Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Violence Vict. Author manuscript; available in PMC 2013 April 18.
Published in final edited form as:
Violence Vict. 2012; 27(5): 793–810.
PMCID: PMC3629810

Correlates of Serious Violent Crime for Recently Released Parolees With a History of Homelessness

Adeline Nyamathi, ANP, PhD, FAAN, Mary Marfisee, MD, MPH, Farinaz Khalilifard, BA, MA, Barbara Leake, PhD, Elizabeth Marlow, PhD, C-FNP, Sheldon Zhang, PhD, Elizabeth Hall, PhD, David Farabee, PhD, and Mark Faucette, RAS, CSAC


This study used baseline data on recently-released homeless paroled men who are homeless (N = 157), residing in a residential drug treatment program, and enrolled in a longitudinal study to examine personal, developmental, and social correlates of parolees who are homeless and parolees who have committed serious violent offenses. Having experienced childhood sexual abuse, poor parental relationships, and early-onset incarceration (prior to 21 years of age) were important correlates of serious violent crimes. These findings highlight the need for interventions that address offenders’ prior adult and childhood victimization, and suggest that policies for reentering violent offenders should encompass an understanding of the broader family contexts in which these patterns of maltreatment often occur.

Keywords: violence, serious violent crimes, childhood history, paroled men

The State of California spends more on its correctional system than any other in the nation and houses over more than 170,000 prisoners; another 123,000 parolees reside in the community (California Department of Corrections and Rehabilitation [CDCR], 2009). Experts contend that effective community reintegration continues to remain elusive for the state; in particular, 96% of inmates will eventually be paroled back into the community (Petersilia, 2003; Travis & Waul, 2004). However, previously incarcerated populations are at heightened risk for homelessness (Greenberg & Rosenheck, 2008; McNiel, Binder, & Robinson, 2005; Rodriguez & Brown, 2003); in particular, up to 50% of those on parole supervision in San Francisco and Los Angeles have become homeless (Travis, Solomon, & Waul, 2001) because of challenges with locating appropriate housing (Roman & Travis, 2004; Travis et al., 2001), lack of employment opportunities (Zhang, Roberts, & Callanan, 2003), weakened family and community ties (Travis et al., 2001), as well as, limited education and substance abuse (Roman & Travis, 2004). Being homeless is similarly a risk factor for involvement in the criminal justice system (Kushel, Hahn, Evans, Bangsberg, & Moss, 2005), and may be rooted in the need to survive on limited resources.

Currently, relatively few studies have assessed the personal, developmental, and social correlates of homeless parolees convicted of committing serious violent crimes. The purpose of this cross-sectional study is to describe characteristics associated with a history of committing serious violent offenses among a sample of male parolees enrolled in a residential drug treatment facility. Serious violent crime was defined by the United States Department of Justice (USDOJ) and the Federal Bureau of Investigation (FBI) as murder and non-negligent manslaughter, rape or other sexual assault, robbery, and aggravated assault. Knowledge to be gained from this study is critical as an understanding of correlates of violent behaviors may lead to improved interventions and treatment protocols designed to decrease violent crimes and the experience of incarceration.


Impoverished living conditions are a known risk factor for violent crimes (Smith, Ireland, & Thornberry, 2005; Turner, Hartman, & Bishop, 2007); however, the literature that ties homelessness to serious violent crimes is very limited. In one study, 88% of those ever imprisoned had been homeless, and 41% had been marginally housed in the last year (Kushel et al., 2005). While on the streets, many persons who are homeless persons become involved with prostitution, survival sex (Miller et al., 2011), as well as illicit drug use (Riley et al., 2007); and in one study of recently discharged inmates who are homeless inmates, findings revealed that these homeless ex-offenders who are homeless were less likely to receive public assistance than the more veteran homeless parolees who are homeless (Hudson et al., 2009).

For homeless populations, data suggest that current drug use is predicted by less positive coping (Galaif, Nyamathi, & Stein, 1999), which may potentiate further violence (Fagan, 1993; Haggard-Grann, Hallqvist, Langstrom, & Moller, 2006). In addition, low self-esteem and depression are also common among men who are homeless men with a history of incarceration (Nyamathi et al., 2011). Drug use is also related to interpersonal violence; for example, alcohol may lead to an intoxication–violence relationship, and withdrawal from cannabis may lead to aggressive tendencies (Hoaken & Stewart, 2003). These factors may lead to violent crimes and subsequent incarceration as because data suggest that 56% of men were under the influence of drugs or alcohol when the crime was committed (National Crime Victims Center [NCVC], 2011).

For successful community reintegration, clinicians and researchers require knowledge of the personal, developmental, and social challenges ex-offenders have faced, particularly for those who are homeless and have committed serious violent offenses. In an effort to understand these factors, the Comprehensive Health Seeking and Coping Paradigm (CHSCP), - a framework which has been utilized used for over more than two decades with vulnerable homeless populations (Nyamathi, 1989; Nyamathi, Flaskerud, Bennett, Leake, & Lewis, 1994; Nyamathi et al., 2012), - will identify the challenges parolees have grappled with in an effort to enable providers to design targeted interventions to circumvent the obstacles homeless ex offenders who are homeless face in reentering the community. Based on this framework, antecedent (gender, race, age, and history of homelessness, abuse, and incarceration) and, mediating (coping, social support, mental health) variables may influence outcome variables. In this article, the outcome is serious violent crime. These findings may enable providers to design targeted interventions in order to circumvent obstacles of reentering into the community.


Violent offenders, particularly those with prior records of violence, have long attracted attention from the research community. Violent offenders frequently have a history of experiencing a myriad of adverse personal, social, and developmental challenges (Huebner, Varano, & Bynum, 2007; Phillips, Gleeson, & Waits-Garrett, 2009).

Although less is known about correlates of homeless persons with a history of committing serious violent crimes, general non-modifiable personal predictors of violent behaviors are gender, race, and age. Gender has long been identified as one of the most significant predictors for violence—men are much more likely than females to commit violent crimes (Allen, Swan, & Raghavan, 2009), and being African- American has been found related to committing violent crimes (Warren et al., 2002). However, when prior violent victimization (Burnette et al., 2008) was controlled, the effect of race disappeared.

Developmental and social hardships, such as childhood and adolescent trauma including physical and sexual abuse (Christofferson, Soothill, & Francis, 2010; Felson & Lane, 2009), interparental violence (Hill & Nathan, 2008), poor and unstable parenting (Loeber et al., 2005), impoverished living conditions (Smith et al., 2005; Turner et al., 2007), and gang membership (Bellair & McNulty, 2009; Bjerregaard, 2010; Huebner et al., 2007), are also associated with the commission of violent crimes. Additionally, the use of drugs and alcohol (Darke, Torok, Kaye, Ross, & McKetin, 2010) and access to guns (Curfman, Morrissey, & Drazen, 2008; Stolzenberg & D’Alessio, 2000) increases the risk of interpersonal violence. Although the added challenge of homelessness has not been considered in terms of serious violent crimes, it is clear that adults who are homeless adults report significant histories of childhood abuse and poor parental support (Stein, Leslie, & Nyamathi, 2002; Stein, Nyamathi, & Zane, 2009); - all of which impact ongoing drug and alcohol use and criminal activity.


Although there is evidence of the linkage of homelessness, mental illness, and history of incarceration, (Metreaux, Culhane, & Michaels, 2002), the linkage to serious violent crimes is scant. Moreover, there is considerable disagreement about the relationship between mental illness and violent behavior, which may stem from differences in psychiatric diagnostic conventions and measures, definitions of criminality, heterogeneity in the group (Sirotich, 2008), and rates of substance use disorders (SUDs) among groups who are mentally ill and criminally -involved groups (Darke et al., 2010; Volavka & Swanson, 2010).

Successful community reintegration depends on addressing underlying mental health and social problems. In one large study of prisoners (N = 22,790), 10% met the criteria for major depression, 3.7% of men had a psychotic illnesses, and 47% had antisocial personality disorder (Fazel & Danesh, 2002). Dutch researchers found that an individual’s criminal behavior tended to grow increasingly violent over time; this finding was attributed to a combination of psychopathology and poor social skills (Geest, Blokland, & Bijleveld, 2009). Schizophrenia and bipolar disorder, particularly in combination with SUDs, are also thought to contribute to violent behaviors (Fazel, Lichtenstein, Grann, Goodwin, & Langstrom, 2010; Sacks et al., 2009).


This study uses baseline data on men from a larger, ongoing study in which recently -released parolees who were designated as homeless were randomized to one of three programs focused on hepatitis/HIV education, hepatitis B vaccination (for all eligible), and coach facilitated mentoring. The University of California at Los Angeles (UCLA) Human Subjects Protection Committee approved the study.

Sample and Site

The sample included 157 parolees who were released from prison and jails within a month before recruitment. These men all met the following criteria: (a) had a history of drug use prior to entry into the prison system; (b) aged 18–60 years; (c) entered a participating drug treatment program; and (d) labeled as homeless on their prison exit form. Participants were excluded if they did not speak English or Spanish or b) judged to be cognitively impaired, as measured by the Short -Blessed Screener (National Institute on Drug Abuse [NIDA], 1993). Participants were recruited from a residential drug treatment facility in Southern California. Parolees who reside at State-funded- or Federally -funded residential drug treatment program were designated as homeless by the correctional department based on the Stewart B. McKinney-Vento Act of 1994 (42 USC Sec. 11302), because these individuals resided in a supervised publicly or privately operated shelter designated to provide temporary living arrangements.


Potential participants were made aware of the study by posted flyers and a presentation by trained staff at the drug treatment site. Research staff met with drug treatment participants who expressed interest in a private area of the facility. For those meeting basic eligibility criteria, and after informed consent for the study had been discussed, read, and signed, the staff administered a brief 2-minute structured screening questionnaire asking about socio demographic characteristics, hepatitis-related information, and locator information. For those eligible and interested in participating in the full study, the research staff read and discussed the final informed consent; after signing, the baseline questionnaire was administered.


Sociodemographic information was collected by a structured questionnaire that assessed age, birthdate, country of birth, race/ or ethnicity, education, partnership status, number of parents in family, family socioeconomic status (SES), history of victimization, drug and alcohol use, and criminal history.

Childhood Relationships were assessed by the following questions: Relationship with parents (measured on a 6-point Likert scale ranging from “excellent relationship” to “no relationship”), closeness of family (measured on a 6-point Likert scale ranging from “very close” to “don’t know”), supportiveness of the mother (measured on a 5-point Likert scale ranging from “very supportive” to “not around”), and supportiveness of the father (measured on a 5-point Likert scale ranging from “very supportive” to “not around”).

Crime was assessed by the Texas Christian University (TCU) Correctional Assessment intake crime grid, which records information including number of arrests, types of crimes, and incarceration history (Simpson & Knight, 1998). This study focused on three categories of violent crime, namely kidnapping/hostage taking, homicide/manslaughter/attempted homicide, and violent sex offenses (rape/aggravated assault of a minor). For purposes of this study, we refer to crime in any of these three groups as “serious violent crime.” It is important to note that domestic violence (DV) and intimate partner violence (IPV) and stalking, as well as, incest were not included in this definition.

Self-Esteem was measured using the revised 23-item Self-Esteem Inventory (SEI) (Coopersmith, 1967). The internal consistency of this scale with homeless males was .83 (Nyamathi, Flaskerud, & Leake, 1997). Internal consistency in this study, as measured by Cronbach’s alpha, was .84. Additionally, parolees were asked how they had felt about themselves as teenagers on a 5-point Likert scale ranging from “liked yourself a great deal” to “disliked yourself a great deal.”

Coping Behaviors were measured by a 12- item Brief COPE (Carver, 1997). These items represented six separate subscales, with two items each. The items are rated on a 4-point Likert scale from “not at all” to “a lot.” Internal consistencies for these subscales in this study were .61 for self-blame coping, .66 for denial coping, .74 for disengagement coping, .76 for instrumental support coping, .80 for planning coping, and .84 for religious coping.

Drug and Alcohol Use Behaviors were measured by a modified TCU Drug History form (Simpson & Chatham, 1995). This questionnaire has been tested with men and women with a history of drug addiction and homelessness. The modified form recorded the frequency of use of alcohol, tobacco, and seven drugs and selected combinations used by injection and orally during the 6 months prior to the last incarceration and also elicited information about lifetime use. Drugs assessed were: heroin, cocaine, crack, methamphetamine, inhalants, marijuana/hashish, and hallucinogens. Participants were also asked if they thought they had a drinking problem.

Depressive Symptoms were assessed with the 10-item short form of the Center for Epidemiological Studies Depression (CES-D) Scale (Radloff, 1977). This self-report instrument is designed to assess depressive symptomology in the past week; it measures the frequency of a symptom on a 4-point response scale from 0 (“rarely or none of the time [(less than 1 day])” to 3 (“all of the time [(5–7 days])”. The internal reliability of the scale in this sample was .80. Scale scores ranged from 0 to 30 with higher scores indicating greater symptom severity. Cut -points of 8 and 10 have been used to screen for individuals who may need psychiatric evaluation.

Emotional well-being was measured by the five-item mental health index (MHI-5), which has a well-established reliability and validity (Stewart, Hays, & Ware, 1988). Cronbach’s alpha for the scale in this study was .81. Following convention, scores were linearly transformed to a range of 0–100, with higher scores signifying greater emotional well-being.

Social support was measured by 18 items from the Medical Outcomes Study (MOS) Social Support Survey (Sherbourne & Stewart, 1991). These items contained 4 four subscales: emotional support (8 eight items), tangible support (3 three items), positive support (4 four items), and affective support (3 three items). Reliabilities for these subscales were .95, .86, .91, and .88, respectively.

Data Analysis

Unadjusted associations of serious violent crime with socio-demographic, substance use, and psychosocial measures were examined with chi-square and two -sample t -tests. To delineate important adjusted associations involving both distal and proximal factors, we conducted a series of staged stepwise backward logistic regression analyses. At each stage, the initial model included selected predictors that were related to serious violent crime at the .15 level in preliminary analyses. The retention level was .10. Stage 1 predictors consisted of childhood -demographic and family variables. Stage 2 introduced childhood or adolescent abuse factors into the model produced in Stage 1. In Stages 3 and 4, childhood or adolescent personal and legal history factors and current demographic and psychosocial variables, respectively, were added to models from previous stages. The final model was tested for multicollinearity, and goodness of fit was assessed with the Hosmer-Lemeshow test.


Sociodemographic Characteristics

Parolees on average were 41.9 years of age (SD =: 10.1) and generally had completed less than 12 years of education (SD =: 1.9;) (See Table 1). The majority was either African American (47%) or Latino (26%), and nearly 90% were either never married (63%) or separated/divorced (27%). Almost two -thirds had children. In total, 65 (41.4%) of the sample reported a history of serious violent crime in the past. Specifically, 42 reported a history of homicide, manslaughter, or attempted homicide, 22 reported a history of kidnapping (hostage taking); and one man reported multiple violent sex offenses. On the average, time in prison was 2.7 years (SD = 5.3 years) and the range was from 1 month to 33 years. All residents who have been at the residential drug treatment program for at least 30 days met the criteria for inclusion. We have a 94% response rate, as 6% did not meet the criteria because of being outside the age range.

Table 1
Sample Characteristics

More than half of the parolees had a history of group living or institutional (juvenile hall, jail, or orphanage) setting in childhood. More than 15% reported childhood sexual abuse, 31% reported childhood physical abuse, and verbal abuse was common (47%). About 40% of the sample had been imprisoned before the age of 21, and almost half reported belonging to a gang.

Childhood Relationships

More than half of the sample at baseline reported having fair- to-very poor or no relationship with parents. Yet, more than 70% reported “very supportive” to “somewhat supportive” mothers, whereas less so for the father (42%, data not shown). Seven percent of mothers and 37% of fathers were considered absent. Nearly 11% and 13% of mothers and fathers, respectively, had been treated for alcohol problems. About one third (37%) of parolees liked themselves as a teenager a great deal (data not shown).

Substance Use

Among this cohort of parolees, many substances were reported. Alcohol use was reported by about 90% of parolees prior to the last incarceration. Among drugs ever used, the most prevalent were marijuana (88%), crack (66%), cocaine (65%), methamphetamine (49%) and hallucinogens (48%;) (See Table 2).

Table 2
Substance Use and Psychosocial Resources

Psychosocial Factors

Perceiving drinking as a problem was not uncommon (26%). Prevalence of self-reported depression and anxiety were similar (30% and 28%, respectively); depressive symptomatology was moderately high, and emotional well-being was poor on average. Self-esteem was reported as moderate. Average scores were a little below the midpoint for denial and disengagement coping, and slightly above the midpoint for self-blame, planning, religious, and instrumental support coping. The sample averaged close to the midpoint on each of the four social support subscales.

Associations with Serious Violent Crime

Age and education were not found to be significantly associated with serious violent crime (Table 3), although weak associations were found for being Latino and married. Family SES, coming from a two-parent family, and having children were also not related to serious violent crime (data not shown). However, childhood relationship measures and personal abuse factors were important. These measures and factors included adult physical or sexual abuse (within 6 months preceding the current incarceration); poor relationships with parents; lack of family closeness; and childhood physical, sexual, and verbal abuse. Being on their own before the age of 18, was another important correlate. Moreover, being in juvenile hall as a teenager, in prison before the age of 21, not particularly liking oneself as a teen, and having been in a gang were associated with serious violent crime. Of interest, Latinos were almost twice as likely to have been involved with gangs as their peers (75% vs. 40.5%, respectively, p < .001; data not shown). Trends were also observed for having a mother who was present and treated for alcohol problems and having been in a group care setting in childhood.

Table 3
Associations Between Sample Characteristics and Serious Violent Crime

No significant correlates were found among the substance use variables, although there were trends for use of heroin, hallucinogens, inhalants, and the combination drugs to be associated with serious violent crime. In terms of psychosocial measures and resources, self reported anxiety, self-reported depression, and lower levels of religious coping were found to be related to serious violent crime (p < .05). No important associations were found for any of the other measures in Table 2, although a trend was found for self-reported drinking problems.

Multivariate Results

In Stage 1, using the CHSCP as a guide, stepwise backward logistic regression analysis with candidate predictors from the childhood demographic and family background categories was conducted; being Latino and having a poor relationship with parents were important predictors of serious violent crime. Having a mother who was present and treated for alcohol problems had a weaker association with serious violent crime. When childhood or adolescent abuse variables were allowed to enter the model in Stage 2, only sexual abuse did so, and the previous relationships were maintained. In Stage 3, childhood or adolescent personal and legal history variables were assessed as predictors. The resulting model included having been in juvenile hall as a young teen (aged 13–17 years) and having been imprisoned before the age of 21. Having a mother treated for alcohol problems was no longer important. In the final stage, current sociodemographic characteristics and psychosocial measures were added as potential predictors. Adult physical or sexual victimization (within 6 months prior to the current incarceration), self-reported lifetime problem with anxiety, and being either married or living as married were important additional correlates of serious violent crime; a weaker association with ever having been in a gang was also found. Latino ethnicity and having been in juvenile hall between the ages of 13 and 17 years no longer had important effects. However, having been in prison before the age of 21, childhood sexual abuse, and poor parental relationships remained as independent correlates of serious violent crime (Table 4).

Table 4
Summary of Staged Logistic Regressions on Serious Violent Crime


Findings from this study elucidate the complex associations related to serious violent crime. In this study and as guided by the CHSCP, participants who had committed serious violent crimes were more likely to have experienced childhood sexual abuse, poor parental relationships, and early-onset incarceration (prior to 21 years of age). Past gang membership and physical or sexual victimization within the 6 months prior to their current incarceration were also associated with the commission of serious violent crime. Being married was another factor found to be associated with serious violent crime. This finding may suggest the presence of spousal involvement in violent crime as well. Of note, none of substance use, coping, or current social support measures was related to serious violent crime in the regression models.

Although abstinence from drugs, more adaptive coping, and enhanced social support are laudable goals for re-entry programs, they are unlikely to keep parolees from re-offending in the absence of efforts to address the multidimensional problems confronting them. Further investigation of the impact of marriage on crime may also be of interest.

The impact of childhood exposure to violence, including physical and sexual abuse and inter-parental violence, and of social violence on an adult’s commission of serious violent crime is well documented (Christofferson et al., 2010; Felson & Lane, 2009; Sarchiapone, Carli, Cuomo, Marchetti, & Roy, 2009). A large cohort study of young men found that those convicted of a lethal violent crime were 4.5 times more likely to have been physically abused and neglected than those not convicted (Christofferson et al., 2010). In addition to homicide, childhood physical abuse has been linked to an adult’s sexual or physical assault on other adults (Felson & Lane, 2009). As for childhood sexual trauma, Sarchiapone et al. (2009) found that childhood neglect and emotional, physical, and sexual abuse were associated with high levels of aggression, conviction at an early age and more than once, violence in prison, and continued violence on release. Finally, physical and sexual maltreatment during adolescence increases the risk of violent offending in both late adolescence and early adulthood (Smith et al., 2005).

Our findings support what is documented in current research, namely, that childhood trauma is strongly associated with violent crime and has a deleterious effect on a young adult’s maturation. Therefore, interventions for violent offenders must address their personal histories of trauma and victimization as both children and adults. For example, assessment and treatment of mental illness, including depression, anxiety, and post-traumatic stress disorder (PTSD), should be integrated into substance abuse treatment (SAT) programs for this population (Ford, Russo, & Mallon, 2007). Given the high-rates of SUDs, physical and sexual trauma, and mental illness in the correctional population (James & Glaze, 2006; Mumola & Karberg, 2006), the SAT setting is an appropriate place to address the manifold problems that violent offenders experience. Coordinated efforts could improve parolees’ mental well-being, increase successful community reintegration, and decrease violent crime.

Family dynamics, in particular, parental relationships, also influenced participants’ commission of serious violent crimes. Of those participants committing such crimes, two-thirds had poor parental relationships and close to one-half described their families as not being close. Parental engagement and child-rearing practices, as well as parental substance use and incarceration, have been described as predictors of violent behavior in the literature (Phillips et al., 2009; Turner et al., 2007; Woldoff & Washington, 2008). A longitudinal, multiple cohort study of delinquency in boys from adolescence to young adulthood found that boys with four or more violence risk factors were six times more likely to commit violence and 14 times more likely to commit homicide as young adults than those with less than four risk factors (Loeber et al., 2005). Parental risk factors included unstable child rearing, (i.e., more than one caregiver before the age of 10), physical punishment, and poor supervision and communication (Loeber et al., 2005; Turner et al., 2007). Each of these factors can set the stage for long-term parent child disaffection. Other factors that may negatively influence the child-parent dynamic include parental substance abuse (Phillips et al., 2009; Woldoff & Washington, 2008). Parental risks increase a child’s exposure to poor living conditions, decrease access to education, and foster emotional and behavioral problems; these factors place the youth at risk for criminal behaviors, including violent criminality (Phillips et al., 2009; Turner et al., 2007). Results from this study provide further evidence that poor parental relations may negatively influence violent criminal activity into adulthood. Policies, programs, and interventions for reentering violent offenders must begin to consider their problematic behaviors across the varied domains in which they occur, including the family (Huey, Henggeler, Brondino, & Pickrel, 2000).

One such program,—multisystemic therapy (MST),—is a psychotherapeutic intervention that considers the family essential to long-lasting behavior change (Swenson, Schaeffer, Henggeler, Faldowski, & Mayhew, 2010). MST has demonstrated efficacy in treating adolescents with serious antisocial behaviors (Borduin, Schaeffer, & Heiblum, 2009) and may be appropriate for addressing violent adult offenders’ problems, particularly around parenting, resolving family conflict, and establishing meaningful relationships with children, partners, and other family members (Huey et al., 2000; Swenson et al., 2010). MST’s holistic and individualized approach for addressing problematic behaviors across multiple social arenas provides a model for improving relations within the violent offender’s familial context, as well as potentially preventing intergenerational incarceration. Future research should consider residential drug treatment programs that incorporate the MST approach into other treatment modalities parolees receive, such as anger management training, and parenting courses.

Gang membership was found to be a correlate of violent crime in the final model. Gang membership is associated with violent crime and increased involvement in crime generally (Bellair & McNulty, 2009; Bjerregaard, 2010). Although gang membership increases exposure to violent crime, it also reflects the economic disadvantage experienced by many communities where gangs are prevalent (Bellair & McNulty, 2009). For many gang members, criminality, in particular selling drugs, is a primary form of income generation (Bellair & McNulty, 2009). Involvement in drug sales is associated with increased firearm use and, therein, increased opportunity to commit a violent crime (Bjerregaard, 2010). Cutting gang ties may be an important component of successful rehabilitation. Regarding homicides, both gang- and non-gang-related homicides were committed by young, African American males who killed their victims with guns (Decker & Curry, 2002). Thus, both groups were likely to engage in and experience violence, but being in a gang increased both risks (Decker & Curry, 2002). Our findings concur that being a perpetrator and victim was correlated with serious violent crimes.

Finally, current research describes the connection between violence and socioeconomic disadvantage, and our findings support this observation as well-64% of the violent offenders in this sample were either poor or working class (Bellair & McNulty, 2009; Bjerregaard, 2010). Thus, programs that encourage education, employment, and financial equity are critical for this population. Education and employment are associated with decreased criminal activity and increased community stability (Lochner & Morette, 2004). Work-place readiness, computer literacy, general education courses, and employment assistance should be incorporated into all programs serving the parole population. Future research should also examine the relationship between stable employment and reductions in crime, including serious violent crimes.


This study is limited by a small sample whose characteristics may reflect the urban residential drug treatment program where the sample was recruited (e.g., older than average for California parolees, larger percentage of African Americans) and, therefore, may differ from other homeless parolee populations. We also do not know participants’ ages when they committed their crimes. Another limitation is the restricted geography of the current sample. However, California is the largest state in the nation and a logical place to begin interventions to improve community reintegration of parolees. This study is also derived from baseline data; thus, no causal associations can be made. Other limitations include self-report data, such as substance use, although parolees must have had a substance use problems to qualify for treatment. Moreover, the sample had limited ranges in education and family SES, making it difficult to detect associations. Nevertheless, many of the associations found in this sample concur with those found in other studies of parolees. Furthermore, there was a relatively large subgroup of Latinos, reflecting California’s diverse ethnic population.


This study sheds light on the environmental and social factors that are associated with the commission of violent crime. These factors include deficits in family structure and childhood support, early exposure to and experience of sexual violence that can continue into adulthood, easy access to gang membership, early exposure to the correctional system, and socioeconomic disadvantage. It is important for clinicians, other service providers, and policy makers to be aware that homeless violent offenders who are homeless bring complex backgrounds and marginalized circumstances with them to the community on release. These individuals’ family and social histories and mental health status must be considered if programs and policies are to effectively address their reintegration needs, and, in turn, prevent future violent crime.


This study is funded by the National Institute on Drug Abuse, 1R01DA27213-01.

Contributor Information

Adeline Nyamathi, University of California, Los Angeles, School of Nursing.

Mary Marfisee, University of California, Los Angeles, School of Nursing.

Farinaz Khalilifard, University of California, Los Angeles, School of Nursing.

Barbara Leake, University of California, Los Angeles, School of Nursing.

Elizabeth Marlow, Department of Family Health Care Nursing.

Sheldon Zhang, San Diego State University, San Diego.

Elizabeth Hall, University of California, Los Angeles, Integrated Substance Abuse Program.

David Farabee, University of California, Los Angeles, Integrated Substance Abuse Program.

Mark Faucette, Amity Foundation.


  • Allen CT, Swan SC, Raghavan C. Gender symmetry, sexism, and intimate partner violence. Journal of Interpersonal Violence. 2009;24(11):1816–1834. [PubMed]
  • Bellair PE, McNulty TL. Gang membership, drug selling, and violence in neighborhood context. Justice Quarterly. 2009;26:644–669. [PMC free article] [PubMed]
  • Bjerregaard B. Gang membership and drug involvement: Untangling the complex relationship. Crime and Delinquency. 2010;56:3–34.
  • Borduin CM, Schaeffer CM, Heiblum N. A randomized clinical trial of multisystemic therapy with juvenile sexual offenders: Effects on youth social ecology and criminal activity. Journal of Consulting and Clinical Psychology. 2009;77(1):26–37. [PubMed]
  • Burnette ML, Ilgen M, Frayne SM, Lucas E, Mayo J, Weitlauf JC. Violence perpetration and childhood abuse among men and women in substance abuse treatment. Journal of Substance Abuse Treatment. 2008;35(2):217–222. [PubMed]
  • California Department of Corrections and RehabilitationCDCR. Annual report. 2009 Retrieved from
  • Carver CS. You want to measure coping but your protocol’s too long: Consider the Brief COPE. International Journal of Behavioral Medicine. 1997;4(1):92–100. [PubMed]
  • Christofferson MN, Soothill K, Francis B. Violent life events and social disadvantage: A systematic study of social background of various kinds of lethal violence, other violent crime, suicide, and suicide attempts. Journal of Scandinavian Studies in Criminology and Crime Prevention. 2010;8:157–184.
  • Coopersmith S. The antecedents of self-esteem. San Francisco, CA: W.H. Freeman; 1967.
  • Curfman GD, Morrissey S, Drazen JM. Handgun violence, public health, and the law. The New England Journal of Medicine. 2008;358(14):1503–1504. [PubMed]
  • Darke S, Torok M, Kaye S, Ross J, McKetin R. Comparative rates of violent crime among regular methamphetamine and opioid users: Offending and victimization. Addiction. 2010;105(5):916–919. [PubMed]
  • Decker SH, Curry DG. Gangs, gang homicides, and gang loyalty: Organized crimes or disorganized criminals. Journal of Criminal Justice. 2002;30:343–352.
  • Fagan J. Interactions among drugs, alcohol, and violence. [Review] (Millwood) Health Affairs (Project Hope) 1993;12(4):65–79. [PubMed]
  • Fazel S, Danesh J. Serious mental disorder in 23000 prisoners: A systematic review of 62 surveys. Lancet. 2002;359(9306):545–550. [pii]10.1016/S0140-6736(02)07740-1. [PubMed]
  • Fazel S, Lichtenstein P, Grann M, Goodwin GM, Langstrom N. Bipolar disorder and violent crime: New evidence from population-based longitudinal studies and systematic review. Archives of General Psychiatry. 2010;67(9):931–938. [PubMed]
  • Felson RB, Lane KJ. Social learning, sexual and physical abuse, and adult crime. Aggressive Behavior. 2009;35(6):489–501. [PubMed]
  • Ford JD, Russo EM, Mallon SD. Integrating treatment of posttraumatic stress disorder and substance use disorder. Journal of Counseling and Development. 2007;85:475–490.
  • Galaif ER, Nyamathi AM, Stein JA. Psychosocial predictors of current drug use, drug problems, and physical drug dependence in homeless women. [Research Support, U.S. Gov’t, P.H.S.] Addictive Behaviors. 1999;24(6):801–814. [PubMed]
  • Geest V, Blokland A, Bijleveld C. Delinquent development in a sample of high risk youth: Shape, content, and predictors of delinquent trajectories from 12–32. Journal of Research in Crime and Delinquency. 2009;46:111–143.
  • Greenberg GA, Rosenheck RA. Jail incarceration, homelessness, and mental health: A national study. Psychiatric Services (Washington, D C) 2008;59(2):170–177. [PubMed]
  • Haggard-Grann U, Hallqvist J, Langstrom N, Moller J. The role of alcohol and drugs in triggering criminal violence: A case-crossover study*. [Multicenter Study Research Support, Non-U.S. Gov’t] Addiction. 2006;101(1):100–108. http// [PubMed]
  • Hill J, Nathan R. Childhood antecedents of serious violence in adult male offenders. Aggressive Behaviors. 2008;34:329–338. [PubMed]
  • Hoaken PN, Stewart SH. Drugs of abuse and the elicitation of human aggressive behavior. Addictive Behaviors. 2003;28(9):1533–1554. [PubMed]
  • Hudson AL, Nyamathi A, Slagle A, Greengold B, Griffin DK, Khalilifard F, Reid C. The power of the drug, nature of support, and their impact on homeless youth. Journal of Addictive Diseases. 2009;28(4):356–365. [PMC free article] [PubMed]
  • Huebner BM, Varano SP, Bynum TS. Gangs, guns, and drugs: Recidivism among serious young offenders. Crime and Public Policy. 2007;6:187–222.
  • Huey SJ, Jr, Henggeler SW, Brondino MJ, Pickrel SG. Mechanisms of change in multisystemic therapy: Reducing delinquent behavior through therapist adherence and improved family and peer functioning. Journal of Consulting and Clinical Psychology. 2000;68(3):451–467. [PubMed]
  • James DJ, Glaze LE. Mental health problems of prison and jail inmates. Washington, D.C: U.S. Department of Justice, Bureau of Justice Statistics; 2006.
  • Kushel MB, Hahn JA, Evans JL, Bangsberg DR, Moss AR. Revolving doors: Imprisonment among the homeless and marginally housed population. [Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov’t Research Support, U.S. Gov’t, P.H.S.] American Journal of Public Health. 2005;95(10):1747–1752. [PubMed]
  • Lochner L, Morette E. The effect of education on crime: Evidence from prison inmates, arrests, and self-reports. The American Economic Review. 2004;94:155–189.
  • Loeber R, Pardini D, Homish DL, Wei EH, Crawford AM, Farrington DP, Rosenfeld R. The prediction of violence and homicide in young men. Journal of Consulting and Clinical Psychology. 2005;73(6):1074–1088. [PubMed]
  • McNiel DE, Binder RL, Robinson JC. Incarceration associated with homelessness, mental disorder, and co-occurring substance abuse. [Research Support, Non-U.S. Gov’t] Psychiatric Services (Washington, D C) 2005;56(7):840–846. [PubMed]
  • Metreaux S, Culhane D, Michaels D. Homelessness and indicators of mental illness among inmates in New York City’s correctional system. Hospital and Community Psychiatry. 2002;43:150–155. [PubMed]
  • Miller CL, Fielden SJ, Tyndall MW, Zhang R, Gibson K, Shannon K. Individual and structural vulnerability among female youth who exchange sex for survival. [Research Support, Non-U.S. Gov’t] The Journal of Adolescent Health. 2011;49(1):36–41. [PMC free article] [PubMed]
  • Mumola CJ, Karberg JC. Drug use and dependence, state and federal prisoners. Washington, D.C: Bureau of Justice Statistics; 2006.
  • National Crime Victims Center. Drug related crime. 2011 Retrieved from
  • National Institute on Drug Abuse. Prevalence of drug use in the Washington D.C. metropolitan area homeless and transient population. Rockville, MD: Department of Health and Human Services; 1993.
  • Nyamathi A. Comprehensive health seeking and coping paradigm. [Review] Journal of Advanced Nursing. 1989;14(4):281–290. [PubMed]
  • Nyamathi A, Flaskerud J, Bennett C, Leake B, Lewis C. Evaluation of two AIDS education programs for impoverished Latina women. [Clinical Trial Comparative Study Randomized Controlled Trial Research Support, U.S. Gov’t, P.H.S.] AIDS Education and Prevention. 1994;6(4):296–309. [PubMed]
  • Nyamathi A, Flaskerud J, Leake B. HIV-risk behaviors and mental health characteristics among homeless or drug-recovering women and their closest sources of social support. Nursing Research. 1997;46(3):133–137. [PubMed]
  • Nyamathi A, Leake B, Albarran C, Zhang S, Hall E, Farabee D, Faucette M. Correlates of depressive symptoms among homeless men on parole. Issues in Mental Health Nursing. 2011;32(8):501–511. [PMC free article] [PubMed]
  • Nyamathi A, Marfisee M, Slagle A, Greengold B, Liu Y, Leake B. Correlates of depressive symptoms among homeless young adults. Western Journal of Nursing Research. 2012;34(1):97–117. [PMC free article] [PubMed]
  • Petersilia J. When prisoners come home -Parole and prisoner reentry. New York, New York: Oxford, University; 2003.
  • Phillips SD, Gleeson JP, Waits-Garrett M. Substance-abusing parents in the criminal justice system: Does substance abuse treatment improve their children’s outcomes. Journal of Offender Rehabilitation. 2009;48:120–138.
  • Radloff L. The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurements. 1977;1:385–401.
  • Riley ED, Weiser SD, Sorensen JL, Dilworth S, Cohen J, Neilands TB. Housing patterns and correlates of homelessness differ by gender among individuals using San Francisco free food programs. [Research Support, N.I.H., Extramural] Journal of Urban Health. 2007;84(3):415–422. [PMC free article] [PubMed]
  • Rodriguez N, Brown B. Preventing homelessness among people leaving prison: Vera Institute of Justice. 2003 Retrieved from
  • Roman C, Travis J. Taking stock: Housing, homelessness, and prisoner reentry. The Fannie Mae Foundation; Washington, DC: Urban Institute, Justice Policy Center; 2004. p. 126.
  • Sacks S, Cleland CM, Melnick G, Flynn PM, Knight K, Friedmann PD, Coen C. Violent offenses associated with co-occurring substance use and mental health problems: Evidence from CJDATS. Behavioral Sciences & The Law. 2009;27(1):51–69. [PMC free article] [PubMed]
  • Sarchiapone M, Carli V, Cuomo C, Marchetti M, Roy A. Association between childhood trauma and aggression in male prisoners. Psychiatry Research. 2009;165(1–2):187–192. [PubMed]
  • Sherbourne CD, Stewart AL. The MOS social support survey. Social Science & Medicine. 1991;32(6):705–714. [PubMed]
  • Simpson D, Chatham L. TCU/DATAR forms manual. Fort Worth, TX: Institute of Behavioral Research, Texas Christian University; 1995.
  • Simpson D, Knight K. TCU data collection forms for correctional residential treatment. Fort Worth, Texas: Texas Christian University, Institute of Behavioral Research; 1998.
  • Sirotich F. Correlates of crime and violence among persons with mental disorder: An evidence based review. Treatment and Crisis Intervention. 2008;8:171–194.
  • Smith CA, Ireland TO, Thornberry TP. Adolescent maltreatment and its impact on young adult antisocial behavior. Child Abuse & Neglect. 2005;29(10):1099–1119. [PubMed]
  • Stein JA, Leslie MB, Nyamathi A. 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. 2002;26(10):1011–1027. [PubMed]
  • Stein JA, Nyamathi AM, Zane JI. Situational, psychosocial, and physical health-related correlates of HIV/AIDS risk behaviors in homeless men. American Journal of Men’s Health. 2009;3(1):25–35. [PMC free article] [PubMed]
  • Stewart AL, Hays RD, Ware JE., Jr The MOS short-form general health survey. Reliability and validity in a patient population. Medical Care. 1988;26(7):724–735. [PubMed]
  • Stewart B. McKinney-Vento Act, 42 U.S.C. § 11302 (1994).
  • Stolzenberg L, D’Alessio SJ. Multisystemic therapy for child abuse and neglect: A randomized effectiveness trial. Journal of Family Psychology. 2000;24:497–507. [PMC free article] [PubMed]
  • Swenson CC, Schaeffer CM, Henggeler SW, Faldowski R, Mayhew AM. Multisystemic therapy for child abuse and neglect: A randomized effectiveness trial. Journal of Family Psychology. 2010;24(4):497–507. [PMC free article] [PubMed]
  • Travis J, Solomon A, Waul M. From prison to home: The dimensions and consequences of prisoner reentry. Washington, DC: Urban Institute: Justice Policy Center; 2001. p. 64.
  • Travis J, Waul M. Prisoners once removed: The impact of incarceration and reentry on children, family and communities. Washington, D.C: The Urban Institute; 2004.
  • Turner MG, Hartman JL, Bishop DM. The effects of prenatal problems, family functioning, and neighborhood disadvantage in predicting life course persistent offending. Criminal Justice and Behavior. 2007;34:1241–1261.
  • Volavka J, Swanson J. Violent behavior in mental illness: The role of substance abuse. The Journal of American Medical Association. 2010;304(5):563–564. http// [PMC free article] [PubMed]
  • Warren JI, Hurt S, Loper AB, Bale R, Friend R, Chauhan P. Psychiatric symptoms, history of victimization, and violent behavior among incarcerated female felons: An American perspective. International Journal of Law & Psychiatry. 2002;25(2):129–149. [PubMed]
  • Woldoff RA, Washington HM. Arrested contact: The criminal justice system, race, and father engagement. The Prison Journal. 2008;88:179–206.
  • Zhang S, Roberts R, Callanan V. An evaluation of the California preventing parolee crime program. San Marcos, CA: California State University San Marcos (CSUSM); 2003. p. 80.