Description
Article published in Victims & Offenders
Using the 2019 National Child Abuse and Neglect Data System, the prevalence and predictors of substantiated reports of child sex trafficking (CST) (n = 1,286) and other maltreatment (n = 705,778) (e.g., physical abuse, sexual abuse, but not sex trafficking) were explored. Descriptive statistics were used to present the profile of children who were victims of CST, and negative binomial regression modeling explored the individual/ontological-, microsystem-, and exosystem-level risk factors for CST compared to other forms of maltreatment. Findings showed that multiracial identity, substance use, disability status; caregiver type – most notably living in a group home – and residing in more rural areas; and professionals as the reporting source were associated with CST versus other maltreatment. Implications for research, policy, and practice for CST are discussed.
Child sex trafficking (CST), or the “sex trafficking of minors,” is defined by the Trafficking Victims Protection Act (TVPA) (2000) as the recruitment, harboring, transportation, provision, or obtaining of a person for the purpose of a commercial sex act … in which a commercial sex act is induced by force, fraud, or coercion, or in which the person induced to perform such an act has not attained 18 years of age (22 USC §7102; 8 CFR §214.11(a)). Thus, unlike other forms of human trafficking, no proof of force, fraud, or coercion is needed for CST because children cannot consent to commercial sex (Boxill & Richardson, 2007). CST can take multiple forms including child pornography (e.g., selling sexual images involving children) and prostitution of children, or the physical exchange of sex with children for money or something else of value (e.g., drugs) (TVPA, 2000). Individuals who traffic children for sex range from non-relative acquaintances and strangers to adults close to and trusted by the child, such as family members, friends, or romantic partners (Reid & Piquero, 2014). As noted by Reid and Piquero (2014), “a key barrier to detection of [CST] is the youthfulness of the victims and their lack of understanding that they are being groomed or exploited” (p. 370).
Figure 1. Sample selection for child maltreatment reports in NCANDS 2019.
Given the clandestine nature of CST, identifying victims of CST is often difficult (Reid & Jones, 2011). At the same time, there has been a growing focus on CST, resulting in considerable research on the context of CST. Identified risk factors have been used to create screening tools and training for personnel likely to encounter victims: first responders, medical providers, teachers, and youth justice and child welfare system staff. These efforts have been significant in identifying trafficked children and further understanding the factors that contributed to their risk for victimization; however, the scope of such initiatives varies widely from state to state and jurisdiction to jurisdiction, leading to disparate opportunities to identify and respond to victims of CST. Specifically, although histories of child maltreatment and child welfare involvement have been identified as key risk factors for sex trafficking, few studies have examined CST as a form of child maltreatment within child welfare populations (for exceptions, see Gibbs et al., 2018; Havlicek et al., 2016). One likely reason underpinning the paucity of attention to CST within the child welfare context is the lack of available data identifying trafficking as a type of maltreatment experience alongside other forms of maltreatment.
To better detect victims of CST in the United States, the Justice for Victims of Trafficking Act of 2015 mandated the collection of data and reporting on the number of children determined to be victims of sex trafficking. In 2018, this reporting requirement was implemented in the National Child Abuse and Neglect Data System (NCANDS) data providing the first national-level information on reports of child maltreatment, including sex trafficking. Using the 2019 NCANDS, the present study aims first to identify the prevalence and profile of children identified as substantiated victims of sex trafficking. Then, individual/ontological-, microsystem-, and exosystem-level characteristics are examined to identify factors that distinguish substantiated victims of CST from substantiated victims of other maltreatment. Finally, the findings are discussed in the context of future research, policy, and practice for children who experience sex trafficking victimization.
CST is frequently associated with other forms of child maltreatment. Indeed, one of the most consistent risk factors for CST is prior trauma and victimization, as evidenced by research linking ACE scores, child abuse, cumulative child victimizations, and CST (e.g., sexual and physical abuse; Cobbina & Oselin, 2011; Reid et al., 2017; Turner et al., 2013; De Vries & Goggin, 2020; C. Williamson & Prior, 2009). For example, Reid et al. (2017) found that among a sample of more than 68,000 youths arrested in Florida from 2007 to 2015, youths who had trafficking reports had significantly higher ACE scores and greater numbers of ACEs indicative of child maltreatment: emotional abuse, physical abuse, sexual abuse, emotional neglect, physical neglect, and family violence, than youths who did not have trafficking reports.
Relatedly, research has consistently noted high levels of child welfare system involvement among victims of CST (Anderson et al., 2019; Gragg et al., 2007; Wright et al., 2021), and particularly, involvement in foster care (Gragg et al., 2007; Havlicek et al., 2016; Landers et al., 2017). Early research by Gragg et al. (2007) found that 86% of victims of CST identified in New York State had prior child welfare system involvement, while 71% of CST victims had experienced placement in foster care. More recent research estimating the prevalence of human trafficking in Ohio found that 82% of identified victims—most of whom were victims of sex trafficking and were minors—had experiences with the child welfare system (Anderson et al., 2019).
Despite these associations, few studies have examined CST as a form of child maltreatment or compared the prevalence and risk factors associated with CST and other maltreatment reported to child welfare agencies. One notable exception, Havlicek et al. (2016) analyzed administrative data from the Illinois child welfare agency to describe the 419 alleged victims of human trafficking (i.e., either sex or labor trafficking) between 2011 and 2015. Findings showed that alleged child trafficking victims were 14.5 years old on average at the time of their first alleged trafficking victimization; most alleged victims were female (90.1%) and identified as Black (53.0%). Nearly 60% had a prior substantiated allegation of child maltreatment. In addition, nearly half of alleged trafficking victims had experience with foster care either before, during, and/or after their alleged trafficking victimization.
More recently, Gibbs et al. (2018) investigated both the prevalence and context of trafficking allegations among all maltreatment allegations received by Florida’s Division of Children and Families (DCF) between January 1, 2011 and December 31, 2015, for children who were at least 10 years old at the time of the allegation. Results found that trafficking allegations comprised 0.5% of all maltreatment allegations during the study period: 53.8% of trafficking allegations were for sex trafficking, 5.7% were for labor trafficking, and 40.5% were for “unspecified human trafficking allegations” (a DCF definition used primarily before 2013). Comparisons between children who were alleged victims of trafficking compared to other forms of child maltreatment showed that alleged trafficking victims were significantly more likely to be older, female, nonwhite, and non-Hispanic. In addition, alleged trafficking victims were significantly more likely to have prior involvement with child welfare than alleged victims of other forms of child maltreatment.
While these seminal studies have helped shed light on the prevalence and context of trafficking among maltreatment allegations, they have done so with limited samples and examined differences among a limited set of risk factors. Further, these studies examined sex and labor trafficking and focused on allegations, not substantiated cases. Taken together, there is still much to be understood regarding CST as a form of child maltreatment.
The paucity of prior studies focused on trafficking, specifically sex trafficking, among child welfare populations is likely due partly to system-wide failures to formally recognize trafficking as a form of child maltreatment. Indeed, it was not until 2014 that federal legislation defined the role of the child welfare system in response to the human trafficking of children. The Preventing Sex Trafficking and Strengthening Families Act of 2014 (Public Law 113–183) first amended the Child Abuse Prevention and Treatment Act (CAPTA) (Public Law 93–274 and later revisions) to require state child welfare agencies to establish policies and procedures to identify, document, screen, and define services for children under their supervision who are victims of, or at risk of, sex trafficking. Additionally, and most relevant to the current study, the 2015 Justice for Victims of Trafficking Act (Public Law 114–82) further amended CAPTA by requiring states to define sex trafficking as a form of child maltreatment and to report trafficking victimization to the National Child Abuse and Neglect Data System (NCANDS), the federal system documenting reports of child maltreatment. The inclusion of sex trafficking victimization as a type of child maltreatment within NCANDS provides an opportunity to examine the prevalence and predictors of sex trafficking victimization among the entire United States’ child welfare population.
According to Ecological Systems Theory (Bronfenbrenner, 1977, 1989), people’s experiences are shaped by a set of nested settings and structures: their social-ecological environment. Prior research has used this framework to better understand the multidimensional risk factors across different types of interpersonal violence, including child abuse and neglect (Belsky, 1980), domestic violence (Heise, 1998), sexual assault (Moylan & Javorka, 2020), and sexual harassment (Richards et al., 2021). Here, we use a socio-ecological framework to conceptualize the interrelated risk factors for CST across three ecological levels: individual/ontological, microsystem, and exosystem.
First, several individual/ontological factors have been linked to increased risk for CST. For example, some research has posited that identifying as female is a risk factor for CST given the dynamics of power and control exerted over victims by predominantly male traffickers (Reid & Piquero, 2016); however, evidence suggests that both girls and boys are at risk for CST (O’Brien et al., 2017) and that boys may be at an increased risk of CST compared to girls (Edwards et al., 2006). In addition, studies have found that nonwhite children (Fedina et al., 2020) and specifically Black children (Reid & Piquero, 2014) are at increased risk of CST compared to their White counterparts. Williamson and Flood (2021) hypothesized that racial disproportionalities in CST victimization were due to the systemic traumas and oppressions racial/ethnic minorities experience. For example, prior studies have suggested that racial disparities in risk for CST may be connected to concentrated economic disadvantage experienced by Black and Brown youth (Reid & Piquero, 2014). At the same time, others suggest that the hypersexualization of Black/African American girls leads to their increased vulnerability to CST compared to their White counterparts (Blake & Epstein, 2019).
In addition, mental health problems such as depression and suicidal attempts, ideation, and self-harming behaviors have been associated with CST victimization (Edwards et al., 2006; Middleton et al., 2018), as has substance use (Franchino-Olsen et al., 2021). This relationship may, in part, be due to a variety of biological, genetic, and psychosocial factors coupled with prior victimization (such as polyvictimization) that, thereby, create persons vulnerable to the allure of traffickers (Levine, 2017; Yakushko, 2009). For example, using data from the National Longitudinal Study of Adolescent Health, Edwards et al. (2006) found that victims of CST were significantly more likely to report lifetime and prior 30-day use of marijuana, cocaine, injection drugs, inhalants, and other illegal drugs compared to children who did not endorse CST victimization. Furthermore, given the linkage between disability status and child maltreatment (Fisher et al., 2008; De La Sablonnière-Griffin et al., 2021), children with disabilities may be at an increased risk for CST (Franchino-Olsen et al., 2020). One study by Reid (2018) specifically linked intellectual disability with CST victimization.
Next, the microsystem consists of immediate settings and social relationships, where people live their daily lives and engage in interpersonal relationships. Given that the present study is focused on children, here we concentrate on the family and household as the microsystem. The household dynamics and characteristics of caregivers impact the child’s environment in manners that may be protective or risk-inducing regarding CST. For example, the household composition has been repeatedly linked to child maltreatment risk (Radhakrishna et al., 2001; Turner et al., 2013, 2007; Zerr et al., 2019): Children living with single parents or a combination of single parents with stepparents or unrelated adult figures (e.g., romantic partners) have a greater risk of child maltreatment (Daly & Wilson, 1985; Margolin, 1992; Radhakrishna et al., 2001; Turner et al., 2007; Zerr et al., 2019), and households including unrelated males are especially risky (e.g., boyfriends of parent/caregiver; Margolin, 1992; Radhakrishna et al., 2001). In contrast, research shows that children living in two-parent households are at a reduced risk of child maltreatment (Turner et al., 2013; Zerr et al., 2019). Regarding CST specifically, some research demonstrates that strains on caregivers (e.g., lack of support, domestic violence, substance abuse, and financial hardships) are causally linked to the risk of victimization (Chohaney, 2016; Reid, 2011, 2012a, 2012b). Comparatively, as previously noted, children in foster care are at a heightened risk for sex trafficking victimization (Landers et al., 2017). Some scholarship suggests that congregate care settings may pose a particular risk for CST (Gibbs et al., 2018).
Relatedly, children’s caregivers serve an essential role in ensuring their safety and well-being. Consequentially, the characteristics of caregivers impact risk for CST. For instance, caregiver substance abuse and caregiver disabilities are associated with the risk of child physical and sexual abuse (English et al., 2002; Fix & Nair, 2020). Additionally, research shows a strong co-occurrence between domestic violence and child maltreatment within households, suggesting that where one exists, the other is more likely (Appel & Holden, 1998; Dong et al., 2004; Fantuzzo et al., 1997; Guille, 2004; Hartley, 2002; Herrenkohl et al., 2008; Holt et al., 2008; G. Margolin & Gordis, 2003).
Furthermore, household financial hardship and living in a disadvantaged neighborhood are prominent risk factors for child maltreatment (Feely et al., 2020; McLeigh et al., 2018). Regarding CST specifically, the absence of resources has been identified as a critical leverage point for traffickers in the grooming process as traffickers exploit children’s need for food, clothes, and money to gain their compliance in initiating them into sex trafficking (Bigelsen & Vuotto, 2013). Household location, specifically, residing in rural areas—compared to metropolitan areas—may be a risk factor for CST as rural areas have less surveillance, training, recordkeeping, and resources for identifying and serving trafficking victims (Castañeda, 2000; Edwards et al., 2006). Taken together, factors associated with the household and community environment in which a child resides are likely associated with experiencing CST (Bigelsen & Vuotto, 2013; Dank et al., 2017; Fedina et al., 2020; Franchino-Olsen et al., 2021).
Finally, microsystems are embedded within the exosystem, which includes broader social settings and structures. Regarding the exosystem, here we focus on one case characteristic that likely reflects differences in the structure of CST compared to other forms of maltreatment: reporting source. Given the clandestine nature of CST and that most training on detecting CST has been directed at first responders, it is likely that professionals (e.g., teachers, doctors) are more likely to report CST than family members, friends, and direct caregivers (e.g., parents).
Identifying victims of CST is often challenging (Reid & Jones, 2011). One avenue of research has identified histories of child maltreatment and child welfare involvement as a key risk factor for sex trafficking (Gibbs et al., 2018). The Justice for Victims of Trafficking Act of 2015 mandated the collection of sex trafficking as a form of maltreatment in NCANDS beginning in 2018, resulting in national-level data on CST among maltreated children in the United States. The current study used 2019 NCANDS to examine the prevalence of substantiated reports of CST and to identify factors that distinguish victims of CST and victims of other maltreatment among substantiated reports. In this regard, three research questions guided our analysis:
Among the population of substantiated reports of child maltreatment in the United States, what is the prevalence of sex trafficking?
Among the population of substantiated reports of child maltreatment in the United States, what is the profile of victims of substantiated reports of CST in the United States?
Are there significant differences regarding (1) individual/ontological-level characteristics (e.g., race/ethnicity), (2) microsystem-level characteristics (e.g., caregiver disability), and (2) exosystem-level characteristics (e.g., report source) for victims of CST compared to victims of other maltreatment among substantiated reports in the United States?
Data for the current study were drawn from the 2019 National Child Abuse and Neglect Data System (NCANDS). NCANDS is a voluntary annual data collection effort on all investigated reports of maltreatment to state child protective service agencies and child welfare outcomes for all 50 states, the District of Columbia, and Puerto Rico conducted by the Children’s Bureau, a federal agency under the United States Department of Health and Human Services’ Office of the Administration for Children and Families. States submit data during the reporting period of October 1 to September 30 of the reporting year. For the 2019 NCANDS file, all states reported. Data collection occurred from October 1, 2018 to September 30, 2019.
While each state has its own definition of child abuse and neglect, federal legislation provides a foundation for state law by identifying a set of acts or behaviors that define child abuse and neglect. For example, the Child Abuse Prevention and Treatment Act (CAPTA) (P.L. 100–294), as amended by the CAPTA Reauthorization Act of 2010 (P.L. 111–320), defines child abuse and neglect as, at a minimum:
Any recent act or failure to act on the part of a parent or caretaker which results in death, serious physical or emotional harm, sexual abuse or exploitation []; or an act or failure to act, which presents an imminent risk of serious harm.
In addition, the Justice for Victims of Trafficking Act of 2015 included an amendment to CAPTA to collect and report the number of children determined to be victims of sex trafficking. In 2018, this reporting requirement was implemented in NCANDS. The key distinction between “sexual abuse” and “sex trafficking” of the child here is that sex trafficking involves the exchange of something of value (e.g., money, drugs) in compensation for the sex.
In 2019, there were 4,255,946 reports of child maltreatment in NCANDS (see Figure 1). Of these, 4,657 reports included allegations of sex trafficking. To maintain consistency with the TVPA’s (2000) definition of minor sex trafficking (e.g., individuals under the age of 18) and to account for varying state laws regarding who is considered a “minor,” maltreatment reports involving individuals 18 years and older were excluded. In addition, we were interested in substantiated reports of maltreatment, and thus unsubstantiated/unconfirmed reports were removed. After excluding reports of maltreatment where (1) victims were 18 years and older as well as (2) unsubstantiated reports, the working sample comprises 707,064 substantiated maltreatment reports; 1,286 include substantiated sex trafficking maltreatment. Most of the 707,064 substantiated reports of child maltreatment involved a unique victim of child maltreatment (i.e., the child was only associated with a single substantiated maltreatment report); however, 7.7% of substantiated maltreatment reports involved a child with more than one substantiated report. Substantiated reports stemmed from all 50 states, the District of Columbia, and Puerto Rico; in 0.2% of reports, the location was unknown (see Appendix).
Each NCANDS report of maltreatment could include up to four types of maltreatment across the following eight categories: (1) physical abuse, (2) neglect or deprivation of necessities, (3) medical neglect, (4) sexual abuse, (5) psychological or emotional maltreatment, (6) sex trafficking, and (7) other. A child was identified as a victim of CST if any of the four maltreatment allegations were sex trafficking (0 = other form/s of child maltreatment only, 1 = sex trafficking was at least one form of maltreatment).
To begin, we captured individual/ontological-level variables. Consistent with prior work, we include the child’s age, sex, race, and ethnicity (Edwards et al., 2006; Fedina et al., 2016; Sprang & Cole, 2018). Age was measured as a continuous variable in years with two notable exceptions: “0” refers to children who are less than one-years-old and “77” refers to children who are unborn. Children who were coded as 77 (i.e., were unborn) were excluded as they could not be trafficked. Sex is coded in NCANDS as a dichotomous variable (0 = boy, 1 = girl). Race is coded categorically (0 = White, 1 = Black/African American, 2 = Asian, 3 = Native Hawaiian/Other Pacific Islander, 4 = Native American/Alaskan Native, 5 = Multiracial). In this regard, we use “White” as our reference category. Hispanic ethnicity is dichotomously coded (0 = Not Hispanic/Latino, 1 = Hispanic/Latino).
In addition, prior research shows that a child’s disability and medical conditions may create additional vulnerability for abuse, and sexual exploitation (Carrellas et al., 2021; Franchino-Olsen et al., 2020; Reid, 2018). NCANDS includes nine categories of child disability that are dichotomously coded (0 = no, 1 = yes): (1) alcohol abuse, (2) drug abuse, (3) intellectual disability, (4) emotionally disturbed,Footnote1 (5) visual/hearing impairment, (6) learning disability,Footnote2 (7) physical disability, (8) behavior problem,Footnote3 and (9) other medical conditions. If a child had a “0” for all of the above, they were coded as having “no diagnosis.” Here, child disability was recoded as (0 = no diagnosis, 1 = substance abuse (alcohol or drug), 2 = any disability, and 3 = any medical condition). In this regard, “no diagnosis” was the reference category.
Further research suggests that prior maltreatment is one of the most consistent risk factors for sex trafficking (Cobbina & Oselin, 2011; Kulig & Pratt, 2021; Reid et al., 2017; De Vries & Goggin, 2020; C. Williamson & Prior, 2009). Prior maltreatment indicates whether, at the time of the current report of maltreatment, the child had prior substantiated or indicated reports for any type of maltreatment in the state’s information system (0 = no, 1 = yes).
Furthermore, five microsystem-level variables were included. Caregiver type indicates the type of caregiver/s with whom the child resided at the time of the maltreatment report. To be clear, the caregiver was not necessarily the perpetrator. Though NCANDS included 13 categories regarding the household type, we condensed these into eight based on the relationship each parent had with the child. For example, a household consisting of two parents who are biologically or adoptively related to the child was combined with the same relationship with the parental figures being unmarried. In contrast, two parent households where one parent is either stepparent or a cohabitating adult figure was coded with when two parental figures have an unknown marital status. The final coding of caregiver type was as follows: 0 = Two parent-household, both related to child, 1 = Two-parent household, one related to child, 2 = Single mother household, 3 = Single father household, 4 = Single-parent household, with other adult relative(s), 5 = Nonparent, relative caregiver, 6 = Nonparent, non-relative caregiver, 7 = Group homes or residential treatment settings, 8 = Other household setting (e.g., hospitals, secure facilities). In this regard, two-parent household in which both were biologically/adoptively related to the child was the reference category.
In addition, a measure of caregiver disability or medical conditions was included. Like the variable for a child’s disability or medical conditions, NCANDS includes a series of dichotomous measures (0 = no, 1 = yes) which reflects whether the child’s primary caregiver had any of the following disabilities or conditions: (1) alcohol abuse, (2) drug abuse, (3) intellectual disability, (4) emotionally disturbed, (5) visual/hearing impairment, (6) learning disability, (7) physical disability, (8) other medical condition. Here, it was recoded as (0 = no diagnosis, 1 = substance abuse (alcohol or drug), 2 = any disability, and 3 = any medical condition). We used “no caregiver diagnosis” as the reference category.
Additionally, domestic violence captured whether during the child maltreatment investigation a child’s caregiver(s) had been identified as a potential perpetrator or victim of domestic abuse (0 = no, 1 = yes). NCANDS defines domestic violence as, “Any abusive, violent, coercive, forceful, or threatening act or word inflicted by one member of a family or household on another.” Financial hardship was measured dichotomously based on whether the family experienced any of the following: (1) inadequate housing, (2) financial problems, and (3) public assistance (0 = no, 1 = yes). First, inadequate housing measured whether the caregiver’s housing facilities were, “substandard, overcrowded, unsafe, or otherwise inadequate” for a child. This included caregiver experiencing homelessness. Second, financial problems indicated whether the caregiver had an inability to financially provide the minimum needs of the family. Third, public assistance indicates that the family was participating in social service programming such as Supplemental Nutrition Assistance Program (SNAP), Supplemental Security Income (SSI), Temporary Assistance for Needy Families (TANF), and/or general assistance.
Finally, evidence shows that children may be at greater risk for trafficking in rural areas, compared to large metropolitan areas (Sprang & Cole, 2018). Rural areas are more isolated (Castañeda, 2000), and may have lower awareness, practitioner training, and recordkeeping regarding human trafficking (Cole & Sprang, 2015; Newton et al., 2008). Rurality of the child’s household was derived from the USDA (2013) rural–urban continuum. Following prior work by Sprang and Cole (2018), the nine categories on the continuum were collapsed into the following: 0 = metropolitan areas, 1 = micropolitan areas, and 2 = rural counties. In this regard, “metropolitan areas” was the reference category.
Finally, one exosystem-level measure was included. The report source refers to the individual(s) who made the maltreatment report (0 = parent, 1 = victim, 2 = other relative or friends/neighbors, 3 = alleged perpetrator, 4 = anonymous, 5 = professional entity [e.g., doctor, teacher], 6 = other). This would also indicate which entity was the primary source of information for the maltreatment report. In this regard, the reference category was “parent.”
As is often observed with administrative data generally and with NCANDS data specifically (e.g., see, Palusci et al., 2005), some case information is coded as missing, unavailable, or unknown. Regarding the independent variables included here, six variables had 0–5% missing data, three variables had 5–10% missing data; and five variables had 10% or more missing data (maximum was 61.40%). These missing data were addressed through multiple imputations using SPSS version 28.
First, descriptive statistics are presented for the full sample of children with substantiated reports of maltreatment as well as for the CST sample and other maltreatment sample (Table 1) and the profile of CST victims is discussed. Then, a series of nested negative binomial regression models were conducted to assess the effect of individual/ontological-, microsystem-, and exosystem-level factors on substantiated maltreatment type. Negative binomial regression modeling is appropriate given that there was substantial positive skew on the dependent variable, indicating that CST was a relatively rare form of child maltreatment (i.e., there was an excess of zeros for the dependent variable). Second, nested modeling was suitable given the three layers of nesting in the data—individual/ontological factors, nested in microsystem factors, and nested in exosystem factors. Odds ratios (OR) are calculated by exponentiating the coefficient and can be interpreted as a one-unit change in any independent variable associated with an increase or decrease in the dependent variable. Robust standard errors were utilized to account for any clustering in cases by state. Alpha was set at p < .05.
Table 1 presents the descriptive statistics for all substantiated maltreatment reports in NCANDS in 2019. On average, children were nearly 7 years old (SD = 5.27), most were girls (51.55%), White (65.60%), non-Hispanic (75.72%), and had no disabilities or medical conditions (57.35%). More than one-third of children (34.13%) had a prior substantiated report of maltreatment. In terms of household characteristics, most children resided with single mothers (28.75%), in two-parent households where both parents were biologically/adoptively related to the child (25.25%), or in two-parent households where one parent was a stepparent or cohabitating partner (24.10%). Half of caregivers had no disabilities or medical conditions, and nearly 29% had a history of domestic violence. Most households experienced financial hardship (53.00%) and most resided in metropolitan areas (80.04%). Finally, 77.96% of maltreatment reports originated with professional entities (e.g., teachers, doctors).
Table 1. Descriptives for Maltreated Children and Comparisons across Sex Trafficking and Other Maltreatment
Variable | M (SD)/% | M (SD)/% | |
---|---|---|---|
Total Sample N = 707,064 | Sex Trafficked n = 1,286 | Other Maltreatment n =705,778 | |
Individual/Ontological Factors | |||
Age | 6.74 (5.27) | 14.06 (3.35) | 6.72 (5.27) |
Girl | 51.55% | 88.34% | 51.48% |
Child’s Race | |||
White | 65.60% | 59.72% | 65.61% |
Black/African American | 24.28% | 28.77% | 24.27% |
American Indian/Alaskan Native | 1.83% | 1.17% | 1.83% |
Asian | 1.13% | 0.78% | 1.13% |
Hawaiian/Other Pacific Islander | 0.31% | 0.39% | 0.31% |
Multiracial | 6.86% | 9.18% | 6.85% |
Hispanic/Latino/a | 25.64% | 25.66% | 25.64% |
Child’s Disability or Medical Conditions | |||
No diagnosis | 57.35% | 44.09% | 57.38% |
Substance abuse | 13.05% | 8.94% | 13.06% |
Any disability | 12.83% | 31.26% | 12.80% |
Any medical condition | 16.76% | 15.63% | 16.76% |
Prior maltreatment | 34.13% | 56.45% | 34.09% |
Microsystem Factors | |||
Caregiver Type | |||
Two parent household, both related to child | 25.25% | 5.68% | 25.28% |
Two parent household, one related to child | 24.10% | 5.75% | 24.13% |
Single mother household | 28.75% | 12.83% | 28.78% |
Single father household | 3.82% | 2.10% | 3.82% |
Single-parent household, with other adult relative(s) | 9.91% | 3.65% | 9.92% |
Nonparent, relative caregiver | 3.91% | 2.41% | 3.91% |
Nonparent, nonrelative caregiver | 2.36% | 3.42% | 2.35% |
Group homes or residential treatment settings | 0.85% | 47.59% | 0.76% |
Other | 1.06% | 16.56% | 1.03% |
Caregiver Disability or Medical Conditions | |||
No diagnosis | 50.44% | 55.37% | 50.43% |
Substance abuse | 36.43% | 29.00% | 36.44% |
Any disability | 10.18% | 12.29% | 10.17% |
Any medical condition | 2.96% | 3.42% | 2.96% |
Domestic Violence | 28.61% | 6.84% | 28.65% |
Financial Hardship | 53.00% | 53.97% | 53.00% |
Rurality | |||
Metropolitan areas | 80.04% | 88.72% | 80.03% |
Micropolitan areas | 14.15% | 8.40% | 14.16% |
Rural counties | 5.81% | 2.88% | 5.81% |
Exosystem Factor | |||
Report source | |||
Parent | 3.52% | 1.56% | 3.53% |
Victim | 0.19% | 0.39% | 0.19% |
Other relative, friends, or neighbors | 7.55% | 2.64% | 7.56% |
Alleged perpetrator | 0.08% | 0.39% | 0.08% |
Anonymous | 3.88% | 2.18% | 3.89% |
Professional entity | 77.96% | 85.77% | 77.95% |
Other | 6.80% | 1.56% | 3.53% |
The first aim of the present study was to identify the prevalence of CST among substantiated cases of child maltreatment: CST comprised 0.18% of all substantiated reports of child maltreatment. Victims of CST were approximately 14 years old and majority female. Less than 60% identified as White, while nearly 29% identified as Black/African American and more than 9% identified as Multiracial. More than one-third of victims of CST had a disability, and more than 55% had a prior substantiated report of child maltreatment. Victims of CST were most likely to live in group homes (47.59%), other care settings (e.g., hospitals) (16.56%), or with a single mother (12.83%). Most CST caregivers had no disability or medical condition, no history of domestic violence, and no financial hardship, but this is likely confounded by the high rates of victims residing in the non-familial settings (i.e., group homes, other care settings). Nearly 90% of victims of CST resided in a metropolitan area and almost 86% of reports were made by a professional entity.
The second aim of the present research was to present a profile of CST victims among substantiated cases of child maltreatment. CST victims were likely an emerging adolescent girl who identified as Black/African American or Multiracial, had a disability, and had previously experienced maltreatment. CST victims were also more likely to reside in non-familial settings and in metropolitan areas. Likewise, their CST victimization was most often reported by a professional.
The final aim of the paper was to assess the independent effects of individual/ontological-, microsystem-, and exosystem-level characteristics on the risk of sex trafficking versus other maltreatment. To do so, a series of nested negative binomial logistic regression models using robust standard errors were estimated (see, Table 2). In the first model, individual/ontological-level characteristics were assessed exclusively. Here, the most prominent risk factor was gender in which girls were nearly six times more likely to experience sex trafficking compared to boys (OR = 5.42, p < .001). Additionally, for each year older a child was, there was an associated 39% increase in the odds of being sex trafficked (p < .001). Regarding race/ethnicity, compared to White children, Black/African American (OR = 1.45, p < .001) and Multiracial (OR = 1.72, p < .001) children were significantly associated with an increased risk of sex trafficking. CST risk was twice as likely for children with a substance abuse issue (OR = 2.12, p = .01). Having a disability diagnosis was associated with a 63% increase in the risk of CST (p < .001), while prior maltreatment increased the risk of sex trafficking by 60%, compared to children without prior maltreatment (p < .001).
Table 2. Negative Binomial Regression Models Predicting Sex Trafficking among Maltreated Children
Model 1 (child-level) | Model 2 (child- and household-level) | Model 3 (child-, household-, and case-level) | |||||||
---|---|---|---|---|---|---|---|---|---|
B (SE) | LL-UL | OR | B | LL-UL | OR | B | LL-UL | OR | |
Age | 0.33 (0.01) | 0.31-0.35 | 1.39*** | 0.19-0.27 | 1.26*** | 0.23 (0.02) | 0.19-0.27 | 1.26*** | |
Female | 1.69 (0.09) | 1.52-1.86 | 5.42*** | 1.63 (0.11) | 1.40-1.87 | 5.10*** | 1.63 (0.11) | 1.40-1.87 | 5.11*** |
Child’s Race | |||||||||
Black/African American | 0.37 (0.07) | 0.23-0.51 | 1.45*** | 0.02 (0.08) | -0.13-0.17 | 1.02 | 0.02 (0.08) | -0.13-0.16 | 1.02 |
American Indian/Alaskan Native | -0.29 (0.37) | -1.06-0.49 | 0.75 | -0.47 (0.51) | -1.63-0.68 | 0.63 | -0.47 (0.50) | -1.61-0.68 | 0.63 |
Asian | -0.33 (0.36) | -1.06-0.39 | 0.72 | -0.02 (0.37) | -0.77-0.72 | 0.98 | -0.05 (0.38) | -0.79-0.70 | 0.95 |
Hawaiian/Other Pacific Islander | 0.19 (0.49) | -0.78-1.15 | 1.21 | 0.17 (0.59) | -1.00-1.35 | 1.19 | 0.16 (0.60) | -1.04-1.36 | 1.17 |
Multiracial | 0.54 (0.10) | 0.33-0.74 | 1.72*** | 0.33 (0.12) | 0.10-0.57 | 1.39** | 0.33 (0.12) | 0.10-0.57 | 1.40** |
Hispanic/Latino/a | 0.04 (0.08) | -0.13-0.21 | 1.04 | 0.03 (0.11) | -0.20-0.26 | 1.03 | 0.02 (0.11) | -0.21-0.24 | 1.02 |
Child’s Disability or Conditions | |||||||||
Substance abuse | 0.75 (0.21) | 0.26-1.24 | 2.12* | 0.81 (0.22) | 0.31-1.31 | 2.25** | 0.79 (0.22) | 0.29-1.30 | 2.21** |
Any disability | 0.49 (0.09) | 0.31-0.68 | 1.63*** | 0.49 (0.11) | 0.25-0.72 | 1.63*** | 0.47 (0.11) | 0.24-0.70 | 1.60*** |
Any medical condition | 0.03 (0.14) | -0.29-0.35 | 1.03 | 0.31 (0.16) | -0.05-0.66 | 1.36 | 0.29 (0.16) | -0.06-0.64 | 1.34 |
Prior maltreatment | 0.47 (0.06) | 0.36-0.59 | 1.60*** | 0.15 (0.09) | -0.04-0.33 | 1.16 | 0.15 (0.09) | -0.03-0.33 | 1.16 |
Caregiver Type | |||||||||
Two parent household, one related to child | — | — | — | -0.22 (0.17) | -0.55-0.10 | 0.80 | -0.22 (0.17) | -0.55-0.10 | 0.80 |
Single mother household | — | — | — | 0.41 (0.14) | 0.13-0.69 | 1.51*** | 0.41 (0.14) | 0.13-0.69 | 1.51** |
Single father household | — | — | — | 0.29 (0.23) | -0.17-0.74 | 1.34 | 0.30 (0.23) | -0.16-0.76 | 1.35 |
Single parent household, with other adult relative(s) | — | — | — | 0.23 (0.19) | -0.14-0.61 | 1.26 | 0.25 (0.19) | -0.13-0.62 | 1.28 |
Nonparent, relative caregiver | — | — | — | 0.43 (0.22) | 0.01-0.85 | 1.54* | 0.45 (0.22) | 0.02-0.87 | 1.56* |
Nonparent, nonrelative caregiver | — | — | — | 1.36 (0.19) | 0.98-1.74 | 3.90*** | 1.36 (0.19) | 0.98-1.74 | 3.90*** |
Group homes or residential treatment settings | — | — | — | 4.04 (0.92) | 1.53-6.56 | 56.83* | 4.03 (0.92) | 1.51-6.55 | 56.09* |
Other care setting | — | — | — | 3.18 (0.94) | 0.63-5.74 | 24.05* | 3.17 (0.94) | 0.62-5.73 | 23.86* |
Caregiver Disability or Conditions | |||||||||
Substance abuse | — | — | — | -0.01 (0.11) | -0.23-0.21 | 0.99 | 0.0 (0.11) | -0.22-0.22 | 1.00 |
Any disability | — | — | — | 0.14 (0.17) | -0.25-0.53 | 1.15 | 0.15 (0.17) | -0.24-0.54 | 1.16 |
Any medical condition | — | — | — | -0.30 (0.25) | -0.82-0.22 | 0.74 | -0.28 (0.24) | -0.80-0.23 | 0.75 |
Domestic Violence | — | — | — | -0.99 (0.15) | -1.29— -0.69 | 0.37*** | -0.99 (0.15) | -1.29 – (-0.69) | 0.37*** |
Financial Hardship | — | — | — | 0.16 (0.08) | -0.01-0.32 | 1.17 | 0.16 (0.08) | -0.01-0.32 | 1.17 |
Rurality | — | — | — | ||||||
Micropolitan areas | — | — | — | -0.42 (0.11) | -0.64— -0.21 | 0.66*** | -0.42 (0.11) | -0.63- (-0.20) | 0.66*** |
Rural counties | — | — | — | -0.43 (0.18) | -0.77— -0.08 | 1.54* | -0.41 (0.18) | -0.76- (-0.07) | 0.66* |
Report source | — | — | — | — | — | — | |||
Victim | — | — | — | — | — | — | -0.26 (0.54) | -1.31-0.80 | 0.77 |
Relative/friend/neighbor | — | — | — | — | — | — | 0.01 (0.30) | -0.57-0.59 | 1.01 |
Alleged perpetrator | — | — | — | — | — | — | 0.69 (0.77) | -0.85-2.23 | 1.99 |
Anonymous | — | — | — | — | — | — | 0.44 (0.32) | -0.19-1.06 | 1.55 |
Professional entity | — | — | — | — | — | — | 0.52 (0.24) | 0.06-0.99 | 1.69* |
Other | — | — | — | — | — | — | 0.41 (0.27) | -0.12-0.94 | 1.51 |
Intercept | -11.72 (0.17) | -12.07 — -11.41 | < .001 | -11.20 (0.26) | -11.76 — -10.65 | < .001 | -11.66 (0.35) | -12.37 — -10.95 | < .001 |
*p < .05; **p < .01; ***p < .001. Reference categories: White, No child diagnosis, Two parent household (both related), No caregiver diagnosis, Metropolitan areas, and Parent report. |
In the second model, microsystem-level characteristics were nested within individual/ontological-level characteristics. Once microsystem-level factors were added to the model, identifying as Black/African American and prior maltreatment were no longer significant predictors of CST; however, identifying as Multiracial continued to be associated with risk for CST, but the strength of the association decreased (OR = 1.39, p < .01). In contrast, the link between child substance abuse and CST was stronger in model 2 with child substance abuse increasing the risk for CST by 2.25 times (p = .006). Having a disability diagnosis continued to increase the risk of CST by 63% (p < .001). In addition, several microsystem-level factors were significantly related to CST, primary among them was caregiver type. Residing in a single-mother household increased the risk of CST by 1.52 times, compared to residing in a two-parent household (both biologically/adoptively related to the child) (p < .001). Similarly, residing in a household with non-parent, relative caregivers increased the risk for CST by 54% (p = .05), while residing with non-parent, non-relative caregivers increased the risk for CST by 4-fold (OR = 3.90, p < .001). The most impactful household type regarding risk for CST was residing in a non-familial setting, whereas residing with “other” caregivers (e.g., hospitals, secure facilities) increased the risk for CST by 24.05 times (p = .03), while residing in group homes and residential treatment settings increased the risk by 56.83 times (p = .01). Unlike household settings, evidence of domestic violence was associated with a 63% decrease in the likelihood of CST. Finally, compared to metropolitan areas, residing in micropolitan areas were associated with a 34% decrease in the likelihood of CST (p < .001), whereas residing in rural counties was associated with a 54% increase in the likelihood of CST (p = .02).
In the third model, individual/ontological- and microsystem-level characteristics were nested within one exosystem-characteristic: report source. First, the child’s age, gender, identification as Multiracial, substance abuse, and disabilities continued to be related to CST risk, net of all other variables. Specifically, each additional year older a child was, there was a 26% increase in the CST risk (p < .001), and girls were more than five times more likely to experience CST compared to boys (p < .001). Additionally, children who identified as Multiracial were 40% more likely to experience CST, compared to children who identified as White (p = .006). The link between substance abuse and CST remained significant as did having a diagnosed disability, although the strength of both relationships was slightly reduced, OR = 2.21, p = .007 and OR = 1.60, p < .001, respectively.
Second, most microsystem characteristics remained significant, net of all other variables, except for two parent households in which only one parent was biologically/adoptively related to the child, single-parent households with one other related adult figure (e.g., grandparents), and single fathers. Residing with a single mother increased CST risk by 51% (p = .004); residing with nonparent, relative caregivers increased CST risk by 56% (p = .04); and residing with nonparent, non-relative caregivers continued to be associated with nearly four times the risk of CST (p < .001), compared to residing in a two-parent household (both biologically/adoptively related). Notably, residing with “other” caregivers (e.g., hospitals) and in group homes or residential treatment settings still yielded the strongest associations with CST among all household types. Specifically, CST risk was increased by 23.86 times for children residing with an “other” caregiver type (p = .03) and 56.09 times in group homes or residential treatment settings (p = .01).Footnote4 Like the prior model, domestic violence was still associated with a decrease in CST risk (OR = 0.37, p < .001). In contrast to the previous model, both micropolitan and rural counties were associated with a decreased risk of CST (34% and 34%, respectively). Finally, in regard to the exosystem factor, professional entities (e.g., teachers, doctors) were 69% more likely (p = .03) to be the source of a report of CST compared to a report of other maltreatment.
The current exploratory study used the 2019 NCANDS to identify the prevalence of CST as a substantiated form of child maltreatment, explore the profile of CST victims, and examine factors that distinguish victims of CST and victims of other maltreatment. We used the socio-ecological framework to organize individual/ontological-, microsystem-, and exosystem-level factors associated with the risk of CST. First, the current study revealed that there were approximately 181.88 substantiated cases of CST per 100,000 substantiated maltreatment allegations. The general profile of a victim of CST comprised a girl, often nonwhite, who had a diagnosed disability and had previously experienced maltreatment. Furthermore, CST victims often lived in group homes, treatment settings, and “other” settings (e.g., hospitals) as opposed to familial settings and resided in metropolitan areas. Relatedly, professional entities were most often the reporting source for CST to child welfare agents.
Taken together, these findings suggest that at present, substantiated reports of CST as maltreatment are rare compared to substantiated reports of other forms of maltreatment. Furthermore, the profile of CST victims suggests that most substantiated reports of CST stem from professionals working with vulnerable youth (e.g., youth living in group homes and youth who have experienced previous maltreatment). This finding highlights the critical importance of screening vulnerable youth for CST.
Regarding predictors of CST, older children, girls identified as Multiracial, had a diagnosed disability, abused substances, and experienced prior maltreatment were more likely to be victims of CST versus other maltreatment. These findings are consistent with much of the previous work on individual/ontological-level factors that increase vulnerability for CST (see meta-analysis by Franchino-Olsen et al., 2021), and that victims of CST within the child welfare context are often girls, teens, and nonwhite align with the two previous studies of child maltreatment rolls in Illinois (Havlicek et al., 2016) and Florida (Gibbs et al., 2018). Implications of these findings are consistent with those noted by previous scholars regarding the importance of prevention and intervention programs that address factors that can create vulnerability to sex trafficking, particularly among youth with racial and ethnic backgrounds other than White (e.g., Fedina et al., 2020).
Findings also identified that neither domestic violence among caregivers nor caregiver disability or medical conditions were predictive of CST in this child welfare population. These findings are inconsistent with prior literature on the impact of family violence and caregiver strain on CST (see, Franchino-Olsen et al., 2021); however, they must be understood within the context of the predominant living situations for victims of CST in this population: group homes and other treatment settings. In other words, most victims of CST were not living with familial caregivers and, as such, would not be impacted by familial dysfunction at the time of their substantiated CST. Furthermore, given the prevalence of CST victims living in non-familial settings, we can infer that this population has experienced significant prior family dysfunction.
Our findings also revealed that several microsystem factors were significant predictors of CST victimization over and above individual/ontological factors. For example, residing with non-relative caregivers versus caregivers related to the child has implications for CST victimization. In this study, children residing with non-relative caregivers were nearly four times more likely to be sex trafficked compared to two-parent households (biologically/adoptively related). This finding is consistent with recommendations to prioritize kinship care over non-relative foster care because the former allows for children to remain in a familiar environment and maintain familial bonds, in turn resulting in less exposure to experiences (e.g., substance use) that increases vulnerability to CST (Andersen & Fallesen, 2015; Maclean et al., 2016; Winokur et al., 2018). In addition, children living in group homes had a 50-fold increase in their risk of CST victimization, while children living in “other” settings (e.g., hospitals) had a 24-fold increase in CST risk. Some media reports have suggested that non-familial settings such as group homes may be a particularly risky setting for CST recruitment (e.g., Fonrouge, 2018; Menzel, 2012; Spicuzza, 2018).
These non-familial settings often house youth who have cumulative risk factors for CST (Chow et al., 2014; Lee & Thompson, 2009), and rates of running away from group homes are significant (Branscum & Richards, 2022)—as such additional research is needed to unpack these relationships. For example, a greater understanding of the impact of microsystem-level factors (e.g., caregivers and/or family structure) on vulnerability for CST could potentially improve screening tools as existing screens focus primarily on individual/ontological factors such as experiences with sexual abuse, drug use, running away, etc. (e.g., Panlilio et al., 2019, see, also O’Brien et al., 2017; De Vries & Goggin, 2020). Further, these findings point to the growing need for trauma-informed foster care settings tailored to the specific needs of CST victims (see, Landers et al., 2017).
Furthermore, inconsistent with prior research (e.g., Castañeda, 2000; Edwards et al., 2006), rurality was not associated with an increased CST risk in this child welfare population. We would caution against interpreting these findings as an indication that “resource deserts” that exist in rural counties—less training, recordkeeping, and resources serving crime victims—are not a critical point of vulnerability for CST victimization (Satcher, 2022). Instead, it may be that these very resource limitations result in less detection (e.g., less reporting, fewer resources for investigation), which is reflected in these findings.
Finally, we found that reports of CST substantiations predominantly stemmed from professional entities. This finding is unsurprising given the focus on training first responders and professionals who are likely to encounter vulnerable youth regarding CST. These results highlight the need for broad adoption of validated screening tools that are easy for practitioners to administer (e.g., Dank et al., 2017; Middleton et al., 2018) within a range of agencies and contexts that work with vulnerable children.
Taken together, the present findings highlight the utility of the socio-ecological framework for organizing and understanding risk factors for CST. For example, racial/ethnic minority status has been identified as an individual/ontological risk factor for CST (e.g., Fedina et al., 2020; Reid & Piquero, 2016); however, the myriad ways in which the larger socio-ecological levels, the macrosystem, exosystem, and microsystem, support vulnerabilities to CST that are concentrated within minority communities have received less attention. For example, at the macrosystem level, racism and the hypersexualization of Black people likely impact how systems—such as the child protective services and youth justice systems—view and respond to Black children (see, Williamson & Flood, 2021). Indeed, prior research suggests that Black children are perceived as “more mature/more adult” (Blake & Epstein, 2019; Epstein et al., 2017; Goff et al., 2014), and Black girls are perceived as sexually active at younger ages than their White peers (Blake & Epstein, 2019; Epstein et al., 2017). In addition, Black youth are generally viewed as more culpable for criminal offending than White children (Bridges & Steen, 1998). Thus, Black children may be more likely to be perceived by system actors as “delinquents” rather than victims (i.e., exosystem-level factors) (see, Epstein et al., 2017), thereby increasing their likelihood of being sent to group care and other treatment settings (i.e., microsystem-level factors). Therefore, examining factors across the full spectrum of the socio-ecological environment calls attention to the inter-related mechanisms that produce vulnerabilities for CST and suggests larger systems change to address risk for CST.
While the present study uses novel data to examine the prevalence and predictors of substantiated reports of CST compared to other forms of child maltreatment, it is not without limitations. One limitation of the NCANDS data is that submitting case-level data is voluntary. Even though all 50 states, the District of Columbia, and Puerto Rico reported case-level data in 2019, the detail of data submitted may vary state by state. Additionally, certain key relationships, such as the victim/trafficker relationship (e.g., parent, peer, romantic partner, etc.), were not possible for these analyses. At the same time, NCANDS is the most comprehensive dataset on child maltreatment reports to child protective services (Fallon et al., 2010).
In addition, as sex trafficking was only added to NCANDS as a form of child maltreatment in 2018, these data are quite new. Indeed, we are unaware of any other studies using these data. As such, future research is needed to continue to examine trends in substantiated cases of CST in these data. Further, unsubstantiated cases were not examined in this study; thus, we leave the examination of all reports of CST to future research. In addition, while beyond the scope of the current paper, macrosystem factors such as laws and policies related to training first responders regarding how to screen/identify victims of CST and at-risk children are also important avenues for future research. For example, a recent report found that four large states were not screening missing foster care youth for CST upon their return to care (Dilanian, 2022). It is likely, therefore, that reporting rates of CST in NCANDS will increase as states continue to adopt screening tools and train a greater number of professionals to screen children for CST. Any net widening in detection may also identify additional predictive factors for subsets of CST victims that are not yet being identified.
While previous research has identified histories of child maltreatment and child welfare involvement as key risk factors for CST (Gibbs et al., 2018), few studies have examined CST victimization alongside other forms of child maltreatment. The present exploratory research leveraged newly available, national-level data on CST among maltreated children in the United States (i.e., NCANDS) to explore the prevalence and predictors of substantiated reports of CST. Findings show that substantiated reports of CST are rare, but a range of individual/ontological-, microsystem-, and exosystem-level factors distinguish CST victims from other maltreatment victims. The findings highlight the need to consider factors across the socio-ecological spectrum (i.e., namely, microsystem-level factors) within screening tools used to predict risk for CST. Furthermore, continued research using the NCANDS data is paramount to developing reporting and substantiation trends and to provide a longitudinal analysis of time-varying risk factors and prevalence and risks for recurrent CST victimization.
1. Emotional disturbance is a formal diagnosis based on the Diagnostic and Statistical Manual of Mental Disorders (DSM) (NCANDS Codebook, 2019).
2. Learning disabilities include, but are not limited to, perceptual disability, brain injury, minimal brain dysfunction, dyslexia, and developmental aphasia (NCANDS Codebook, 2019).
3. Behavior problems refer to behavior in the school or community that adversely impacts the child’s overall development such as involvement in the juvenile justice system and/or running away (NCANDS Codebook, 2019).
4. The large confidence intervals around the odds ratios here are likely due to the fact that residing in either of these settings are highly correlated with the outcome variable—allegation of sex trafficking.
Administration for Children and Families (2018). Child abuse, neglect data from 2015 released. @ACFHHS https://www.acf.hhs.gov/media/press/2017/child-abuse-neglect-data-released
Andersen, S. H., & Fallesen, P. (2015). Family matters? The effect of kinship care on foster care disruption rates. Child Abuse & Neglect, 48, 68–79. https://doi.org/10.1016/j.chiabu.2015.06.005
Anderson, V. R., Kulig, T. C., & Sullivan, C. J. (2019). Estimating the prevalence of human trafficking in Ohio, 2014–2016. American Journal of Public Health, 109(10), 1396–1399. https://doi.org/10.2105/AJPH.2019.305203
Appel, A. E., & Holden, G. W. (1998). The co-occurrence of spouse and physical child abuse: A re and appraisal. Journal of Family Psychology, 12(4), 578–599. https://doi.org/10.1037/0893-3200.12.4.578
Belsky, J. (1980). Child maltreatment: An ecological integration. American Psychologist, 35(4), 320–335. https://doi.org/10.1037/0003-066X.35.4.320
Bigelsen, J., & Vuotto, S. (2013). Homelessness, survival sex, and human trafficking: As experienced by the youth of. Convenant House: New York. https://humantraffickinghotline.org/sites/default/files/Homelessness%2C%20Survival%20Sex%2C%20and%20Human%20Trafficking%20-%20Covenant%20House%20NY.pdf
Blake, J. J., & Epstein, R. (2019). Listening to Black women and girls: Lived experiences of adultification bias. Georgetown Law, Center on Poverty and Inequality. https://endadultificationbias.org/wp-content/uploads/2019/05/Listening-to-Black-Women-andGirls-v7.pdf
Boxill, N. A., & Richardson, D. J. (2007). Ending Sex Trafficking of Children in Atlanta. Affilia, 22(2), 138–149. https://doi.org/10.1177/0886109907299054
Branscum, C., & Richards, T. N. (2022). An Updated Examination of the Predictors of Running Away from Foster Care in the United States and Trends over Ten Years (2010–2019). Child Abuse & Neglect, 129(105689), 0145–2134, https://doi.org/10.1016/j.chiabu.2022.105689.
Bridges, G. S., & Steen, S. (1998). Racial disparities in official assessments of juvenile offenders: Attributional stereotypes as mediating mechanisms. American Sociological Re, 63(4), 554–570. https://doi.org/10.2307/2657267
Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. American Psychologist, 32(7), 513–531. https://doi.org/10.1037/0003-066X.32.7.513
Bronfenbrenner, U. (1989). Ecological systems theory. In R. Vasta (Ed.), Annals of child development (pp. 187–248). JAI Press.
Carrellas, A., Resko, S. M., & Day, A. G. (2021). Sexual victimization and intellectual disabilities among child welfare involved youth. Child Abuse & Neglect, 115, 104986. https://doi.org/10.1016/j.chiabu.2021.104986
Castañeda, D. (2000). The close relationship context and HIV/AIDS risk reduction among Mexican Americans. Sex Roles, 42(7/8), 551–580. https://doi.org/10.1023/A:1007095221993
Chohaney, M. L. (2016). Minor and adult domestic sex trafficking risk factors in Ohio. Journal of the Society for Social Work and Research, 7(1), 117–141. https://doi.org/10.1086/685108
Chow, W.-Y., Mettrick, J. E., Stephan, S. H., & Von Waldner, C. A. (2014). Youth in group home care: Youth characteristics and predictors of later functioning. The Journal of Behavioral Health Services & Research, 41, 503–519. https://doi.org/10.1007/s11414-012-9282-24
Cobbina, J. E., & Oselin, S. S. (2011). It’s not only for the money: An analysis of adolescent versus adult entry into street prostitution*: Adolescent versus adult entry into street prostitution. Sociological Inquiry, 81(3), 310–332. https://doi.org/10.1111/j.1475-682X.2011.00375.x
Cole, J., & Sprang, G. (2015). Sex trafficking of minors in metropolitan, micropolitan, and rural communities. Child Abuse & Neglect, 40, 113–123. https://doi.org/10.1016/j.chiabu.2014.07.015
Daly, M., & Wilson, M. (1985). Child abuse and other risks of not living with both parents. Ethology and Sociobiology, 6(4), 197–210. https://doi.org/10.1016/0162-3095(85)90012-3
Dank, M., Yahner, J., Yu, L., Vasquez-Noriega, C., Gelatt, J., & Pergamit, M. (2017). Pretesting a human trafficking screening tool in the child welfare and runaway and homeless youth systems. Urban Institute. https://www.urban.org/research/publication/pretesting-human-trafficking-screening-tool-child-welfare-and-runaway-and-homeless-youth-systems
De La Sablonnière-Griffin, M., Paquette, G., Hélie, S., & Dion, J. (2021). Child maltreatment investigations and substantiations in child protection services: Factors distinguishing children with intellectual disabilities. Disability and Health Journal, 14(4), 101128. https://doi.org/10.1016/j.dhjo.2021.101128
De Vries, I., & Goggin, K. E. (2020). The impact of childhood abuse on the commercial sexual exploitation of youth: A systematic re and meta-analysis. Trauma, Violence, & Abuse, 21(5), 886–903. https://doi.org/10.1177/1524838018801332
Dilanian, K. (2022July6). Most missing foster kids in four major states weren’t screened to see whether they’d been sex trafficked, says watchdog. NBC News. https://www.nbcnews.com/news/us-news/missing-foster-kids-four-major-states-werent-screened-see-sex-traffick-rcna35937
Dong, M., Anda, R. F., Felitti, V. J., Dube, S. R., Williamson, D. F., Thompson, T. J., Loo, C. M., & Giles, W. H. (2004). The interrelatedness of multiple forms of childhood abuse, neglect, and household dysfunction. Child Abuse & Neglect, 28(7), 771–784. https://doi.org/10.1016/j.chiabu.2004.01.008
Edwards, J. M., Iritani, B. J., & Hallfors, D. D. (2006). Prevalence and correlates of exchanging sex for drugs or money among adolescents in the United States. Sexually Transmitted Infections, 82(5), 354–358. https://doi.org/10.1136/sti.2006.020693
English, D. J., Marshall, D. B., Coghlan, L., Brummel, S., & Orme, M. (2002). Causes and consequences of the substantiation decision in Washington state child protective services. Children and Youth Services Re, 24(11), 817–851. https://doi.org/10.1016/S0190-7409(02)
Epstein, R., Blake, J., & González, T. (2017). Girlhood interrupted: The erasure of black girls’ childhood. Center on Poverty and Inequality, Georgetown Law. girlhood-interrupted.pdf (jjie.org).
Fallon, B., Trocmé, N., Fluke, J., MacLaurin, B., Tonmyr, L., & Yuan, -Y.-Y. (2010). Methodological challenges in measuring child maltreatment. Child Abuse & Neglect, 34(1), 70–79. https://doi.org/10.1016/j.chiabu.2009.08.008
Fantuzzo, J., Boruch, R., Beriama, A., Atkins, M., & Marcus, S. (1997). Domestic violence and children: Prevalence and risk in five major U.S. cities. Journal of the American Academy of Child & Adolescent Psychiatry, 36(1), 116–122. https://doi.org/10.1097/00004583-199701000-00025
Fedina, L., Backes, B. L., Sulley, C., Wood, L., & Busch-Armendariz, N. (2020). Prevalence and sociodemographic factors associated with stalking victimization among college students. Journal of American College Health, 68(6), 624–630. https://doi.org/10.1080/07448481.2019.1583664
Fedina, L., Williamson, C., & Perdue, T. (2016). Risk factors for domestic child sex trafficking in the United States. Journal of Interpersonal Violence, 34(13), 2653–2673. https://doi.org/10.1177/0886260516662306
Feely, M., Raissian, K. M., Schneider, W., & Bullinger, L. R. (2020). The social welfare policy landscape and child protective services: Opportunities for and barriers to creating systems synergy. The ANNALS of the American Academy of Political and Social Science, 692(1), 140–161. https://doi.org/10.1177/0002716220973566
Fisher, M. H., Hodapp, R. M., & Dykens, E. M. (2008). Child abuse among children with disabilities: What we know and what we need to know. International Re of Research in Mental Retardation, 35, 251–289. https://doi.org/10.1016/S0074-7750(07)
Fix, R. L., & Nair, R. (2020). Racial/ethnic and gender disparities in substantiation of child physical and sexual abuse: Influences of caregiver and child characteristics. Children and Youth Services Re, 116, 1–9. https://doi.org/10.1016/j.childyouth.2020.105186
Fonrouge, G. (2018, April 17). The sick tactics traffickers use to find victims. New York Post. https://nypost.com/2018/04/17/how-sex-traffickers-hunt-for-victims-and-brainwash-them/
Franchino-Olsen, H., Martin, S. L., Halpern, C. T., Preisser, J. S., Zimmer, C., & Shanahan, M. (2021). Adolescent experiences of violence victimizations among minors who exchange sex/experience minor sex trafficking. Journal of Interpersonal Violence 37(17–18). https://doi.org/10.1177/08862605211021967
Franchino-Olsen, H., Silverstein, H. A., Kahn, N. F., & Martin, S. L. (2020). Minor sex trafficking of girls with disabilities. International Journal of Human Rights in Healthcare, 13(2), 97–108. https://doi.org/10.1108/IJHRH-07-2019-0055
Gibbs, D., Kluckman, M., Tueller, S., & Latzman, N. (2018). Assessing the under-reporting of minor victim sex trafficking (NCJ: 253456). Office of Justice Programs. https://www.ojp.gov/library/publications/assessing-under-reporting-minor-victim-sex-trafficking
Goff, P. A., Jackson, M. C., Di Leone, B. A. L., Culotta, C. M., & DiTomasso, N. A. (2014). The essence of innocence: Consequences of dehumanizing black children. Journal of Personality and Social Psychology, 106(4), 526–545. https://doi.org/10.1037/a0035663
Gragg, F., Petta, I., Bernstein, H., Eisen, K., & Quinn, L. (2007). New York prevalence study of commercially sexually exploited children. Westat.
Guille, L. (2004). Men who batter and their children: An integrated re. Aggression and Violent Behavior, 9(2), 129–163. https://doi.org/10.1016/S1359-1789(02)
Hartley, C. C. (2002). The co-occurrence of child maltreatment and domestic violence: Examining both neglect and child physical abuse. Child Maltreatment, 7(4), 349–358. https://doi.org/10.1177/107755902237264
Havlicek, J., Huston, S., Boughton, S., & Zhang, S. (2016). Human trafficking of children in Illinois: Prevalence and characteristics. Children and Youth Services Re, 69, 127–135. https://doi.org/10.1016/j.childyouth.2016.08.010
Heise, L. L. (1998). Violence against women: An integrated ecological framework. Violence Against Women, 4(3), 262–290. https://doi.org/10.1177/1077801298004003002
Herrenkohl, T. I., Sousa, C., Tajima, E. A., Herrenkohl, R. C., & Moylan, C. A. (2008). Intersection of child abuse and children’s exposure to domestic violence. Trauma, Violence, & Abuse, 9(2), 84–99. https://doi.org/10.1177/1524838008314797
Holt, S., Buckley, H., & Whelan, S. (2008). The impact of exposure to domestic violence on children and young people: A re of the literature. Child Abuse & Neglect, 32(8), 797–810. https://doi.org/10.1016/j.chiabu.2008.02.004
Kulig, T. C., & Pratt, T. C. (2021). Meta analysis. In J. C. Barnes & D. R. Forde Eds., The encyclopedia of research methods in criminology and criminal justice (1st ed., pp. 497–502). Wiley. https://doi.org/10.1002/9781119111931.ch102
Landers, M., McGrath, K., Johnson, M. H., Armstrong, M. I., & Dollard, N. (2017). Baseline characteristics of dependent youth who have been commercially sexually exploited: Findings from a specialized treatment program. Journal of Child Sexual Abuse, 26(6), 692–709. https://doi.org/10.1080/10538712.2017.1323814
Lee, B. R., & Thompson, R. (2009). Examining externalizing behavior trajectories of youth in group homes: Is there evidence for peer contagion? Journal of Abnormal Child Psychology, 37(1), 31–44. https://doi.org/10.1007/s10802-008-9254-4
Levine, J. A. (2017). Mental health issues in survivors of sex trafficking. Cogent Medicine, 4(1), 1278841. https://doi.org/10.1080/2331205X.2017.1278841
Maclean, M. J., Sims, S., O’Donnell, M., & Gilbert, R. (2016). Out‐of‐home care versus in‐home care for children who have been maltreated: A systematic re of health and wellbeing outcomes. Child Abuse Re, 25(4), 251–272. https://doi.org/10.1002/car.2437
Margolin, L. (1992). Child abuse by mothers’ boyfriends: Why the overrepresentation? Child Abuse & Neglect, 16(4), 541–551. https://doi.org/10.1016/0145-2134(92)90070-8
Margolin, G., & Gordis, E. B. (2003). Co-occurrence between marital aggression and parents’ child abuse potential: The impact of cumulative stress. Violence and Victims, 18(3), 243–258. https://doi.org/10.1891/vivi.2003.18.3.243
McLeigh, J. D., McDonell, J. R., & Lavenda, O. (2018). Neighborhood poverty and child abuse and neglect: The mediating role of social cohesion. Children and Youth Services Re, 93, 154–160. https://doi.org/10.1016/j.childyouth.2018.07.018
Menzel, M. (2012, July 6). DCF examining group homes after sex trafficking. Jacksonville Record. https://www.jaxdailyrecord.com/article/dcf-examining-group-homes-after-sex-trafficking
Middleton, J. S., Gattis, M. N., Frey, L. M., & Roe-Sepowitz, D. (2018). Youth experiences survey (yes): Exploring the scope and complexity of sex trafficking in a sample of youth experiencing homelessness. Journal of Social Service Research, 44(2), 141–157. https://doi.org/10.1080/01488376.2018.1428924
Moylan, C. A., & Javorka, M. (2020). Widening the lens: An ecological re of campus sexual assault. Trauma, Violence, & Abuse, 21(1), 179–192. https://doi.org/10.1177/1524838018756121
Newton, P. J., Mulcahy, T. M., & S. E. (2008). Martin Finding victims of human trafficking (Doc. In No. 224393). U.S. Department of Justice.
O’Brien, J. E., White, K., & Rizo, C. F. (2017). Domestic minor sex trafficking among child welfare–involved youth: An exploratory study of correlates. Child Maltreatment, 22(3), 265–274. https://doi.org/10.1177/1077559517709995
Palusci, V. J., Smith, E. G., & Paneth, N. (2005). Predicting and responding to physical abuse in young children using NCANDS. Children and Youth Services Re, 27(6), 667–682. https://doi.org/10.1016/j.childyouth.2004.12.004
Panlilio, C. C., Miyamoto, S., Font, S. A., & Schreier, H. M. C. (2019). Assessing risk of commercial sexual exploitation among children involved in the child welfare system. Child Abuse & Neglect, 87, 88–99. https://doi.org/10.1016/j.chiabu.2018.07.021
Radhakrishna, A., Bou-Saada, I. E., Hunter, W. M., Catellier, D. J., & Kotch, J. B. (2001). Are father surrogates a risk factor for child maltreatment? Child Maltreatment, 6(4), 281–289. https://doi.org/10.1177/1077559501006004001
Reid, J. A. (2011). An exploratory model of girl’s vulnerability to commercial sexual exploitation in prostitution. Child Maltreatment, 16(2), 146–157. https://doi.org/10.1177/1077559511404700
Reid, J. A. (2012a). A girl’s pathway to prostitution: Linking caregiver adversity to child susceptibility. LFB Scholarly. Domestic Sex Trafficking Risk Factors in Ohio 13.
Reid, J. A. (2012b). Exploratory re of route-specific, gendered, and age-graded dynamics of exploitation: Applying life course theory to victimization in sex trafficking in North America. Aggression & Violent Behavior, 17(3), 257–271. http://dx.doi.org/10.1016/j.avb.2012.02.005
Reid, J. A. (2018). Sex trafficking of girls with intellectual disabilities: An exploratory mixed methods study. Sexual Abuse, 30(2), 107–131. https://doi.org/10.1177/1079063216630981
Reid, J. A., Baglivio, M. T., Piquero, A. R., Greenwald, M. A., & Epps, N. (2017). Human trafficking of minors and childhood adversity in Florida. American Journal of Public Health, 107(2), 306–311. https://doi.org/10.2105/AJPH.2016.303564
Reid, J. A., & Jones, S. (2011). Exploited vulnerability: Legal and psychological perspectives on child sex trafficking victims. Victims & Offenders, 6(2), 207–231. https://doi.org/10.1080/15564886.2011.557327
Reid, J. A., & Piquero, A. R. (2014). Age-graded risks for commercial sexual exploitation of male and female youth. Journal of Interpersonal Violence, 29(9), 1747–1777. https://doi.org/10.1177/0886260513511535
Reid, J. A., & Piquero, A. R. (2016). Applying general strain theory to youth commercial sexual exploitation. Crime & Delinquency, 62(3), 341–367. https://doi.org/10.1177/0011128713498213
Richards, T. N., Holland, K., Kafonek, K., & Navarro, J. (2021). Sex-based harassment in the United States’ k-12 schools: Rates and predictors of allegations, student reporting, and student discipline. Journal of School Violence, 20(4), 402–416. https://doi.org/10.1080/15388220.2021.1920423
Satcher, L. A. (2022). Multiply-deserted areas: Environmental racism and food, pharmacy, and greenspace access in the Urban South. Environmental Sociology 8(3), 1–13. https://doi.org/10.1080/23251042.2022.2031513
Spicuzza, M. (2018, December 14). Kids missing from group homes common thread in trafficking. Milwaukee Journal Sentinel. https://www.jsonline.com/story/news/politics/2018/12/14/sex-trafficking-group-homes-common-thread-milwaukee-cases/2305584002/
Sprang, G., & Cole, J. (2018). Familial sex trafficking of minors: Trafficking conditions, clinical presentation, and system involvement. Journal of Family Violence, 33(3), 185–195. https://doi.org/10.1007/s10896-018-9950-y
Turner, H. A., Finkelhor, D., Hamby, S. L., & Shattuck, A. (2013). Family structure, victimization, and child mental health in a nationally representative sample. Social Science & Medicine, 87, 39–51. https://doi.org/10.1016/j.socscimed.2013.02.034
Turner, H. A., Finkelhor, D., & Ormrod, R. (2007). Family structure variations in patterns and predictors of child victimization. American Journal of Orthopsychiatry, 77(2), 282–295. https://doi.org/10.1037/0002-9432.77.2.282
U.S. Department of Health and Human Services, Administration for Children and Families, Administration on Children, Youth and Families, Children’s Bureau. 2020. Child maltreatment 2018. Available from https://www.acf.hhs.gov/cb/research-data-technology/statistics-research/child-maltreatment
Williamson, E., & Flood, A. (2021). Systemic and structural roots of child sex trafficking: The role of gender, race, and sexual orientation. In D. Victimization, M. Chisolm-Straker, & K. Chon (Eds.), The historical roots of human trafficking: Informing primary prevention of commercialized violence (pp. 191–216). Springer International Publishing. https://doi.org/10.1007/978-3-030-70675-3_11
Williamson, C., & Prior, M. (2009). Domestic minor sex trafficking: A network of underground players in the Midwest. Journal of Child & Adolescent Trauma, 2(1), 46–61. https://doi.org/10.1080/19361520802702191
Winokur, M. A., Holtan, A., & Batchelder, K. E. (2018). Systematic re of kinship care effects on safety, permanency, and well-being outcomes. Research on Social Work Practice, 28(1), 19–32. https://doi.org/10.1177/1049731515620843
Wright, E. R., LaBoy, A., Tsukerman, K., Forge, N., Ruel, E., Shelby, R., Higbee, M., Webb, Z., Turner-Harper, M., Darkwa, A., & Wallace, C. (2021). The prevalence and correlates of labor and sex trafficking in a community sample of youth experiencing homelessness in metro-Atlanta. Social Sciences, 10(2), 32. https://doi.org/10.3390/socsci10020032
Yakushko, O. (2009). Human trafficking: A re for mental health professionals. International Journal for the Advancement of Counselling, 31(3), 158–167. https://doi.org/10.1007/s10447-009-9075-3
Zerr, A. A., Newton, R. R., Litrownik, A. J., McCabe, K. M., & Yeh, M. (2019). Household composition and maltreatment allegations in the US: Deconstructing the at-risk single mother family. Child Abuse & Neglect, 97, 104123. https://doi.org/10.1016/j.chiabu.2019.104123
Substantiated Maltreatment by State (Order of frequency) | ||
---|---|---|
State | Frequency | Percent |
NY | 77,835 | 11.0 |
CA | 68,288 | 9.7 |
TX | 65,734 | 9.3 |
IL | 37,587 | 5.3 |
MI | 35,491 | 5.0 |
FL | 34,420 | 4.9 |
MA | 27,947 | 4.0 |
OH | 27,786 | 3.9 |
IN | 24,788 | 3.5 |
KY | 22,363 | 3.2 |
SC | 19,169 | 2.7 |
OK | 15,903 | 2.2 |
OR | 14,669 | 2.1 |
IA | 13,726 | 1.9 |
AZ | 13,554 | 1.9 |
CO | 13,036 | 1.8 |
AL | 11,963 | 1.7 |
UT | 11,157 | 1.6 |
GA | 10,259 | 1.5 |
MS | 10,178 | 1.4 |
TN | 9,996 | 1.4 |
NM | 9,296 | 1.3 |
AR | 8,787 | 1.2 |
LA | 8,747 | 1.2 |
CT | 8,625 | 1.2 |
MD | 8,216 | 1.2 |
WV | 7,117 | 1.0 |
MN | 7,061 | 1.0 |
VA | 6,174 | 0.9 |
NC | 5,726 | 0.8 |
NJ | 5,307 | 0.8 |
NV | 5,291 | 0.7 |
PR | 4,987 | 0.7 |
PA | 4,892 | 0.7 |
WA | 4,882 | 0.7 |
MO | 4,833 | 0.7 |
ME | 4,782 | 0.7 |
WI | 4,745 | 0.7 |
MT | 3,951 | 0.6 |
AK | 3,564 | 0.5 |
RI | 3,416 | 0.5 |
KS | 3,111 | 0.4 |
NE | 2,908 | 0.4 |
DC | 2,002 | 0.3 |
ID | 1,927 | 0.3 |
ND | 1,842 | 0.3 |
SD | 1,603 | 0.2 |
Unknown | 1,512 | 0.2 |
HI | 1,366 | 0.2 |
DE | 1,248 | 0.2 |
NH | 1,228 | 0.2 |
WY | 1,131 | 0.2 |
VT | 938 | 0.1 |
Total | 707064 | 100 |