable that sums up the total number of earlier trauma experiences based on the 10 questions that assessed specific earlier traumas.

Early compound sexual trauma

The created complex trauma variable “early compound sexual trauma” represents four groups. One group consisted of individuals who reported none of the earlier traumas assessed. The second group consisted of those participants who reported a single trauma (excluding those who only reported sexual abuse). The third group consisted of those who reported multiple types of non-sexual trauma (two or more). The fourth group consisted of those who experienced multiple types of earlier trauma that included sexual abuse.

Trauma since becoming homeless

The survey also assessed a variety of traumatic events participants might have experienced since becoming homeless, including being robbed, being a victim of harassment, and being a victim of different forms of intimate partner violence. Aggregate categories were also created to represent different types of trauma that occurred since becoming homeless (i.e., any experience of any physical assault, any harassment, etc.). Physical assault was represented by participants’ endorsement to one of the following: being robbed, being physically assaulted, or attacked with a weapon. Sexual assault was indicated when participants had been either sexually assaulted or raped. Harassment was determined when participants reported one of the following: being intimidated, harassed, or attacked based on their ethnicity/race, sexual orientation, or gender identity. Intimate partner violence (IPV) was determined when participants experienced one of the following with any one of their partners: being physically assaulted, being forced to have sex when they did not want to, or being put down repeatedly, or spoken to using extremely hurtful language, or being demeaned.

Depression

Depression was measured by the 4-item version of the Centers for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977). This scale uses a 4-point Likert-type scale ranging from 0 = rarely or none of the time; less than 1 day to 3 = most of all of the time; 5 to 7 days and asks participants to rate how they have felt in the past week. The CES-D yields total scores ranging from 0 to 12; a cutoff score of 4 or higher used to indicate depression. Cronbach’s alpha of this scale in the current study is .88.

PTSD

The 22-item Child PTSD Reaction Index (CPTS-RI) was used in this study given the age distribution and reading level of the sample. The CPTS-RI assesses the frequency of some of the diagnostic symptoms for PTSD as well as guilt, impulse control, somatic symptoms, and regressive behaviors using a Likert-type scale that ranged from none to most of the time (Pynoos, Rodriguez, Steinberg, Stuber, & Frederick, 1998). The CPTS-RI yields total scores ranging from 0 to 80 that reflect the frequency of symptoms (Steinberg, Brymer, Decker, & Pynoos, 2004). Cronbach’s alpha of this scale in the present study is .94.

Self-injury

The dichotomous self-injury outcome variable is created by asking participants whether or not they ever had purposefully cut, burned, or injured their body.

Data Analysis

Bivariate analyses (e.g., correlation analyses, t-tests, ANOVAs, and chi-square tests) were conducted to determine associations between variables of interest. These results help inform hierarchical regression models used to test our hypotheses at the multivariate level. In the first set of hierarchical regression analyses, we first entered socio-demographic covariates, followed by specific trauma experiences and covariates that might have occurred prior to becoming homeless (e.g., having been removed from home by CPS, and having been incarcerated). This is followed by traumas experienced while homeless and covariates, including spending a night on the street in the last 30 days and total time spent homeless (in years). The final step included the “early-on” poly-victimization variable that was a sum of all of the earlier traumatic events reported by participants. This enabled us to see whether poly-victimization adds to the prediction of the outcomes beyond those associated with specific earlier trauma types. These procedures followed closely to those used by Cyr et al. (2013) and Finkelhor, Ormrod, and Turner (2007).

A separate set of hierarchical regression analyses investigated how multiple earlier non-sexual traumas and multiple earlier traumas that included sexual abuse would be associated with changes in mental health symptoms. In these analyses, the created-variable, “early compound sexual trauma,” was dummy-coded into four categories, along with specific traumas that occurred since becoming homeless. Age, race, gender identity, experiences of being removed from the home by CPS, incarceration prior to becoming homeless, sleeping at least one night on the street in the last 30 days, and time spent homeless or precariously housed were also included as covariates for these multivariate analyses. For ease of analysis and interpretation, we used the continuous version of age, and dichotomous version of the ethnicity variable (African American vs. Other Racial/Ethnic group), gender (male vs. female), and sexual identity variables (lesbian, gay, or bisexual vs. not lesbian, gay, or bisexual).

In fitting the multiple linear regression models, correlations and collinearity diagnostics were examined to ensure that there is no evidence of multicollinearity among the variables entered. To assess overall model fit for the multiple linear regression model, we used the F test to examine the significance of adjusted R², which measures the amount of variance in the dependent variable that is uniquely or jointly explained by the independent variable(s). We also examined changes in R² to identify the unique contributions of each set of independent variables entered at each step. The Hosmer–Lemeshow goodness-of-fit test was used to assess model fit at each step of the hierarchical logistic regression models. Model chi-square differences at each step were examined to see whether variables within each step contributed significantly to the prediction of each outcome. In determining the final model for each analysis, we retained variables that not only significantly contributed to the prediction of the outcomes, but also variables that were significant at least at the p < .25 level. Model trimming occurred in conjunction with consideration of the stability of the model and the model’s overall fit.

Results

Demographic and background characteristics of the sample are presented in Table 1. Participants’ age ranged from 13 to 25 years, with the average at 19.6 years (SD = 2.78 years). Most youth were from racial/ethnic minority backgrounds, with the largest group who identified as African American/Black, followed by Latino/Hispanic, then White/Caucasian. Among the participants, about 6% identified themselves as multi-ethnic. A total of 40% of the sample reported having been removed from home by CPS. More than a quarter reported having been incarcerated prior to homelessness. About half of our sample reported spending at least one night on the street or other places unsuitable for human habitation. For this population, the average length of time having been on the street or precariously housed (not necessarily continuously) ranged from a few days to almost 13 years, with the average being 2.7 years.

Table 1.

Sample Characteristics (N = 389).

Variables	n (%)
Race/ethnicity
White/Caucasian	66 (17.1)
Black/African American	173 (44.8)
Latino/Hispanic	98 (25.4)
Asian/Asian American/Pacific Islander	10 (2.6)
American Indian/Native American	10 (2.6)
Other/not listed	29 (7.5)
Gender
Female	123 (31.6)
Male	233 (51.9)
Transgender, unsure, or used other terms	33 (8.5)
Age category
13 to 14 years	14 (3.6)
15 to 17 years	82 (21.1)
18 to 20 years	137 (35.2)
21 to 25 years	156 (40.1)
Sexual orientation
Gay, lesbian, bisexual, uncertain	145 (37.2)
Heterosexual or straight	224 (57.6)
Don’t know, refused to answer, missing	20 (5.1)
Removed from home by Child Protective Services (CPS)	157 (40.4)
Incarcerated prior to becoming homeless	104 (26.9)
Slept at least one night on streets in the last 30 days	196 (50.4)
	M (SD)
Time spent on the street or precariously housed in lifetime (in years)^a
Range: a few days to 13 years	2.7 (2.8)
Mental health symptoms
Depressive symptoms
Range: 0 to 12	4.54 (4.01)
PTSD summary score
Range: 0 to 68	22.6 (16.2)
	n (%)
Self-injury, ever	117 (30)

Note. PTSD = posttraumatic stress disorder.

This range represents time spent on the street or precariously housed during one’s lifetime in aggregate.

Trauma Exposure Prevalence and Associations Between Trauma Types

About 70% of the sample reported growing up in an adverse home environment (71.2%). The next most prevalent form of trauma that occurred prior to homelessness is the experience of some form of emotional abuse/neglect (58.9%), followed by physical abuse (51.4%). Childhood sexual abuse was the least prevalent, but still high at 33.2%. Since becoming homeless, participants reported the experience of physical assault most frequently (37.3%), followed by harassment (27.4%), then IPV (22.5%). Sexual assault occurred in about 13% of the sample. Early-on poly-victimization, which is the sum of 10 specific earlier traumas assessed, ranged from 0 to 10, with a mean 3.8. All of the categories of earlier traumas were significantly associated with each other, with physical and emotional abuse/neglect the most highly correlated (r = .62, p < .001). Traumas that occurred since becoming homeless were also highly associated with each other, with the greatest associations between harassment and physical assault (r = .52, p < .001), followed closely by sexual assault and IPV (r = .41, p < .001), and then physical assault and IPV (r = .40, p < .001).

Specific Traumas Experienced Prior to and Since Becoming Homeless and Mental Health

Hierarchical regression analyses examined the relative importance of individual trauma types in predicting depression, PTSD, and self-injurious behaviors, while accounting for the effects of psychosocial covariates and socio-demographic variables (see Table 2). In the first step, where we entered socio-demographic variables, being male compared with being female was significantly associated with lower depressive symptoms, being older was nearly significantly associated with greater PTSD symptoms, and being African American relative to other ethnic groups is protective of the likelihood of self-injury.

Table 2.

Hierarchical Regression Models of Specific Traumas as Predictors of Depression Symptoms, PTSD Symptoms, and Self-Injury.

	Step 1		Step 2		Step 3
	β		β		β
Depression symptoms
Gender: male	−.12*		−.07		−.10
Sexual abuse			.15**		.12*
Emotional abuse/neglect			.17***		.09
Harassment					.13*
Intimate partner violence					.12*
Physical assault					.15*
Sexual assault					−.06
One night on the street in the past 30 days					.06
R ² _adj	.01*		.06***		.14***
PTSD symptoms
Age	.11^†		.11*		.07
Sexual abuse			.13*		.10^†
Adverse home life			.17**		.14*
Emotional abuse/neglect			.20***		.15**
Intimate partner violence					.13*
Physical assault					.13*
One night on the street in the past 30 days					.02
Total time spent on the street (in years)					−.03
R ² _adj	.01^†		.14***		.17**
	OR	95% CI	OR_adj	95% CI	OR_adj	95% CI
Self-injurious behaviors
African American	0.49**	[0.31, 0.78]	0.51**	[0.31, 0.82]	0.48**	[0.29, 0.81]
Sexual abuse			1.81*	[1.10, 2.98]	1.42	[0.83, 2.43]
Adverse home life			2.22**	[1.21, 4.09]	1.88*	[1.01, 3.52]
Removed from home			1.45	[0.90, 2.35]	1.41	[0.85, 2.35]
Intimate partner violence					3.87***	[2.09, 7.15]
Sexual assault					0.66	[0.31, 1.42]
One night on the street in the past 30 days					1.57^†	[0.95, 2.60]

Note. For each regression model, Step 1 contains demographic covariates, Step 2 contains trauma variables and covariates prior to homelessness, Step 3 contains trauma variables and covariates that occurred since becoming homeless. Omnibus tests of model coefficients were significant at each step for Self-Injurious Behaviors. Hosmer–Lemeshow goodness-of-fit χ²(8) for Self-Injurious Behavior was not significant, suggesting good or adequate fit. PTSD = posttraumatic stress disorder; CI = confidence interval.

†

p < .10. *p < .05. **p < .01. ***p < .001.

When traumatic experiences that occurred prior to homelessness were entered in the second step, the gender effect on depression was diminished to non-significance, while age was significant for PTSD symptoms, and being African American remained protective for risk of self-injury. Sexual abuse was a significant factor in predicting all of the mental health symptoms in the second step. Emotional abuse/neglect was predictive of depressive and PTSD symptoms, while adverse home life was significantly predictive of PTSD symptoms and likelihood of self-injury. Earlier trauma that occurred prior to homelessness explained 6% of the variance in depressive symptoms and 14% of the variance in PTSD summary score.

When we entered traumas that occurred since becoming homeless in the third step, we found that the only socio-demographic variable that remained significant was the protective effect of being African American for self-injury. Among the earlier traumas, we found that sexual abuse remained a significant predictor of depression, but emotional abuse/neglect was not. Conversely, the effect of sexual abuse was diminished for both PTSD and self-injury, but emotional abuse/neglect remained important for PTSD. Adverse home environment continued to be important for predicting PTSD and self-injury. Among the traumas experienced since becoming homeless, intimate partner violence was a significant predictor for all of the mental health symptoms. The experience of physical assault was predictive of both depressive and PTSD symptoms. Harassment was predictive of depressive symptoms. Having spent one night on the street in the last 30 days was also nearly significantly associated with self-injury. The addition of this set of variables contributed to an additional 9% of the variance in depressive symptoms and 4% to the PTSD summary scores.

Finally, we added the early-on poly-victimization variable to each of the mental health models in the fourth step. Unlike what was observed by previous researchers (e.g., Cyr et al., 2013), we found that poly-victimization did not add to the prediction of these models after accounting for specific trauma types, suggesting that much of the same variance in the outcomes has been accounted for by specific trauma types. Because of this, results with poly-victimization were not reported in Table 2. Additional investigations where poly-victimization was entered in place of specific trauma variables revealed that poly-victimization did contribute significantly to the prediction of mental health symptoms. However, its explanatory power did not match those of the set of specific trauma types used in the prediction of each mental health outcomes.

Early Compound Sexual Trauma and Mental Health

A compound variable was used to investigate whether the experience of multiple earlier trauma types inclusive of sexual abuse would be more detrimental to one’s mental health than the experience of a single trauma type (of non-sexual nature) or the experience of multiple earlier traumas not including sexual abuse. The distributions of participants within each category of the early compound variable with different level of depressive and PTSD symptoms, and the percentage of those having ever engaged in self-injurious behaviors are presented in Table 3. As shown, there is a linear increase in mean scores across the different levels of the compound sexual trauma variable for depressive symptoms and PTSD summary score. Results from ANOVAs confirmed these differences to be significant. Among those who engaged in self-injurious behaviors, 47% reported experiences of multiple types of trauma including sexual abuse prior to homelessness. Pearson chi-square tests also revealed significant differences between each level of the compound variable and self-injury. These relationships were investigated further at the multivariate level.

Table 3.

Distribution of Early Compound Sexual Trauma Variable and Mental Health Symptoms.

		Depressive Symptoms (n = 364)	PTSD Summary Score^a (n = 376)	Self-Injurious Behaviors (Lifetime)^b (n = 357)
Early Compound Sexual Trauma	n (%)	M (SD)	M (SD)	%
0 = No reported trauma	71 (18.3)	3.03 (3.65)	15.73 (15.04)	7.8
1 = Reported single trauma (inclusive of non-sexual types only)	79 (20.3)	3.2 (3.42)	19.14 (15.56)	16.4
2 = Multiple types of non-sexual trauma	115 (20.4)	4.93 (4.07)	23.78 (14.83)	29.1
3 = Multiple types of trauma inclusive of sexual trauma	122 (31.5)	5.83 (4.10)	27.62 (17.0)	46.6

Note. PTSD = posttraumatic stress disorder.

For Depressive symptoms summary score, F(3) = 9.91; p < .001.

For PTSD summary score, F(3)=10.01; p < .001.

For self-injury, Pearson χ²(3) = 23.5; p < .001.

Early Compound Sexual Trauma, Trauma Since Homelessness, and Mental Health

Following a similar modeling approach as in the previous analysis, two separate regression models examined the relationship between different levels of the early compound sexual trauma variable and mental health outcomes. In the first model, “single non-sexual trauma” served as the reference group, which contrasted the experience of one type of earlier trauma to multiple types of traumas without sexual abuse or multiple types of traumas that occurred with sexual abuse. In the second model, “multiple types of non-sexual trauma” served as the reference group, which contrasted the effects of multiple non-sexual traumas to multiple traumas that included sexual abuse.

Single versus multiple earlier traumas on mental health

With single trauma as the reference group, we found that those who experienced multiple traumas without sexual abuse had significantly higher PTSD scores (β = .15, p = .03). We found a similar relationship for those who experienced multiple traumas with sexual abuse compared with those who reported a single trauma in predicting depression, PTSD, and increased odds of having ever attempted self-injury (β = .22, p < .01; β= .30, p < .001; odds ratio [OR] = 2.2, 95% confidence interval [CI] = [1.15, 4.20], respectively). After accounting for traumas since becoming homeless and covariates in Step 3 (the same ones as shown in Table 2), the effect between multiple traumas without sexual abuse and single trauma for PTSD was no longer significant. The effect between single and multiple traumas with sexual abuse remained for PTSD (β = .22, p < .01) and marginally so for depression (β = .13, p = .06), but not so for self-injury.

Multiple earlier traumas of non-sexual nature versus multiple earlier traumas inclusive of sexual abuse on mental health

Using multiple non-sexual traumas as the reference group, participants who experienced earlier multiple traumas that included sexual abuse were significantly more likely to have higher PTSD scores (β = .16, p = .01) and had significantly higher odds of having ever engaged in self-injury (OR = 1.96, 95% CI = [1.06, 3.30]). This was not the case for depression. After accounting for traumas that occurred since homeless in the third step, the effect between multiple traumas without sexual abuse and multiple traumas with sexual abuse remained for PTSD summary scores (β = .13, p < .05) and marginally so for self-injury (OR = 1.68, 95% CI = [0.93,3.05]), but not for depression. In all, the compound sexual trauma variable contributed to 6% of the variance for depressive symptoms and 12% of the variance in PTSD symptoms.

Discussion

Our findings clearly show that homeless youth experience a high rate of traumatic experiences; over 80% of the sample reported at least one earlier trauma and a little over half of our sample (~52%) reported experiencing multiple earlier traumas. As expected, sexual abuse was found to be a significant predictor of depressive symptoms and nearly significantly for PTSD symptoms, after accounting for the influence of traumas experienced since becoming homeless, socio-demographic characteristics, and factors that have been previously linked to mental health among homeless youth, such as incarceration. The results are consistent with Putnam’s (2003) review, which found depression to be the most documented mental health outcome in adults who had experienced sexual abuse as a child. Despite evidence suggesting that sexual abuse would be an important predictor of self-injury (Weierich & Nock, 2008), this was not the case in the present study. Unlike some of the previous cited studies, for example, Cyr et al. (2013), poly-victimization did not add to the prediction of any of the mental health symptoms after accounting for the effects of specific traumas. Even though it may be important to look at the overall impact of multiple traumas, it is also important to consider the relative impact of specific trauma types, and even incorporate both concepts in a compound trauma variable as we have done here.

The use of the early compound sexual trauma variable enabled us to distinguish the effects of different multiple trauma experiences that occurred prior to homelessness on mental health outcomes among homeless youth. Our findings confirmed our hypothesis that, in general, experience of multiple trauma types regardless of whether it includes sexual abuse, to be more detrimental to mental health than the experience of zero reported or any single trauma type. Furthermore, for both PTSD and self-injury, results confirmed our hypothesis that experiences of multiple traumas that occur in conjunction with sexual abuse predicted worse outcomes than the experience of multiple traumas of non-sexual nature for PTSD symptoms. These findings help substantiate the usefulness of alternate ways to conceptualize complex trauma among homeless youth and provide support for our current approach.

Results also illustrated the magnitude of influences between earlier and later trauma experiences on mental health. Both earlier traumas and traumas that occurred since becoming homeless explained nearly equal amounts of variance in predicting levels of depression, which may suggest that trauma that occurred across the life span, regardless of timing, can contribute to depressive symptoms among homeless youth. In contrast, earlier trauma that occurred prior to homelessness explained more than three times the variance than trauma experiences since becoming homeless in predicting PTSD symptoms. This suggests that traumatic experiences that occurred earlier on may be more important drivers of PTSD symptoms.

In addition, growing up in a chaotic home environment and the experience of emotional abuse/neglect exerted significant influence on homeless youth’s PTSD symptoms and self-injurious behaviors. These findings add to the growing importance of examining not only the consequences of abusive experiences, but also other contextual factors that contribute to a difficult upbringing. For example, Kaess and colleagues (2013) found maternal antipathy (defined as showing hostility, coldness, or rejection) to be the only factor significantly associated with self-injury after accounting for effects of physical, sexual abuse, maternal and paternal neglect, and paternal antipathy. Disruptions in parental care, familial conflicts, witness of violence, and parental substance use have all been implicated in self-injurious behaviors and PTSD symptoms (Asgeirsdottir, Sigfusdottir, Gudjonsson, & Sigurdsson, 2011; Hunt et al., 2011)

Among the later traumas that occurred since becoming homeless, IPV was clearly important in predicting each of the mental health symptoms examined, which is consistent with previous studies with both homeless and non-homeless youth, whereby self-injury, PTSD, and depression have all been significantly predicted by IPV (Cisler et al., 2012; Levesque, Lafontaine, Bureau, Cloutier, & Dandurand, 2010). Physical assault, which is the most frequently reported trauma that occurred since becoming homeless, was also an important predictor for depressive and PTSD symptoms for homeless youth. These findings provide additional empirical support of the cycle of violence, danger, and vulnerability that homeless youth may contend with on a regular basis, and their impact on their current mental health.

Limitations, Future Directions, and Implications

Limitations of this current study include the self-report nature of the data and the use of single items to assess physical abuse, sexual assault, and self-injury. We also did not assess the chronicity (i.e., how long each trauma lasted) or severity of specific trauma types as has been done in previous studies, for example, Kisiel, Conradi, Fehrenbach, Torgersen, and Briggs (2014), which would have been important dimensions to consider in the conceptualization of complex trauma. Nevertheless, the empirical findings of the current study support an alternate and meaningful way to conceptualize complex trauma. The geographic location of the study, limited to the Hollywood area of Los Angeles County, California, may not generalize to homeless youth in other areas. This article focused specifically on interpersonal traumas but homeless youth are impacted by a wider range of trauma, such as loss of a parent or being diagnosed as HIV infected (Luz et al., 2011). In addition, the PTSD assessment asked youth to focus on the worse trauma they experienced, and we do not know if they chose a pre-homeless trauma or a trauma they experienced while homeless. When assessing the impact of multiple traumas on PTSD outcomes, we still lack a feasible PTSD assessment of psychological symptoms related to multiple over single traumas, and are in need of an assessment of responses to both multiple instances of the same type of trauma, as well as multiple types of traumas.

Future studies with homeless youth should examine the complex reactions to trauma beyond depression, PTSD, and self-injury, including their impact on comorbidities involving substance use. Studies should also consider how complex traumatic experiences may lead to the development of complex mental health symptoms, particularly the comorbidity between depression and PTSD (Campbell et al., 2007). In addition, more work is needed to address the role of complex trauma histories in homeless youth’s ability to make use of existing services to exit homelessness. Future studies should explore longitudinally and in greater depth, the role of adverse home environment, co-occurring sexual trauma with other types of trauma, and IPV on mental health outcomes. In addition, our findings suggest that for African American youth, mental health was less affected by earlier or later trauma experiences, suggesting the need for culturally competent mental health care and services. Future research should address the most efficacious treatment modalities that can realistically be provided to culturally and ethnically diverse homeless youth.

While there is a general push toward family reunification in reports of abuse that requires social service intervention, this is an area that warrants caution due to the detrimental nature of an adverse home environment. Traditional models of treatment for traumatized youth usually rely on strengthening support through parents or other family members, to offer reassurance to help youth understand and process the trauma experienced and to help mitigate its effects (American Academy of Child and Adolescent Psychiatry, 1998). Homeless young people usually do not have support from parents, extended family members, or school personnel, and effective models of intervention need to be designed for these youth. Many homeless youth are disenfranchised from, and distrustful of, traditional mental health services; as a result, trauma treatment needs to be embedded within service systems that have already demonstrated engagement with these youth.

Footnotes

Acknowledgements

We are profoundly grateful to our partner agencies of the Hollywood Homeless Youth Partnership for all of their thoughtful contributions, and most importantly, we thank the youth who shared their experiences with us by completing this survey.

Authors’ Note

We are indebted to key people on this project: Susan Ravinovitz, the original PI; Mona Desai and Arlene Schnier, coauthors of the final report; Donna Lopez, field supervisor; and project staff Leah Molaiepour, Leilani Martinez, and Xenia Martinez for all of their hard work in recruiting youth to complete the survey. This project was completed in partial fulfillment of the requirements of Dr. Lauren Marlotte’s participation in the Clinical Child Psychology Postdoctoral Fellowship at the USC University Center for Excellence in Developmental Disabilities, and in the California Leadership Education in Neurodevelopmental Disabilities (LEND) Interdisciplinary Training Program.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, 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: This work was funded by an award from The California Endowment award (Grant #20052902) and The California Wellness Foundation (Grant # 2007-095).

Authors Biographies

Carolyn F. Wong, PhD, is an assistant professor at the University of Southern California (USC), Department of Pediatrics, Keck School of Medicine. She is a member of the Community, Health Outcomes and Intervention Research Program and also part of the Division of Adolescent and Young Adult Medicine at Children’s Hospital Los Angeles. Her research interests include examining psychosocial risk and resilient factors of adolescent and young adults and the way these factors are related to mental health and behavioral risk-taking, including substance use/abuse and HIV risk-related behaviors.

Leslie F. Clark, PhD, MPH, is the director of research for the Division of Adolescent and Young Adult Medicine, Department of Pediatrics, USC Keck School of Medicine. As a social psychologist, her research interests include risk and protective factors for low-income, high-risk populations including ethnic and sexual minority adolescents, homeless youth, HIV+ youth, gang-involved youth, and teen mothers. She has developed and evaluated risk behavior change interventions for addressing sexual and drug use risk behaviors in these adolescent populations.

Lauren Marlotte, PsyD, is a licensed clinical psychologist and postdoctoral fellow in the Division of Adolescent and Young Adult Medicine, Children’s Hospital Los Angeles. She earned a doctorate in clinical psychology from the University of La Verne and completed her predoctoral internship at San Bernardino County Department of Behavioral Health.

References

American Academy of Child and Adolescent Psychiatry. (1998). Summary of the practice parameters for the assessment and treatment of children and adolescents with posttraumatic stress disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 37, 997-1001.

Anda

R. F.

Whitfield

C. L.

Felitti

V. J.

Chapman

D. P.

Edwards

V. J.

Dube

S. R.

Williamson

D. F.

(2002). Adverse childhood experiences, alcoholic parents, and later risk of alcoholism and depression. Psychiatric Services, 53, 1001-1009. doi:10.1176/appi.ps.53.8.1001

Asarnow

J. R.

Porta

Spirito

Emslie

Clarke

Wagner

K. D.

. . . Brent

D. A.

(2012). Suicide attempts and nonsuicidal self-injury in the treatment of resistant depression adolescents: Findings from the TORDIA study. FOCUS: The Journal of Lifelong Learning in Psychiatry, 10, 380-388.

Asgeirsdottir

B. B.

Sigfusdottir

I. D.

Gudjonsson

G. H.

Sigurdsson

J. F.

(2011). Associations between sexual abuse and family conflict/violence, self-injurious behavior, and substance use: The mediating role of depressed mood and anger. Child Abuse & Neglect, 35, 210-219. doi:10.1016/j.chiabu.2010.12.003 S0145-2134(11)00054-8 [pii]

Auerswald

C. L.

Puddefoot

(2012). Comparing white and African American homeless youth in San Francisco: Research findings and policy implications. Sacramento: The California Wellness Foundation.

Bender

K. A.

Ferguson

Thompson

Komlo

Pollio

(2010). Factors associated with trauma and posttraumatic stress disorder among homeless youth in three U.S. cities: The importance of transience. Journal of Traumatic Stress, 23, 161-168. doi:10.1002/jts.20501

Bender

K. A.

Thompson

S. J.

Ferguson

K. M.

Yoder

J. R.

Kern

(2014). Trauma among street-involved youth. Journal of Emotional and Behavioral Disorders, 22, 53-64. doi:10.1177/1063426613476093

Briere

Gil

(1998). Self-mutilation in clinical and general population samples: Prevalence, correlates, and functions. American Journal of Orthopsychiatry, 68, 609-620.

Briere

Hodges

Godbout

(2010). Traumatic stress, affect dysregulation, and dysfunctional avoidance: A structural equation model. Journal of Traumatic Stress, 23, 767-774. doi:10.1002/jts.20578

10.

Briere

Jordan

C. E.

(2009). Childhood maltreatment, intervening variables, and adult psychological difficulties in women: An overview. Trauma Violence Abuse, 10, 375-388.

11.

Briere

Kaltman

Green

B. L.

(2008). Accumulated childhood trauma and symptom complexity. Journal of Traumatic Stress, 21, 223-226.

12.

Campbell

D. G.

Felker

B. L.

Liu

C.-F.

Yano

E. M.

Kirchner

J. E.

Chan

. . . Chaney

E. F.

(2007). Prevalence of depression-PTSD comorbidity: Implications for clinical practice guidelines and primary care-based interventions. Journal of General Internal Medicine, 22, 711-718.

13.

Chapman

D. P.

Whitfield

C. L.

Felitti

V. J.

Dube

S. R.

Edwards

V. J.

Anda

R. F.

(2004). Adverse childhood experiences and the risk of depressive disorders in adulthood. Journal of Affective Disorders, 82, 217-225.

14.

Chen

Cai

Cong

Liu

Gao

. . . Flint

(2014). Childhood sexual abuse and the development of recurrent major depression in Chinese women. PLoS ONE, 9, e87569. doi:10.1371/journal.pone.0087569

15.

Cisler

J. M.

Begle

A. M.

Amstadter

A. B.

Resnick

H. S.

Danielson

C. K.

Saunders

B. E.

Kilpatrick

D. G.

(2012). Exposure to interpersonal violence and risk for PTSD, depression, delinquency, and binge drinking among adolescents: Data from the NSA-R. Journal of Trauma Stress, 25, 33-40. doi:10.1002/jts.21672

16.

Classen

C. C.

Palesh

O. G.

Aggarwal

(2005). Sexual revictimization: A review of the empirical literature. Trauma Violence Abuse, 6, 103-129. doi:6/2/103 [pii] 10.1177/1524838005275087

17.

Cloitre

Stolbach

B. C.

Herman

J. L.

van der Kolk

Pynoos

Wang

Petkova

(2009). A developmental approach to complex PTSD: Childhood and adult cumulative trauma as predictors of symptom complexity. Journal of Traumatic Stress, 22, 399-408. doi:10.1002/jts.20444

18.

Cyr

Clement

M.-E.

Chamberland

(2013). Lifetime prevalence of multiple victimizations and its impact on children’s mental health. Journal of Interpersonal Violence, 29, 616-634. doi:10.1177/0886260513606220

19.

D’Andrea

Ford

Stolbach

Spinazzola

van der Kolk

B. A.

(2012). Understanding interpersonal trauma in children: Why we need a developmentally appropriate trauma diagnosis. American Journal of Orthopsychiatry, 82, 187-200. doi:10.1111/j.1939-0025.2012.01154.x

20.

Dyer

K. F.

Dorahy

M. J.

Hamilton

Corry

Shannon

MacSherry

. . . McElhill

(2009). Anger, aggression, and self-harm in PTSD and complex PTSD. Journal of Clinical Psychology, 65, 1099-1114. doi:10.1002/jclp.20619

21.

Favazza

A. R.

(1998). The coming of age of self-mutilation. Journal of Nervous and Mental Disease, 186, 259-268.

22.

Finkelhor

Ormrod

R. K.

Turner

H. A.

(2007). Poly-victimization: A neglected component in child victimization. Child Abuse & Neglect, 31, 7-26.

23.

Finkelhor

Ormrod

R. K.

Turner

Hamby

S. L.

(2005). The victimization of children and youth: A comprehensive, national survey. Child Maltreatment, 10, 5-25.

24.

Follette

V. M.

Polusny

M. A.

Bechtle

A. E.

Naugle

A. E.

(1996). Cumulative trauma: The impact of child sexual abuse, adult sexual assault, and spouse abuse. Journal of Traumatic Stress, 9, 25-35.

25.

Ford

J. D.

Elhai

J. D.

Connor

D. F.

Frueh

B. C.

(2010). Poly-victimization and risk of posttraumatic, depressive, and substance use disorders and involvement in delinquency in a national sample of adolescents. Journal of Adolescent Health, 46, 545-552. doi:10.1016/j.jadohealth.2009.11.212 S1054-139X(09)00636-3 [pii]

26.

Ford

J. D.

Grasso

Greene

Levine

Spinazzola

van der Kolk

(2013). Clinical significance of a proposed developmental trauma disorder diagnosis: Results of an international survey of clinicians. Developmental Trauma Disorder Clinician Survey, 74, 841-849. doi:10.4088/JCP.12m08030

27.

Gibb

B. E.

Abela

J. R. Z.

(2008). Emotional abuse, verbal victimization, and the development of children’s negative inferential styles and depressive symptoms. Cognitive Therapy Research, 32, 161-176. doi:10.1007/s10608-006-9106-x

28.

Gilbert

Widom

Browne

Fergusson

Webb

Janson

(2009). Burden and consequences of child maltreatment in high-income countries. The Lancet, 373, 68-81.

29.

Gustafsson

P. E.

Nilsson

Svedin

C. G.

(2009). Polytraumatization and psychological symptoms in children and adolescents. European Child & Adolescent Psychiatry, 18, 274-283. doi:10.1007/s00787-008-0728-2

30.

Hunt

K. L.

Martens

P. M.

Belcher

H. M.

(2011). Risky business: Trauma exposure and rate of posttraumatic stress disorder in African American children and adolescents. Journal of Traumatic Stress, 24, 365-369. doi:10.1002/jts.20648

31.

Kaess

Parzer

Mattern

Plener

Bifulco

Resch

Brunner

(2013). Adverse childhood experiences and their impact on frequency, severity, and the individual function of nonsuicidal self-injury in youth. Psychiatry Research, 206, 265-272.

32.

Kipke

M. D.

Simon

T. R.

Montgomery

S. B.

Unger

J. B.

Iversen

E. F.

(1997). Homeless youth and their exposure to and involvement in violence while living on the streets. Journal of Adolescent Health, 20, 360-367. doi:S1054-139X(97)00037-2 [pii] 10.1016/S1054-139X(97)00037-2

33.

Kisiel

Conradi

Fehrenbach

Torgersen

Briggs

E. C.

(2014). Assessing the effects of trauma in children and adolescents in practice settings. Child and Adolescent Psychiatric Clinics of North America, 23, 223-242. doi:10.1016/j.chc.2013.12.007

34.

Lauritsen

J. L.

Sampson

R. J.

Laub

J. H.

(1991). The link between offending and victimization among adolescents. Criminology, 29, 265-292. doi:10.1111/j.1745-9125.1991.tb01067.x

35.

Levesque

Lafontaine

M. F.

Bureau

J. F.

Cloutier

Dandurand

(2010). The influence of romantic attachment and intimate partner violence on non-suicidal self-injury in young adults. Journal of Youth and Adolescence, 39, 474-483. doi:10.1007/s10964-009-9471-3

36.

Luz

M. P.

Mendlowicz

Marques-Portella

Gleiser

Berger

Neylan

T. C.

. . . Figueira

(2011). PTSD criterion A1 events: A literature-based categorization. Journal of Traumatic Stress, 24, 243-251. doi:10.1002/jts.20633

37.

Milburn

N. G.

Ayala

Rice

Batterham

Rotheram-Borus

M. J.

(2006). Discrimination and exiting homelessness among homeless adolescents. Cultural Diversity & Ethnic Minority Psychology, 12, 658-672. doi:2006-20708-005 [pii] 10.1037/1099-9809.12.4.658

38.

Moskowitz

Stein

J. A.

Lightfoot

(2013). The mediating roles of stress and maladaptive behaviors on self-harm and suicide attempts among runaway and homeless youth. Journal of Youth and Adolescence, 42, 1015-1027.

39.

National Alliance to End Homelessness. (2006). Fundamental issues to prevent and end youth homelessness (Youth Homelessness Series [Vol. Brief No. 1]). Retrieved from http://www.endhomelessness.org/library/entry/fundamental-issues-to-prevent-and-end-youth-homelessness

40.

Nesmith

(2006). Predictors of running away from family foster care. Child Welfare, 85, 585-609.

41.

Nooner

K. B.

Linares

L. O.

Batinjane

Kramer

R. A.

Silva

Cloitre

(2012). Factors related to posttraumatic stress disorder in adolescence. Trauma, Violence, & Abuse, 13, 153-166. doi:10.1177/1524838012447698

42.

Putnam

F. W.

(2003). Ten year research update review: Child sexual abuse. Journal of the American Academy of Child & Adolescent Psychiatry, 42, 269-278. doi:10.1097/01.CHI.0000037029.04952.72

43.

Pynoos

Rodriguez

Steinberg

Stuber

Frederick

(1998). UCLA PTSD index for DSM-IV. Retrieved from http://chipts.ucla.edu/wp-content/uploads/downloads/2012/02/UCLA-Adolescent-PTSD-Index.pdf

44.

Rabinovitz

Desai

Schneir

Clark

L. F.

(2010). No way home: Understanding the needs and experiences of homeless youth in Hollywood. Los Angeles, CA: Hollywood Homeless Youth Partnership.

45.

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.

46.

Resick

P. A.

Bovin

M. J.

Calloway

A. L.

Dick

A. M.

King

M. W.

Mitchell

K. S.

. . . Wolf

E. J.

(2012). A critical evaluation of the complex PTSD literature: Implications for DSM-5. Journal of Traumatic Stress, 25, 241-251. doi:10.1002/jts.21699

47.

Scott

Chant

Andrews

Martin

McGrath

(2007). Association between trauma exposure and delusional experiences in a large community-based sample. British Journal of Psychiatry, 190, 339-343. doi:190/4/339 [pii] 10.1192/bjp.bp.106.026708

48.

Scott

S. T.

(2007). Multiple traumatic experiences and the development of posttraumatic stress disorder. Journal of Interpersonal Violence, 22, 932-938.

49.

Steinberg

A. M.

Brymer

M. J.

Decker

K. B.

Pynoos

R. S.

(2004). The University of California at Los Angeles Post-Traumatic Stress Disorder Reaction Index. Current Psychiatry Reports, 6, 96-100.

50.

Stevens

T. N.

Ruggiero

K. J.

Kilpatrick

D. G.

Resnick

H. S.

Saunders

B. E.

(2005). Variables differentiating singly and multiply victimized youth: Results from the National Survey of Adolescents and implications for secondary prevention. Child Maltreatment, 10, 211-223. doi:10/3/211 [pii] 10.1177/1077559505274675

51.

Streeck-Fischer

van der Kolk

(2000). Down will come baby, cradle and all: Diagnostic and therapeutic implications of chronic trauma on child development. Australian & New Zealand Journal of Psychiatry, 34, 903-918.

52.

Tansill

E. C.

Edwards

K. M.

Kearns

M. C.

Gidycz

C. A.

Calhoun

K. S.

(2012). The mediating role of trauma-related symptoms in the relationship between sexual victimization and physical health symptomatology in undergraduate women. Journal of Traumatic Stress, 25, 79-85. doi:10.1002/jts.21666

53.

Turner

H. A.

Finkelhor

Ormrod

R. K.

(2010). Poly-victimization in a national sample of children & youth. American Journal of Preventive Medicine, 38, 323-330.

54.

Tyler

K. A.

Hoyt

D. R.

Whitbeck

L. B.

Cauce

A. M.

(2001). The impact of childhood sexual abuse on later sexual victimization among runaway youth. Journal of Research on Adolescence, 11, 151-176. doi:10.1111/1532-7795.00008

55.

van der Kolk

B. A.

Perry

J. C.

Herman

J. L.

(1991). Childhood origins of self-destructive behavior. American Journal of Psychiatry, 148, 1665-1671.

56.

Walsh

Danielson

C. K.

McCauley

J. L.

Saunders

B. E.

Kilpatrick

D. G.

Resnick

H. S.

(2012). National prevalence of posttraumatic stress disorder among sexually revictimized adolescent, college, and adult household-residing women. Archives of General Psychiatry, 69, 935-942. doi:archgenpsychiatry.2012.132 [pii] 10.1001/archgenpsychiatry.2012.132

57.

Weierich

M. R.

Nock

M. K.

(2008). Posttraumatic stress symptoms mediate the relation between childhood sexual abuse and nonsuicidal self-injury. Journal of Consulting and Clinical Psychology, 76, 39-44. doi:10.1037/0022-006X.76.1.39 2008-00950-006 [pii]

58.

Whitbeck

L. B.

Simons

R. L.

(1990). Life on the streets: The victimization of runaway and homeless adolescents. Youth & Society, 22, 108-125.

59.

Yates

T. M.

Carlson

E. A.

Egeland

(2008). A prospective study of child maltreatment and self-injurious behavior in a community sample. Development and Psychopathology, 20, 651-671. doi:10.1017/S0954579408000321S0954579408000321 [pii]