Importance of Leadership and Employee Engagement in Trauma-informed Organizational Change at a Girls’ Juvenile Justice Facility












Importance of Leadership and Employee Engagement in Trauma-informed Organizational Change at a Girls’ Juvenile Justice Facility



TITLE: Importance of Leadership and Employee Engagement in Trauma-informed Organizational Change at a Girls’ Juvenile Justice Facility

Juvenile justice facilities have historically struggled with creating and maintaining safe, therapeutic environments. The need to maintain order has often led to practices that diminish and traumatize residents rather than enable healing and development. Efforts to create change in institutional settings face a range of implementation challenges and constraints that include lack of leadership, follow through, and entrenched staff practices and attitudes(Aarons, 2006). Using descriptive information gathered at a secure juvenile justice facility for girls in Pennsylvania that implemented a trauma-informed organizational change model, the Sanctuary Model®, this study seeks to understand what changed in the facility duringand immediately after the implementation of the model. We also seek to understand what factors, other than the model itself, were perceived as critical to model implementation and positive change.

Juvenile Justice System

Since the inception of the juvenile justice system in 1899, juvenile courts have struggled with offering care and rehabilitation versus control and punishment to juvenile offenders (for historical reviews, see Griffin, Germain, & Wilkerson, 2012; Siegel & Welsh, 2011). During this time, there have been many attempts to improve the rehabilitative potential of juvenile justice institutions and programs (Sedlak & McPherson, 2010). Reducing recidivism is of course a key outcome of juvenile justice programs, but reconciling the complex and contradictory organizational goals of juvenile justice systems can undercut efforts to promote consistent positive change.In a meta-analysis of 548 study samples that included, but was not restricted to, residential facilities, Lipsey (2009) found that several over-riding factors predicted a reduction in recidivism in treatment groups compared to control groups. One of the relevant factors that predicted improved outcome was what Lipsey terms the “therapeutic intervention philosophy.” He found that interventions involving counseling and its variants - skill building, restorative justice programs or multiple coordinated services, all predicted a substantive decrease in recidivism; whereas interventions involving surveillance or deterrence (e.g., prison visitation) did not. Additionally, interventions involving discipline (e.g., para-military regimens in boot camps) actually predicted an increase in recidivism. Another critical factor was the quality of implementation of the intervention.

Lipsey’s findings are consistent with Griffin et al.’s (2012) conclusion that punishment-oriented programs are not as effective at rehabilitation as a mental health model that, recognizing the high prevalence of diagnosable mental illness in secure juvenile justice facilities,provides assessment, behavioral management, and clinical treatment (e.g., medication, therapy) to residents(Skowyra & Cocozza, 2007). However, Griffin and colleagues also identify a critical problem with the mental health model in residential facilities, where therapeutic services are provided by professional clinical staff, but not by direct care staff who handle day-to-day behavior management. The separation of therapeutic services from the daily environment has been a historic problem not only in juvenile justice facilities, but in other residential settings that treat, for example, individuals with psychiatric problems. An hour a day of clinical counseling may promote growth and rehabilitation, but may be counteracted by the other 15 hours of interaction with peers and direct care staff responsible for behavior management involving punitive, and sometimes traumatizing, strategies such as restraints and seclusion. These interactions may introduce new traumas as well as mirror and reinforce traumas that residents might have experienced in their earlier lives (Hennessey, Ford, Mahoney, Ko and Siegfried, 2004)

As early as the 1940s, Bettelheim and others promoted the concept of a therapeutic milieu where every interaction involving every staff position in a residential setting was designed to be therapeutic(Bettelheim, 1974). Similarly, in the 1960s, researchers finding that the impact of peer dynamics on adolescents in residential settings could be a much more powerful influence on resident behavior than adult interventions, developed tools and programs designed to harness and redirect peer influence in prosocial ways (Mullen, 1999; Polsky, 1963). However, perhaps due both to political and financial constraints, and the difficulties involved in implementing and sustaining institutional change, therapeutic milieu-based programs have never become the norm, particularly in juvenile justice facilities.

Trauma-Informed Care

  More recently, however, spurred by research on the impact of traumatic and adverse experiences, a new model of intervention - trauma-informed care - has been developed and implemented in a range of institutional settings, presenting a promising alternative for juvenile justice facilities (Griffin et al., 2012). Many youth detained in juvenile justice facilities have extensive histories of exposure to trauma (Ford & Blaustein, 2013). A recent report summarizing data across states suggested that at least 80% of girls in the juvenile justice system have experienced physical or sexual abuse (Saar, Epstein, Rosenthal, & Vafa, 2015). Documentation of extensive trauma histories, particularly among female delinquents, has highlighted the need to address trauma in this population. Trauma-informed models view angry, aggressive, and emotionally reactive behaviors as resulting from the impact of traumatic experiences (Griffin et al., 2012). For example, when a youth feels threatened by a reminder of prior abuse (a trigger), they may become aggressive or respond in other inappropriate ways. A staff response based on simply maintaining order, such as a physical restraint, will likely exacerbate the situation and reinforce the maladaptive behavior; whereas a trauma-informed response will help the youth use tools to self-regulate and calm down, and will de-escalate the situation. Such a response will also educate the youth on the reasons for their own behavior, ultimately increasing their pro-social competence. While trauma-focused treatment may be beneficial, and can be used in conjunction with trauma-informed care, the advantage to trauma-informed models in residential settings is that they explicitly target the entire organization and its culture and, thus, the day-to-day environment where crises and negative behaviors are most likely to occur. Direct care and ancillary staff members, as well as clinical staff,become responsible for the therapeutic response.

Although this study examines the Sanctuary Model (detailsbelow), other trauma-informed interventions have been implemented in juvenile justice facilities, as well as other residential settings for youth, and preliminary research findings showing promise for these models are beginning to be published. For example, Hodgdon and colleagues(2013)found a reduction in negative behaviors, and the use of restraints, following implementation of the Attachment, Regulation and Competency (ARC) model at two residential treatment programs.   Similarly, an adaptation of the Trauma Affect Regulation: Guide for Education and Therapy (TARGET) intervention to a milieu model, resulted in a decrease of disciplinary incidents and use of seclusion in juvenile detention centers, and suggested a potential decrease in recidivism(Ford & Hawke, 2012).

However, not all research on implementation of trauma-informed models in juvenile justice or other residential facilities for youth, have shown positive findings. Some of the reasons for this appear to be problems with the implementation process in the facilities, including staff resistance to change, conflict between managerial and line staff, and other factors which stymie organizational change efforts (Cox, 2013). This should not be surprising given the complexities of initiating organizational change, especially in public institutions (Dadich, Stout, & Hosseinzadeh, 2015). As mentioned earlier, one of the main factors which Lipsey (2009) identified in his meta-analysis that predicted reductions in recidivism from juvenile justice programs, was the quality of implementation of the intervention. This is consistent with evidence from other fields relating to implementation science. Indeed, analysis of many years of implementation outcomes by the National Implementation Research Network (NIRN; Bertram, Suter, Bruns, & O'Rourke, 2011; Fixsen, Blase, Naoom, & Wallace, 2009) has pointed to the importance of studying and addressing implementation issues such as organizational context, organizational readiness, organizational culture, and facilitative administration.

Current Study

This article presents a descriptive exploratory examination,based on retrospective accounts by program staff, of the process of implementation of the Sanctuary Model at the North Central Secure Treatment Unit Girls Program (NCSTU) in Pennsylvania. It grew out of a study designed more generally to determine the effectiveness of Sanctuary Model implementation at NCSTU, including an evaluation of an objective of the Sanctuary Model – changes in institutional safety(Elwyn, Esaki, & Smith, 2015). Findings suggested that the facility was a safer place for both residents and staff after implementation of the model. For example, there was a substantial decrease in restraints, assaults on staff, and in the percentage of interviewed youth who reported that they feared for their safety. Facility safety indicators also compared very favorably to those of the juvenile justice correctional field in general.Additionally, staff turnover decreased substantially. The percentage of staff members vacating positions over the course of a year, out of the total number of positions at the NCSTU Girls Program, declined from 97% in 2008 when the model was introduced, to 17% in 2012, two years post-implementation.

In the course of theprior study, we became intrigued by descriptions of the Sanctuary Model implementation process and perceived changes in the facilitythat staff described, as well as the implicationsthis might have for implementation of organizational-level, trauma-informed models more generally. This article explores this qualitative data and asks two descriptive research questions: (1) what changed at the facility in the period between the implementation of the Sanctuary Model and two years after the facility received certification in the model and(2), in addition to the model itself, what were critical factors in the change process?

Sanctuary Model

Building on the concept of therapeutic communities, in which staff and clients collectively participate in creating a system of healing (Jones, 1953, 1968; Lees, Manning, Menzies, & Morant, 2004; Main, 1946), Bloom and her colleagues, Foderaro and Ryan, formed The Sanctuary, a program for adult trauma survivors. The Sanctuary Model, an outgrowth of The Sanctuary and co-developed with the Andrus Sanctuary Institute, is an evidence-supported blueprint for clinical and organizational change that promotes safety and recovery from adversity through the active creation of a trauma-informed community(National Child Traumatic Stress Network, 2008; Rivard, Bloom, McCorkle, & Abramovitz, 2005).

The four core elements of the Sanctuary Model include Trauma Theory and the Seven Sanctuary Commitments - namely Non-Violence, Emotional Intelligence, Democracy, Open Communication, Social Responsibility, Commitment to Social Learning, and Growth and Change. The third core element is S.E.L.F. – an acronym for the organizing categories of safety, emotion management, loss, and future, which is used to formulate plans for client services or treatment as well as for interpersonal and organizational problem solving; and lastly, the Sanctuary Tool Kit includes a set of ten practical applications of trauma theory.

Typical implementation consists of an initial 5-day training on the model for key leaders in an organization. The leaders, referred to as the Steering Committee, are then tasked with returning to their agency and forming a Core Team, a representative group of employees from all levels and departments, who are the primary change agents to work with colleagues to implement the model. The Core Team is provided technical assistance from a trained AndrusSanctuary Institute faculty during a three year implementation period. After implementation, agencies may choose to pursue “certification” which involves qualified Andrus Sanctuary Institute faculty and peer certifiers assessing the agency’s practice of the model. Full details of the model and its intermediate objectives and processes are provided in a logic model article (Esaki et al., 2013).


This study was conducted using qualitative interviews and focus groups with staff members at NCSTU. The period of focus for the study is from 2008, when the Sanctuary Model was first introduced, through 2012, when the model had been fully implemented for two years.  All human subject research protections were observed during the study which was monitored by the Institutional Review Board of the University at Albany – State University of New York.


Data were collected from the NCSTU Girls Program which is a secure facility for adjudicated delinquent females in Pennsylvania. Located in a separate building on a campus that also contains three different programs for adjudicated delinquent males, the Girls Program building is surrounded by a high fence with razor wire, internal doors are always locked, and the girls are monitored at all times and required to wear handcuffs when attending outside appointments. Girls are referred by their home counties, often after failing in other placements, and because NCSTU is state-operated the facility cannot deny admission to any referral. Their capacity is 30 girls but there are times when their census is slightly higher. NCSTU accepts girls between the ages of 13 and 21. Charges range from harassment, absconding, drug possession, and probation violation, to a host of serious, violent offenses including assault, arson, sex offenses, terroristic threats, drug trafficking, gang-related activities, firearm offenses, attempted murder, and homicide. Typical length of stay at NCSTU is 9 to 12 months but can be as long as 2 to 3 years.


In October 2012, semi-structured interviews and focus groups with a representative cross-section of staff (n = 17; 45%) that included Youth Development Aides, Aide Supervisors, Counselors, as well as clinical, community transition and nursing staff, were carried out on the NCSTU premises by two researchers. Key informant interviews included the NCSTU Campus Director, the Manager of the Girls Program, and two staff members (one Youth Development Counselor and one Unit Supervisor) who had been with the facility prior to Sanctuary Model implementation. Staff had been with the facility an average of 4.5 years, but this ranged from less than a year to 13 years. The focus group interviews were recorded and transcribed.

Data Analysis

The interview protocol asked about implementation of the model and impressions of changes in youth outcomes.  During and immediately following data collection the researchers realized that topics such as changes in employee relationships and the role of leadership were recurring in the data and were of particular interest.  Thus topics related to implementation were prioritized in the subsequent coding.  Focus group and interview transcripts were coded line by line first by one researcher, then reviewed by a second researcher, allowing for comparison to enhance reliability.  An iterative process was then used to identify sub-categories such as change in line-staff attitudes toward managers and then connections between sub-categories resulting in categories such as change in relationships between staff members generally (Bertram et al., 2011; Strauss, 1987).


            In the Results Section we first describe the process of implementing the Sanctuary Model in the NCSTU Girls Program over several years to provide a context for addressing our two research questions.  We then address Research Question 1:  What changed at the facility during the process of Sanctuary Model implementation? and Research Question 2:  In addition to the model irself, what were critical factors in the change process?

Process ofImplementation of the Sanctuary Model

Introduction of the Sanctuary Model at the NCSTU Girls Program started in 2008 at a time when, according to staff members who were there, the facility was in disarray. Two years previously the state had ended a contractfor operation of the NCSTU Girls Programwith a private agency that had experienced 9 management changes in 6 years. There weremajor safety issues with “people getting really seriously hurt.” The state decided to take over the program.However, the transition was also difficult and safety issues and other problems continued. “It was the first girls’ facility run by the state and there were a lot of hands wanting to go different ways.” There were chronic struggles with simply maintaining “structure, normalcy, and schedule.” In addition, staff on the larger campus did not want to work with the girls who were seen as “off the hook, very aggressive, very self-harming.” The organizational atmosphere wasalso “… blaming, dysfunctional, sick.” “No one was supportive of each other; every individual looked out only for themselves.” There was chronic difficulty filling staff shifts because staff would call off and go on injury leave, in turn causing burn-out for remaining staff and it was difficult to get staff to work in the Girls Program.Thus, there were major problems with the running of the organization which had triggered introduction of a new organizational model.

Introduction of the Sanctuary Model coincided with changes in organizational leadership both at the Girls Program and the entire NCSTU campus that were arguably necessary for but not intrinsic to model implementation. In 2006, the current (as of 2012) campus Director was hired first as a Consulting Director, and then as Director of the Girls Program and one of the Boys Programs. In 2008, the current Girls Program Manager was hired as a Counselor, and then promoted to Counselor Supervisor and then Manager within a six month period. Introduction to the Sanctuary Model was initiated by a state administrator. There was initial disinterest and resistance by staff to implementation of the model, which was viewed by many as “yet another training” in a long line of new initiatives. “Initially people said‘this is crazy, Sanctuary is soft.’”  However, some managerial and line staff recognized that Sanctuary might give an identity to what they were already trying to practice: “We were thinking it was too soft, but that was the philosophy we have been preaching, things like social commitment; it gave us a label, something to hang our hats on; feedback from staff was that ‘we’re already doing this’; it just gave us something to identify with.”

            Training provided by faculty from the AndrusSanctuary Institute included three to five day off-site trainings for managerial and selected personnel. In turn these staff trained other staff at the facility along with quarterly on-site trainings conducted by the AndrusSanctuary Institute. Steering and Core Teams were formed to oversee implementation. In the first year, they started by posting signs of the Seven Commitments. They also had all staff and residents make Safety Plan cards (written plans of how to handle triggers and other safety measures) and made sure that everyone wore them. Managerial and invested staff began to discuss concepts of the model while also role modelingbehavior for other staff and residents.In year two, they integrated the Sanctuary Model into the resident handbook, which new residents are expected to read and pass an orientation test on, thus introducing new girls to the concepts and expectations related to Sanctuary. The staff “got into it more” and began talking frequently about what the Sanctuary signs meant. They also developed incentives for use of the model principles through formal programs such as theSanctuary Resident and Staff of the Month. In year three, all documentation, such as treatment plans,was changed to reflect the language and concepts of the model. By 2012, residents were receiving mental health, substance abuse, trauma treatment, transition planning, and other services based on treatment plans that used the Seven Commitments and the Safety, Emotions, Loss, Future (SELF) language of Sanctuary. The SELF model was also incorporated into daily programming: “The girls are in charge of the ‘SELF Board:’ they put up different things to resemble aspects of the SELF model; it’s part of their program on day one.”  Other Sanctuary Model Tools such as Community Meetings, where the “community” of staff and girls would gather regularly, and Red Flag meetings, to defuse a potentially serious incident, were held consistently. Sanctuary was also incorporated into aftercare planning and in work with the girls’ families. 

Implementation was neither easy nor smooth: an experienced staff member on the Steering Committee noted that it was “hard to get people in the right direction and using the terminology; we had a Steering Committee meeting every other week for 2 hours to discuss everything, what’s going on, where do we want to be in 2 weeks; but would have hiccups in actually getting the stuff in and what’s the best way to do it slowly so it is not too much with one shot.”  Training, however, continued with staff both on-site and off-site, and there was continued emphasis on implementation and gaining staff buy-in. “When we started, we had had huge amounts of training [in different things over the years]. When we implemented the Sanctuary Tools was when things got better and kids and staff saw things get better. Safety cards and Community Meetings became a meaningful process. It started to come together and made sense to people and kids, and staff really started buying in.”

Research Question 1:  What Changed at the Facility During the Process of Sanctuary Model Implementation?


You need to “have basic safety and structure first, like the Maslow hierarchy of needs; if staff feel safe, they’ll carry out the message for you.” Prior to implementation of the Sanctuary Model, according to staff who were there at the time, the NCSTU Girls Program was an unsafe environment. “Before we might have 70 or 80 restraints in one month; this year (2012) in Augustwe had 0 restraints, in May we had 0 restraints.” “When I was first here it was hectic, with 9, 10, 13 restraints a day to 9 a month now.” “When I came [in early 2008] it was totally chaotic, just got to get through the day safe, very uneasy…I could see staff cringe because they were afraid; they wouldn’t come out of the staff station. There was no treatment.” Now “everything’s better, it was a rough place, but the kids started to invest and bought in and understood; not sure who came first staff or kids but somehow we all got there together;the safety feature, understanding what safety is and feeling safe because of the organization; everything got better, needs being met, kids adapting and feeling better, everybody’s able to do what we need…”“Sanctuary, our whole philosophical approach is that you don’t build relationships when you do a restraint, you build relationships when a girl sees you talking to another girl in a certain way. It makes you feel positive.”

Staff member attitudes and relationships

A fundamental change was in the attitudes of staff to their jobs, management and other staff, as well as to the residents.“People were out sick, injured --- changed to people who want to be here, people who are committed.” Before Sanctuary, many employees in the Girls Program confined their job responsibility to specific roles and tasks, were unwilling to take on other jobs, and did not necessarily follow through on daily routines; they also did not consider the impact they had on other staff, the residents and the community as a whole when they called off or took leave. Subsequent to Sanctuary Model implementation, there is shared responsibility and everyone is willing to take on other roles or tasks as necessary. For example, supervisors will cover the shifts of aides, counselors will provide transportation, and everyone will take on tasks assigned to other staff or residents as necessary. Team meetings at the beginning of each shift use open communication between staff and supervisors to review the program schedule and cover the daily responsibilities. Although full staffing of the program still remains a challenge, employees reported consistently and willingly working extra hours: “Four years ago there was a lack of staff and the quality of staff was not what it was now. Now people care more. Being on the same page, they’re coming to work because they care, not just for a pay check.”

Supervisors take care of, and promote development of line staff. “I sit down with a worker, ask how they’re doing today, we get into each other’s lives. If I need to get them a day off, I will; if I need to be an aide, willdo.”Relationships between staff and managers have become more open and democratic: “Our supervisors and managers allow us a lot of insight into how we can change things and make things better; they take our input as line staff on work with kids day to day; they are very open to us.”Relationships have also changed between direct care, clinical, and ancillary staff positions: “One of the greatest things is that the commitment to democracy created a team; it’s not residential against clinical, workers against managers.”  One clinical staff member described how the relationships that now exist in the Girls Program differ from other programs and from the past. In other programs the “clinicalstaff is perceived as touchy/feely; in the Girls Program it’s not us and them, we can talk to the aides. In other programs there is more of a separation and less approach. There is more openness in this program, everyone’s opinion matters. We can listen to everyone’s perspective and then resolve and come up with a solution.” Another staff member noted the support of open discussion between staff members in different roles. “Anybody can call a Red Flag meeting; you can be a Dietary Worker and call a Red Flag meeting. Your opinion is important and you’re encouraged to communicate openly about what you really think. There’s not a party line that you have to follow. Having all the philosophies helps you feel empowered; empowers girls but helps you feel empowered.” This development of a community has resulted in staff members understanding the effect their attitude and mood has on residents and other staff: “Your attitude will lead off to other people: if you come in smiling, it will transfer to others; so will coming in with a frown.”


The atmosphere in the facility has changed. “The tension is gone. There would be times when you walked into a Unit and felt the negative tension and thought ‘Oh God, this is going be a rough shift.’”  “In the old days, I remember sitting in my car for some time trying to get myself to get out of the car and go into the building. Now I’m very happy to go into work; the stress level has declined very noticeably.” The commitment to open communication has contributed to this change. “Now, if there’s an issue we’re going to talk about it and we’re going to shift it, which makes everyone not carry secrets, not carry things day-to-day that are going to explode. If there’s an issue we are going to talk about it; it’s part of the day.”


Not only residents, but staff members too,are held accountable to the model components. There is “no shift without a resident or staff member referring to the Seven Commitments; if staff picks up a piece of paper from the floor, that’s social responsibility.” In fact, staff members are held accountable to the model by the residents. “We have to role model. The kids know if you’re not buying in and they say something; for example, ‘you’re not using emotional intelligence.’”  “I hate to say the girls know it better than the staff, but sometimes I think they do. They’re like four year olds, they just absorb this information and they take it and they live it and I think they love it.” Girls who have been at the facility longer and are higher on the level system also mentor newer girls. “The girls come from a different culture; we work with them until they get it; some buck it, fight.”  “I’ve only seen it happen here. The residents when they first come in have heard horror stories about lock downs, no free time; they come in with their guard up. We use their mentor and high level residents to disarm all that.” Sanctuary gives the girls “something to believe in,” “a sense of belonging,” “validates them as an individual; they’re the biggest reason for their success or failure.”

Relationships with residents

The framework within which staff and residents relate has also changed. In the past there was no attempt to understand the reasons for behaviors, to listen, discuss or explain the reasons for staff decisions: the “girls even brushed an eyebrow and they hit the floor.” Now there is much more listening, discussion and explanation: “I’m the authority figure, but now I listen to why and try to explain myself, not treat them like inmates, that they’re not better than you and I’m not going to respect their thoughts.” “Not something we were used to. It’s hard to have 16 residents who all have different perspectives and [have to discuss] what are the pros and cons [of a decision]. [It requires] a lot of group time, a lot of bonding, a lot of getting together. It makes people understand why they do what they do, make the decisions they make. We have everyone talk about it and come to common ground; it eliminates conflict.”

Residents can facilitate changes in rules and policies. For example, staff members tell the story of a resident who would perspire heavily after physical activity and become concerned about her body odor. She would become upset that she could not take a shower, but facility policy was that showers were only allowed at a specific time of day. When speaking with a counselor, she started crying. In the past, employees report they would have looked at this as “being manipulative” but now are “looking at strengths and how they feel; we can bend the rule and that’s the right thing to do.” With the change in staff perspective, showers after physical activity were incorporated into the schedule.“Sanctuary has definitely impacted the girls; they have a sense of ownership, empowerment, they can express themselves. They have more of a voice; they may not like staff decisions, but they can bring up an idea and it may get implemented.”

Staff also provided examples of residents considering the effect of their behaviors on staff members and other residents. “They actually evaluate is this a good thing to bring up now; they take into consideration their surroundings. The model makes everyone think about others’ feelings.” “We have a resident who has seizures all the time and it can get really scary. I was the only staff in the room with 14 girls and she showed signs of seizure. The girls ran and got the mats and pillows; they don’t want to make things more difficult. We didn’t see this in the past.”  “We have had assaults on staff. The other girls get really mad at girls for doing that. At a Large Group Meeting, they gave the girl who assaulted a really hard time.”Sometimes the girls will use the principles against each other: “kids can use in negative ways; have to make sure that we’re not letting them use it negatively or to pick on other kids.” Staff report, however, that in general this is not a problem: “I haven’t seen a dynamic where they feel too powerful; there is a good balance between safety, empowerment, growth and change.”

            In 2010, the NCSTU Girls Program was certified by the AndrusSanctuary Institute. Since that time they have become nationally, and even internationally, renowned as a model secure juvenile justice facility for girls. They regularly host tours for visitors from all over the U.S. and beyond. Both staff members and residents say that many of the girls apply the principles of Sanctuary to their lives outside the facility: “The girls are able to translate how they apply [Sanctuary] to their futures. We had a day for presentations by the girls to other departments and programs on campus for each girl to talk about a Sanctuary Commitment. One girl talked about how she is using non-violence andis from a violent family; there was no dry eye in the house.” According to staff, the components of Sanctuary (the Seven Commitments, SELF, the Tools) and the meaning of Sanctuary are well integrated throughout the facility and beyond. Sanctuary “affects how you talk to people, how you say things, all around the building.” Staff and residents all wear Safety Plan cards all the time and the terminology of Sanctuary is reflected in all documentation regarding residents, and in employee reviews. It also extends to work with the families of residents. Some employees reported using it in their homes and in other aspects of their lives. The Seven Commitments “apply in my everyday life, not just to my job; I really live it.”    

In summary, according to the retrospective reports of staff members in 2012, there were substantive improvements in many domains at the NCSTU Girls Program between 2008 and 2012; including physical and psychological safety, staff morale, accountability and attitudes toward their work, and the relationships of staff members with administrators, other staff members and residents. The general climate at the facility changed from negative, chaotic and dangerous, to one where problems and conflicts were openly discussed and resolved, social responsibility by individuals to the community was the norm, and processes were in place to maintain safety and promote positivity. Although these changes can be linked to the Sanctuary Model principles and components, there were other environmental changes occurring at NCSTU during the same time period that are hard to separate from model implementation. Therefore, we next explore the reports by staff and administrators on changes in the make-up of leadership and staff at the NCSTU Girls Program during the implementation of the Sanctuary Model.

Question 2: In addition to the model itself, what were critical factors in the change process?


            Changes in leadership at NCSTU and the NCSTU Girls Program are inextricably linked with implementation of the Sanctuary Model. The initial impetus for Sanctuary implementation at the Girls Program came from a state level administrator at the Pennsylvania Department of Public Welfare and support at this level continued throughout the process. The Campus Director, who began working at the NCSTUGirls Program in 2007,and the Girls Program Manager,who started in 2008, had both previously worked at the same privately run youth facility in another part of Pennsylvania. They had both been trained in other models designed to improve conditions and outcomes for youth in correctional facilities and continued to use tools such as Polsky’s Diamond(see Mullen, 1999).

            In addition to prior experience in residential facilities, the Campus Director had also worked with youth in the community and had an educational background in social work and sociology. He had also played team sports for many years and ascribed to this the importance he placed on building a team. It is “so important to have the players in place when implementing [the model]….you can’t be a football player without the rest of the team; so we’re sitting down regarding employees and determining who fills the niche.” He also recognized the importance of leadership in implementing the Sanctuary Model. “Leadership is a big part of this; not so much talking about myself but about [the Girls Program manager]; he is a good salesman and you need people to sell it and you need people to buy-in.”

            The Girls Program Manager similarly described himself as a salesman. “I’m a people person, I’m a salesman; I listen to people and understand their concerns and issues and come to a solution.” He said that some of this ability was innate, some learned from his father, some learned from leadership at the facility where he had previously worked: “there were strong leaders there, but there were also things I didn’t like, and I developed my own style of leadership… I learned more of the political stuff from some of the bosses. I was always observant, wanted to learn the job ahead of me before I moved up, that’s just me, always wanted to strive to be better, to grow and change.” In the job of implementing the Sanctuary Model and turning around the Girls Program, he also felt supported by the Campus Director:  he “is very direct, gave me the power, gave me the support to say just make it run, implement Sanctuary, make it good; the most important thing to me in leadership is someone who has my back; we made it run; made it work.”

Employee Engagement

            The Girls Program Manager also saw his main focus as building a team, but there were enormous challenges. Initially, he was seen as an outsider who had moved up rapidly from counselor, and was therefore viewed negatively by some employees who had been at the facility long-term: “I was an outsider who took other people’s positions so people hated me, wanted to see me fail.” In addition, there were chronic staff shortages and the toxic atmosphere of negative staff attitudes inherited from the past. Building a working team and gaining staff investment in a new way of thinking thus required a process over a few years that included recruitment and selection out, as well as changes in the attitudes of the staff members who remained.

In 2008, there were many open positions, presenting a problem but also an opportunity: we had “so many vacancies… lots of room for growth and change.” Several staff had come to NCSTU from the private facility where the Director and Manager had previously worked, some directly to the Girls Program. As changes in leadership and safety at the Girls Program became evident, more staff came over or, in some cases, returned from the Boys Programs. At the same time, staff members who were not comfortable with the changes being made at the Girls Program selected out.This process was not easy. “The main focus was dual: we had to build the team of the building who were here already, and try to get some of the people we didn’t have. It was a weird dynamic because the people who came over were seen as favorites so [managerial] recruiting stopped after two employees who had worked at the private facility came over; but then the [new employees] kept on talking and recruiting. It was more important to work on the people who were here; one was very challenging, but once he bought in, it was very powerful.”

The Manager used a number of approaches to develop a team of staff with changed perspective including role modeling: “I role modeled what I wanted to see on the units. I didn’t ask staff to do something I wouldn’t do; I took other shifts… I just continued to push through and push through.” He also focused on a staff member who had been there long-term and was respected: “He had leadership skills. I spent a lot of time with him, coaching and developing him. It wasn’t a one person thing; I knew he had relationships with people, had leadership abilities, had worked up through the ranks. I looked to make him my right hand man.”

As the organizational culture began to shift, it was sometimes difficult to incorporate staff from other programs. A staff member who transferred to the Girls Program commented: “[I knew someone] who worked at NCSTU for 29 years. It went from small to large, more like a military type setting. It was a lot quicker to be hands on, very hands on. Coming from being trained how to do it that way, I slowly had to make changes during the years to get away from that. I could always talk to [the residents], but discipline was different. I have to spend more time talking with them, and give them more leeway as to how to do things, even threatening behavior.” Another staff member commented that “trying to change staff coming from other programs is difficult; they are coming from the old school, more structure oriented, not caring, but ‘do as a I say’ attitude.” However, over a few years there was a coalescence of a team of staff who were invested in the model and the new practices. As new hires came in, they were screened based on the Sanctuary Model: “we use situational questions like ‘if a resident is in her room crying due to a phone call, what would you do?’ We feel them out.” Once new employees were hired, they were trained in the Sanctuary Model.

Authentic commitment to the principles of the model by management and staff is essential to successful implementation. The Director of the Girls Program stated: “I am not going to ask of the kids what I wouldn’t expect of myself. My expectations are high for myself, staff and kids; if I can’t embrace the model myself, I shouldn’t ask them to.” Role modeling is practiced continuously by managers and staff members. Making the changes involved “a lot of role modeling; not sitting behind a desk, but being out on the floor, showing them how things should be done.”  Like staff members, the resident girls were initially resistant to implementation of the model. But authentic practice and active modeling of the Sanctuary Model principles made a difference: “We adhere to democracy and they have a voice; they see staff practice it, see staff cleaning up; they are more likely to step up because staff models it.”

In summary, in 2008, the NCSTU Girls Program faced many challenges, but the introduction of Sanctuary Model principles and tools combined with investment by leadership and eventually staff and resident buy-in,resulted in a change of culture in the facility.Importantly, changes occurred and were maintained in a juvenile justice context with serious levels of resident behavioral disorder and trauma.


            These findings indicate to us that the Sanctuary Model, although anecessary component, was not sufficient by itself, to bring about the changes reported at the NCSTU Girls Program.  One administrator described the Sanctuary Model as something that provided a name or identity to what they wanted and were trying to practice anyhow. Successful implementation – implementation that changed the climate, attitudes, relationships and day-to-day practices of the facility – also required the commitment and ability of leadership at several levels to sell the model to staff and make concomitant changes in every aspect of organizational function.  Successful implementation also required staff investment in the model and the changes it called for. This, in turn, required a long-term process of staff recruitment and selection in, and out, that resulted in genuine staff commitment to the theory and principles of the model.

These findings are consistent with research suggesting that there are a number of aspects to solidifying real world change in practicesto address trauma amidthe complexity of juvenile justice settings. Changes that were made over the span of the study consistent with the Sanctuary Model shifted the culture of the organization in the direction of less traumatizing practices. Examples of these changes include specific practices consistent with a trauma informed model such as more respectful attitudes, increased communication, and more supportive relationships between staff members with each other and with residents. In addition, concurrent changes in leadership, facilitated buy in of other staff and subsequent employee engagement permitted changes in perspective on the part of staff that undergirded the potential for more therapeutic and democratic responses to residents.

Findings are also consistent with literature on fostering organizational change in general.  For example, illustrations in the qualitative data above of ‘exemplary leadership’ practices includingrole modeling and inspiring others are consistent with literature that identifies the critical role of leaders in fostering organizational change(Kouzes & Posner, 1995). Employee acceptance of change is affected by characteristics of the change process (Dent & Goldberg, 1999; Oreg, 2006). For change efforts to succeed, employees must have confidence in the reliability and integrity of management and accept management’s vision (Li, 2005). Change leaders have to understand how their employees perceive change and ensure they are ready for, and accept the change (Al-Haddad & Kotnour, 2015). Employee acceptance of change and investment is linked in several vignettes presented above,with confidence in management and management’s vision, and with team building practices. There are also echoes of literature suggesting that positive orientation toward a change agent influences employees’ responses to change (Oreg & Sverdlik, 2011). Worker perceptions have been identified as “integral components of organizational social context”(Patterson Silver Wolf, Dulmus, Maguin, Keesler, & Powell, 2014, p. 215).

The growing field of implementation science (NIRN; Fixsen et al., 2009)has also pointed to the importance of evaluating not only key intervention components, but implementation issues such as organizational context, organizational readiness, and facilitative administration. Research on the implementation process is key to identifying mediating mechanisms and pathways to more successful client outcomes, but is still undeveloped particularly in the juvenile justice field (Dadich et al., 2015). Increasing the uptake of evidence-based practices will also depend on implementation interventions that focus on improving climate and culture.


There are significant limitations in the ability of these data to speak both to the effectiveness of the Sanctuary Model at NCSTU, or to provide empirical support for our conclusions regarding the importance of leadership and staff investment. The data are basically descriptive and retrospective, and although almost half of the NCSTU staff members were interviewed, they are not necessarily representative of all staff members.Also in part because this study used data that was collected originally to address somewhat different questions, we did not have sufficient information on the power dynamics and complex interactions involved in organizational level change in a juvenile justice facility. In addition, generalizations from this one small girls’ facility to the larger residential juvenile population, and even to other girls’ corrections facilities, may not be appropriate


It is important to keep in mind however, that interventions that work well under controlled conditions, but are very difficult to implement in the real world, have little to offer to real world problems (Anderson, Lindsay, & Bushman, 1999). Findings as a whole point to positive changes in a number of aspects of institutional, staff and resident functioning. These qualitative data enrich our understanding of the ground-level process of implementation of a trauma-informed model in a juvenile justice facility, and give voice to the lived experience of staff members at varying levels and in varying roles. Their experience points to the need to explicitly focus on the process and realization of organizational change when implementing new intervention models. Hopefully these findings will take the field in the direction of further research, and ultimately promote more healing environments and post-institutional success for residents of juvenile justice facilities.



Aarons, G. A. (2006). Transformational and transactional leadership: association with attitudes toward evidence-based practice. Psychiatric Services, 57(8), 1162-1169.

Al-Haddad, S., & Kotnour, T. (2015). Integrating the organizational change literature: a model for successful change. Journal of Organizational Change Management, 28(2), 234-262. doi: doi:10.1108/JOCM-11-2013-0215

Anderson, C. A., Lindsay, J. J., & Bushman, B. J. (1999). Research in the Psychological Laboratory: Truth or Triviality? Current Directions in Psychological Science 8(1), 3-9.

Bertram, R., Suter, J., Bruns, E., & O'Rourke, K. (2011). Implementation Research and Wraparound Literature: Building a Research Agenda. Journal of Child & Family Studies, 20(6), 713-725. doi: 10.1007/s10826-010-9430-3

Bettelheim, B. (1974). A home for the heart. New York: Knopf.

Cox, A. (2013). Juvenile facility staff responses to organizational change  Retrieved July 14, 2015, from

Dadich, A., Stout, B., & Hosseinzadeh, H. (2015). Reacting to and managing change within Juvenile justice. Journal of Organizational Change Management, 28(2), 315-328. doi: 10.1108/jocm-11-2013-0228

Dent, E. B., & Goldberg, S. G. (1999). Challenging “resistance to change”. Journal of Applied Behavioral Science, 35(1), 25-41. doi: 10.1177/0021886399351003

Elwyn, L. J., Esaki, N., & Smith, C. A. (2015). Safety at a girls secure juvenile justice facility. Therapeutic Communities: The International Journal of Therapeutic Communities, 36(4), 209-218. doi: doi:10.1108/TC-11-2014-0038

Esaki, N., Benamati, J., Yanosy, S., Middleton, J., Hopson, L. M., Hummer, V., & Bloom, S. L. (2013). The Sanctuary Model®: Theoretical Framework  Families in Society: The Journal of Contemporary Social Work  94(2), 87-95.

Fixsen, D. L., Blase, K. A., Naoom, S. F., & Wallace, F. (2009). Core Implementation Components. Research on Social Work Practice, 19(5), 531-540.

Ford, J., & Blaustein, M. (2013). Systemic Self-Regulation: A Framework for Trauma-Informed Services in Residential Juvenile Justice Programs. Journal of Family Violence, 28(7), 665-677. doi: 10.1007/s10896-013-9538-5

Ford, J. D., & Hawke, J. (2012). Trauma Affect Regulation Psychoeducation Group and Milieu Intervention Outcomes in Juvenile Detention Facilities. Journal of Aggression, Maltreatment & Trauma, 21(4), 365-384. doi: 10.1080/10926771.2012.673538

Griffin, G., Germain, E. J., & Wilkerson, R. G. (2012). Using a Trauma-Informed Approach in Juvenile Justice Institutions. Journal of Child & Adolescent Trauma, 5(3), 271-283. doi: 10.1080/19361521.2012.697100

Hennessey, M., Ford, J. D., Mahoney, K., Ko, S. J., & Siegfried, C. B. (2004). Trauma Among Girls in the Juvenile Justice System. Retrieved from

Hodgdon, H., Kinniburgh, K., Gabowitz, D., Blaustein, M., & Spinazzola, J. (2013). Development and Implementation of Trauma-Informed Programming in Youth Residential Treatment Centers Using the ARC Framework. Journal of Family Violence, 28(7), 679-692. doi: 10.1007/s10896-013-9531-z

Holman, B., & Ziedenberg, J. (2006). The Dangers of Detention: The Impact of Incarcerating Youth in Detention and Other Secure Facilities. Retrieved from

Jones, M. (1953). The Therapeutic Community: A New Treatment Method in Psychiatry. New York: Basic Books.

Jones, M. (1968). Beyond the Therapeutic Community: Social Learning and Social Psychiatry. New Haven, CT: Yale University Press.

Kouzes, J. M., & Posner, B. Z. (1995). The leadership challenge: how to keep getting extraordinary things done in organizations (2nd ed.): Jossey-Bass.

Lees, J., Manning, N., Menzies, D., & Morant, N. (2004). A Culture of Enquiry: Research Evidence and the Therapeutic Community. London: Jessica Kingsley.

Li, L. (2005). The effects of trust and shared vision on inward knowledge transfer in subsidiaries' intra- and inter-organizational relationships. International Business Review, 14(1), 77-95. doi: 10.1016/j.ibusrev.2004.12.005

Lipsey, M. W. (2009). The Primary Factors that Characterize Effective Interventions with Juvenile Offenders: A Meta-Analytic Overview. Victims & Offenders, 4(2), 124-147. doi: 10.1080/15564880802612573

Main, T. F. (1946). The hospital as a therapeutic institution Bulletin Of The Menninger Clinic (Vol. 10, pp. 66-70). New York, NY: Guilford

Mullen, J. (1999). Working in a diamond mind: Polsky's insight on youth subcultures. Reclaiming Children & Youth, 8(3), 169-180.

National Child Traumatic Stress Network. (2008). Trauma-Informed Interventions. Sanctuary Model: General Information, from

Oreg, S. (2006). Personality, context, and resistance to organizational change. European Journal of Work & Organizational Psychology, 15(1), 73-101. doi: 10.1080/13594320500451247

Oreg, S., & Sverdlik, N. (2011). Ambivalence toward imposed change: The conflict between dispositional resistance to change and the orientation toward the change agent. Journal of Applied Psychology, 96(2), 337-349. doi: 10.1037/a0021100

Patterson Silver Wolf, D., Dulmus, C., Maguin, E., Keesler, J., & Powell, B. (2014). Organizational Leaders' and Staff Members' Appraisals of Their Work Environment Within a Children's Social Service System. Human Service Organizations: Management, Leadership & Governance, 38(3), 215-227. doi: 10.1080/23303131.2014.884032

Polsky, H. W. (1963). Cottage six: the social system of delinquent boys in residential treatment. New York: Russell Sage.

Rivard, J. C., Bloom, S. L., McCorkle, D., & Abramovitz, R. (2005). Preliminary results of a study examining the implementation and effects of a trauma recovery framework for youths in residential treatment. Therapeutic Community: The International Journal for Therapeutic and Supportive Organizations, 26(1), 83-96.

Saar, M. S., Epstein, R., Rosenthal, L., & Vafa, Y. (2015). The sexual abuse to prison pipeline: The girls' story. Washington DC: Georgetown Law Center on Poverty and Inequality.

Sedlak, A. J., & McPherson, K. S. (2010) Juvenile Justice Bulletin: Conditions of confinement: Findings from the survey of youth in residential placement. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.

Siegel, L. J., & Welsh, B. C. (2011). Juvenile Delinquency: Theory, Practice, and Law (11th ed.). Belmont, CA: Wadsworth.

Skowyra, K. R., & Cocozza, J. J. (2007). A Blueprint for Change: Improving the System Response to Youth with Mental Health Needs Involved with the Juvenile Justice System. Retrieved from

Strauss, A. L. (1987). Qualitative analysis for social scientists. United States of America: Cambridge University Press.