A Treatment-to-Prison-Pipeline? Scoping Review and Multimethod Examination of Legal Consequences of Residential Treatment among Adolescents

Abstract

Objective:

Toward the overall goal of interrogating systems that contribute to racial inequity in child and adolescent psychology, we examine the role and function of Residential Treatment Centers (RTCs) in creating or exacerbating race and gender inequities using the language of mental health and the logic that treatment intentions justify children’s confinement.

Methods:

In Study 1, we conduct a scoping review to investigate the legal consequences of RTC placement, attending to race and gender in 18 peer-reviewed articles, encompassing data for 27,947 youth. In Study 2, we use a multimethod design focusing on RTCs in one large mixed-geographic county to examine which youth are formally charged with a crime while in RTCs, and the circumstances under which these charges occur, attending to race and gender (N= 318, 95% Black, Latine, Indigenous youth, mean age= 14, range= 8-16).

Results:

Across studies, we find evidence for a potential treatment-to-prison pipeline through which youth in RTCs incur new arrests and are charged with crimes during and following treatment. This pattern is pronounced for Black and Latine youth and especially girls, for whom use of physical restraint and boundary violations are recurring challenges.

Conclusions:

We argue that the role and function of RTCs via the alliance between mental health and juvenile legal systems, however passive or unintentional, provides a critical exemplar of structural racism; and thus invite a different approach that implicates our field to publicly advocate to end violent policies and practices and recommend actions to address these inequities.

There is widespread agreement among mental health scholars and practitioners that sending young people to treatment is more effective than sending them to traditional juvenile justice or child welfare facilities like detention or placement (Olsson et al., 2021; Teplin et al., 2021). This endorsement makes great sense in light of robust evidence that children and adolescents impacted by these systems experience disproportionately high mental health burden compared to their non-system-impacted counterparts (Underwood & Washington, 2016). As such, placing youth who experience significant mental health burden in “residential treatment” centers (RTCs¹) likely does not alarm mental health professionals and scholars aware of the treatment and service access needs of system-impacted youth (Annie E. Casey KidsCount Data Center, n.d.; Hockenbery & Puzzanchera, 2020).

Indeed, RTC placement is a common outcome representing nearly a quarter of all confinements for the 2.2+ million children and adolescents involved in the family court system (henceforth referred to as “system-impacted”). Although there is wide diversity in the types of settings that fall under the RTC umbrella (Government Accountability Office [GAO], 2008; Whittaker et al. 2016), there is general agreement that RTCs are populated by youth with significant mental health challenges who have been or may be unsuccessful in less restrictive settings (e.g., Development Services Group, 2019; Schauss et al., 2020). RTCs are defined by international consensus as “purposefully constructed, multi-dimensional living environment(s) designed to enhance or provide treatment” (Whittaker et al., 2016, p. 94). By some estimates, RTCs represent the most frequently referred-to settings for system-impacted youth with trauma and mental health needs (Sawyer & Wagner, 2022; Puzzanchera, 2020). Whether RTCs typically or effectively reduce youths’ mental health burden is an empirical question in need of re-examination; and one which requires and has lacked a racial equity lens to fully understand (Galán et al., 2021).

Family and juvenile courts overseeing system-impacted youth are typical sources of referral to RTCs, suggesting that court-involved youth in general – who are disproportionately youth of color – are more likely to be referred to RTCs (e.g., Schauss et al., 2020). In fact, RTCs house a disproportionate number of Black, Latine and Indigenous children, who constitute about 70% of the RTC population in the United States (Gatwiri et al., 2021; Hockenberry, 2020; Sickmund et al., 2019). Proponents of residential treatment suggest that youth of color are likely to be referred to RTCs given the greater structural barriers that impede their access to mental health services, despite similar overall rates of mental health challenges as compared to their white peers (Baglivio et al. 2017). Family court judges and legal stakeholders suggest that RTCs are often the only option for youth without a delinquency charge who are in need of mental health treatment because of a lack of housing stability and economic resources to secure intensive community-based or private mental health services (GAO, 2008).

Once placed in RTCs, however, the type or efficacy of treatment youth receive is largely unknown despite estimates that states spend approximately $7.1 million a day keeping youth in residential treatment (Justice Policy Institute, 2009). The treatment goals of RTCs may also clash with their institutional realities, with data suggesting that over 70% of youth in RTCs are in locked settings, over 80% of RTCs report using physical restraint and seclusion, and only 1% of youth are voluntary residents (Sickmund et al., 2019). Qualitative research adds to this narrative, describing some children’s experiences of punitive and sexual victimization by staff and peers (e.g., Collins & Bilge, 2020; Milne & Collin-Vézina, 2014). Some studies also suggest that youth in RTCs report worse access to mental health and substance use treatment than those in long-term secure detention facilities in the same jurisdictions – potentially a result of stricter oversight of detention facilities compared to RTCs (Bowser et al., 2018). Understood in this light, RTCs may be functioning as another setting of confinement for youth of color, but in this case, youth confinement does not require an arrest or conviction presumably because treatment is the stated goal. Whereas public safety goals can justify formal incarceration, youth treatment goals justify referrals to RTCs.

Residential Treatment through a Structural and Racial Equity Lens

In the spirit of reckoning with the systemic racism instantiated within our social institutions, we take a new look at RTCs to question the assumption that RTCs have a neutral influence on youth mental health (at worst) and a positive influence (at best). That is, regardless of their stated intentions to treat adolescent mental health, we interrogate whether RTC placement is associated with harm – specifically through entrapping young people in the legal system by increasing risk for arrests and delinquency. This is a necessary question when interrogating settings with a racial equity lens given the role of mass incarceration in racialized oppression and patriarchy (Alexander, 2012). Further, legal consequences are deemed to be critically important to youth wellbeing and opportunity access according to system-impacted youth and youth-led community organizations (Melendrez, 2021; Ozer, 2016).

It is well-documented that legal outcomes (i.e., being arrested or charged with a legal offense) have deleterious effects on adolescent mental health. Arrests are uniquely and prospectively associated with negative mental health outcomes (Sugie & Turney, 2017). Even childhood police contact not resulting in an arrest is linked with accelerated aging and can promote vicarious mental health risk through maternal anxiety and depression (Das, 2022). Prospective studies find that an arrest beginning between age 12 and 16 years is associated with poor health outcomes, especially for Black girls (Christie-Mizell, 2022) and youth living in poverty (Sugue & Turney, 2017). Similarly, legal system contact leading to incarceration uniquely predicts worse mental health and educational outcomes during late adolescence or early adulthood (Javdani, 2019; Powell, 2022). Not surprisingly, the importance of legal risk is corroborated by youth’s lived experiences. Young people describe their legal system involvement as traumatic (Desai, 2019) and name legal outcomes as a key issue for community change (Christens et al., 2014). In participatory work, system-impacted youth consistently identify arrest and confinement as chronic barriers to supporting youth wellbeing and safety (Gardner et al., 2022; Rose et al., 2022; Melendrez, 2021).

Given these documented patterns and the historical interplay between racial injustice and mass incarceration (Alexander, 2012), this paper focuses on the legal consequences of RTC placement. Broadly, we aim to investigate the ways in which mental health and juvenile legal system processes and practices have converged to promote the disproportionate confinement of children of color under the guise of psychological treatment – an argument previously seeded by psychologists (e.g., Simon, 1956). Defined as “a system in which public policies, institutional practices, cultural representations, and other norms work in various, often reinforcing ways to perpetuate racial inequity” (Aspen Institute, 2016), structural racism within treatment contexts is increasingly acknowledged but more rarely the focus of empirical investigation (see Shim, 2021). We argue that investigating the legal consequences of RTCs represents an interrogation of one specific process through which structural racism may be instantiated – via a treatment-to-prison pipeline² justified by the language and logic of psychology.

Although we do not claim to directly investigate the complex and interwoven set of systems, structures, and processes of systemic racism, we view RTCs with a structural and historical lens and interrogate them as sites of racial injustice (Shim, 2021). Key to this is an understanding of how RTCs came to exist and grow as settings of mental health care for system-impacted youth. Specifically, the juvenile legal system was on a path toward less restrictive models of care following key policy shifts in the late 70s and 90s – in other words, towards fewer and shorter confinements for young people (see Javdani et al., 2011). On the mental health side, there emerged widespread recognition that over 70% of system-involved young people had a diagnosable mental health challenge, and public awareness that secure detention facilities were being used to “warehouse” young people because no other alternatives existed. On the juvenile legal side, federal policy made it illegal to incarcerate youth for engaging in a category of behaviors called “status offenses,” which are behaviors for which children but not adults can get in trouble with the law (e.g., running away from home). These combined forces could have created the social and organizational foundation to significantly reduce the numbers of young people – particularly youth of color, girls, and youth with minor offenses – confined in any setting through family courts (see Hinton, 2017).

It was within this policy and practice landscape, however, that RTCs emerged and grew as a link to family courts. This is because youth without an arrest, offense, or conviction could not be detained in secure settings, but they could be mandated to treatment in RTCs (Developmental Services Group, 2019). In this way, RTCs represent new spaces of confinement legitimized through the language of treatment and mental health. Structurally, RTCs do not fall squarely within the juvenile legal or child welfare systems; they can house youth across these child-serving systems, as well as youth who have never been charged with a crime, so long as there are mental health justifications to do so. Given their broad net, RTCs exist in every state in the US and tend to serve system-impacted youth of color living in economic precarity (GAO, 2008; Sickmund et al, 2019).

However, among the few studies that provide a synthesis or empirical overview of RTCs, none investigate race as a central question even though children of color are disproportionately likely to be housed in RTCs (Hockenberry, 2020; Sickmund et al., 2019). Though part of this challenge lies in the lack of a standard definition for RTCs (Daly et al., 2018; Whittaker et al. 2016), this lack of specificity has not curtailed their existence and growth. Indeed, one common characteristic of RTCs is their treatment agenda, which is a key mechanism to justify their presence and rationalize the child separation and mandatory confinement processes they routinely employ (Holmes et al. 2018). Despite their stated treatment agenda, it is important to note that empirical studies of RTC effectiveness in reducing mental health symptoms of youth are rare, mixed, and rely on whether an RTC implements an evidence-based model of care to study (Development Services Group, Inc. 2019; James et al., 2017). Regardless of these mixed findings, there is enough emergent evidence to question the risks associated with these treatment settings, which cast a wide net that most typically catches and confines youth of color – even when they have not committed a crime. Based on their link to juvenile and family courts, we investigate whether and how RTC placement is legally risky for youth through creating a treatment-to-prison pipeline, and attend to race and gender given the legacy of racial and gendered injustice in the courts (Chesney-Lind & Morash, 2013).

Present Study

We use a two-study design to understand both the scope of the potential treatment-to-prison pipeline as well as the processes that undergird or sustain. First, we establish the scope and nature of the problem, asking whether and for whom RTC placement poses legal risk (Study 1); second, we examine the processes and justifications undergirding all instances in which a formal legal charge was made against adolescents while they lived in residential treatment (Study 2). Specifically, Study 1 employs scoping review methodology to provide a synthesis of findings around the question: to what degree is there a link between RTC placement and being arrested or formally charged with a legal offense during or following RTC placement, with attention to race and gender? Study 2 provides depth to the question of whether and how RTCs represent a treatment-to-prison pipeline by empirically investigating when and why young people referred to treatment via RTCs experience legal consequences using a mixed-method case study methodology developed by the first author.

Methods: Study 1

The purpose of the scoping review was to use systematic methods to provide an overview of the available peer-reviewed research evidence regarding the degree to which RTC placement is associated with legal system risk, such as arrest or being charged with a crime. Scoping review methodology is well-suited to synthesize the evidence on RTCs as legally risky placements and identifying important knowledge gaps (Tricco et al., 2018). We detail our systematic process and chart included studies in supplementary Figure 1 and supplementary Table 1 per best practices for scoping reviews (Sucharew & Macaluso, 2019; Tricco et al., 2018). Specifically, the review focused on peer-reviewed literature published after 2002, corresponding with ratification of the Juvenile Justice Delinquency Prevention Act (JJDPA), and used systematic search and reporting methods from the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA; Moher, et al., 2009) framework adapted for Scoping Reviews (PRISMA-ScR; Tricco et al., 2018). Included studies were selected through keyword searches in PsychINFO, Google Scholar, and the National Criminal Justice Reference Service Abstract Database. Studies initially screened were selected by searching databases for English language empirical studies published through September of 2021 that included a common variant of key terms describing RTCs (e.g., residential treatment, residential facility) and legal outcomes (e.g., court, juvenile justice, arrest, incarceration, delinquency). All abstracts of identified articles were reviewed, and all articles that provided empirical data on legal outcomes for youth during or after engagement in residential treatment were included, yielding a total of 18 studies. Studies were excluded if they were not peer reviewed, did not present data on legal system outcomes, did not include residential treatment, or pertained solely to medical drug rehabilitation or sexual offender programs. Studies using quantitative and qualitative methods were included.

Results: Study 1

The review included 18 studies informed by 27,947 young people and 77 staff or stakeholders (see supplementary Table 1). Of these 18 studies, the majority (n = 15) used data from young people who were currently in care or had previously been in care, three examined the perspectives of residential care workers, and six used administrative data or case file reviews. Eleven studies used quantitative, five used qualitative, and two used multiple or mixed methods. Studies were conducted in the U.S. (n = 9), England (n = 4), Australia (n =4), and Canada (n = 1). International studies are included given that these countries are named as part of the international consensus workgroup on RTCs as a global matter of relevance for children (Whittaker et al., 2016). Data regarding race/ethnicity and/or gender are discussed when available, in 50% of all studies and 78% of U.S.-based studies.

Rates of Offending within Care

The findings on rates of youth offending during residential treatment are somewhat mixed. Overall, there is variability across studies, with findings suggesting a range: between 10% and 35% of young people are formally arrested or referred to the legal system during their stay in RTCs. Examining offending rates alone, studies informed by British (Hayden, 2010), Australian (Baidawi, 2020), and U.S.-based (Robst et al., 2011) samples suggest that 10%, 13%, and 15% of youth in residential treatment, respectively, were charged with a legal offense during their time in care. In one study, this rate was comparatively lower than youths’ rate of offending six months prior to care (33%) and 6 months after care (24%) (Robst et al., 2011). Yet, another study examining a sample of 20,000 youth in care in the United States found that 8% of youth had their very first arrest while in a care setting, and 40% of these arrests occurred in group settings (Ryan et al., 2008). Similarly, a U.S. study examining types of out of home treatment found that the highest rates of offending were seen in residential treatment groups: 28%, compared to 10% of youth in state inpatient psychiatric treatment and 14% of youth in therapeutic foster care (Robst et al., 2013). This rate is comparable to an Australian study examining criminal case files of 160 youth, which found that 35% of RTC youths’ offenses were related to youths’ placement (e.g., property damage, assault on staff or residents) and 14% of these cases also led to additional charges related to assault on police or resisting arrest (McFarlane (2017). Only one study finds no evidence for greater legal risk among RTC youth (compared to non-residential / outpatient counterparts; Pullman, 2010), and this study had the largest white sample among U.S. based studies (77%).

Studies examining offense severity and gender contextualize these patterns, suggesting that youth are more likely to be charged with less serious offenses during treatment compared to when they are residing in the community (Hayden & Graves, 2018). Specifically, a pattern of peer-involved charges and low-level assault is corroborated by other studies, which find that RTC are most often charged for property or peer-related offenses (Baidawi, 2020) and threats and misdemeanor violence (Ryan et al., 2008). Two studies specifically examined gender differences, and found patterns of offending by gender interplay. In the first, although boys offended 2.5 times more frequently, girls spent longer time in care for less serious offenses compared to boys Hayden & Graves, 2018). Second, girls in care were more likely to obtain legal records for offending compared to boys, even for similar offenses (Hayden, 2010).

Factors Associated with Offending During Care

Offending during care often occurs within the context of child stressors, like family challenges, rejection by peers or staff, changes to routines or placement, and conflict with peers or staff (Baidwai & Sheehan, 2020). Young people interviewed after leaving care reported offending in care due to disagreements with staff, who they perceived as disrespectful and confrontational (Shaw, 2014). They also reported feeling disempowered in care because of rigid rules and struggled to adjust to RTC placement (Shaw, 2014). In a qualitative study, youth reported that feelings of powerlessness, lack of trust with social workers, and perceiving that the system had given up on them were factors that influenced their offending during care (Day, 2017).

When youth work professionals were asked why they believe youth might offend while in care, social workers expressed that youth may offend for instrumental purposes (e.g., to get one’s way) and that residential settings may be too limited in their ability to properly monitor and control youths’ behavior, ultimately resulting in police involvement (Shaw, 2011). Frontline care workers reported believing that it is necessary to respond punitively to disincentivize future offending (Shaw, 2011). Legal and court professionals, however, believed care settings unnecessarily criminalized young people for behaviors that would likely not be considered illegal if youth were in their homes (Shaw, 2011). In another study, staff in RTCs perceived residential care as a method of last resort for young people who are particularly difficult to manage (Hayden, 2010). This suggests young people may enter residential settings with stigmatized identities, characterizing them as difficult or unable to be helped.

Additionally, RTCs may create a context that more readily calls upon law enforcement to respond to perceived problem behavior (Ryan et al., 2008). In a mixed methods analysis in England, Hayden (2010) found that RTC staff frequently relied on police call-outs to manage youth behavior. While the majority (62%) of police calls were for missing persons, these calls acted as mechanisms to place youth on police’s radar (Hayden, 2010). In a qualitative study with youth work professionals in Australia, criminal justice professionals described the RTC setting as criminogenic, reporting that young people were charged with criminal offenses for behaviors that would not be formally punished in other settings (Gerard et al., 2019). Their reports suggest that dysfunctional dynamics of the care environment itself, the overuse of police for behavior management, and insufficient staff training promoted police and subsequent legal system contact. RTC staff reported that poor training, low support, and limited pay contributed to staff’s decision to call police even for minor problems, especially in response to youth of color (Gerard et al., 2019).

Ineffective peer dynamics were also associated with offending during and after care (Robst et al., 2011). Qualitative work highlights that youth in RTCs perceived that the introduction of new peers often disrupts positive environments and trust (Shaw, 2014), creates pressure for youth to offend to gain status or identify with peers (Day, 2017), and contributes to a sense of powerlessness and inability to adapt to the unstable rules and culture of RTCs (Shaw, 2011). Youth work professionals similarly pointed to the presence of ‘delinquent peers’ as antithetical to treatment goals, suggesting that young people with a history of system contact require different programming (Shaw, 2011). These patterns show how housing system-impacted youth with varying needs and treatment goals in the same residential setting can be contraindicated and create myriad legal consequences for youth (Shaw, 2011).

Youth Legal Outcomes After Residential Care

Although fewer studies focused on outcomes following residential care, several studies illustrate that risk for legal system contact increases following residential treatment, especially for youth of color and youth with trauma or prior arrest histories. Young people arrested during treatment had quadruple the risk of arrest following treatment (Robst et al., 2013); and RTC youth had a 2.5 increase in risk of post-treatment delinquency compared to their foster care youth counterparts (Ryan et al., 2008). Similarly, a qualitative analysis of former RTC youth in Canada reveals that 24% of youth were incurred legal challenges or charges within two years of treatment (Grosset et al., 2018). Post-treatment legal risk increased by 80% if a youth was African American, and by 32% if a youth was Hispanic (Ryan et al., 2008); a finding corroborated by another study suggesting that Black youth and boys were more likely than youth of other races to have police contact after treatment (Robst et al., 2011), and severe trauma histories predicted youth re-arrest more so for youth in RTCs than in treatment foster care (Robst et al., 2017).

The only exceptions to this pattern come from two related studies that examined youth outcomes after an evidence-based treatment including therapy and case management was implemented in RTCs for girls. Specifically, in one study, girls with histories of delinquency who had received Multidimensional treatment Foster Care (MTFC) in RTCs spent fewer days in locked settings, had fewer criminal referrals, and less caregiver-reported delinquency at one year follow-up compared to girls in traditional group care (Leve et al., 2005). These effects persisted at two-year follow up (Chamberlain et al., 2007). This suggests that evidence-based models are rare but promising in reducing legal risk among RTC girls, though racial differences were not examined.

Synthesis of Findings: Study 1

The review provides an initial foundation for understanding whether residential treatment placement is related to legal risk for youth with mental health needs. Though limited and somewhat mixed, findings from the literature indicate that RTC placement is associated with increased legal risk for young people; a pattern that holds across U.S. and non-U.S.-based studies. This may be due to a variety of reasons, including factors related to needs and challenges among individual youth (e.g., youths’ mental health needs, trauma responses, relational challenges), RTC staff (e.g., lack of training and support, over-reliance on police as a form of behavioral management) and the RTC setting (e.g., rigid rules, high surveillance). Taken together, these challenges may create criminogenic instead of treatment-focused processes in residential care setting (Gerard et al., 2019; Hayden, 2010; Ryan et al., 2008). Indeed, as a function of the restrictions and surveillance in these settings, RTCs might be especially legally risky and punitive for youth without prior justice system contact (Robst et al., 2011; Shaw, 2014), youth exhibiting externalizing behaviors during mental health crises (Baidawi & Sheehan, 2020; Day, 2017), and Black youth and girls exhibiting low level and normative adolescent behaviors (Shaw, 2011). There were two exceptions to this pattern linking RTC placement with legal risk. First, one study found no elevation of legal risk among RTC youth during treatment, compared to their counterparts in outpatient treatment settings (Pullman, 2010). We note that this study had the largest proportion of white children reported among U.S. studies, and underscore the need to further examine race differences in risk patterns. A second study finds that youth’s legal risk is relatively lower while they are in RTCs compared to the time window immediately before or after treatment (Robst et al., 2011). However, this protective effect is reversed for Black youth and youth with trauma histories (Robst et al., 2013; 2017). Finally, youth in RTCs are at increased risk of legal system contact following release from treatment; except when an evidence-based model (i.e., MTFC) had been implemented and studied in the RTC (Chamberlain et al., 2007; Leve et al., 2005).

The review also highlights specific gaps in the current literature. First, the limited number of peer-reviewed studies that examine the relationship between residential treatment settings and legal system involvement indicates that this is an overlooked mechanism that may present a significant risk for youth with mental health needs. Given the limited literature, our review included studies from different jurisdictions, with 50% of the included studies conducted with non-U.S. samples. These studies were included because they supported our goal of understanding the scope and degree to which RTCs represent a setting through which young people may come into contact with the legal system. We note that all four countries represent Western first-world nations with large scale child-serving systems and laws that govern juvenile crime. Although there is variability in juvenile law across countries, there is international consensus informed by working groups including representatives from included countries about the vision of RTCs to “enhance or provide treatment” and at minimum provide “safe care” (e.g., Daly et al., 2018; Whittaker et al., 2016). International studies also include diversity within their samples. Of the nine total internationally-based studies, three examine race or gender differences and two suggest that indigenous youth fare worse in the treatment to prison pipeline (Gerard et al., 2019; McFarlane, 2017), whereas one study from the same country (Australia) does not (Baidawi, 2020).

This review raises important questions about how race and gender influence the legal consequences of RTC placement for youth. While our review provides some evidence that girls are arrested for less severe offenses while in RTC than boys, and that Black and Latine youth face greater legal risks than white youth in RTCs, this literature is mostly silent on whether and how race and gender influence legal risk while youth are still in treatment settings. Additionally, extant studies tend to focus greater attention on individual characteristics of youth and less attention on the organizational characteristics and processes of RTCs. These questions are a focus of Study 2.

Methods: Study 2

The overarching purpose of the second study was to identify and interrogate all instances in which youth residing in one large mixed geographic county in the Northeast United States were charged with delinquency offenses while they were in RTCs. Data informing these analyses were made available through a multi-year research-court partnership, which allowed for access to every juvenile court petition filed in the county during an 18-month period prior to the COVID-19 pandemic (January 2016-June 2017). This design eliminates selection or self-report biases because the full population of petition data were made available for analyses. The 18-month time window also maximizes variability and reduces selection confounds. All identifying information from court petitions was redacted for research and the study was reviewed and approved by a University institutional review board.

Petitions are defined as accusatory instruments that can allege that a young person has committed a crime, and include quantifiable indicators (e.g., offense type), and narrative documents that describe the nature of allegation– specifically police reports, witness statements (informed by youth or staff), and victim statements. Court petitions are the primary source for documenting details surrounding the nature and context of an alleged crime.

Data Analytic Approach

We conduct a mixed-methods analysis was conducted to, first, identify and quantify delinquency charges filed against all youth during the study period (n = 318) and code for youth gender, race/ethnicity, setting of offense (i.e., RTC versus community), type of offense (e.g., assault), and offense severity (e.g., misdemeanor vs. felony). This allowed us to examine the proportion of legal petitions filed while youth were in RTC (compared to the community) and whether this varied by race, gender, and offense characteristics. Next, we used an innovative case file methodology to identify and qualitatively code all instances where charges were filed against youth while they were in RTCs (n = 83). Specifically, archival documents contained in all 83 RTC petitions were extracted. These documents included police, witness, and victim statements describing why a delinquency offense was alleged and the context surrounding the type of charge filed. Once extracted, all archival documents were transcribed and systematically coded using a staged open thematic coding approach (Fereday & Muir-Cochrane, 2006; Gibbs, 2007). In the deductive stage, the first and second authors developed a coding manual that defined a priori themes, outlined inclusion and exclusion criteria, and described unique elements of each theme. Next, these themes were applied to an initial set of raw data for refinement of definitions. In the inductive stage, the second and third authors reviewed 24% of all archive transcripts to summarize data and identify cross-cutting themes from the “ground up” from each transcript. This set of themes was integrated with those identified a priori to generate final themes, which were reviewed, corroborated in reference to the new data, and further refined by the first and second authors. All transcripts were coded according to final themes and all inconsistencies in coding were discussed and reconciled until consensus was reached (Hill et al., 1997).

Results: Study 2

Quantitative analyses found that about one-quarter (26%) of total delinquency charges (i.e., petitions) were made against young people while they were mandated to treatment at RTCs. The majority of delinquency charges for RTC youth were for simple and attempted assault, criminal mischief, and minor offenses. Notably, attempted assault charges were almost exclusively filed for youth in RTCs (versus the community), suggesting that a higher proportion of more minor delinquency charges were made for RTC youth. Further, youth of color in RTCs were more likely to have petitions filed against them; indeed, 99% of the petitions filed in RTCs were for youth of color, despite the fact that white youth make up approximately 28% of the youth RTC population in this jurisdiction (Sickmund et al., 2019).

Examination of gender and race patterns also revealed differences for girls of color compared to boys of color. Of the 83 petitions filed among youth in RTCs, 74% were against girls compared to boys, and nearly all girls’ RTC allegations were against girls of color (99%). Girls of color were also considerably more likely to be charged while residing in RTCs (73%), compared to when they were receiving services in the community (22%). The opposite pattern emerged for boys of color, who were less likely to be charged while residing in RTCs compared to when they were in the community (χ2 (1) = 66.80, p < .001). Girls of color were also more likely to be charged for less serious offenses such as misdemeanor simple assault (64%) or “other” category offenses subject to discretionary judgment, whereas boys’ offenses were for more serious and felony charges such as sexual offenses and robbery (23%) (χ2 (6) = 32.354, p < .001). Findings suggest that youth of color were more likely to be charged with delinquency in this RTC, and that girls of color are specifically at risk of these legal consequences despite engaging in more minor offenses.

Court Petition Themes

To interrogate the broader context and processes that placed youth of color at heightened risk for system-involvement once placed in RTCs, the second goal of study 2 was to transcribe and analyze each petition filed against young people in RTCs to understand why youth are charged with crimes during treatment. In this sample, witness and victim statements could include RTC youth and staff, depending on who filed the petition (e.g., RTC staff) and who was present during the allegation (e.g., other RTC youth witnesses). Analyses revealed three overarching themes: lack of safety in care; limited strategies for and standards of care triggering escalated responses and police contact; and residential treatment centers as a high surveillance care setting (see supplementary Table 2). As we elaborate below, these findings contextualize the criminal allegations made against RTC youth and highlight the interpersonal and organizational dynamics contributing to the legal risk of residing in these treatment settings. All names used below are pseudonyms generated to help convey themes and connect narratives where relevant. The term “respondent” is used to refer to the young person being charged with delinquency below.

Lack of Safety in Care

Our first theme speaks to the lack of safety reported and experienced among RTC young people and staff. Safety encompasses physical threats and harms to bodily integrity as well as delayed or ineffective responses that exacerbated safety risks, disaggregated in three sub-themes.

Physical Threat to Safety and Criminalizing Safety-Seeking Behaviors.

Both staff and youth cited feeling unsafe as a primary reason for filing a petition (n=21, 25% of RTC petition sample). This included youth and staff feeling unsafe after experiencing assault while in care. For example, Jacob [youth victim] “explained that he does not feel safe with Will [youth respondent] in the…same program [because Will] has anger issues. Jacob wants [Will] arrested for making it harder for him to breathe”. Similarly, Uma [youth victim] describes feeling “like [other RTC youth] were going to do something to me and I got scared. I asked [staff] to walk me to my room because I wanted to … tell [them] that I think those girls are gonna jump me tonight. [When I told staff, they] … told me to mind my business and stay in my room”. RTC Staff also report safety fears, stating things such as “I still feel I am afraid for my safety. Layla [your respondent] has violence tendencies and has attacked [other] staff in the past.”

These examples show how both youth and staff feel unsafe despite the controlled RTC environment. Further, while most petitions about safety were directed against youth (e.g., young people or staff feeling unsafe from another resident), one youth felt unsafe with RTC staff. In a petition against this youth, a Black girl, a frontline staff member accused her of attempted assault for declining to leave the staff office when she was not provided with a train ticket she was expecting. Police shared:

Upon arrival, Vicky [youth respondent] expressed that she did not feel safe with … any staff [at this facility]. Vicky …wanted to give a statement [even though] I reminded her that it …would not help her case. She [felt] angry that no one …cared [about] … her side of the story. …Vicky tried to explain that she was punched and kicked by staff [after] purposely spitting in [staff victim’s] face …[and] resisting … and trying to get away.

These quotes speak to the ongoing perceptions of threat, heightened states of vigilance, and fear for one’s safety that young people and staff in this RTC experienced. Given that both youth and staff reported ongoing feelings of threat and vigilance, it may be that distinct aspects of the RTC setting itself contributes to staff and youth feeling unsafe; this setting, therefore, may work to further criminalize or perpetuate trauma - rather than treat - the needs of youth given the frequency that both staff and youth reported feeling unsafe.

Ineffective and Delayed Staff Response.

Nearly one-third of petition narratives described an ineffective or delayed staff response to experiences of violence or feeling unsafe among youth (n=26, 31%) that culminated in a delinquency charge filed against a young person. The of these types of charges were against girls (n=25) compared to boys. For instance, in one petition filed by a parent of a resident assaulted by several RTC girls, she states:

Victim [Parent]: It then seemed like 15 minutes later staff showed up and finally helped. But first a staff member jumped on me and pushed me down. They did not know I was a parent and the one being assaulted. Staff later told me that they thought that it was just residents fighting and that’s why they did not do anything.

Importantly, once staff did respond, they “tackled” the mother and sister who were being assaulted to the ground and then left the incident before de-escalating; this led to the mother and sister of the resident being attacked for a second time following the first staff intervention. In a similar incident, a young resident describes how she was assaulted four times over the course of several hours by the same person (a peer) after staff failed to effectively de-escalate the issue and maintain physical separation between the young people:

Staff broke up [our] fight, and took [us] out of unit [but] … 20 minutes later I was returned to the unit [and when] … I walked towards my room to take a shower [she] punched me in the face and I punched her back.

Across a handful of petitions, staff and youth report staff “doing their best” to de-escalate and protect youth in times of harm; for instance, as one youth reported in a victim statement: “the staff member tried to shield (me) from the four girls but, was unable to completely keep (me) from being injured”. This suggests that staff may not have the most effective tools for addressing interpersonal and physical challenges even with the intent of keeping youth safe. This may be particularly salient among girls given that no petition filed against boys in this setting described a similar pattern, whereby a youth was assaulted several times over a short period of time following staff intervention.

Several implications were also documented as a result of delayed and ineffective staff responses to threat and violence, including exacerbating feelings of threat and survival responses among youth in their care that contributed to their criminalization. Indeed, a quarter of petitions (n=17, 20.4%), documented incidents where youth left RTC campuses (i.e., Absent Without Leave or AWOL), with the bulk of these being for girls (n=14) in care. While some youth left RTC for no specified reasons, other petitions also described youth leaving RTCs after experiencing an assault or feeling unsafe in care. For instance, one girl shared: “I walked out of the unit and went AWOL to get away from these girls because I am afraid they are going to attack me again.” This suggests that, when faced with ongoing experiences of harm in the RTC setting, young people may leave RTCs altogether to protect themselves. Three other petitions corroborate the serious extent of safety risks and describe requests for orders of protection. Of these petitions, two were filed by staff while the third was filed by another young person to ensure her own safety; in the petition she describes how another resident “is always starting problems with [her]” and asks for “an order of protection so she cannot come near me again.” This theme not only emphasizes the ineffectiveness of the RTC care setting to prevent basic safety violations, but also signals how safety concerns can usurp the treatment intentions of these care settings and create a culture of fear and mistrust.

Unmet Needs of Youth During Care.

Several petitions also showed how young people asserting their right to access resources to meet their needs escalated conflict and limited safety (n=10, 12%, of which 80% were girls), resulting in assault or theft allegations. Multiple petitions described one incident in which several girls fought amongst themselves because they wanted to watch television. Another petition was filed after a girl used a staff member’s computer without their permission: “Tina [Respondent] was using a computer belonging to [facility] staff to watch movies. Tina has done this before, and has been told several times that she is not to use the computer to watch movies.” Another incident described a boy who ripped an oven out because he wanted to use it to cook food because he was hungry but was told he was not allowed to. Notably, one incident described a girl stealing pain relievers and a pregnancy test while in the community, suggesting a healthcare resource need that was not met by the treatment setting. Particularly among black girls, resource needs are a key correlate of sexual and reproductive health and should trigger a healthcare (not a legal) response (Berezin et al., 2022). In another petition a Black girl named Zara was held down for not moving away from a door when trying to obtain her train tickets from staff; in response, she spit on and pushed the staff member. Following this, Zara was transported to the hospital for a psychological evaluation. Notably, this was the only instance in which any RTC petition cites referral for psychological evaluation prior to police contact. However, Zara was charged with attempted assault upon returning from the psychological evaluation for raising her voice to a staff member. These examples highlight power differentials between youth and staff, and suggest that youth’s resource access and agency seeking strategies can trigger police contact.

Limited Strategies for and Standards of Care Trigger Escalated Responses and Police Contact

The second theme contends with the high number of therapeutic holds used in the RTC setting and the ways in which this clinical tool can become a direct pipeline to system involvement for youth - especially girls of color.

Over-reliance on Therapeutic Holds as a Response to Mental Health Needs.

Therapeutic holds were documented in over a quarter (n = 21) of RTC, and of these, 86% were filed against girls of color. Therapeutic holds were used to respond to a variety of behaviors and crises, including perceived noncompliance (e.g., not following rules, staff feeling disrespected), during escalated arguments between youth or between staff and youth, and as an intervention to “calm” young people experiencing mental health crises. For instance, one petition filed against a girl who was engaging in self-harm behaviors describes how she was met with a therapeutic hold from staff:

Chelsea [a resident; respondent] was in crisis and I was attempting to calm her down along with other staff members. At one point [she] attempted to choke herself with a shirt and bra. At this time I intervened and as a staff member had Chelsea by the legs, I bent down to try to get the shirt and bra away.

Even if used as a “last resort” for this young person, her response to being held culminated in her being charged with a criminal allegation. This suggests that, even if staff have good intentions for utilizing therapeutic holds, the use of such an intervention even in a residential treatment setting, can lead to a young person’s introduction into the legal system. In other petitions, staff appeared to use therapeutic holds when disagreements between staff and youth quickly escalated from perceived disrespect of staff to physical altercations between staff and youth. Here, a staff member describes using a therapeutic hold for a girl experiencing a mental health crisis in school:

Julia [Respondent] was, at that time, standing around watching; as this staff member (witness) went to go break up the girls, Julia … then tried to open the rescue window and tried to jump out of it. As more people showed up [Julia] became more agitated and started swinging and hit me in my stomach…Once in the classroom me and another staff member got Julia down and held her. The third staff member had the legs and I had the top part. After that Julia calmed down. She said she was sorry, that she had lost it.

Other petitions document the use of therapeutic holds in response to perceived noncompliance; for instance, in one allegation, a girl experiencing a prolonged crisis picked up a small rock and threw it at a staff member’s parked car. Following this, “the staff eventually put her in a hold for her safety”. Another petition documented how 3 staff held down a girl who had shoved a staff member and spit in her face, resulting in escalation of the girls’ behavior.

The wide range of behaviors through which holds are used – as well as the high frequency of holds used among girls – suggests that staff rely on physical restraint to respond to girls’ mental health needs and crises that arise during treatment. This is especially notable given the low relative rate of holds against boys despite their more violent allegations, including violent physical and sexual assault charges with weapons.

Indeed, although this sample only includes instances where a delinquency charge was filed, we systematically examined all petitions and supporting documents (e.g., witness statements) for evidence that staff used or referenced any other clinical strategies used prior to a therapeutic hold. We found no such evidence, except for five instances where staff transported girls for additional psychological evaluation following a therapeutic hold and in conjunction with filing a legal allegation. Thus, while these data do not reveal the full scope of treatment available to youth within the RTC, they do suggest that therapeutic holds are a frequently used tool commonly used to respond to the mental health needs of children in this setting. This is not only legally consequential, but of significant clinical concern as therapeutic holds, if required, are intended as a “last resort” to de-escalate youth in crisis (Hem et al., 2018). In this sample, we find corroboration with other studies suggesting that holds may be more commonly used than intended, especially with girls of color, with the function of punitive behavioral management that prioritizes compliance over treatment (Braun et al. 2020). Such compliance-focused settings serve to escalate crises, whereby adolescents may not follow rules or engage in arguments with adults they do not feel safe around (theme 1), and staff respond with heightened vigilance (theme 3 below), commonly resulting in use of therapeutic holds (Hem et al. 2018). This can deteriorate the psychological health and well-being of adolescents – particularly RTC girls of color who experience high rates of complex trauma prior to RTC involvement (Baumle, 2018), and report elevated fear, distress and dehumanization, and increased depression and anxiety following therapeutic holds (Hem, 2018). This further violates young people’s bodily autonomy and integrity which can perpetuate the feelings of unsafety reported among youth in this setting. Thus, therapeutic holds are neither effective nor indicated responses to mental health needs; yet, they are commonly linked to legal consequences for youth in this sample as we describe below.

Therapeutic holds: A Bridge from Treatment to Prison.

In addition to the over-reliance on therapeutic holds to respond to mental health needs and crises, these data also suggests that therapeutic holds can become a mechanism through which youth in treatment sustain legal consequences and become system involved. This pattern is especially salient for alleging assault and attempted assault charges against girls of color. The following petition exemplifies this, documenting how Chelsea was charged with assault when she was actively suicidal:

Chelsea was in crisis and I [staff] was attempting to calm her down along with other staff members. At one point [she] attempted to choke herself with a shirt and bra. At this time I intervened and as a staff member had [her] by the legs, I bent down to try to get the shirt and bra away. Chelsea then bit me on my right bicep causing an injury. She then pulled a key that was [around my neck]…[and] eventually gave me my key back.

The previously described incident involving Julia also underscores the consistency of staff’s responses to and interpretation of youth crises as requiring police intervention. In Julia’s case, several additional staff filed assault charges against her following her therapeutic hold: “Me and another staff member tried to detain her to put her in a standing hold. Julia was still amped up and she decked me in the face with a closed fist;” another narrative notes: “I grabbed Julia to put her in the classroom and that’s when she kicked [the alleged victim].” Girls’ physical response to being held aligns with existing clinical research pointing to the triggering impact therapeutic holds can have on all adolescents, and especially those with complex trauma (Sweeney et al. 2018). Thus, the use of therapeutic holds to respond to mental health needs and crises can escalate youth behavioral problems and result in police involvement, suggesting that using holds may be a bridge from treatment to delinquency in this RTC.

Residential Treatment as a High Surveillance Care Setting

The final theme points to the ways in which this RTC functioned as a high surveillance setting and includes three sub-themes that collectively show how legal system risk is elevated through the enforcement of punitive rules via behavioral management practices, reliance on police as a primary intervention in response to youth non-compliance and staff safety concerns, and the differential treatment of girls of color in this setting.

Punitive Behavioral Control that Prioritizes Compliance over Treatment.

Residential care was described frequently across petitions as a highly punitive setting in which staff used their discretionary power to control youths’ behaviors in response to perceived non-compliance. Specifically, 28 petitions in this sample (~34%) included narratives characterizing punitive behavioral control towards youth that staff perceived as non-compliant. For instance, many assault charges were filed after youth were reportedly being disrespectful to staff or not following rules. These include examples such as youth being in the wrong unit or out of their room past curfew, even if for a reasonable purpose such as using facility resources like computers or the kitchen outside of specifically designated hours. Importantly, these behaviors are not illegal, yet they triggered legal consequences in this RTC setting. For example, in one petition, a girl was charged with a misdemeanor for pulling a fire alarm even though “it didn’t cause any damage” because she “continued to refuse to follow any of the directions.” In another petition, a group of girls were socializing past visiting hours. One girl was charged for refusing to leave and cursing when staff attempted to remover her friends. Even though “the girls eventually left the building through the fire exit”, the supervisor locked out the girls completely even though some were residents there.

These petitions allude to differential power dynamics between staff and youth in this setting, whereby staff expect a high level of unwavering compliance from youth and leverage their power in response to even low level transgressions. For instance, disrespecting staff, even in the absence of threat or assault, also led to the criminalization of youth: “I want to press charges against Angela [respondent] for pushing me and purposely spitting in my face. Angela has anger management issues and I am concerned for my safety and the safety of the staff and other residents.” Although pushing a staff member might be viewed as disrespectful or hostile, this behavior may not rise to the legal standard of assault elsewhere. Other petitions highlight how staff come to rely on the physical handling of youth to get them to comply with rules, such as moving from one common area to another. These practices violate the bodily autonomy of youth in care and create a culture of dehumanization, particularly triggering for youth with trauma and abuse histories (Zelechoski et al., 2013). It further suggests limited implementation of trauma-informed strategies through delivering strengths-based and relationally focused services (Oudshoorn, 2016). Moreover, the lower threshold for filing an allegation against a young person in this RTC is corroborated in other studies investigating power dynamics between staff and RTC youth (James, 2011). This scholarship suggests that, when physical handling and/or police threats are used to respond to perceived problematic behavior in RTCs, youth can be pushed out of treatment and into the legal system. This can create an ineffective treatment culture and undermine trust, boundary setting, and relationship building.

Staff also employed more formal strategies to control youth behavior. Specifically, 9 petitions (~11% of the total sample) had a documented pre-petition detention (PPD) filed, meaning that youth were immediately confined following an allegation on the basis of being perceived either as “too dangerous” or likely to miss court. Staff’s ability to remove youth from a restrictive care setting through the PPD process suggests a tendency towards behavioral control and a failure of the treatment setting to prioritize treatment and restorative approaches over and above legal punishment and removal. The existence of these strategies may contribute to a climate of fear rather than wellness. Such fear strategies were also documented within petitions describing staff calling an administrator on duty when youth “aren’t in their place”. In one instance, a boy took a staff member’s phone for fear they would call the administrator on duty. In another, a call to the administrator triggered a youth to run away from treatment. In both instances, the youth were charged with delinquency because of how they responded to the threat of having an administrator called, instead of for the behavior resulting in the threat in the first place.

Different Threshold of Offenses Based on Gender.

Staff appeared to have a much lower threshold for reporting behaviors and filing charges against girls of color compared to boys of color in this setting. Indeed, quantitative data demonstrated the significantly higher proportion of girls of color who received petitions for low-level offenses (most often simple assault) compared to boys of color who were charged with more severe violent offenses. However, petition narratives suggested that, even when comparing similar offenses by gender, the threshold for girls’ receiving an assault charge was lower than that applied to boys. For example, many of the girls’ petitions noted that there were no visible injuries resulting from girls’ alleged assaults. One petition filed against a girl referred to a fight as a “tussle,” although this led to a subsequent juvenile delinquency charge for assault. Another petition was filed against a girl for attempted assault after pushing a staff member to get out of her way. Comparatively, five of the seven assault charges against boys involved a homemade weapon (shank, belt, box cutter, lock). While assault was the most frequently filed allegation against girls, no assault charges for girls involved the use of weapons. Additional charges among boys included assaults involving severe choking and attempted rape and sexual assault. This suggests a lower threshold for girls to incur legal risk.

Gendered Dynamics around Behavioral Control.

Distinct gendered pathways emerged suggesting that the context of allegations is different for girls compared to boys, that staff respond differently to girls, and this appears to place girls at higher risk for justice system contact once placed in care. Two aspects of girls’ allegations were particularly salient in this regard. The first were the number of assaults that occurred specifically as a result of challenging interpersonal dynamics. Indeed, most petitions filed against girls involved more than one youth resident whereas most of the allegations filed against boys occurred due to one-on-one conflict and appeared to reflect one-off incidents rather than stem from relational loyalties or a buildup of interpersonal tensions or. For instance, a staff member filed an assault charge against a youth who confronted her for publicly “talking shit” about her. Moreover, many girls described fighting, or being punished, on behalf of a friend: “Vida stated to one of the other females, ‘if they lock her up, they have to lock me up too.” Another petition described an incident where three girls began to punch and kick staff members who were attempting to put a fourth girl in a therapeutic hold:

At that time myself and other staff members tried to put Nina [co-respondent] in a therapeutic hold to calm her. Before we were even able to hold Nina, I was attacked by [4 other girls who] … were punching and kicking.

A similar incident occurred when a girl was being handcuffed by officers; upon seeing their friend handcuffed, three girls attempted to prevent the arrest from occurring. Ultimately, all girls were charged with either assault, resisting arrest, and/or obstructing government administration. These examples whereby girls fought against staff on behalf of other residents suggest that girls may view each other as allies when faced with potentially threatening or harmful situations. This allyship seemed to work both against girls (e.g., when girls themselves were assaulted for rumors or interpersonal challenges) and be a source of safety against perceived threat from other staff. Considering the quantity of petitions that named interpersonal problems to justify allegations (n=34, ~41%) – most of which are against girls (94%) – staff seem ill-equipped to respond to girls’ mental health challenges in gender- and trauma-responsive ways (Keris & Schindler, 2013), and instead may over-rely on law enforcement.

The second gendered aspect identified across petitions were the use of girls’ personality traits and characteristics to justify allegations in petitions, unlike petitions filed for boys, which most commonly focused only on the behaviors they were charged for. Many of the girls’ petitions (n=13, ~21% of girls’ petitions) used language to describe personality traits or individual characteristics in addition to the alleged behavior, whereas boys’ petitions solely focused on boys’ behavior (with only n=2 (9%) boys naming personality traits or stable individual characteristics). For instance, staff reports included descriptions of girls as “rowdy”, “disruptive”, and “out of control”. Other petitions sexualized girls and focused on their clothing or the way girls were “touching each other”. This suggests surveillance of girls’ bodies and judgmental evaluations around girls’ morality or respectability, which appeared irrelevant to the specific allegation of assault cited in the actual petition. This aligns with research on how the legal system may punish girls who deviate from expectations of white girlhood (Chesney-Lind & Morash, 2013).

Discussion

The analysis of court petitions reported in Study 2 provides a detailed empirical examination of the legal consequences of RTC placement with attention to both race and gender, expanding the scoping review described in Study 1, which identifies a clear paucity of equity-focused research on RTCs despite concrete yet descriptive evidence that RTC placement is legally risky and socially precarious. Study 2’s quantitative findings also paint a clear picture: if it were not possible to charge young people with a crime during the course of their treatment, youth crime would drop by over 25% in the sample county; almost all (99%) of this decriminalization would affect youth of color; and over three-quarters of total crime would be eliminated for girls of color. This finding has high clinical significance and is corroborated by the results of Study 1, which show that over half of youth do not have a criminal record upon entering treatment, but between ~8% and 79% are arrested or charged with a crime for the first time in care.

Qualitative results from Study 2 extend and elaborate on these patterns. Findings evidenced a pathway whereby youth of color and especially girls of color encounter physical restraint, surveillance, and problematic power dynamics with staff as recurring challenges; and where neither staff nor youth involved in formal petitions experienced basic levels of safety. Juxtaposed with the literature synthesized in Study 1, these dynamics may be heightened by staff feeling disrespected and under-resourced, relying on law enforcement to set a precedent through arrest in order to maintain control of youth behavior, which is provoked by traumatic and stressful interactions with staff and peers. These dynamics likely contribute to youth and staff feeling unsafe, and to youth being legally punished for mostly minor behaviors not considered criminal outside of these high surveillance settings. In fact, across petitions, and especially among those filed against girls of color, when youth mental health challenges or crises are described, the most typical responses of the care setting are to initiate therapeutic holds or regulate youth behavior through punitive behavioral management. In our data, both strategies are linked to police contact and legal allegations against youth. As such, one interpretation is that treatment is required as a justification to confine youth but relegated to a lower priority once police become involved. This introduces the possibility that residential treatment placement itself a specific risk factor for entrapment of youth of color in carceral systems. We suggest that these patterns warrant interrogation of a potential treatment-to-prison pipeline whereby youth with mental health needs can be placed in RTCs without a legal record but leave treatment with one.

Although data are limited on the causal impact of residential treatment, the potential for serious negative consequences for system-impacted youth of color is evident. More research is clearly needed to understand whether and under what contexts RTCs are able to provide sustainable and effective treatment for their residents, especially studies that disaggregate by race and gender. Further, despite agreement in “our field… [that] the debate is now about how evidence should inform clinical practice, not whether it should” (Chorpita & Daleiden, 2014, p. 323; Stuart & Lilienfeld, 2007), our service systems have not moved beyond this in practice; and we risk leaving the most vulnerable youth behind if we do not address inequitable systems even (and especially) when evidence-based care exists. There is a clear role for our field to play in advancing evidence-based care through examining effectiveness and implementation processes through intervention science and engaging in translational and dissemination science. Yet, these endeavors are likely insufficient if we do not face the possibility that mental health treatment is part of the contemporary justification for child separation and confinement – and do something about it.

Leveraging the Current Moment: Directions Forward

We hope this study provides one important step forward in understanding systemic racism in the alliance between mental health and juvenile justice by reviewing literature on the legal consequences of RTC placement, and presenting rarely attainable court data to unpack the black box of what happens when youth are in RTCs. Next, we turn to the question: how might we push ourselves as a field to examine and change these systemic policies and practices, which create and maintain racial inequities in the name of psychology? In the following sections, we draw from the guidance and framework of foundational community-facing (e.g., Sarason, 1981; Humphreys & Rappaport, 1992), structurally-oriented (e.g., Metzl & Hansen, 2014), and more recent antiracist child and adolescent psychology (Castro-Ramirez et al., 2021; Galán et al., 2021; Shim, 2021) literature to recommend action steps forward.

Natural Extensions of Clinical Paradigms: Provider Bias and Structural Treatment Targets

First, there is a clear need to understand and address bias among the supervisory and frontline staff within RTCs – and we are well-positioned as a field to attenuate provider bias through training, coaching, and prevention and intervention science. In doing so, we can advance educational models that directly address the role of racial and gender inequity in order to increase staff’s critical consciousness so that they better understand the role of oppression in youths’ lives (e.g., Heberle et al., 2020). Acknowledging our findings that suggest RTC staff themselves often do not feel a basic sense of physical or psychological safety in RTC settings, models that promote staff’s mental health and skill development (e.g., engagement in trauma-informed care training) would also be promising and may further reduce violent interactions among youth and staff by changing criminogenic setting-level dynamics (Baetz et al., 2017; Barnett et al., 2018) – especially since lack of sufficient training and support may bias staff to view RTC youth as criminals (McElvaney & Tatlow-Golden, 2016; Shaw, 2011). Similarly, youth may internalize staff’s perceptions of them in ways that increase their likelihood of using violence while in care (Van den Tilaart et al., 2018). Treatment models for youth in RTCs can explicitly address internalized oppression and acknowledge the institutional betrayals in the lives of system-impacted youth, thus incorporating a sociopolitical lens into care approaches (e.g., Heberle et al., 2020). Further integrating advocacy to identify and support youth’s self-determined goals is also an effective individual modality that targets structural change and improves youth mental health (e.g., Javdani, 2020). This also highlights the value of contextual competence, or the development of skills and strategies to effectively navigate systems with and on behalf of youth (e.g., Pedersen et al., 2005).

Graduate Training and Social Problem Definition

Graduate program training is a ripe context for engaging with a structural competencies approach and can be applied to myriad programs in clinical psychology and allied disciplines (Metzl & Hansen, 2014; Ali & Sichel, 2020). Ensuring that we attend to and evaluate structural competencies in graduate programs in disciplines such as clinical psychology, psychiatry, counseling, nursing, and social work is of critical importance given the number of mental health practitioners graduated by such programs. Such education and training will foster cultural competencies as part of their foundation through clinical training that sharpens trainee’s attunement to the ways in which client interactions can be influenced by subjective and perceived identity characteristics, prior experiences of discrimination, and intergenerational oppression (Heilbrun et al., 2021). Assessment training, for instance, can underscore the importance of allowing structural factors to drive case conceptualization (Wright, 2021) and understand the biases inherent in specific tools when applied to non-white individuals. Clinical trainees can similarly be trained to interface with child-serving systems through making specific recommendations, informing treatment plans, and evaluating youth engagement and progress (Otto et al., 2012).

Toward structural competence, graduate programs can intentionally value and promote curiosity over compliance; scaffold opportunities for students to imagine and intervene in systems as well as individual lives (e.g., case study of an RTC setting); develop partnership skills to engage with system stakeholders who deploy resources around mental health often with little guidance; and teach policy analysis skills to unpack and leverage the sociopolitical landscape and its influence on mental health treatment access and delivery. All of this requires intentional and direct engagement with our social problem definitions (e.g., broken systems, not broken people), and reflection around the racial justice narratives inherent in the contributions of our field. If we adhere only to progress narratives suggesting that racial justice tends to be on a linear path toward improvement, we may unintentionally ally with oppressive systems without scrutinizing their mandate to provide treatment as much as those systems scrutinize the behavior and morality of the system-impacted families they are meant to serve (Santiago, 2019). If we instead challenge trainees to reflect on the interconnection between historical and contemporary injustice to identify the self-serving function of systems, we can intervene when systems create new settings with familiar logic that move us backwards in the path to racial justice (Alexander, 2012).

Public Advocacy and Structural Touchstones

Moving into policy-focused recommendations, we argue that it is time for the mental health field to advocate against RTCs that do not demonstrate a reasonable effort to implement evidence-based treatment through public advocacy and partnerships with communities and coalitions (Castro-Ramirez et al., 2021). This effort may start with education and self-reflection for ourselves individually and as a field, to understand that systems are challenging and intentionally complex. We invoke the ethical touchstones advanced by O’Neill (1989) as he casts the questions: who are we responsible to, and what are we responsible for; and answers the former by invoking power (we are responsible to the most vulnerable group) and the latter by drawing upon equity-focused ethics (we are responsible for even unintended negative consequences stemming from or justified by our work). In this case, our greatest responsibility lies in advocating for the actual treatment and wellness of youth in RTCs, and acknowledging and advocating against the perhaps unintended pipeline from treatment to prison.

Public advocacy is part of the foundational touchstones of public psychology (Sarason, 1981) and the urgent and emerging equity-focused vision for clinical science (e.g., Shim, 2021); and imperative to implementing a structural competency approach that invites us to redefine social problems in structural terms (Metzl & Hansen, 2014). In the case of RTCs, this might mean removing the label of ‘delinquent’ from the youth who ‘disrupt’ treatment, and applying a label of ‘non-compliant’ or ‘ineffective’ to the treatment settings themselves, which are funded to provide care to children. Further, the invitation to demonstrate competence through developing structural humility also requires the field of mental health to acknowledge our relative silence and complicity as carceral systems continue to co-opt the label of psychological treatment to separate children from their families during developmentally critical windows; thus, we contend that clinical scientists and practitioners should forge a path out of this paradoxical alliance and instead stand against it. This may mean developing partnerships with legal system and court stakeholders to understand how treatment decisions are made and by whom, and how universities and mental health centers can partner to deploy resources to young people and their families. Indeed, such partnerships may be necessary to access and examine data about system-impacted youth, as was the case for the current study, implicating a role for mental health researchers to pursue data-driven treatment decisions.

Imagining structural interventions that target systems’ change may be a particularly important set of competencies to develop; and may require expanding our traditional practice and research paradigms through collaboration and partnership building. For instance, exemplary grassroots organizations have been on the frontlines of dismantling persistently inequitable systems, and can provide roadmaps that we can and should illuminate and study as a field. The Young Women’s Freedom Center (YWFC) Freedom Housing model is one such exemplar that creates structural alternatives to traditional congregate care (see Avil Moreta, 2021). Recognizing the potential harms of settings such as RTCs, the Freedom Housing model recommends connecting youth to non-traditional family and kinship networks able to provide housing and care who are provided financial support to do so. These basic needs of safe housing are considered integral to treatment goals; once met, youth coaches connect young people with trusted community-based mental health supports alongside universal basic income stipends. Formal residential or group housing is reserved only for young people who choose this option or have no other community-based alternative.

Limitations

The scoping review findings (Study 1) are shaped by the broad and idiosyncratic definitions of residential treatment; include primarily, but not all, U.S.-based samples; and are as conclusive as the existing literature allows, which is largely descriptive and limited in its focus on race and gender. Many studies on RTCs did not systematically track legal outcomes and were not included in our review. We are also limited by publication bias. Consistent with the goals of a scoping review, we included only peer-reviewed articles in our review to bound the scope in favor of including fewer studies that have met a higher standard of scientific rigor (Munn et al., 2018). Further RTCs from within and outside of the U.S. may function differently to funnel young people into the justice system in ways that were beyond the scope of interrogation in this study. Nevertheless, key patterns are consistent across U.S. and international studies. Notwithstanding these limitations, use of scoping review methodology is appropriate as it captures the widest net of peer-reviewed studies over a 20-year policy-relevant period to provide the most data-driven synthesis around the nature and degree of legal outcomes associated with RTC placement. Findings from study 2 were limited to youth from one county and focused on youth who had a formal petition filed, limiting conclusions about youth who were not charged while in RTCs. That said, the sample county was not exemplary and data for the entire population of RTC youth charged with a crime over an 18-month period were subject to analysis. Qualitative petition findings may not generalize to other RTCs. However, they provide detailed information about the contexts and characteristics of youth crime while in treatment, with attention to race and gender.

Conclusion

These studies and the impetus for investigating RTCs as a specific site of the treatment-to-prison pipeline are inspired by the call to acknowledge the lies on which racial injustices are built and maintained historically and writ large (Baldwin, 2012), and motivated by the need to call on the field of psychology to contend with and critically act against the inequities promulgated in the name of psychological treatment (APA, 2021; Shim, 2021). Psychology’s part in passively allowing and actively shaping policies and practices that maintain persistent inequities are evident in our role in naming homosexuality as a disease and remaining largely silent during the discriminatory and violent practices of the 1980s AIDS epidemic that benefited from the logic of our disease-based models (Halkitis, 2010); using psychological science to legitimize the inhumane torture of thousands of Muslim and Brown people following 9/11 (Pope, 2016); and the continued proliferation of scholarship on and about Black and Brown people devoid of a structural lens (APA, 2021). At the same time, we know our field holds instrumental power to, for instance, convince the highest courts that adolescents are not developmentally positioned to be legally tried as adults (Steinberg, 2013); shift policy to universalize early education based on sound science and clear values (Yoshikawa et al., 2013); and shape movements that deinstitutionalized adult mental health and named treatment as a human right (Wolff, 2014).

This study aimed to elucidate a current example of systemic racism by interrogating a potential pipeline from treatment to arrest and incarceration for youth of color and especially girls. Through RTCs – which represent a broad umbrella of settings and facilities encompassed within congregate care – we can confine youth without an arrest or allegation of a crime. Once confined, children can and are likely to be charged with a legal offense. RTCs are perhaps an example of a bigger and more insidious issue – that as a field we need to acknowledge that the confinement and incarceration of children is the business of psychology, especially when done in the name of psychological treatment. It is imperative that psychologists be attuned to the functions of treatment settings regardless of their intentions, especially as such residential facilities become more popular alternatives to traditional youth detention centers and prisons. Efforts to reduce or end children’s incarceration must consider these residential settings as potential spaces of confinement in their own right and as pipelines to prison. We name this as a critical issue of public advocacy for Black, Latine, and Indigenous children who inherit and carry the harm and burdens of mass incarceration from birth.