Type of Victimization Exposure,Perpetrator Type,and Timing of Victimization: The Impact on Behavioral Health Outcomes

Abstract

The effect of abuse victimization in correctional samples has been researched previously, particularly with an eye toward these experiences on justice-involved youth and prison samples’ offending and recidivism behavior. The role of this type of victimization, including physical abuse, sexual abuse, and polyvictimization, is less studied in jail populations. The effect of abuse victimization is also less researched among other outcomes, including behavioral health disorders (BHDs) and substance use disorder (SUD). While the effect of abuse, generally, has been examined, less is known about how abuse perpetrator type and timing of abuse impact justice-involved individuals’ outcomes. Using logistic regressions, we examined the influence of abuse perpetrator type (non-stranger or stranger) and timing (before childhood, after childhood, or before and after childhood) in a population of jailed adults from one state (n = 4,713). Outcomes studied included internalizing BHDs, externalizing BHDs, and severe SUD. Results indicated that abuse perpetrated by a non-stranger yielded a greater impact on mental illness compared to abuse perpetrated by a stranger. In contrast to abuse experienced as an adult, childhood abuse was more consistently associated with internalizing and externalizing disorders but was not related to severe SUD, with an exception of physical abuse. Further, BHDs and SUD were strongly associated with each other. Overall, polyvictimization had the strongest effect on the outcomes compared to either physical abuse or sexual abuse alone. Our findings suggest that screening for abuse experiences as a potential destabilizing factor in justice-involved populations could improve case management and interventions for people incarcerated in jails. Results also highlight the importance of distinguishing between the perpetrator type of abuse and timing of abuse.

Keywords

physical abuse child abuse alcohol and drugs mental health and violence

Introduction

Past experiences of victimization¹ are prevalent in justice-involved populations (Givens & Cuddeback, 2021; Karlsson & Zielinski, 2020). Research has explored the effect of victimization on justice-involved youth and has documented the damaging effects of these experiences on youth’s mental and behavioral outcomes (Charak et al., 2019; Duron et al., 2021). Although research addressing the negative outcomes of victimization exists for adults under community supervision (Givens & Cuddeback, 2021) and individuals in the prison system (Cain et al., 2016; Maschi et al., 2019; Meade & Steiner, 2013), little is known about the influence of victimization on outcomes for adults incarcerated in jails (for an exception, see Dalbir et al., 2022). Research is even more limited with respect to the effect of various aspects of victimization, such as the type of abuse experienced, the perpetrator type, or the timing of abuse.

Victimization is strongly interrelated with various other maladies, such as behavioral health disorders (BHDs)² or substance use disorder (SUD) (Dalbir et al., 2022; Radatz & Wright, 2017), and each of these are potentially destabilizing or responsivity factors that may influence a person’s likelihood of success in the community (Taxman & Caudy, 2015). Further, these attributes are common in justice-involved populations (Givens & Cuddeback, 2021; Karlsson & Zielinski, 2020) and may be particularly salient among those in jail settings. However, our understanding of these potentially destabilizing factors is limited in several ways, including most notably, a focus of research on individuals incarcerated in prison, while largely ignoring jail populations. We also know considerably less about the impact of specific aspects of victimization experiences (such as when the victimization was experienced and who perpetrated it), and how these experiences are related to BHD and SUD among those incarcerated in prison or in jail. We attempt to address some of these limitations in the current study.

Abuse Victimization Among Incarcerated Persons

Much of the literature in the criminal justice field focuses on recidivism as a primary outcome (Ostermann et al., 2015). However, the link between victimization or abuse and recidivism is unclear and likely indirect, working its way through mental illness and/or SUD (Salisbury & Van Voorhis, 2009; see Figure 1). This makes other potentially intervening outcomes (such as BHDs and/or SUD, see Dalbir et al., 2022; Salisbury & Van Voorhis, 2009) important to consider since they may determine whether an incarcerated person will recidivate following release. Important to the current study, these intervening problems may be especially salient for people who have been exposed to victimization. Also relevant are experiences of polyvictimization and how they are related to BHDs and SUD. Victimization tends to cluster in terms of type of victimization experienced, where victims often experience a range of abuses instead of just one (Finkelhor et al., 2011), and timing of polyvictimization (e.g., young victims are often victims later in life; Widom, 2008). Consequently, polyvictimization may be detrimental beyond any singular type of victimization (Ford et al., 2013).

Figure 1.

Relationship among abuse, substance use disorder, behavioral health disorders, and criminal behavior.

Pathways Theory

Pathways theory is useful to explain the interconnectedness of an abuse history, mental illness, substance use, and criminal behavior. According to this theoretical perspective, girls/women and boys/men enter into the justice systems via different pathways, necessitating gender-specific explanations for criminal justice contact (Chesney-Lind & Pasko, 2013). In one of the earliest studies establishing the pathways theory, Daly (1992) described five pathways: (a) “harmed and harming” women who experienced abuse or neglect as children, and who may have mental health and substance use issues; (b) “street” women who experienced abuse as children, who subsequently ran away from home, turned to petty crimes to survive, became addicted to drugs, and committed crime to support their drug habit; (c) “battered” women who became justice-involved after experiencing an abusive relationship; (d) “drug-addicted” women who became involved in the system as a result of selling or using drugs with a significant other or family member; and (e) “other” women who appeared to engage in criminal behavior as a result of greed (no history of abuse, mental illness, or substance abuse). The current study focuses on the associations between abuse victimization, mental illness, and SUD, primarily testing the “harmed and harming” pathway among a justice-involved sample of men and women in jail.

As indicated, pathways theory suggests that victimization gives rise to maladaptive coping strategies, such as SUD and risky behaviors (Chesney-Lind & Pasko, 2013), or contributes to mental health problems (MHPs)—all of which can eventually contribute to increased criminal behavior (Dalbir et al., 2022; Salisbury & Van Voorhis, 2009). Accordingly, this could explain why abuse, BHDs, and SUD are so prevalent in correctional populations (Givens & Cuddeback, 2021; Karlsson & Zielinski, 2020). For instance, Salisbury and Van Voorhis (2009) found childhood victimization was connected to mental illness and substance abuse. Further, Lynch et al. (2012) discovered that women in jail who had a serious mental illness indicated higher rates of victimization experiences and criminal behavior compared to women without a serious mental illness. They also found that women with a serious mental illness were at a higher risk for substance use.

The above-cited research studies were conducted on samples of females, but scholars recognize that boys’/men’s experiences of victimization may also help to explain male offending and have, therefore, begun to explore how these pathways are relevant to men’s criminal behavior (Belknap, 2016). For instance, among incarcerated men in prison, Wolff and Shi (2012) found that childhood physical abuse was associated with a greater likelihood of substance abuse problems, while childhood sexual abuse was associated with higher levels of depression and anxiety symptoms as well as increased substance abuse problems. Physical abuse experienced as an adult was related to greater depression symptoms, anxiety symptoms, and substance abuse, while sexual abuse experienced as an adult was also related to increased substance abuse problems. Regarding externalizing symptoms, physical abuse experienced as a child or adult was associated with higher levels of interpersonal, self-regulation, and aggression problems. Sexual victimization as a child or adult was associated with more interpersonal and self-regulation problems. This study therefore demonstrates that type of abuse (sexual, physical, or polyvictimization), as well as the timing of abuse (during childhood and/or adulthood) are related to SUDs and BHDs among incarcerated men.

A few studies have examined pathways theory among samples of both males and females and have demonstrated some gender-specific pathways. Jones et al. (2014) tested pathways theory in a sample of female and male youth under community supervision and found a gendered pathway for girls only, which included abuse, mental illness, and substance abuse. In a study conducted with adult male and female pretrial defendants, Gehring (2018) found relationships between pretrial failure and key pathways theory predictors—childhood abuse, history of mental illness, and substance use—for women only. However, her research also showed that physical or sexual abuse as a child was correlated with mental health among women and men. A history of childhood physical or sexual abuse, as well as a history of mental illness, were also associated with substance use for women and men. Thus, pathways theory appears to be relevant for men and women who are justice involved; however, questions remain, which we describe below.

Abuse: Variations in What, By Whom, and When

Although research on victimization experiences among justice-involved individuals has gained much attention over the past few decades, there remain limitations in our understanding of its impact on various maladaptive outcomes. While early research in the area typically examined whether victimization had simply “ever” occurred in one’s lifetime, more recent research has demonstrated the importance of the type of abuse experienced, the relationship and characteristics of the perpetrator, and when the abuse occurred (Cain et al., 2016; Dunn et al., 2017; Meade & Steiner, 2013; Salisbury & Van Voorhis, 2009; Tripodi & Pettus-Davis, 2013). Regarding the perpetrator of abuse, it appears that victimization by a non-stranger likely yields greater effects on offenders’ outcomes (Cain et al., 2016; Meade & Steiner, 2013; Salisbury & Van Voorhis, 2009), but there is limited evidence exploring this issue. The extant research has detailed the impact of non-stranger victimization on women’s mental status and behavior (Tripodi & Pettus-Davis, 2013), but less is known about this effect on male offenders’ outcomes. An exception to this is Meade and Steiner’s (2013) work, which assessed the impact of physical and sexual victimization by a known assailant or a stranger. They examined the association between exposure to violence prior to incarceration and prison adjustment in a national sample of individuals in prison. Physical and sexual assault by a non-stranger were strong predictors of maladjustment, including poor behavioral and mental health outcomes. Further, Cain and Colleagues (2016) explored the impact of non-stranger victimization on prison maladjustment among men and women, demonstrating that victimization by a non-stranger was more common among females. Yet, experiences of non-stranger victimization impacted both male and females’ misconduct and mental health issues, contributing to heightened mental health issues for both sexes. Taken together, these studies demonstrate that non-stranger victimization appears to be a risk factor for incarcerated women and men.

The timing of when abuse occurred in one’s lifetime is also a key component of understanding the impact of victimization on maladaptive coping and later problem behaviors. Type and timing of abuse are interrelated (Claessens et al., 2011) and can complicate research. As an example, childhood maltreatment impacts the severity of later internalizing and externalizing BHDs (Dunn et al., 2017; Manly et al., 2001)—yet, is it the type of abuse experienced, or the timing of abuse that is primarily linked to BHDs? Research has yet to determine these answers, especially as they pertain to correctional populations, primarily because most research does not examine when victimization occurred. Studies that have explored this issue have reported that the effect of adversity during childhood may be more detrimental compared to abuse experienced solely during adulthood. Meade and Steiner (2013) investigated the effect of victimization on incarcerated adults’ maladjustment and found that childhood abuse and physical assault in adulthood predicted drug use and mental health concerns. Their findings also suggest that the effects of victimization timing vary in magnitude, where childhood abuse had a greater influence on drug use and MHPs than sexual victimization as an adult. Similarly, Cain and Colleagues (2016) reported that childhood abuse and assault experienced as an adult resulted in greater manic and depressive symptoms for men and women, but child abuse led to a greater influence on the number of depressive symptoms experienced by men. Additionally, experiences of childhood abuse and victimization in adulthood may contribute to a greater likelihood of detrimental outcomes as prolonged exposure to stress can result in destructive psychological and behavioral changes (Cicchetti & Toth, 2005), including greater levels of externalizing behavior (Howes et al., 2000). Based on the extant literature, one could reasonably expect that persons in jail who enter the criminal justice system with a history of abuse may also be more likely to experience BHDs and SUD.

Research on Jail Populations

Research on the above topics has primarily been conducted on prison populations, with studies from jail settings limited in both number and scope. While scholars have studied the associations between victimization, BHDs, and SUD among those incarcerated in prisons (Cain et al., 2016; Maschi et al., 2019; Meade & Steiner, 2013), this research is nonetheless limited, especially when understanding the influence of various forms of victimization, the impact of the perpetrator of abuse, or when the abuse was experienced (for exceptions see Cain et al., 2016; Meade & Steiner, 2013; Salisbury & Van Voorhis, 2009; Tripodi & Pettus-Davis, 2013). Further, inquiries on these topics are virtually nonexistent regarding jail populations.

These shortcomings in the research base are surprising given that individuals in jail indicate a higher prevalence of BHDs and SUD than those incarcerated in prisons (Bronson & Berzofsky, 2017) and the possibility that BHDs and SUD may be even more problematic among those in jail than in prison. As Dalbir and colleagues (2022) note, the prevalence of BHDs and SUD may be even higher among persons incarcerated in jail than prison because jails may be more likely to house individuals whose mental illness and substance use problems have brought them to the attention of the police, but whose offenses may not be serious enough to result in a prison sentence (e.g., nuisance offenses). Therefore, not only do those in jail represent a larger sample of individuals who better represent the risk and needs of all justice-involved individuals, but they also represent a subsample with, arguably, more needs and destabilizing factors than those in prison. Accordingly, further research is needed to examine the effect of destabilizing factors, including trauma and BHDs, among this population. Since these destabilizing factors can be related to continued deviant behavior, individuals may benefit from programming for successful reentry into their respective communities. Moreover, additional research in this area has potential implications for case planning and/or potential divergence of these individuals away from the criminal justice system and into the health or mental health systems.

Current Study

As far as we are aware, no studies to date have explored differences in abuse perpetrator type or abuse timing on jailed individuals’ mental health and behavioral outcomes. This exploration is timely, as experiences of abuse, BHDs, and SUD are common among jail populations (Givens & Cuddeback, 2021; Karlsson & Zielinski, 2020). Additionally, those incarcerated in jail return more quickly to the community compared to prison populations. Therefore, research that can assist with reentry of persons from jail may result in improving the mental and behavioral health outcomes of these individuals, ultimately benefiting the communities to which they return. We seek to contribute to the literature by examining a population of individuals incarcerated in jail to: (a) assess the effects of different types of abuse exposure, (b) test the impact of different types of perpetrators, (c) evaluate differential effects of the timing of victimization, and (d) examine whether victimization is associated with internalizing BHDs, externalizing BHDs, and/or SUD.

Methods

Data and Population

Data were obtained from a large Midwestern jail wherein administrative admissions data were supplemented by assessments conducted at intake. Individuals admitted to the jail between February 21, 2017, and September 12, 2017, were screened by intake staff who, for the most part, had backgrounds in social work. Assessments were administered at intake or within a few days of admission. Assessors were provided an interview guide and trained on a 79-question inventory developed by jail administrators and a research team of faculty research partners. Over the 203 days of data collection, 4,713 individuals were admitted to the jail.

Independent Variables

Primary predictor variables are indicators of physical abuse and/or sexual abuse, including timing of the victimization and characteristics of the perpetrator. At intake, people were asked whether they were abused prior to arrest. If an individual indicated that they were physically or sexually abused, they were also asked whether they knew the person who abused them. Options included: spouse/partner, other family member, friend/acquaintance, or no—I did not know the person who abused me. To examine the effect of abuse perpetrator type on SUD and BHD outcomes, we created binary measures for physical abuse by a non-stranger (1 = yes), sexual abuse by a non-stranger (1 = yes), polyvictimization by a non-stranger (1 = experienced both physical and sexual abuse by a non-stranger), physical abuse by a stranger (1 = yes), sexual abuse by a non-stranger (1 = yes), and polyvictimization by a stranger (1 = experienced both physical and sexual abuse by a stranger). No history of abuse served as the reference category.

If an individual reported a history of abuse, they were asked when it occurred—before they were 18 years old, after they were 18 years old, or both before and after they were 18 years old. To examine the effect of the timing of abuse on SUD and BHD outcomes, we created binary measures for physical abuse before 18 years old (1 = yes), sexual abuse before 18 years old (1 = yes), polyvictimization before 18 years old (1 = experienced both physical and sexual abuse before 18 years old), physical abuse after 18 years old (1 = yes), sexual abuse after 18 years old (1 = yes), polyvictimization after 18 years old (1 = experienced both physical and sexual abuse after 18 years old), physical abuse before and after 18 years old (1 = yes), sexual abuse before and after 18 years old (1 = yes), and finally, polyvictimization before and after 18 years old (1 = experienced both physical and sexual abuse before and after 18 years old). No history of abuse served as the reference category.

Dependent Variables

We explored how the abuse measures influenced BHD and SUD outcomes. For all mental health models, SUD was a control variable, while the mental health variables were controls in SUD models. These outcomes were derived from the full questionnaire administered at intake, which included the Global Appraisal of Individual Needs-Short Screen (GAINS). The GAINS is a clinical, biopsychosocial assessment measuring behavioral and substance abuse issues (Dennis et al., 2006). The instrument can be used as a quick method to identify individuals who may be flagged as having one or more BHDs, suggesting a need to be referred to the behavioral health treatment system, and it also provides count measures for disorders across four domains: crime and violence, internalizing behaviors, externalizing behaviors, and substance abuse. Current, recent, and past symptoms are also captured in the instrument. Both the internalizing and externalizing disorder domains include five questions, and each of them can be given a score of 0 (“never”) to 3 (“in the past month”). An example of a question from the internalizing domain is: “When was the last time that you had significant problems with feeling very trapped, lonely, sad, blue, depressed, or hopeless about the future?” An example question from the externalizing domain is: “When was the last time that you did the following things two or more times? Started physical fights with other people?” We recoded the internalizing and externalizing symptoms into two dichotomous variables: high internalizing BHDs and high externalizing BHDs (coded 0 = 0–2 symptoms, 1 = 3+ symptoms).³

We measured SUD with an 11-item scale: the TCU drug screen, which is an evidence-based assessment used in correctional populations (Knight et al., 2002, 2018). An example question from this instrument is, “Did you have a strong desire or urge to use drugs?” A mild disorder constitutes 2 to 3 points (symptoms), moderate 4 to 5 points (symptoms), and severe 6 or more points (symptoms). The screen ranges from 0 to 11 points. We dichotomized this measure to no, mild, or moderate disorder (0) or presence of severe SUD (1). We elected to separate out severe disorder, versus mild or moderate, to better identify people with the highest substance use need who, per the Risk-Needs-Responsivity model, should be prioritized for programming and other resources (Bonta & Andrews, 2017).

Covariates

We included an ordinal age variable to indicate age at time of booking and this is measured as (0) greater than 35 years of age, (1) 26 to 35 years old, and (2) less than 26 years old (reference category).⁴ Age is associated with psychological and other health problems that may constitute risk factors for substance misuse (Gossop & Moos, 2008). However, research on peak onset of mental illness, BHDs, or SUD is mixed, with no definitive answer on when these issues may arise (Solmi et al., 2022). We include age as a covariate to control for potential variations in age of onset for BHD or SUD. A person’s sex refers to a binary, biological measure with male coded as 1. As articulated, research has shown that victimization by a non-stranger is more common among women (Cain et al., 2016), and that males tend to be more likely to engage in externalizing behaviors whereas females are more likely to express internalizing symptoms following abuse (Wright & Schwartz, 2021). Additionally, there may be important gender differences in substance use—where men demonstrate higher rates of drug use and dependence (Compton et al., 2007). We also included a binary race measure (0 = White, 1 = racial/ethnic minority). Individuals who identified as Black, Hispanic, Asian/Pacific Islander, and “other race” were included in the racial/ethnic minority group. Research shows that rates of mental illness and SUD tend to be lower for people of color, as compared to White individuals, although this may be a result of underdiagnosis in the former population (Panchal et al., 2022).

Additionally, we included various criminogenic needs factors. These covariates include: education (high school or more = 1), unemployed (yes = 1), homeless (yes = 1), and single (yes = 1). Lower education has been associated with poorer mental health (Niemeyer et al., 2019) but higher substance use issues (Substance Abuse and Mental Health Services Administration, 2017). Further, unemployed individuals may exhibit higher levels of mental distress compared to those who are employed (Paul & Moser, 2009). Employment may also help to reduce use of illicit substances, which can help to decrease likelihood of criminal behavior (Joe et al., 1990). Moreover, rates of mental illness and SUD are high among homeless individuals, as compared to the general population (Krausz et al., 2013). In terms of relationship status, Grundström et al. (2021) found that being single, divorced, or widowed was consistently associated with poorer mental health. Additionally, marriage, depending on the quality of the relationship, can be a protective factor against substance abuse (Sinha, 2018).

We identified traumatic brain injuries (TBIs) via the Ohio State University Traumatic Brain Injury-Identification Method (Corrigan & Bogner, 2007), which produces a scale of responses: no history, mild without loss of consciousness, mild with loss of consciousness, moderate, or severe TBI. Research has shown an association between TBIs and later development of mental illness or SUD (Dams-O’Connor et al., 2013). We created a binary TBI variable to indicate whether an individual experienced any TBI (e.g., moderate or severe) before their current period of incarceration (yes = 1). We included post-traumatic stress disorder (PTSD) as a binary measure (yes = 1), which was derived from the Abbreviated PTSD Checklist—Civilian version. Examples of items from this screen include feeling irritable or angry outbursts, feeling distant or cut off from other people, and difficulty concentrating. Responses range from one (not at all) to five (extremely). We included this measure for the SUD models because research has shown that PTSD and SUD are related, where substances may be used to lessen the effect of PTSD symptoms (Jacobsen et al., 2001).

Finally, acute MHPs and acute physical health problems (PHPs) were treated as binary measures to indicate whether the person reported having a mental or physical health symptom in the 30 days prior to intake. We include the former because people may turn to illicit substances to cope with mental health symptoms (National Institute of Mental Health, 2023). For the latter measure, PHPs may contribute to mental illness or exacerbate existing mental health symptoms (Evans et al., 2005). Additionally, many physical health conditions are also associated with SUD (Hser et al., 2017).

Procedures

We first conducted a simple descriptive analysis between women and men to check the rate of abuse between the two groups. This check was intended to identify whether both groups could, equally, fit the “harmed and harming” path proposed by Daly (1992). Regression diagnostic checks were conducted prior to running the logistic regressions to test for the possibility of multicollinearity among the predictor variables. Variance inflation factors were also produced and were less than four, indicating that multicollinearity was not a problem. Finding no violations, we regressed our abuse measures on internalizing BHDs, externalizing BHDs, and SUD. Although past work has demonstrated gender-specific and gender-neutral effects of non-stranger victimization on offenders’ well-being and criminal behavior (Chesney-Lind & Pasko, 2013; Salisbury & Van Voorhis, 2009), we do not address gender-specific differences due to a low prevalence of victimization broken out by abuse timing and perpetrator type for men and women. Further, no information was available regarding non-binary gender/sex identities. Therefore, we examined a pooled population of jailed males and females.⁵

For each outcome, we conducted two binary logistic models to examine the effects of perpetrator type and abuse timing. This resulted in a total of six models. While all covariates were included in the SUD models, PTSD and acute MHPs were not included in the internalizing and externalizing BHD models as a result of conceptual overlap between the controls and the outcomes. However, SUD was included in these models as the association between abuse, BHDs, and SUD is unclear; meaning BHDs may contribute to SUD or vice versa.

Results

Descriptive Statistics

As demonstrated in Table 1, a minority of the sample (27.9%) reported a history of abuse. People who experienced physical abuse, sexual abuse, or polyvictimization were more likely to be abused by a non-stranger (90.3%, 84.9%, and 89.6%, respectively). Respondents who experienced abuse were most likely to be abused as a child for physical abuse (47.1%), sexual abuse (69.8%), and polyvictimization (50.1%) relative to adulthood abuse (33.5%, 17.9%, and 31.4%, respectively). Approximately 44% of respondents indicated a high internalizing BHD while about 19% reported a high externalizing BHD. Around 18% of the population were identified as having a severe SUD. A slight majority of respondents (35.9%) were 35 years or older while most were male (73.0%). About 44% of individuals self-identified as a racial/ethnic minority, 81% reported having achieved at least a high school education, 42% were unemployed, 11% were homeless, and 70% were single. Approximately 38% of individuals indicated a TBI, 27% PTSD, 34% acute MHPs, and 22% acute PHPs. Women (28.3%) and men (27.5%) reported similar levels of abuse prior to arrest, with no significant difference between the groups (χ² = 0.28, df = 1). In other words, approximately 28% of our population has the potential to meet the “harmed and harming” pathway (Daly, 1992) to justice-system involvement (without consideration of any covariates).

Table 1.

Descriptive Statistics—Total Sample.

Item	Total Sample
Item	%	M (SD)
History of physical and/or sexual abuse (all respondents)	27.9	0.28 (0.45)
Abuse perpetrator (for respondents reporting abuse)
Physical abuse perpetrator type		1.90 (0.30)
Stranger	9.7
Non-stranger	90.3
Sexual abuse perpetrator type		1.85 (0.36)
Stranger	15.1
Non-stranger	84.9
Polyvictimization abuse perpetrator type		0.90 (0.31)
Stranger	10.4
Non-stranger	89.6
When abused (for respondents reporting abuse)
Physical abuse timing		1.72 (0.77)
Before 18 years old	47.1
After 18 years old	33.5
Before and after 18 years old	19.4
Sexual abuse timing		1.42 (0.70)
Before 18 years old	69.8
After 18 years old	17.9
Before and after 18 years old	12.3
Polyvictimization abuse timing		0.68 (0.77)
Before 18 years old	50.1
After 18 years old	31.4
Before and after 18 years old	18.4
High internalizing BHDs	43.6	0.44 (0.50)
High externalizing BHDs	19.4	0.19 (0.40)
Severe SUD	18.1	0.18 (0.39)
Controls
Age		0.94 (0.81)
>35	35.9
26–35	33.8
<26	30.3
Male	73.0	0.73 (0.44)
Racial/ethnic minority	43.5	0.44 (0.50)
High school or more	80.9	0.81 (0.39)
Unemployed	41.9	0.42 (0.49)
Homeless	10.7	0.11 (0.31)
Single	69.9	0.70 (0.46)
TBI	37.9	0.38 (0.49)
PTSD	27.1	0.27 (0.44)
Acute MHPs	34.0	0.34 (0.47)
Acute PHPs	22.2	0.22 (0.42)
N	4,713

Note. BHDs = behavioral health disorders; MHPs = mental health problems; PHPs = physical health problems; SUD = substance use disorder.

Multivariate Results

Table 2 shows the effects of perpetrator type. Model 1 indicates that, compared to no abuse history, physical (OR = 2.02, p < .001) and sexual (OR = 2.26, p < .001) abuse by a non-stranger was associated with two times higher odds of an internalizing BHD. Polyvictimization by a non-stranger was associated with over four-and-a-half times the odds of an internalizing BHD (OR = 4.71, p < .001) compared to people with no abuse history. Severe SUD was associated with over three times the heightened risk of an internalizing disorder (OR = 3.46, p < .001). Model 2 reveals that, compared to people with no history of abuse, those physically abused by a non-stranger (OR = 1.61, p < .001) or stranger (OR = 1.86, p < .05), or who were polyvictimized by a non-stranger (OR = 1.76, p < .001) demonstrated greater odds of externalizing symptoms. Severe SUD was related to over three times increased odds of an externalizing BHD (OR = 3.49, p < .001). Model 3 demonstrates that polyvictimization by a stranger, compared to no abuse history, was related to over three times heightened odds of severe SUD (OR = 3.43, p < .05). Both internalizing and externalizing BHDs were associated with over two times greater odds of severe SUD (OR = 2.19, p < .001 and OR = 2.38, p < .001, respectively).

Table 2.

Effect of Abuse Perpetrator Type on Outcomes.

Item	Internalizing Disorder		Externalizing Disorder		Severe Substance Use Disorder
	Model 1		Model 2		Model 3
	b (SE)	OR	b (SE)	OR	b (SE)	OR
Physical abuse—Non-stranger	0.731 (0.099)	2.02***	0.477 (0.110)	1.61***	0.201 (0.117)	1.22
Sexual abuse—Non-stranger	0.769 (0.234)	2.26***	0.091 (0.283)	1.10	−0.063 (0.291)	0.94
Physical abuse—Stranger	0.335 (0.289)	1.45	0.620 (0.302)	1.86*	0.522 (0.323)	1.69
Sexual abuse—Stranger	0.921 (0.562)	2.79	0.541 (0.624)	1.72	0.234 (0.629)	1.26
Polyvictimization—Non-stranger	1.550 (0.146)	4.71***	0.562 (0.136)	1.76***	0.102 (0.143)	1.11
Polyvictimization—Stranger	0.739 (0.658)	2.09	−0.064 (0.640)	0.94	1.232 (0.609)	3.43*
SUD	1.241 (0.094)	3.46***	1.25 (0.092)	3.49***
Internalizing disorder					0.784 (0.107)	2.19***
Externalizing disorder					0.869 (0.098)	2.38***
Controls
Age (<26 as ref.)
>35	−0.294 (0.087)	0.75**	−0.443 (0.105)	0.64***	0.432 (0.117)	1.54***
26–35	0.052 (0.085)	1.05	−0.210 (0.102)	0.81*	0.449 (0.115)	1.57***
Male	−0.051 (0.076)	0.95	0.098 (0.093)	1.10	−0.020 (0.098)	0.98
Racial/ethnic minority	−0.185 (0.070)	0.83**	−0.232 (0.086)	0.79**	−0.842 (0.096)	0.43***
High school or more	−0.154 (0.086)	0.86	−0.290 (0.100)	0.75**	0.043 (0.112)	1.04
Unemployed	0.012 (0.069)	1.01	0.025 (0.082)	1.03	0.358 (0.087)	1.43***
Homeless	0.617 (0.115)	1.85***	0.415 (0.118)	1.52***	0.461 (0.122)	1.59***
Single	0.121 (0.074)	1.13	0.071 (0.090)	1.07	0.140 (0.097)	1.15
TBI	0.532 (0.071)	1.70***	0.269 (0.085)	1.31**	0.251 (0.091)	1.29**
PTSD					0.610 (0.105)	1.84***
Acute MHPs					0.046 (0.091)	1.05
Acute PHPs	0.871 (0.085)	2.39***	0.447 (0.094)	1.56***	0.106 (0.101)	1.11
Constant	−0.939 (0.130)	0.39***	−1.739 (0.154)	0.18***	−2.982 (0.184)	0.05***
N	4,395		4,395		4,393

Note. MHPs = mental health problems; PHPs = physical health problems; SUD = substance use disorder.

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

Our next set of analyses tested the effect of abuse type by timing (see Table 3). Model 4 shows that all of the abuse measures, compared to no abuse history, were associated with greater odds of internalizing disorders (ORs = 1.41–7.13, p < .05 to p < .001). Severe SUD was related to over three times greater odds of an internalizing BHD (OR = 3.42, p < .001). Model 5 demonstrates that, compared to no abuse history, physical abuse as a child (OR = 1.83, p < .001), physical abuse before and after the age of 18 (OR = 2.17, p < .001), and sexual abuse before and after the age of 18 (OR = 3.60, p < .05) were significant predictors of heightened odds of an externalizing disorder. Polyvictimization as a child, compared to no abuse history, was related to two times greater odds of an externalizing BHD (OR = 2.07, p < .001). Severe SUD was associated with over three times greater odds of externalizing symptoms (OR = 3.53, p < .001). Model 6 shows that, compared to no abuse history, people who were physically abused as a child demonstrated greater odds of severe SUD (OR = 1.36, p < .05). Internalizing and externalizing BHDs were associated with two times greater odds of severe SUD (OR = 2.15, p < .001 and OR = 2.41, p < .001, respectively).

Table 3.

Effect of Abuse Timing on Outcomes.

Item	Internalizing Disorder		Externalizing Disorder		Severe Substance Use Disorder
	Model 4		Model 5		Model 6
	b (SE)	OR	b (SE)	OR	b (SE)	OR
Physical abuse—Before 18 years old	0.631 (0.133)	1.88***	0.603 (0.139)	1.83***	0.309 (0.149)	1.36*
Sexual abuse—Before 18 years old	0.588 (0.257)	1.80*	−0.075 (0.324)	0.93	0.153 (0.319)	1.17
Physical abuse—After 18 years old	0.342 (0.150)	1.41*	−0.069 (0.190)	0.93	0.239 (0.185)	1.27
Sexual abuse—After 18 years old	1.323 (0.527)	3.76*	−0.326 (0.687)	0.72	−0.078 (0.613)	0.93
Physical abuse—Before and after 18 years old	1.203 (0.211)	3.33***	0.773 (0.198)	2.17***	0.088 (0.214)	1.09
Sexual abuse—Before and after 18 years old	1.277 (0.616)	3.59*	1.282 (0.570)	3.60*	−0.459 (0.689)	0.63
Polyvictimization—Before 18 years old	0.968 (0.205)	2.63***	0.726 (0.202)	2.07***	0.318 (0.217)	1.38
Polyvictimization—After 18 years old	1.681 (0.386)	5.37***	−0.182 (0.344)	0.83	0.584 (0.309)	1.79
Polyvictimization—Before and after 18 years old	1.964 (0.340)	7.13***	0.054 (0.258)	1.06	0.419 (0.241)	1.52
SUD	1.228 (0.076)	3.42***	1.262 (0.093)	3.53***
Internalizing disorder					0.767 (0.107)	2.15***
Externalizing disorder					0.880 (0.099)	2.41***
Controls
Age (<26 as ref.)
>35	−0.296 (0.087)	0.74**	−0.428 (0.105)	0.65***	0.422 (0.117)	1.53***
26–35	0.064 (0.084)	1.07	−0.183 (0.102)	0.83	0.442 (0.115)	1.56***
Male	−0.038 (0.076)	0.96	0.105 (0.093)	1.11	−0.023 (0.098)	0.98
Racial/ethnic minority	−0.201 (0.069)	0.82**	−0.230 (0.086)	0.80**	−0.835 (0.096)	0.43***
High school or more	−0.149 (0.086)	0.86	−0.273 (0.100)	0.76**	0.040 (0.112)	1.04
Unemployed	0.009 (0.068)	1.01	0.041 (0.082)	1.04	0.351 (0.087)	1.42***
Homeless	0.576 (0.116)	1.78***	0.422 (0.119)	1.52***	0.453 (0.122)	1.57***
Single	0.111 (0.074)	1.12	0.079 (0.091)	1.08	0.139 (0.097)	1.15
TBI	0.527 (0.071)	1.70***	0.272 (0.085)	1.31**	0.241 (0.091)	1.27**
PTSD					0.602 (0.106)	1.83***
Acute MHPS					0.045 (0.091)	1.05
Acute PHPs	0.907 (0.085)	2.48***	0.472 (0.094)	1.60***	0.089 (0.101)	1.09
Constant	−0.900 (0.130)	0.41***	−1.763 (0.155)	0.17***	−2.972 (0.185)	0.05***
N	4,395		4,395		4,393

Note. MHPs = mental health problems; PHPs = physical health problems; SUD = substance use disorder.

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

Discussion

The effects of victimization among prison populations have been well-established (Cain et al., 2016; Maschi et al., 2019; Meade & Steiner, 2013), but less is known about jail populations, and research examining the influence of abuse timing and perpetrator type in jail populations has been even more limited. However, studying jailed individuals is vital because jailed persons return to the community more quickly and also indicate a higher incidence of victimization, BHDs, and SUD (Givens & Cuddeback, 2021; Karlsson & Zielinski, 2020). Prevalence of these factors in jail populations is notable, as facilities generally lack the resources to address trauma experiences or BHDs (Bonta & Andrews, 2017). With more evidence of the relationships between these factors available, policymakers may make resource provision a priority. In brief, our results showed that abuse is damaging to the mental health of people who are incarcerated in jail. It also appears that abuse inflicted by a non-stranger may be more damaging than victimization from a stranger. Somewhat unexpectedly, we found that abuse was mostly unrelated to severe SUD, but that internalizing and externalizing BHDs (which are strongly related to victimization) were associated with severe SUD and vice versa. While our findings demonstrate a complicated relationship between abuse histories, BHDs, and SUD, they also indicate that abuse perpetrator type and timing could be considered key response factors, particularly when addressing mental illness. We discuss these findings below.

In line with prior research, our findings demonstrate the interrelatedness of abuse, BHDs, and SUD. While we found strong evidence for the pathway between abuse and internalizing and externalizing BHDs, as well as between BHDs and SUD, our findings did not reveal a significant pathway between abuse and SUD, with the exception of polyvictimization by a stranger (compared to no abuse history). This finding is contrary to research conducted with individuals incarcerated in prison, which has shown that non-stranger victimization tends to have a larger impact on justice-involved individuals’ outcomes, including SUD, when compared to victimization perpetrated by a stranger (Cain et al., 2016; Meade & Steiner, 2013; Tripodi & Pettus-Davis, 2013). The other exception was physical abuse as a child. This finding comports with past research done with prison samples, which have shown that childhood abuse is related to substance use (Meade & Steiner, 2013). The, generally, nonsignificant relationship between abuse and severe SUD in our population is curious, as this information helps to disentangle the relationship between abuse, BHDs, SUD, and justice-system involvement. Rather than being an intervening variable between abuse and criminal behavior, which appears to be the case for BHDs in our study, SUD may not act as an indirect pathway between abuse and offending but may still impact criminal behavior by way of mental illness (or vice versa).

Regarding the impact of perpetrator type on outcomes, the pattern of results indicates that physical abuse and polyvictimization by a non-stranger, compared to no abuse history, were most consistently related to greater BHDs (both internalizing and externalizing) but were not associated with SUD. Sexual abuse by a non-stranger was also related to internalizing BHD, but not externalizing BHD or SUD. Interestingly, compared to no abuse history, stranger victimization was not significantly associated with internalizing BHDs, only physical abuse by a stranger for externalizing BHDs, and only polyvictimization by a stranger for severe SUD. Thus, the pattern of results suggests that abuse by a non-stranger—regardless of the type of abuse (physical or sexual) experienced—was predictive of BHDs among persons incarcerated in jail. Our findings lend credence to the argument that experiencing abuse may negatively impact mental health outcomes beyond internalizing BHDs (which is often the argument for women with trauma histories), to include externalizing BHDs as well. Overall, abuse perpetrated by a non-stranger, compared to no abuse history, contributed to greater associations with outcomes of interest, which is similar to findings from prison samples (Cain et al., 2016; Meade & Steiner, 2013). Additionally, like Meade and Steiner’s (2013) study, our findings showed that neither perpetrator type, compared to no abuse history, was significantly related to SUD, with the exception of polyvictimization perpetrated by a stranger.

The abuse timing models show that, compared to no abuse history, abuse during childhood appears to be more consistently related to internalizing and externalizing BHDs than abuse experienced only during adulthood. Abuse before and after the age of 18, compared to no abuse history, was also consistently related to the BHD outcomes. Evidence that childhood victimization was related to greater odds of mental illness is in line with other research that has examined the mental health outcomes of youth samples (Duron et al., 2021; Ford et al., 2010), as well as Cain et al.’s (2016) and Meade and Steiner’s (2013) studies on prison samples that revealed childhood victimization was associated with a greater and stronger likelihood of mental health symptoms. Unlike Meade and Steiner’s (2013) finding that childhood victimization resulted in a greater likelihood of drug/alcohol misconduct, however, childhood victimization in our population was unrelated to increased odds of severe SUD, with the exception of physical abuse.

Finally, we should note that polyvictimization, compared to no abuse history, appeared to have a particularly strong relationship with internalizing BHDs, regardless of abuse timing, but again, polyvictimization was not associated with SUD. Similarly, sexual abuse was consistently related to internalizing BHDs—regardless of timing—but it was not significantly associated with externalizing BHDs or SUD (with the exception of abuse experienced before and after the age of 18 on externalizing symptoms). Past research has shown that childhood victimization and abuse as an adult are associated with greater externalizing BHDs (Howes et al., 2000). However, this effect did not hold for our sexual abuse model (with the exception of before and after 18 years old), indicating that type of abuse may be critical when examining the relationship between abuse histories and mental health outcomes.

Policy and Practice Implications

Overall, findings for the relationship between abuse and mental health issues align with pathways theory, in that abuse was significantly related to BHDs. Yet, results regarding the association between abuse and SUD are somewhat surprising and do not fully align with expectations. We did find, however, consistent evidence that BHDs and SUDs are significantly interrelated across all models and regardless of what characteristics of abuse (perpetrator type or timing) were considered. Findings from the current study have implications for service delivery and treatment needs of the jail population. Although some research has addressed trauma interventions for incarcerated women, a gap exists for incarcerated men, with the effect of traumatic experiences on recidivism likelihood having been largely ignored (Pettus-Davis et al., 2019). We did not examine recidivism as an outcome, and we used a pooled sample of men and women; however, if true that mental health acts as an intervening variable between abuse exposure and reoffending, then interventions implemented to target both destabilizing factors (abuse and BHDs) may help to reduce recidivism likelihood for both groups.

Additionally, screening for and/or assessment of victimization as part of criminal justice procedures could result in a more cost-effective implementation of trauma-informed care services to people most in need of evidence-based rehabilitative interventions (Ford et al., 2016). As indicated by our findings, perpetrator type and abuse timing could be considered to optimize any intervention, service, or referral to additional community services. Our findings suggested that earlier exposure to abuse contributed to a larger effect on BHDs than did more recent exposure. Therefore, and as previously suggested by Farrell and Zimmerman (2018), screening for early experiences of victimization, instead of solely recent exposure, is critical when providing an intervention. Such work may aid in diverting individuals with abuse histories and BHDs from the criminal justice system so their victimization histories can better be addressed in an environment that supports healing rather than punishment (Wolf & Prabhu, 2021).

Our results and recommendations suggest one predominate theme: trauma histories in the form of physical and/or sexual abuse, as well as BHDs, could be reconsidered as responsivity or destabilizing factors to be targeted within or outside of the justice system. Because trauma exposure can be a responsivity factor (Bates-Maves & O’Sullivan, 2017), individuals with a trauma history may be unsuccessful in any intervention intended to facilitate successful reentry if trauma reactions (and any mental health issues resulting from exposure) are not first addressed. Alternatively, efforts that are preventative in nature may be one solution to halt individuals with abuse histories from entering the criminal justice in the first place. Generally, however, our results indicate that incarcerated persons in jail with victimization histories warrant further administrative, clinical, and research attention, particularly because other research has suggested that these factors are related to recidivism (Bonta & Andrews, 2017; Dalbir et al., 2022; Salisbury & Van Voorhis, 2009).

Limitations and Future Directions

Findings from the current study should be considered within the context of some limitations. We include jailed individuals from one jail serving one large county, limiting external validity considerations. Future research should examine the effect of abuse perpetrator type and timing in other locations, such as a national sample. We were also unable to precisely pinpoint how much abuse occurred, leaving us unable to account for severity of abuse. As such, continued research should examine frequency of abuse exposure. We also could not account for any current or prior interventions, services, or programming participants received. It is possible that participants received such services, reducing their current level of need—thus potentially underestimating the effects estimated here. Further, we could not determine temporal order between BHDs and SUD as these measures in our study indicate whether a person ever experienced the disorder. For example, while we could not establish whether mental illness preceded SUD, this point buttresses other research showing that these experiences are interrelated. Moreover, our abuse measures do not account for structural adversity (e.g., racism and sexism under an intersectional approach) and other types of trauma exposure (e.g., witnessing violence) that may impact behavioral outcomes (Mersky et al., 2017). Additional forms of trauma can cumulate or interact with experiences of abuse and result in a larger impact on BHDs, SUD, and offending outcomes.

Past research has shown a strong effect of victimization on male prisoners’ externalizing outcomes, but we did not find strong victimization effects on this outcome. It is possible that our jail population was older, more educated, and consisted of more White and female individuals than the prison samples used in other studies. The higher prevalence of women in this study may have inflated the relationship between victimization and internalizing BHDs.⁶ Although we included men in our analyses, further research is needed regarding the effect of victimization on male-only samples and their outcomes as victimization, particularly sexual abuse, is often treated as a gender-based problem involving a female victim (Depraetere et al., 2020). Additionally, future research could involve a deeper dive into distinctions among additional racial/ethnic groups (vs. a binary measure) as well as interactions between gender and race (e.g., Black women, Native American women, and White women). Research has shown that experiences of abuse can vary across different racial/ethnic groups in incarcerated persons (Wolff & Shi, 2012).

Lastly, data restrictions led us to rely on cross-sectional data for our analyses. As a result, we cannot claim causality between our abuse measures and the outcomes (or between BHDs and SUD or vice versa). Despite examination of the impact of abuse on BHDs and SUD, a bidirectional relationship may exist. Other research has shown that BHDs and SUD predict abuse (Azimi & Daigle, 2021; Walters, 2021). Further research could involve use of longitudinal data to establish temporal order between abuse and BHDs, as well as between abuse and SUD, to help clarify these relationships. Yet, short jail stays may hinder such efforts.

Conclusion

An extensive body of evidence illustrates the harmful effects of abuse on mental health, behavioral, and criminal behavior among justice-involved youth and prison samples (Charak et al., 2019; Duron et al., 2021). Other work has explored the effect of perpetrator type and abuse timing in samples of incarcerated individuals (Cain et al., 2016; Meade & Stenier, 2013). We add to this research by examining the effects of abuse perpetrator type and timing in a jail population. In summary, results showed that experiences of physical, sexual, or polyvictimization, primarily abuse perpetrated by a non-stranger, childhood abuse, or abuse that ever occurred, were damaging for our population’s mental health. While abuse was mostly unrelated to SUD, internalizing and externalizing BHDs were associated with SUD and vice versa. We encourage continued examination of these effects to understand how those in jail may be better served. We especially argue that abuse exposure and mental illness should be considered destabilizing factors that may enhance responsivity to programming.

Early attention to these experiences can contribute to a more preventative, rather than reactive, approach to offenders exposed to abuse, and can also result in greater collaboration with systems outside of the criminal justice system. The end result may be an approach that considers multiple aspects of a person’s behavior and health rather than focusing solely on their present criminal behavior. Not only does the current work emphasize inclusion of abuse and mental health experiences as potential responsivity and destabilizing factors, but it also emphasizes the importance of considering the effect of perpetrator type and abuse timing when providing an intervention to address incarcerated individuals’ unique abuse experiences. Attempts to address abuse and mental illness can aid development of evidence-based practices, as well as promote public safety and reduce costs associated with jails.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research and/or authorship of this article.

ORCID iDs

Melissa A. Kowalski

Emily M. Wright Steiner

Notes

Author Biographies

Melissa A. Kowalski, PhD, is a Research Analyst—Advanced for the Wisconsin Department of Corrections. Her research areas include juvenile justice, risk/needs assessment, mental health, trauma, dynamic needs, responsivity, and gender-specific offending and programming. Her work aims to improve the case management and reentry process for individuals involved in the criminal justice system through evidence-based practices.

Michael Campagna, PhD, conducts research and evaluation activities for justice agencies and non-profit organizations. His work seeks to enhance the reentry process by improving the use of risk and needs assessment to drive evidence-based interventions and case management. His research topics span reentry services, correctional case management, risk/needs assessment, community supervision, life-course theory, and criminogenic thinking patterns.

Emily Wright, PhD, is a Principal Research Associate in the Justice Policy Center at the Urban Institute. Her research focuses on victimization among marginalized populations. She has published over 90 articles, chapters, and reports on the topic, and has received funding from the National Institute of Justice and the National Institutes of Health to conduct research.

Ryan Spohn, PhD, is the Director of the Nebraska Center of Justice Research at the University of Nebraska Omaha where he performs research and evaluation activities targeted at improving the performance of criminal justice, reentry, and corrections agencies. Dr. Spohn is a Fellow of the BJA Innovations Suite Research-Practitioner Academy and a Fellow of Georgetown University’s Center for Juvenile Justice Reform.

References

Azimi

A. M.

Daigle

L. E.

(2021). Mental health and victimization: Does risky lifestyle matter? Journal of Interpersonal Violence, 36, 103–131.

Bates-Maves

J. K.

O’Sullivan

(2017). Trauma-informed risk assessment in correctional settings. International Journal of Criminology and Sociology, 6, 93–102.

Belknap

(2016). Feminist theory. In Piquero

A. R.

(Ed.), The handbook of criminological theory (pp. 290–300). John Wiley & Sons, Inc.

Bonta

Andrews

D. A.

(2017). The psychology of criminal conduct (6th ed.). Routledge, Taylor & Francis Group.

Bronson

Berzofsky

(2017). Indicators of mental health problems reported by prisoners and jail inmates, 2011–2012. U.S. Department of Justice.

Cain

C. M.

Steiner

Wright

E. M.

Meade

(2016). Nonstranger victimization and inmate maladjustment: Is the relationship gendered? Criminal Justice and Behavior, 43, 992–1017.

Charak

Ford

J. D.

Modrowski

C. A.

Kerig

P. K.

(2019). Polyvictimization, emotion dysregulation, symptoms of posttraumatic stress disorder, and behavioral health problems among justice-involved youth: A latent class analysis. Journal of Abnormal Child Psychology, 47, 287–298.

Chesney-Lind

Pasko

(2013). The female offender: Girls, women, and crime (3rd ed.). SAGE.

Cicchetti

Toth

S. L.

(2005). Child maltreatment. Annual Review of Clinical Psychology, 1, 409–438.

10.

Claessens

S. E. F.

Daskalakis

N. P.

van der Veen

Oitzl

M. S.

de Kloet

E. R.

Champagne

D. L.

(2011). Development of individual differences in stress responsiveness: An overview of factors mediating the outcome of early life experiences. Psychopharmacology, 214, 141–154.

11.

Compton

W. M.

Thomas

Y. F.

Stinson

F. S.

Grant

B. F.

(2007). Prevalence, correlates, disability, and comorbidity of DSM-IV drug abuse and dependence in the United States: Results from the national epidemiologic survey on alcohol and related conditions. Archives of General Psychiatry, 64, 566–576.

12.

Corrigan

J. D.

Bogner

(2007). Initial reliability and validity of the Ohio State University TBI Identification Method. Journal of Head Trauma Rehabilitation, 22, 318–329.

13.

Dalbir

Wright

E. M.

Steiner

(2022). Mental illness, substance use, and co-occurring disorders among jail inmates: Prevalence, recidivism, and gender differences. Corrections: Policy, Practice and Research. Advance online publication. https://doi.org/10.1080/23774657.2022.2090028

14.

Daly

(1992). Women’s pathways to felony court: Feminist theories of lawbreaking and problems of representation. Southern California Review of Law and Women’s Studies, 2, 11–52.

15.

Dams-O’Connor

Spielman

Singh

Gordon

W. A.

Lingsma

H. F.

Maas

A. I.

Manley

G. T.

Mukherjee

Okonkwo

D. O.

Puccio

A. M.

Schnyer

D. M.

(2013). The impact of previous traumatic brain injury on health and functioning: A TRACK-TBI study. Journal of Neurotrauma, 30, 2014–2020.

16.

Dennis

M. L.

Chan

Y. F.

Funk

R. R.

(2006). Development and validation of the GAIN Short Screener (GSS) for internalizing, externalizing and substance use disorders and crime/violence problems among adolescents and adults. American Journal on Addictions, 15, 80–91.

17.

Depraetere

Vandeviver

Vander Beken

Keygnaert

(2020). Big boys don’t cry: A critical interpretive synthesis of male sexual victimization. Trauma, Violence & Abuse, 21, 991–1010.

18.

Dunn

E. C.

Nishimi

Powers

Bradley

(2017). Is developmental timing of trauma exposure associated with depressive and post-traumatic stress disorder symptoms in adulthood? Journal of Psychiatric Research, 84, 119–127.

19.

Duron

J. F.

Williams-Butler

Mattson

Boxer

(2021). Trauma exposure and mental health needs among adolescents involved with the juvenile justice system. Journal of Interpersonal Violence, 37, NP15700–NP15725.

20.

Evans

D. L.

Charney

D. S.

Lewis

Golden

R. N.

Gorman

J. M.

Krishnan

K. R. R.

Nemeroff

C. B.

Bremner

J. D.

Carney

R. M.

Coyne

J. C.

Delong

M. R.

(2005). Mood disorders in the medically ill: Scientific review and recommendations. Biological Psychiatry, 58, 175–189.

21.

Farrell

Zimmerman

G. M.

(2018). Is exposure to violence a persistent risk factor for offending across the life course? Examining the contemporaneous, acute, enduring, and long-term consequences of exposure to violence on property crime, violent offending, and substance use. Journal of Research in Crime and Delinquency, 55, 725–765.

22.

Finkelhor

Turner

Hamby

Ormrod

(2011). Polyvictimization: Children’s exposure to multiple types of violence, crime, and abuse. Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs, U.S. Department of Justice.

23.

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.

24.

Ford

J. D.

Grasso

D. J.

Hawke

Chapman

J. F.

(2013). Poly-victimization among juvenile justice-involved youths. Child Abuse and Neglect, 37, 788–800.

25.

Ford

J. D.

Kerig

P. K.

Desai

Feierman

(2016). Psychosocial Interventions for traumatized youth in the juvenile justice system: Clinical, research, and legal perspectives. Journal of Juvenile Justice, 5, 31–49.

26.

Gehring

K. S.

(2018). A direct test of pathways theory. Feminist Criminology, 13, 115–137.

27.

Givens

Cuddeback

G. S.

(2021). Traumatic experiences among individuals with severe mental illnesses on probation. Criminal Behavior and Mental Health, 31, 1–11.

28.

Gossop

Moos

(2008). Substance misuse among older adults: A neglected but treatable problem. Addiction, 103, 347–348.

29.

Grundström

Konttinen

Berg

Kiviruusu

(2021). Associations between relationship status and mental well-being in different life phases from young to middle adulthood. SSM Population Health, 14, 100774.

30.

Hoeve

McReynolds Wasserman

G. A.

(2015). Comorbid internalizing and disruptive behavior disorder in adolescents: Offending, trauma, and clinical characteristics. Criminal Justice and Behavior, 42, 840–855.

31.

Howes

P. W.

Cicchetti

Toth

S. L.

Rogosch

F. A.

(2000). Affective, organizational, and relational characteristics of maltreating families: A system’s perspective. Journal of Family Psychology, 14, 95–110.

32.

Hser

Y-I.

Mooney

L. J.

Saxon

A. J.

Miotto

Bell

D. S.

Huang

(2017). Chronic pain among patients with opioid use disorder: Results from electronic health records data. Journal of Substance Abuse Treatment, 77, 26–30.

33.

Jacobsen

L. K.

Southwick

S. M.

Kosten

T. M.

(2001). Substance use disorders in patients with posttraumatic stress disorder: A review of the literature. The American Journal of Psychiatry, 158, 1184–1190.

34.

Joe

G. W.

Chastain

R. L.

Simpson

D. W.

(1990). Relapse. In Simpson

D. D.

Sells

S. B.

(Eds.), Opioid addiction and treatment: A 12-year follow-up (pp. 121–136). Krieger.

35.

Jones

Brown

Wanamaker

Greiner

(2014). A quantitative exploration of gendered pathways to crime in a sample of male and female juvenile offenders. Feminist Criminology, 9, 113–136.

36.

Karlsson

M. E.

Zielinski

M. J.

(2020). Sexual victimization and mental illness prevalence rates among incarcerated women: A literature review. Trauma, Violence, & Abuse, 21, 326–34.

37.

Knight

D. K.

Blue

T. R.

Flynn

P. M.

Knight

(2018). The TCU drug screen 5: Identifying justice-involved individuals with substance abuse disorders. Journal of Offender Rehabilitation, 57, 525–537.

38.

Knight

Simpson

D. D.

Hiller

M. L.

(2002). Screening and referral for substance-abuse treatment in the criminal justice system. In Leukefeld

C. G.

Tims

F. M.

Farabee

(Eds.), Treatment of drug offenders: Policies and issues (pp. 259–272). Springer.

39.

Krausz

M. R.

Clarkson

A. F.

Strehlau

Torchalla

Schuetz

C. G.

(2013). Mental disorder, service use, and barriers to care among 500 homeless people in 3 different urban settings. Social Psychiatry and Psychiatric Epidemiology, 48, 1235–1243.

40.

Lynch

DeHart

Belknap

Green

(2012). Women’s pathways to jail: The roles & intersections of serious mental illness & trauma. Bureau of Justice Assistance, Office of Justice Programs.

41.

Manly

J. T.

Kim

J. E.

Rogosch

F. A.

Cicchetti

(2001). Dimensions of child maltreatment and children’s adjustment: Contributions of developmental timing and subtype. Development and Psychopathology, 13, 759–782.

42.

Maschi

Morgen

Leibowitz

Rees

(2019). Exploring the relationship between cumulative trauma and recidivism among older adults: Does race and offense history matter? Traumatology, 25, 11–20.

43.

Meade

Steiner

(2013). The effects of exposure to violence on inmate maladjustment. Criminal Justice and Behavior, 40, 1228–1249.

44.

Mersky

J. P.

Janczwewski

C. E.

Topitzes

(2017). Rethinking the measurement of adversity: Moving toward second-generation research on adverse childhood experiences. Child Maltreatment, 22, 58–68.

45.

National Institute of Mental Health (2023, March). Substance use and co-occurring mental disorders.

46.

Niemeyer

Bieda

Michalak

Schneider

Margraf

(2019). Education and mental health: Do psychosocial resources matter? SSM—Population Health, 7, 100392.

47.

Ostermann

Salerno

Hyatt

(2015). How different operationalizations of recidivism impact conclusions of effectiveness of parole supervision. Journal of Research in Crime and Delinquency, 52, 771–796.

48.

Panchal

Saunders

Ndugga

(2022, September 22). Five key findings on mental health and substance use disorders by race/ethnicity. Kaiser Family Foundation.

49.

Paul

K. I.

Moser

(2009). Unemployment impairs mental health: Meta-analyses. Journal of Vocational Behavior, 74, 264–282.

50.

Pettus-Davis

Renn

Lacasse

J. R.

Motley

(2019). Proposing a population-specific intervention approach to treat trauma among men during and after incarceration. Psychology of Men & Masculinities, 20, 379–393.

51.

Radatz

Wright

E. M.

(2017). Does polyvictimization affect incarcerated and non-incarcerated adult women differently? An exploration into internalizing problems. Journal of Interpersonal Violence, 32, 1379–1400.

52.

Salisbury

Van Voorhis

(2009). Gendered pathways: A quantitative investigation of women probationers’ paths to incarceration. Criminal Justice and Behavior, 36, 541–566.

53.

Sinha

(2018). Effect of marital status on substance abuse—A review. International Journal of Recent Scientific Research, 9, 27012–27015.

54.

Solmi

Radua

Olivola

Croce

Soardo

Salazar

Pablo

Il Shin

Kirkbride

J. B.

Jones

Kim

J. H.

Kim

J. Y.

(2022). Age at onset of mental disorders worldwide: Large-scale meta-analysis of 192 epidemiological studies. Molecular Psychiatry, 27, 281–295.

55.

Substance Abuse and Mental Health Services Administration. (2017, August 15). New report finds that 12th grade aged high school dropouts are more at risk for substance use. Substance Abuse and Mental Health Services Administration.

56.

Taxman

F. S.

Caudy

M. S.

(2015). Risk tells us who, but not what or how: Empirical assessment of the complexity of criminogenic needs to inform correctional programming. Criminology & Public Policy, 14, 71–103.

57.

Tripodi

S. J.

Pettus-Davis

(2013). Histories of childhood victimization and subsequent mental health problems, substance use, and sexual victimization for a sample of incarcerated women in the U.S. International Journal of Law and Psychiatry, 36, 30–40.

58.

Walters

G. D.

(2021). Explaining the drug-crime connection with peers, proactive criminal thinking, and victimization: Systemic, cognitive social learning, and person proximity mechanisms. Psychology of Addictive Behaviors, 35, 366–376.

59.

Wolf

M. R.

Prabhu

(2021). Through the eyes of a child: Witnessing childhood trauma is as impactful as experiencing trauma for polyvictimized survivors of child sexual abuse. Journal of Interpersonal Violence, 36, NP5663–NP5677.

60.

Wolff

Shi

(2012). Childhood and adult trauma experiences of incarcerated persons and their relationship to adult behavioral health problems and treatment. International Journal of Environmental Research and Public Health, 9, 1908–1926.

61.

Widom

C. S.

Czaja

S. J.

Dutton

M. A.

(2008). Childhood victimization and lifetime revictimization. Child Abuse & Neglect, 32, 785–796.

62.

Wright

E. M.

Schwartz

J. A.

(2021). The influence of adverse childhood experiences on internalizing and externalizing problems in early adulthood: Evidence of a gene x environment x sex interaction. Child Abuse & Neglect, 114, 104962.