A GROUNDED THEORY ANALYSIS OF SECLUSION OF STUDENTS WITH DISABILITIES IN SCHOOLS

In the last several years there has been growing concern about the use of restraint and seclusion in school. Little is known about the use of seclusion in schools, particularly with students with disabilities. The purpose of this study was to discover: under what conditions are students with disabilities subject to seclusion in schools? A grounded theory analysis was done of 26 due process hearings containing a complaint about the use of seclusion. The Individuals with Disabilities Education Act provided the framework and the relation between IDEA and seclusion was included in the analysis. For the purpose of this study, seclusion is defined as “the involuntary confinement of a student alone in a room or area from which the student is physically prevented from leaving” (Council of Children with Behavioral Disorders [CCBD], 2009, p. 1). The study identified seven conditions that lead to the seclusion interaction that were analyzed and mapped using Strauss and Corbin’s conditional matrix (1990). These conditions are, in order of lower to higher levels of significance and proximity to the seclusion event: a loose legal boundary, expert recommendation, a special education setting, manifestation of disability, ineffective behavior plan, negative connotation of disability, and the rationale for the seclusion event. The seclusion event is characterized by the rooms used for seclusion, the terms used to define it, the frequency and duration of the seclusion event, the exit criteria, and physical and mental health contraindications. The seclusion event resulted in several short and long term outcomes for students, including a placement trajectory that moved students into more restrictive settings.

appeared to be created for and used by students: the ceiling was painted to resemble a blue sky with white clouds, the floor was painted a bright green and a rug had was taped down on it. Personal items were strewn on the rug, including some articles of clothing, pillows and blankets. The videotape shows parts of the wall where patches of paint appear to have been picked away ("Room No. 20 Controversy," 2008).
What made this room exceptional, and presumably the reason the anonymous tape had been distributed, was the fact that the door knob had been removed and two locks were on the outside portion of the door. The principal of the school, the special education director (who also functioned as the school superintendent), teachers, school committee members, the previous superintendent, and Bradley Hospital, alleged to have been consulted in the construction of the room-all denied knowledge of the room or refused to speak to the media. Some referred to the room as a special education issue and declined to speak for confidentiality reasons (Mulvaney, 2008).
The existence of a seclusion room, in the smallest state in the union, on an island in a school district with 146 students at the time (http://www.ride.ri.gov), begs the question, is this an isolated case? Anecdotal evidence suggests it is not. Recently, there have been several stories about seclusion of students in the national news (Blacker, 2012;Downey, 2010;Fowler, 2012;Hefling, 2012;Kaplan, 2010;Lichtenstein, 2012;Richards, 2012;Turque, 2011). In addition to concern in the media, there has also been increased federal attention on the use of restraint and seclusion in schools. The Government Accountability Office investigated cases of abuse and deaths caused by restraint and seclusion in schools and treatment centers (GAO, 2009) The popular media accounts are harrowing and the call to reduce and eliminate student seclusion has been strong but perhaps ineffective without understanding the nature of seclusion of students with disabilities in schools. Within the psychiatric community the practice of seclusion has proven to be a historically persistent problem (Arthur, 2008;Tovino, 2007), and without a clear understanding it may prove to be more difficult to regulate in schools.
I became interested in the topic of seclusion of students with disabilities through experiencing it first hand as a teacher of students with emotional and behavioral disorders and witnessing it in schools that I have worked in. I wondered about how common the practice is, noticed it was seldom discussed within schools or with parents, and questioned if it may result in a denial of a free and appropriate education, the legal guarantee of the Individuals with Disabilities Education Act. As I began to research this issue, it became clear to me that the discourse about restraint and seclusion is predominantly about restraint, and that while seclusion may not be physically harmful it can certainly potentially be psychologically harmful for children (Ferleger, 2008;Finke, 2001;Westling, Trader, Smith, & Marshall, 2010).
This study seeks to provide a foundation for understanding seclusion in schools by addressing the following question: under what conditions are students with disabilities subject to seclusion in schools? environmental change. It is important to note that, in the applied behavior analysis model, the environmental change effected is the removal of the opportunity for reinforcement. In some literature this is stressed by the term "time-out from reinforcement", instead of the abbreviated "time-out" and in this definition the concept of "time-in" is as important (if not more so) than "time-out" (Alberto & Troutman, 2003;Brantner & Doherty, 1983;Ryan, Sanders, Katsiyannis, & Yell, 2007).
Therefore, if the setting is undesirable, timeout can also act as a reinforcer, helping a student to escape and thus increasing the likelihood of an undesirable behavior (Brantner & Doherty;Gresham, 1979;Yell, 1994).

Definition
Timeout has evolved into an educational practice with four distinct levels on a continuum from least to most restrictive: inclusion timeout, exclusion timeout, seclusion timeout and restrained timeout (Ryan, Sanders, et al., 2007). Inclusion timeout, the least restrictive on the continuum, is defined as the contingent removal of a student from an activity for a given period of time while remaining in the classroom (Ryan, Sanders, et al.). Examples of this would include: planned ignoring (the removal of teacher attention for a brief period of time), removal of reinforcing objects or materials, and contingent observation, during which the student is removed to another location in the classroom and allowed to watch but not participate in the activity (Yell, 1994). There are two benefits of inclusion timeout. First, it is the least restrictive form of timeout. Second, the student is still exposed to instruction and the modeling of appropriate classroom behavior.
Exclusion timeout, next on the continuum from least to most restrictive, removes the student from the activity and the opportunity for reinforcement by requiring the student to enter into an area specifically designed for time-out (Yell, 1994). Some examples of exclusion timeout would be sending a student to another teacher's classroom or into the hall, or sending the student to the principal's office.
The next, and more restrictive form of timeout is seclusion timeout, most commonly known as seclusion. The Council for Children with Behavior Disorders defines seclusion as "the involuntary confinement of a student alone in a room or area from which the student is physically prevented from leaving" (CCBD, 2009, p. 1). The definition of seclusion in schools may hinge on the concept of involuntary confinement: most definitions of seclusion in schools reference it (Ferleger, 2008;Jones & Feder, 2009;Ryan, Peterson, Tetreault & Van Der Hagan, 2007;Ryan, Sanders, et al., 2007). Only one definition found in the literature links seclusion to applied behavior analysis procedures, defining seclusion as a procedure that "calls for the physical separation of a student into another room or area so that no positive reinforcement may occur following an undesirable behavior" (Westling et al., 2010, p. 117).
The CCBD definition of seclusion makes several important distinctions. First, seclusion may or may not occur with the use of physical restraint. Second, seclusion requires the physical prevention of a student from leaving a room or area, not the perception that a student is confined within a school, such as in a detention room or in school suspension setting. Third, seclusion does not include situations in which students willfully choose to isolate themselves and are able to re-enter the classroom when ready. Last, CCBD emphasizes that, "Any time a student is involuntarily alone in a room and prevented from leaving should be considered seclusion regardless of the intended purpose or the name applied to this procedure or the name of the place where the student is secluded" (p. 1). This is an important point, as the practice of seclusion can be masked with many different terms and euphemisms. Within the literature, the practice of seclusion appears under terms such as: think time and cool down (Busch & Shore, 2000); contingent observation (Yell, 1994); the cooling down period (Yell, 1990); and isolation, confinement, extended time out, time in, time away, alone time, separation, remote location, extended quiet time, taking a break, and exclusion (TASH, 2011). The practice of seclusion is often confounded with the physical space in which it occurs, such as: isolation room or cool down room ; comfort room, quiet room, timeout room (Ryan, Sanders, et al., 2007); soft room or freedom room (Day, 2002); control room, ego room, green room (Endres & Goke, 1973); and more derogatory terms by individuals subject to seclusion: the cave (Hollowell, 2009); cooler, lock-up and looney room (Endres & Goke).
Currently, there is no federal definition of seclusion in school settings (Jones & Feder, 2009). Proposed federal legislation, the Keeping All Students Safe Act, defines seclusion narrowly, as "a behavior control technique involving locked isolation. Such a term does not include a time out." Timeout is defined as "a behavior management technique that is part of an approved treatment program and may involve the separation of the resident from the group in a non-locked setting, for the purpose of calming. Timeout is not seclusion" (Public Health Service Act, 42 U.S.C., 290 jj(d)(4).
Therefore, involuntary confinement that does not utilize a lock would not be considered seclusion under this proposed definition, countering the CCBD definition.
The last, and most restrictive form of timeout is restrained timeout, which is when a student is physically restrained in a timeout room or area.

Justification
Time-out procedures have been demonstrated to be effective in reducing inappropriate behaviors across a wide range of student populations as well as in a variety of settings (Brantner & Doherty, 1983;Ryan, Sanders, et al., 2007). This may be part of the reason that time-out has become an extremely popular tool, "flourishing", according to Brantner and Doherty. In 1980, Piersma documented rationales for the use of seclusion as stated in state laws which include, from least to most frequently: as the result of a safety concern; for therapeutic purposes; as part of a behavior modification program; as a result of property damage; to encourage patients in their recovery; due to an attempted suicide; to set limits; due to a danger of elopement; to help a client gain self-control; to help to decrease sensory stimulation and as per physician's order. In the past there has been some theoretical basis for seclusion, such as Redl and Winemans's psychodynamic theory, Cotton's ego deficient theory and Zaslow's attachment theory (Kennedy & Mohr, 2001). Currently, the theory that seclusion serves a therapeutic purpose has been discredited (Ferleger, 2008), however, as Maden points out "seclusion persists in many forensic units because we are reluctant to abandon any intervention, however distasteful and archaic, which is effective as a last resort in the prevention of violence" (Maden, 1999, p. 244). Gutheil (1984) too, in a review of studies found that seclusion is used "principally to contain violence and thus serve a legitimate and irreplaceable purpose on the modern inpatient ward" (p. 137).

Outside of Schools
Seclusion has a lengthy and controversial history, tightly bound to different cultural perceptions of mental illness and influenced by changing opinions about freedom and personal liberty. These constructs and ensuing debates about the efficacy of seclusion have undergone several distinct shifts throughout the historical treatment of people with mental illness, identified by Tovino (2007) as four different causes and correlating types of treatment: physical care, moral treatment, custodial care and community mental health care. It should be noted that the term restraint is the term most often used in historical literature, which appears to refer to both restraint (most often by chains and shackles to a wall), and/or restraint via solitary confinement, and there seems to be some overlap in terms (Soloff, 1984).
The first treatment model of people with mental illness is physical: an attribution of a physical cause for mental illness within the brain and nervous system leading to the use of seclusion, restraints, and the infliction of harsh physical treatments intended to shock the patient back into normal behavior (Tovino, 2007). During the 18 th century, the "therapeutic intent was external control of the patient's will by harsh discipline and constraint" (Soloff, 1984, p. 3). At this time seclusion was thought to be an effective means of calming severely agitated individuals. In 1799, in the first American hospital created for the mentally ill in Williamsburg, Virginia, two underground cells were located under the first floor of the hospital. Patients were secluded in these rooms as well as other facilities for months, years, and some even on a permanent basis (Tovino). In 1815 there was public outrage when the confinement of William Norris became public. Norris was a 55 year old inmate of Bethlem Hospital in London (the origin of the word "bedlam") who was confined in a metal harness to a wall for 14 years. The outrage led to Parliamentary inquiry and eventually the Lunatic Asylum Act in 1842. This act created the Commisioners in Lunacy to supervise the custodial care of the insane and initiated the first reformation in the treatment of the insane: moral treatment (Soloff).
Moral treatment was guided by a belief that people with mental illness are capable of being restored to typical behavior via a soothing and supportive environment, kind treatment and by example. It was a system of "total care" defined by the Commissioners in Lunacy as "all those means which, by operating on the feelings and habits, exert a salutary influence, and tend to restore them to a sound and natural state" (Soloff, 1984, p. 4). The goals of treatment of people suffering from mental illness changed from harsh disciplinary tactics to "internalization of moral standards and self-control" and the beginning of psychological methods of treatment for mental, as opposed to physical derangement (Soloff,p. 3). This stage occurred within the backdrop of enlightenment in the 18 th century and was populated by influential reformers such as Philippe Pinel of France, the Tukes of England, and in America John Conolly, who demonstrated the efficacy of nonrestraint on a large scale at the Hanwell Asylum in 1839 (Suzuki, 1995).
Within the context of humane care and education, attention began to be paid to the practice of restraint and seclusion, leading to initiation of restraint reduction studies (Soloff, 1984). However, seclusion was never completely eliminated and was still used as a therapeutic measure for patients with out of control behavior and as a form of punishment, (Soloff). Many physicians maintained that seclusion was an effective practice: Seclusion is found to have a very powerful effect in tranquilizing, and subduing those who are under temporary excitement or paroxysms of violent insanity. As a temporary remedy, for very short periods, in case of paroxysms and of high excitement, we believe seclusion to be a valuable remedy (Soloff,p. 6).
In fact, a statement made at the first meeting of what was to become the American Psychiatric Association "resolved that it is the unanimous sense of this convention, that the attempt to abandon entirely the use of all means of personal restraint is not sanctioned by the true interests of the insane" (Soloff, 1984, p. 7).
Over time, the institutions practicing moral treatment became overwhelmed with patients, and care became less focused on treatment and therapy and more custodial (Tovino, 2007). Restraint and seclusion began to be used for convenience and in lieu of supervision. Once again, there was an expose in 1950, when the Kansas City Star published a series of articles on the conditions at Osawatomie State Hospital in Kansas. A good example of the evolution of moral treatment to custodial care, this hospital opened with 12 beds in 1866, by 1945 it had a doctor to patient ratio of 1:845.
In order to manage the overwhelming patient needs, up to half of the patients in the hospital were placed in restraints at any given time (Tovino,p. 525).
Eventually, due to what Tovino (2007) identifies as a combination of forces, institutions gave way to the current philosophy of community health care, in which patients are treated in outpatient centers, halfway houses, and community centers.
However, the practice of restraint and seclusion persists, leading to yet another expose in the Hartford Courant in 1998, which found 142 reported deaths over a ten year period in residential psychiatric facilities nationwide, with the likelihood that a the actual number of deaths was higher, estimated at 500 to 1,500 over the 10 year period (Appelbaum, 1999;Weiss, 1998 (Ryan, Peterson, et al., 2008, p. 204). As a result, The Children's Health Act was passed in 2000, which prohibits the use of restraint and seclusion for discipline or convenience (Kaplan, 2010, p. 585). It does not apply to schools.
In conclusion, the use of seclusion has existed and persisted throughout psychiatric history through several different phases and for several different purposes.
It has proven to be controversial, often leading to public outcry and legislation, which has never successfully eliminated it from practice.

Inside of Schools
The practice of seclusion in schools is rooted in the use of timeout. Within schools, timeout has long been present in different forms: from being sent home or to the corner for misbehavior, to the behaviorist movement in the 1960s and 1970s and the creation of timeout rooms and booths (Ryan et al., 2008).  state "timeout is a behavior management procedure that has long been used in the field of education to address a broad range of maladaptive behaviors across educational placement settings" (p. 7).
Seclusion rooms became common in special education settings in the 1970s and 1980s, migrating along with students with emotional and behavioral disorders who were now afforded a public school education via the Education for All Handicapped Children Act of 1975 (P.L. 94-142), the federal statute guaranteeing a public school education to students with disabilities. Unfortunately, the policies and practices regarding seclusion from the mental health field did not travel with the new student population (Arthur, 2008;. At the time, seclusion was considered an acceptable therapeutic practice for children who were considered "ego deficient" (Cotton, 1989) and students were placed in everything from refrigerator boxes, to closets, to time-out booths, to empty rooms (Ryan et al., 2008).
The practice of seclusion led to serious problems such as injuries and suicides, leading to a call for guidelines in their use (Ryan et al, 2008). As early as two years after the passage of the Education for All Handicapped Children Act of 1975 (P.L. 94-142), Gast and Nelson (1977) were appealing for "legal and ethical considerations for the use of timeout in special education settings", stating that "timeout from positive reinforcement has become one of the most frequently used strategies by teachers for suppressing disruptive behavior" (p. 457). The same article included guidelines for the physical structure of timeout rooms in schools, including the size of the space, staff monitoring, and specification that the door should not have a locking mechanism but should have a latch to be used as needed. Soon after, other less aversive behavioral strategies were investigated as potential alternatives to timeout, such as response cost (Gresham, 1979). In 1986, Alberto and Troutman in their manual "Applied Behavior Analysis for Teachers" included seclusionary time out as a level 3 (out of possible 4, on continuum from least to most restrictive and aversive procedures) punishment procedure, intended to arrange consequences that decrease a behavior. It notes that timeout rooms have been misused in the past, and expresses concern regarding the duration of time students are placed in timeout rooms as well as the physical structure of the rooms used for timeout. To address these concerns, the authors list recommendations for the physical structure of the timeout room and guidelines for the use and monitoring of seclusion. The content and cited studies of these guidelines remains the same in more recent editions of this manual, although the newer manual clarifies "this procedure is usually reserved for behaviors such as physical aggression, verbal aggression, and destruction of property" (Alberto & Troutman, 2003, p. 371).

The Practice of Seclusion in Schools
In schools, seclusion occurs for a variety of reasons (CCBD, 2009;Ryan, Peterson, & Rozalski, 2007). It may operate independently of intended rationale or theory (Ryan et al., 2008) and often takes place in the absence of behavior intervention plans (Council of Parent Attorneys and Advocates [COPAA], 2009;Westling et al., 2010).
It appears that, in practice, seclusion often occurs for at least three primary reasons. First, as a punishment procedure intended to decrease a target behavior (Ferleger, 2008;Gast & Nelson, 1977;Yell, 1994). Second, as a therapeutic modality providing appropriate limit setting or decreased stimulation from sensory overload (Busch & Shore, 2000;CCBD, 2009). Finally, in an emergency situation in which the student has lost control behaviorally and is in imminent danger of hurting him or herself or other people, the student may be placed in seclusion until they regain control (Ferleger).
Other functions of seclusion have been identified, such as restoring order to a classroom via the removal of a disruptive student and providing teachers respite from the same (CCBD, 2009), and time to engage in reflection or problem solving or time to "cool down" (Ryan et al, 2008). Teachers also may use seclusion for a variety of purposes, or without a clear purpose in mind (Ryan, Sanders, et al., 2007).  found that teachers reported using seclusion for several nonemergency reasons such as leaving assigned area, noncompliance, disrupting class, misuse of property, disrespect, harassment, and threats. There is widespread consensus and concern that despite the restrictiveness of seclusion and its potential for abuse it is commonly used in schools (CCBD; Gast & Nelson, 1977;Persi & Pasquali, 1999;Yell, 1994).
There is slightly more research available on seclusion in adult and child psychiatric literature, however, there is sparse research on seclusion with children (Allen, 2000). This could be due to the fact that mental health professionals are reluctant to acknowledge and address the practice of seclusion, which Piersma (1980) labeled "the unwanted and undiscussed child of mental health care" (p. 1). Often, restraint and seclusion are confounded with each other, despite the fact that they are widely different interventions, making it difficult to make conclusions about seclusion in and of itself (Beck et al., 2008;Busch & Shore, 2000;De Benedictis et al., 2011). It has also been noted that restraint garners more attention than seclusion (Finke, 2001).
The next section will present an overview of the literature on seclusion, including adults and children in psychiatric settings, as well as seclusion of students in schools.

Literature on Seclusion Seclusion Research with Adults
Much of the literature on the use of seclusion with adults is grounded in the mental health field, is descriptive instead of empirical, and is focused on patients' and caregivers' perspectives on seclusion. Less research has been done on seclusion rates, reducing seclusion, and the effects of seclusion. There is also a small body of literature on anecdotal stories about seclusion and death rates. Little to no empirical research has been done on causes of seclusion, first noted by Gutheil (1984). Moreover, many studies do not differentiate between restraint and seclusion, making it hard to uncover information about seclusion.
Perspectives. Patient perspectives on the use of seclusion are widely negative (Hoekstra, Lendemeijer, & Jansen, 2004;Ray & Myers, 1996;Van Der Merwe, Muir-Cochrane, Jones, Tziggili, & Bowers, 2012), ranging from counterintuitive, unfair, not least restrictive, not compliant with rules, and abusive. As a result of these conditions, patients report the presence of post-traumatic stress disorders. In fact, there is a body of research referred to as "survivor literature" (Amos, 2004;Tovino, 2007). There is minimal evidence on positive effects of seclusion. One study found the ratio of negative to positive statements made by patients about seclusion to be 12:1 (Ray & Myers). The positive effects of seclusion are typically found in staff perceptions regarding seclusion, with staff viewing seclusion as a positive procedure (Allen, 2000) or a necessary tool for running a psychiatric ward (Van Der Merwe et al.).
The use of seclusion has been reported by patients as counterintuitive to treatment, with effects such as feelings of powerlessness, erosion of trust, humiliation, fear and loneliness (Hoekstra et al., 2004). Patients report that they have been subject to the unfair use seclusion in lieu of less restrictive measures, which appears to have a significant negative impact on the experience of seclusion. Patients overwhelmingly report being subject to restraint and seclusion that may be premature, unnecessary, punitive, or the result of provocation from staff (Ray & Myers, 1996). Once a seclusion event is initiated, patients report being ignored, not being allowed to use the bathroom, being very cold in the seclusion room, or being in seclusion for too long (Ray & Myers). Patients may experience feelings or paranoia and hallucinations in seclusion, even when secluded for brief periods of time, especially if already prone to hallucinations (Mason & Brady, 2009), adding to feelings of terror and prolonging behavioral upset (Hoekstra et al.).
It appears that the interaction between staff and patient before and during the seclusion event impacts the patient's experience in positive or negative ways. If patients understand the rationale for the seclusion event they are less likely to view it as a negative experience. (Hoekstra et al. 2004;Ray & Myers, 1996;Veltkamp, et al., 2008). Conversely, patients also report unequal treatment from staff, resulting in more negative outcomes as well as impairment of their ability to process or recover from the seclusion incident (Hoekstra et al.). Negative outcomes range from "sanctuary harm" (a sanctuary experience perceived as harmful) to outright abuse (Ray & Myers;Robins, Sauvageot, Cusack, Suffoletta-Maierle, & Frueh, 2005). In several studies patients reported that the use of seclusion was not compliant with rules and was, at times, abusive. In one study, 78% of patients reported that their care and treatment while being restrained or secluded was not compliant with at least one standard specified in New York state mental hygiene rules and regulations, including not being released every 2 hours, not being permitted to use the bathroom or eat and drink at mealtimes, not being checked by staff every 30 minutes and not being examined by a physician (Ray & Myers). Studies have also uncovered allegations of abuse and injury: including the use of unnecessary force, psychological abuse, physical abuse, physical injuries, and sexual abuse (Ray & Myers; Robins et al.). A significant portion of patients report being subject to childhood sexual and/or physical abuse compounding the negative experience of seclusion Robins et al.).
Due to these factors, in direct contradiction to the therapeutic intent of mental health treatment, some patients' experiences meet the criteria for a traumatic event in that, "the person experienced, witnessed, or was confronted with an event or events that involved actual or threatened death or serious injury or a threat to the physical integrity of self or others, and the person's response involved fear, helplessness, or horror" (Robins et al, 2005(Robins et al, , p. 1134. Frequently, patients report symptoms of posttraumatic stress resulting from having been subjected to seclusion Hoekstra et al., 2004;Robins et al.). These include fear of enclosed spaces and feeling retraumatized when witnessing other patients being secluded. Symptoms of distress may persist long after the seclusion event occurs (Frueh et al.; Hoekstra et al.).
In contrast, staff do not appear to correctly perceive the experiences of patients in seclusion (Allen, 2000). Due to a lack of alternatives when attempting to control violent behaviors (Happell & Koehn, 2010) staff view it as a "necessary evil" and a critical tool for maintaining safety (Allen; Happell & Koehn). Endres and Goke (1973) found a difference in attitudes towards seclusion between experienced and inexperienced staff, with experienced staff feeling more comfortable using the room.
This is the largest body of research in an already under-researched area, and it has some limitations. These include the potential distortion of reality present when patients with mental illness discuss seclusion events, and the accuracy of reporting on events that occurred long ago (Robins et al., 2005). Additionally, staff perspectives are self-reported, with the likelihood that staff are not likely to admit errors in judgment or practice.

Scope of seclusion.
It is difficult to establish the scope of seclusion use, including demographic information. Research in this area is sparse, relying on survey, case studies and record reviews. Seclusion rates appear to be highly variable, and highly dependent on the environment (Gutheil, 1984). Several studies found up to half of a given population of patients being subject to seclusion Ray & Myers, 1996).
Demographics. Again, the lack of research in this area makes it difficult to form broad generalizations, however, one study points towards younger, more acutely ill men, including immigrants, being subject to higher rates of seclusion (Knutzen et al., 2011). In a record review, Beck et al. (2008) found three highly discrete trajectories of patients subject to seclusion, a low, medium, and high trajectory class, with the high-trajectory class of patients presenting with significant unsafe behaviors.
As the previously mentioned study found, patients falling into the medium and high trajectories tended to be younger, however, a difference was noted in that women were over-represented in the high trajectory class. looked at staff related and organizational predictors of the use or restraint and seclusion from the perspective of care providers, finding that team climate, staff perceptions of aggression, and organizational factors were associated with greater use of seclusion and restraint, perhaps more than individual patient characteristics. It appears that a combination of factors, including patient and staff characteristics as well as setting factors may impact seclusion rates.
Another area of research is focused on reduction of the use of seclusion and restraint (Ashcraft & Anthony, 2008;CWLA, 2004;Donat, 2002;Ferleger, 2008;Gutheil & Tardiff, 1984;Huckshorn, 2008). Additionally, guidelines have been recommended for the use of seclusion (Gair, 1984;Huckshorn;Kuehnel & Slama, 1984;Lion & Soloff, 1984;Miller, 2011;Stewart, 2010;Stokowski, 2007;Wexler, 1984 an intervention across three acute inpatient psychiatric units found that while seclusion and restraint were reduced by more than half in a 12 month period, assaults on staff increased as did assaults on other patients. Another way to reduce the use of seclusion is through legislation, although legislation without local environmental changes does not seem to be effective (Keski-Valkama et al., 2007).

Seclusion Research with Children
The medical literature on children subject to seclusion in psychiatric settings follows much of the same path as research done with adults. Research has been done on perspectives, scope and rates, reduction and recommendations, although not as much research has been done overall (Allen, 2000;De Hert, Dirix, Demunter & Correll, 2011;Endres & Goke, 1973;Persi & Pasquali, 1999;Petti, Mohr, Somers, & Sims, 2001). As with adult literature there is overlap between restraint and seclusion (Petti et al.). There is no information available on the efficacy or effects of seclusion.
In lieu of a broad overview of studies, the small amount of studies available will be discussed in greater depth.
Perspectives. In the area of perspectives on seclusion, Petti et al. (2001) compared staff and child perspectives on seclusion directly after the event, finding discrepancies between the two groups on justification for seclusion, preventative measures, and reports of safety. When asked about the rationale for seclusion use, more staff than children reported safety issues. In addition to safety, children also reported not knowing (or not answering) why they were secluded. Staff members provided some ambiguous rationales, and several provided no answer at all. There was more agreement between staff and children on precipitants to the seclusion event.
When asked what took place before the restraint or seclusion event, children identified In another study looking at perspectives, Endres and Goke (1973) sent out questionnaires to 50 residential treatment centers, and asked staff to report their feelings about putting children in seclusion. The majority felt that the use of seclusion enabled the primary therapeutic goals of regaining control and deterring certain behaviors, as well as helping the child gain inner controls. However, they also found a notable difference in how staff experience the use of seclusion, with the majority of experienced staff reported as comfortable with its use, and inexperienced staff feeling hesitant, unsure, and fearful. differences extend to total rates of seclusion, which ranged from 8% of patients (Sourander et al.) to 61% (Donovan). Factors implicated with seclusion include: aggressive or suicidal acts (De Hert; Sourander et al.) and being admitted on an emergency basis (Donovan). Endres and Goke (1973) report duration of seclusion events across 42 residential settings, with most events lasting from 0 minutes to 1 hour, and the maximum amount of time spent in seclusion reported as up to 1 week. It should be noted that this study took place 40 years ago. More current duration information is provided in the CWLA study, with duration rates reported as approximately 25 minutes.
One consistent finding across studies is that seclusion rates can be reduced.
CWLA (2004) reduced rates by 29% across a 3 year period across several different sites. Reduction in seclusion use occurred as per regulation in another study, however, there was a corresponding increase in patient injuries (Donovan et al, 2003). Methods to reduce seclusion include: staff training (Greene, Ablon, & Martin, 2006), incorporation of an intervention model (Martin, Krieg, Esposito, Stubbe, & Cardona, 2008) and technical assistance, training, designing and implementing evaluation systems (CWLA). Gair (1984) provides guidelines for the use of seclusion with children and adolescents.

Seclusion in Schools
In "The Forgotten Room: Inside a Public Alternative School for At-Risk Youth" Hollowell (2009) recounts a day during which she saw a boy named Leon put into a seclusion room after beginning to tantrum in his classroom, Leon banged on the door of his cell, and I winced at every blow. "Let me out! Let me out!" he screamed in a high pitched voice.
Ms. Pearl told me that students were not supposed to have jackets, chairs, or shoelaces inside the cell. They were items that students might use to strangle themselves. is an example of how even good people in alternative schools can stand back and let bad things happen. In extreme conditions, they follow the rules or resort to apathy, just like disaffected youth in their classrooms. And I have no doubt that putting a young child into a cell inside a public school is wrong…" (p.

172)
Much of what is known about seclusion in schools relies on anecdotes like these, more often than not deeply disturbing stories in popular media accounts (Blacker, 2012;Downey, 2010;Fowler, 2012;Hefling, 2012;Kaplan, 2010;Lichtenstein, 2012;Richards, 2012;Turque, 2011). Like other areas of research, it can be difficult to weed out seclusion from restraint, and seclusion in schools from seclusion in other settings. Very few studies specifically target the use of seclusion in schools (Ryan, Peterson, & Rozalski, 2013). Preliminary studies have uncovered some consistencies. Seclusion appears to occur more frequently with elementary and middle school students (COPAA, 2009;Persi & Pasquali, 1999;Villani et al., 2011;Westling et al., 2010), in a special education setting, more often with students with autism and without parental consent or a behavior intervention plan (COPAA; Westling, et al.,). Other studies have uncovered inconsistencies in use across settings (CWLA, 2004), including variability between less and more restrictive settings (Persi & Pasquali). Unlike medical literature, there were no studies found that looked into the perspectives of students subject to seclusion in schools. There is minimal research done on the scope and demographics of seclusion. Most research falls in the area of recommendations or comparison of state laws and regulations.
Additionally, some information on the scope of seclusion use can be gleaned from reduction studies . A recently published book by the Council for Exceptional Children, "Physical Restraint and Seclusion in Schools" does not address the topic of scope, focusing, like much of the research already available, on legal issues, concerns, and reduction (2013).
The only research that comes close to providing an actual rate of seclusion is the survey done by Westling et al., (2010). As mentioned, there are limitations to this survey, which was done with a convenience sample of 1,300 parents and guardians who were referred by advocacy organizations, a non-representative sample. Well over half, 64%, answered the first question affirmatively, "To your knowledge, has your child ever been restrained, secluded, or subjected to aversive procedures while in school, or by school personnel in other locations, or during an after-school program sanctioned or operated by the school?" Of those, 70% of the respondents reported that their child had been subject to seclusion.
Consistent with studies done on seclusion with adults and children in mental health settings, the scope of seclusion use in schools seems to be impacted by student as well as setting factors. Persi and Pasquali (1999) found variability across the continuum of restriction across settings, ranging from a psychiatric inpatient unit, a group home on hospital grounds, a day treatment program in a self-contained school to a similar day treatment program located within a community school. The level of restriction of the program was hypothesized to reflect the severity of need of the child, with the most restrictive setting likely to more often use seclusion and restraint. All settings were within the same institution and therefore followed the same policies and procedures. All settings had a seclusion room, with the exception of the day treatment program. The two most similar settings (in patient characteristic and restrictiveness of setting) reported disparate rates, disproving the hypothesis. The authors conclude that this difference is not the result of staff variables or patient variables such as age and gender. Instead, they question whether it is due to patient misplacement, or widely different patient populations, with most of the challenging patients being placed in a less restrictive setting. Further study found that there was no significant variability in populations. The authors do not consider the possibility that although all of the settings are under the same institutional umbrella there may be widely different cultures of practice within them. The study raises a question about the actual or perceived need for the use of seclusion: "it may be that clinicians use very different standards when deciding whether seclusion and restraint is needed and that these standards are not strongly linked to the level of protection or restrictiveness of the setting or the severity of the child's problems" (Persi & Pasquali,p. 100).
The authors also found that seclusion was used more often than restraint with most patients gradually declining in seclusion incidents over time. Unfortunately the demographic data is reported for restraint and seclusion combined, so there is no demographic information available about seclusion exclusively. Recent OCR (2012) data provides some demographic information, breaking down seclusion data by race, with a notable discrepancy found with Hispanic students without disabilities, comprising 24% of the student population without disabilities, yet 42% of students without disabilities subject to seclusion.
Two surveys done by the Council of Parent Attorneys and Advocates, Inc.
(2009) and Westling et al. (2010) provide comparative demographic information, but both should be interpreted with caution due to limitations in survey design, little information on research methods, and the reporting of information on restraint and seclusion together.
Both studies found that elementary school age students are subject to seclusion and restraint at higher rates primarily in special education settings. Almost half of all students subject to restraint and seclusion in both studies had autism, followed by attention deficit disorder in the COPPA (2009) study and emotional and behavioral disorders in the Westling et al. study (2010). These findings are further supported in a descriptive study Villani et al. (2000), which found that the majority of students subject to seclusion had autism, and that lower/middle school students were secluded more frequently. In both studies, close to 70% of parents had not given consent to the use of seclusion and restraint procedures.
The Westling et al. (2010) study goes into further detail, asking respondents specific information about seclusion incidents. Seclusion (and restraints) were administered primarily by special education teachers, followed by administrators, but also including a wide variety of school personnel with respondents identifying: school counselor, general education teacher, behavior specialist, related service providers, and a wide variety of school support personnel: paraprofessionals, teacher aids, oneto-one assistants, Applied Behavior Analysis assistants, bus drivers, school nurse, school police officer, after-school assistant, and residential staff. Respondents were asked where seclusion events took place, and the majority of respondents reported "in a special seclusion room designed for the purpose of seclusion" (p. 120). The next most frequent response was "in another area of the facility" and these responses include: bathrooms, an old locker room, closets, kitchens, sensory rooms, storage areas, janitor's closet, and a hallway. Questions were also asked relating to the duration of seclusion. Most of the seclusion events lasted from 5-30 minutes (22%), followed closely by durations of 1-3 hours (21%). The responses ranged from 5 minutes to 3 hours, and almost a quarter of respondents did not know the duration of seclusion events. This may due to a lack of communication, parents were asked if and when they were contacted by the schools regarding the incident: 39% responded never, 27% said rarely (less than 50% of the time), 21% said usually and 13% said always. This may be the result of a disparity in the existence and requirements of state policies noted by Ryan, Peterson, and Rozalski (2007).
Consistent with the research done with adults in mental health settings, parents and guardians report a high level of trauma for their children associated with seclusion and restraint: 92% of the 647 parents who responded to this question indicated that their child had experienced emotional trauma, 42% had experienced physical pain, 33% had obvious signs of physical injury, and 39% had other adverse reactions (Westling et al., 2010).
As with adults, there is a body of literature focused on reducing seclusion and restraint, demonstrating that incidences of seclusion can be reduced through changing school culture and increasing capacity (George, 2007;.
Ryan, Peterson et al. had three other relevant findings in their study on reduction.
First, there seems to be a population of students that may incur the most seclusion events and who are not responsive to reduction efforts. Second, a staff survey showed that seclusion was used as a disciplinary strategy and not just in case of emergency, most often for leaving an assigned area and noncompliance. Third, staff were not always following the gated procedures of least to most restriction when using seclusion, and were found skipping through several intermediary steps to preemptively use seclusion. When asked, a staff member stated that the she knew the student, and therefore felt that the intermediary stages would be ineffective. This further supports the finding that the likelihood of a seclusion event is a combination of student, staff, and setting factors (De Benedictis et al., 2011;Mann-Poll et al., 2011, Persi & Pasquali, 1999. As with research done on seclusion outside of schools, there are recommendations in the literature: regarding policy and procedure, the physical structure of timeout rooms, training of staff, and duration of seclusion event (Ryan, Sanders, et al., 2007;Yell, 1994). There is debate regarding putting restrictive measures such as seclusion into a student's individualized education plan (IEP) or behavior plan, with some in favor of it (Yell, 1994) and others against (CCBD, 2009;TASH, 2011). TASH, an organization that advocates on behalf of individuals with significant disabilities, has created a manual for parents that addresses prevention of seclusion and restraint, how to identify warning signs, and how to respond if a parent suspects their child has been subject to seclusion or restraint (2011). Abuse at Public and Private Treatment Centers, 2009). The GAO report also found hundreds of instances of abuse and death, and provided in depth analysis of ten cases in which there were legal or financial consequences for a school district. The following pattern emerged: almost all of the cases involved students with disabilities who were restrained and secluded in nonemergency situations, without parental consent, by untrained staff. In half of the cases involving the injury or death of a student following a seclusion or restraint, staff involved continue to be employed as educators.
The day the GAO report was released, Chairman Miller conducted a hearing before the House Education and Labor Committee. Several parents testified, including parents whose children had died as the result of abusive restraints. In December of the same year, two bills were introduced in the House and the Senate, the Keeping All Students Safe Act, H.R. 4247 and the Preventing Harmful Restraint and Seclusion in Schools Act, S. 2860 (Peterson & Smith, 2013). These laws would close the federal gap identified by the NDRN and GAO reports, applying to all public and private schools, whether school funded or federally funded, and all students, not just those with disabilities. Since then, the bills have been reintroduced and expired at the end of the Congressional year several times. As of this writing, H.R. 1893 was reintroduced on May 8, 2013 and has been referred to committee. There has been no reintroduction to date for S. 2860.
In regards to seclusion, the current iteration of the H.R. 1893 bill states it may only be used in situations of imminent danger, when less restrictive interventions would be ineffective. There are safety precautions built in: staff utilizing seclusion and restraint must be trained in a state approved crisis management program (including first aid and CPR), there must be sufficient staff to student ratios, students must be continuously monitored and the seclusion event must end as soon as the safety threat is over. Parental notification is required, verbally or electronically on the day of the event, and in writing within 24 hours. Federal reporting requirements are also included, disaggregated by number of incidents that result in injury, death, or are administered by untrained personnel, and demographic information including age and disability status. S 2020, the senate version of the bill, last introduced in 2011, expired, and was introduced in March 2014.
Some strengths of the bill include the safety mechanisms and the potential to begin a nationwide database regarding seclusion and restraint (Hoffman, 2011).
Additionally, seclusion and restraint may not be written as a planned intervention for a student in an IEP, a behavior plan or a safety plan, although it may be included as part of a school wide crisis plan that pertains to all students. Supporters include over 200 national, state and local organizations (Butler, 2013).
A major criticism of the bill centers on the definition of seclusion as locked isolation, and not as any situation in which a student is physically prevented from leaving a room or area, as in the CCBD definition (Butler, 2010;Hoffman, 2011).
Therefore, there are many situations that may occur in schools that would not fall under the category of seclusion, such as an adult holding a door shut. Additionally, there are no regulations regarding the physical structure of the rooms used for locked isolation, and no mandated reporting for school personnel breaking the law, with the only potential repercussion for not following the regulation resulting in the loss of federal funding (Butler, 2009). Due to the fact that H.R. 1893 does not provide for a private right of action for individuals, some critics argue that it could lead to poor enforcement (Hoffman;Kaplan, 2010). Nor does it provide distinct training for students with disabilities or protections to students with disabilities, who are a vulnerable population and more at risk for improper use of seclusion and restraint (Hoffman).

The American Association of School Administrators (AASA) (2012) has been
sharply critical of the bill as a whole, but with weak arguments. The first argument is against the mandate that seclusion and restraint cannot be included into behavior plans or IEPs. The claim is that this mandate is counterproductive as it would prevent proactive conversation from taking place regarding a student with significant behavior problems. This argument is somewhat illogical because there is nothing in the law that prevents a conversation from taking place, the restriction is on documentation of seclusion and restraint as a planned intervention for an individual student. The second argument is against the prohibition against manual restraints as it may impede a school resource officer or police officer, however, there is language in the proposed regulation immunizing the use of handcuffs by school resource officers. The remainder of the argument against the bill focuses on the tone of the law as overly negative and implying that educators are out to harm students. However, that argument could be turned against AASA, who make light of seclusion and restraint as a serious educational, and potentially life-threatening issue, by referring to a "mere" In conclusion, "AASA refuses to accept the idea that public school employees are over-using seclusion and restraint and/or using it inappropriately" (2012, p. 8).
Additionally, House Republicans brought up the following concerns regarding proposed legislation: a lack of reliable data on the use of seclusion and restraint in public and private schools; the creation of a one size fits all mandate, when 31 states already have legislation ; the inclusion of traditional private schools, which is unprecedented in educational legislation; and the concern that inclusion of language restricting aversive behavioral interventions that compromise health and safety may expose schools in states that still use corporal punishment to litigation (House Report [2009][2010].
In the interim, state policies on the use of seclusion and restraint have evolved, especially since Arne Duncan issued a letter in response to the GAO report and congressional testimony in 2009, in which he promoted the use of Positive Behavior Interventions and Supports (PBIS), and encouraged school districts to develop, review, and revise policies in preparation for the beginning of the 2009-2010 school year (Duncan, 2009). The number of states with policies in place has increased, from 17 states in 2007 (Ryan, Peterson & Rozalski, 2007) to 33 states as of 2013, with 30 states having made changes (Freeman & Sugai, 2013). Policy seems to have become more cohesive; Freeman and Sugai found a "clear consensus" in an analysis of current state policy that seclusion and restraint should only be used as an emergency intervention and not as a means of discipline, along with the following trends: At this time, and as evidenced by the U.S. Department of Education's resource document, the federal stance on seclusion and restraint remains primarily suggestive, without calling outright for federal mandates regarding these procedures (Peterson & Smith, 2013). The use of seclusion and restraint continues to be monitored by advocacy organization: TASH (2012)  Additionally, claims of abuse and neglect or failure to protect a student from abuse and neglect may be filed with state agencies, such as Protection and Advocacy systems, criminal prosecution may be pursued, tort lawsuits may be filed for personal injury, licensing complaints may be made with the State Education Agency, and, finally, due process hearings under the Individuals with Disabilities Education Act, IDEA (Peterson & Smith, 2013). There is currently no way to definitively determine if and when these complaints or claims have occurred, with the exception of IDEA claims that have gone to a due process hearing. There is some indication that incidents may be under-reported. In the survey done by Westling et al. (2010), 71% of respondents reported never contacting a state agency regarding the seclusion or restraint event. Additionally, Zirkel (2008) noted that most published cases are not based on IDEA, but on constitutional claims and case law. In an analysis of case law regarding the use of aversive interventions for students with disabilities, Lohrmann-O'Rourke and Zirkel (1998) found that, although there were a variety of claims made in addition to IDEA, most claims were 14 th amendment substantive due process and section 504 ADA claims at the federal level, and negligence and infliction of emotional distress at the state level. They conclude that there is a pattern for "qualified support" for the use of timeout, with minimal safeguards for students with disabilities, and a "dramatic disparity" between case law and current best practices in special education (p. 122).
A powerful example of this disparity is the case of 13 year old Jonathon King, who hung himself in a seclusion room with a rope given to him by school staff to hold up his pants. Jonathon had been subject to an average of 88.6 minutes of seclusion for 19 of the previous 29 days of school, and had threatened suicide prior to its commission. In a liability case brought forward by his parents, the court ruled in favor of the school. As Goodmark (2009) argues, "the King decision has set the liability bar for school abuse cases at an unattainable level. If a school is not liable for the mentally ill student that hangs himself with the school's rope, then it is unclear where liability could ever exist in Georgia" (p. 277).

The Individuals with Disabilities Education Act
The Individuals with Disabilities Act, or IDEA provides a clear path for complaints about seclusion of students with disabilities in schools. IDEA is an "educational bill of rights" for students with disabilities (Zettel & Ballard, 1982, p. 15). Special education in America is grounded in the legal foundation of IDEA, which may, or may not, be contraindicated by the practice of seclusion. Much of the framework of IDEA evolved as legal corrections to past educational practice that were then written into regulation. The next section will provide a framework of IDEA, along with a discussion of the legal guarantees of IDEA and how the practice of seclusion may contradict these guarantees.

The Framework of IDEA
Two lawsuits on behalf of children with disabilities set the stage for the Education of All Handicapped Children Act, subsequently reauthorized as IDEA and provided its legal framework; Pennsylvania Association for Retarded Children (PARC) v. Commonwealth of Pennsylvania in 1971, and Mills v. D.C. Board of Education in 1972. These lawsuits established the right to a free and appropriate public education (FAPE) for all students with disabilities, the right to full administrative due process of law, the right to education in the least restrictive environment (LRE) and the right to ethnically and racially nondiscriminatory testing and other assessment procedures (Zettel & Ballard, 1982 Handicapped Children Act, subsequently re-named IDEA, is a unique educational law, in that by "combining an educational bill of rights for handicapped children with the promise of an increased federal fiscal partnership, that mandate became a matter of precise national policy" (Zettel & Ballard, 1982, p. 15). IDEA is an example of a policy mandate, a construction of rules with the intent to produce compliance from individuals and organizations (McDonnell & Elmore, 1987). It has some unusual components, As evidenced by the vague and malleable definitions of these substantive requirements (such as FAPE, LRE, and IEP), the IDEA is a unique statute with and "unconventional, decentralized system of decision making". Rather than detailing the specific services required under the Act, the IDEA outlined "extensive procedural" protections for parents and students and relied heavily on these procedures to effect the Act's purpose. Congress may have chosen this unconventional method of legislating due to a federalism concern; since education had traditionally been the purview of states, Congress was concerned that any federal imposition of substantive requirements would severely limit states' discretion over public education. Another possible reason for the heavy reliance on procedure is that "the immense variety of disabilities and needs made it difficult to formulate universally applicable substantive standards". In the face of numerous physical, mental, and emotional disabilities, it would have been exceedingly difficult for Congress to create a "menu" of educational options to be provided for every disability category." (Phillips, 2008(Phillips, , p. 1818 Although IDEA does not address the use of seclusion or restraint (Office of Special Education Populations [OSEP], 2008;Yell, 1994), the practice of seclusion could violate one, or more, legal guarantees of IDEA. As stated in a letter from the Office of Special Education Populations, the critical inquiry is whether the use of such restraints or techniques can be implemented consistent with the child's IEP and the requirement that IEP teams consider the use of positive behavioral interventions and supports when the child's behavior impedes the child's learning or that of others" (OSEP, p.

5)
The next section will discuss the legal guarantees of a free and appropriate public education (FAPE) in the least restrictive environment (LRE), special education evaluation and eligibility, and discipline of students with disabilities, along with case law regarding seclusion for each of these areas.

Free and Appropriate Public Education
The essential and constant provision of IDEA is a free and appropriate public education for students with disabilities, also known as FAPE (Jones & Toland, 2010 (Yell, Drasgrow, Bradley, & Justesen, 2004). Instead, it mandates the procedural process that schools must follow to ensure that FAPE is provided and individualized for each student with a disability by the IEP team. The steps for ensuring FAPE for students with disabilities is to provide a fair and accurate assessment of students with disabilities, which is in turn used to create specially designed instruction via the IEP, which becomes the "blueprint" of FAPE for each student (Yell, Katsiyannis, & Hazelkorn, 2007).
As with many policies, terms are deliberately left vague in order to be applicable to many different scenarios. "Free" and "public" are easy to provide, however, FAPE hinges on the definition of "appropriate" as well. Despite the broadness of the term "appropriate", Congress has not changed it through several reauthorizations, leaving it up to IEP teams to wrestle with, and the due process procedures to follow if the IEP team cannot come to a consensus. This has generated many due process hearings and court cases; in fact  state FAPE is "one of the most heavily litigated areas in special education law" (p. 25).  . Although some critics have noted that the Supreme Court took this case prematurely, without letting a body of case law develop (Weber, 2012), Rowley is still the standard. Congress has not taken steps to change it through several reauthorizations of IDEA (Kaplan, 2010). The Court also noted that as long as procedural requirements have been met, courts should defer to school decision making authority and refrain from imposing their own views regarding preferable educational methods (Jones & Toland, 2009).
Rowley continues to spawn due process hearings around the two part test. In most circuits, procedural violations alone do not result in a denial of FAPE unless they result in an adverse impact to the student. Substantive harm that leads to a denial of FAPE include a loss of educational opportunity, deprivation of educational benefits or the prevention of parental participation (Kaplan, 2010). What, now, is "meaningful educational benefit? And how much educational benefit should a student get?
Although the Rowley test does not directly address the contents of a FAPE, it does provide guidance for courts to use in deciding, case by case, whether a school has offered a FAPE to a student with disabilities. The degree of benefit provided does not need to result in a student's achieving his or her maximum potential, nor must the FAPE be the best education possible. It must, however, provide the student with an educational program that will result in meaningful and measurable advancement toward goals and objectives that are appropriate for the student given his or her ability and as set forth in the IEP. (Yell et al., 2004, p with students with behavioral disabilities by punishing them for their disability, rather than finding an approach that addresses the problem. We also tacitly approve the district's resort to police intervention for the behavioral problem it helped create by failing to address CJN's unique behavioral disorder" (Jones & Feder,p. 6).

Least Restrictive Environment
Placement of the student in the least restrictive environment (LRE) is another major tenet of IDEA (Russo & Osborne, 2007 The terms mainstreaming and inclusion are often used to in relation to LRE, however, LRE has a legal foundation while mainstreaming and inclusion do not (Jaegar, 2002). The term mainstreaming refers to the placement of students in general education for a portion of the day, while inclusion refers to placement in general education classes typically fulltime, regardless of the student's level of disability or individualized needs, with removal only for special education services (Jaegar; Russo & Osborne, 2007). As Jaegar points out, The term full inclusion implies integration without individualization. The implication of the term partial inclusion is selective integration, which also means selective segregation. IDEA is designed to ensure that students with disabilities are integrated to the maximum extent feasible based upon the individual needs and abilities of each student" (Jaegar, p. 42) Placement decisions are made annually, by the student's IEP team, including parents and the student when appropriate.
The language around LRE changed when IDEA was reauthorized in 1997, leading to a change in presumption and IEP procedure. In IDEA 1997, the preference for integration in general education became a presumption, which can only be overcome by the IEP team documenting that the general education setting will not be appropriate for the student with a disability, even with supplementary aids and services . Essentially, IEP teams should consider the student with a disability in the general education setting first, then consider more restrictive special education settings, not vice versa.
Although there is a clear preference for students with disabilities to be educated in general education classrooms in their neighborhood school when possible, it is not a mandate and placement of each student must be made individually in conformity with what is provided by the IEP. Schools must provide a continuum of services in order to meet the special education and related service needs of students with disabilities (Russo & Osborne, 2007) including, the general education classroom with support, a resource room, a self-contained classroom, a self-contained school, homebound and residential instruction. A common misconception is that placement refers to the physical location of the classroom, rather, it refers to a level of service.
Therefore a change in placement is more accurately described as a change in the level of service. Like FAPE, LRE must be determined on a case-by-case basis by the IEP team (Jaegar & Bowman, 2002;. It is important to note that, per IDEA, students with disabilities should not be placed in general education just for the purpose of inclusion, and in fact segregated settings may be determined to be the least restrictive environment for an individual student (Russo & Osborne).
IDEA includes three procedural provisions around placement, which must be determined annually: first, parents must give informed written consent prior to the child being placed in special education, second, parents must receive prior written notice when a change in placement is proposed and third, parents may challenge a proposal to change placement through mediation or a due process hearing (Yell & Katsiyannis, 2004). IEP teams should consider the educational needs of the student and the goals of the IEP when making placement decisions. However, they may also consider the impact the student with a disability may have on a given classroom, with consideration given to the effects of related services and supports as well (Yell & Katsiyannis). Placement decisions should not be made based on the following: disability category, severity, availability of educational services, related services, or space or for administrative convenience (Yell & Katsiyannis). Another important note: once an IEP has been created and a placement has been determined by the IEP team, including the parent, IDEA has mandated a "stay put" provision for the student. In the event there is disagreement between school and parents regarding the provision of services, the student remains in the current placement until the issue is resolved (Jaegar, 2002).
Like FAPE, LRE is a frequent source of litigation  as well as controversy in the literature (Dorn, Fuchs & Fuchs, 1996). Initially, courts leaned towards the need for specialized services over the need for inclusion, with the exception of Roncker v. Walter, in 1983, when the court noted if services in preferable settings can be offered in less restrictive settings, then the more restrictive setting would be inappropriate (Russo & Osborne, 2007). This became known as the portability standard. Then, in the 1990s, the courts began to lean more towards the side of inclusion, even for students with severe disabilities, viewing LRE as a requirement of IDEA (Russo & Osborne).  .
LRE and seclusion. However, the Supreme Court has decided a case involving "stay put", Honig v. Doe in 1988. This case involved two students with emotional and behavioral disorders who presented significant safety concerns in the school setting, and the case centered on whether or not there could be an exception to stay put due to safety concerns. The answer was no. "Congress very much meant to strip schools of the unilateral authority they had traditionally employed to exclude disabled students, particularly emotionally disturbed students, from school" (Jones & Toland, 2010, p. 9). Although schools do not have unilateral authority to remove dangerous students they may follow several courses of action: work within the IEP team to seek a change in placement, suspend a student for up to 10 days, seek an emergency due process hearing, and as of the 1997 IDEA amendments place a student in an interim setting for 45 days if the student brought illegal drugs or weapons to school, or if the student has caused significant bodily injury to another person while in school (Jones & Toland, 2010). An important note when considering the use seclusion under IDEA, the US Supreme Court in Honig vs. Doe recommended the use of "study carrels, timeouts, detentions, or the restriction of privileges" in lieu of suspension and changes in placement (Honig v. Doe, 1988).

Special Education Evaluation and Eligibility
As with most special education law, special education identification and eligibility as outlined in IDEA is rooted in two important cases: Diana v. State Board of Education in 1970 and Larry P. v. Riles in 1979. Both cases involve discriminatory assessments, resulting in students being placed in special education due to cultural and environmental factors, not in response to the presence of a disability (Zettel & Ballard, 1982). As a result, IDEA has four important requirements regarding special education evaluation and eligibility: tests and evaluations must be provided in the child's native language, tests should be selected and administered in such a way as to gauge the student's actual ability, no single test or procedure should be used to determine eligibility, and the evaluation should be done by a multidisciplinary team or group of persons, including at least one teacher or other specialist with knowledge in the area of suspected disability (Zettel & Ballard).
Special education evaluation, eligibility, and seclusion. Historically, students with emotional and behavioral disorders are under identified, which may lead to misapplication of seclusionary procedures (Smith, Katsiyannis & Ryan, 2011 (Smith et al.).

Discipline
In order to counter decades of exclusion of students with disabilities from public schools, particularly students with disabilities, IDEA contains strict disciplinary requirements. When a student's misbehavior is a manifestation of the student's disability, the IEP team must conduct a Functional Behavioral Assessment (FBA), in order to determine the function of the behavior and replace it with a more socially appropriate one via a behavior intervention plan (BIP). If a behavior intervention plan has already been developed, the IEP team is required to review and revise it as needed (34 C.F.R. section sign 300.530 (f) (1).
If a student's behavior interferes with learning, the IEP team must consider positive interventions and supports (Jones & Feder, 2009), however, language from IDEA (1997) states "The IEP team shall…in the case of a child whose behavior impedes his or learning or that of others, consider, when appropriate, strategies, and supports to address that behavior" (34CFR Section 300.346 (a)(2)(i)). An argument has been made that the use of the term "when appropriate" is vague, and that while positive supports are encouraged aversive supports are not forbidden. Other suggestions have been made, such as conducting FBAS and BIPS proactively as part of an initial evaluation, well before students are facing disciplinary procedures (NDRN, 2010). Additionally, Lohrman O'Rourke and Zirkel (1998) argue that the positive behavioral supports and disciplinary procedures outlined in IDEA are sparse, and the minimum standard that the law requires should not become our maximum practice. "It is our obligation to challenge our own values and subsequent professional practice to ensure that we are providing comprehensive educational supports to students with disabilities, rather than disciplining them because of ineffective educational systems and individual practices" (p. l23).

Individual Education Plans
There is debate regarding the inclusion of seclusion and restraint measures in a student's IEP, with some in favor (Miller, 2011;Yell, 1994) and some opposed (CCBD, 2009). In 2010 a case was decided by the 8 th circuit, in which the use of seclusion and restraints were included in a student's IEP (C.N. v. Willmar Public Schools). The parents argued that seclusion and restraints were used inappropriately and punitively, but the court found in favor of the school because it was part of a predetermined plan agreed to by the IEP team (Smith et al., 2011).
Some preliminary studies have uncovered some consistencies. Restraint and seclusion seem to occur more frequently with elementary and middle school students (COPAA, 2009;Persi & Pasquali, 1999;Villani et al., 2011;Westling et al., 2010) in a special education setting, more often with students with autism and without parental consent or a behavior intervention plan (COPAA; Westling, et al.) Other studies have uncovered inconsistencies in use across settings (CWLA, 2004), including across least to most restrictive settings (Persi & Pasquali).
IDEA is the federal statute for education of students with disabilities. While it references the consideration of positive behavior support plans, it does not address behavior reduction procedures such as seclusion. This has left the courts to become arbiters, leading to a body of case law (Yell, 1994), a patchwork of guidelines and state laws and regulations (Arthur, 2008), and a "dramatic disparity" between best practice in special education and case law on aversive interventions such as seclusion Zirkel, 1998).
It is critical that the special education community gain an understanding of the nature of seclusion of students with disabilities in schools in order to frame regulations such as IDEA and federal restraint and seclusion legislation and prevent abusive seclusion practices. What little research has been done relies on anecdotes (COPAA, 2009;GAO, 2009), convenience surveys (Westling et al., 2010), lists of cases (Lohrmann-O'Rourke & Zirkel, 1998), frequency tables (Zirkel & Lyons, 2011), is limited to one or two schools (George, 2000;Villani et al., 2011) and are inadequate for analysis and understanding. Several studies propose recommendations (Gast & Nelson, 1977;Ryan, Sanders, et al., 2007;Yell, 1994) or are focused on reduction of restraint and seclusion (CWLA, 2004; which may be premature without first understanding the nature of seclusion in schools. There have been calls for more research: (Arthur, 2008;CCBD, 2009;Persi & Pasquali, 1999;Wolf et al., 2006) as well as a recommended canvassing of pertinent litigation (Zirkel & Lyons, 2011). There have been calls for amending IDEA to include restrictions on the use of seclusion and restraint (Miller, 2011) and arguments against (Kaplan, 2010).
The conditions under which students with disabilities are secluded in school are unknown at this time. Additionally, the practice of seclusion in schools could violate IDEA in the following ways: denial of FAPE in the LRE, evaluation and eligibility, and discipline. This study will determine the conditions under which students with disabilities are subject to seclusion as well as the relation of seclusion to IDEA by canvassing another tenet of IDEA, due process hearings.
Due to a history of unilateral decision making and exclusion, IDEA guarantees due process procedures for students with disabilities and their parents (Zettel & Ballard, 1982). Procedural safeguards include the notice of rights, mediation, resolution sessions, and due process procedures (Jones, 2010). Essentially, Congress envisioned a true partnership between parents and schools when planning the education of students with disabilities. With the 1997 amendments, parents were given the right to participate in meetings regarding identification, evaluation, and placement of students with disabilities (Daniel, 2000). In the event that the parties don't agree, or when schools are not following IDEA procedurally or substantially, parents may file a due process complaint with the state education authority. IDEA complaints may address identification, evaluation, placement, or the provision of a free and appropriate public education in the least restrictive environment (Kaplan, 2010).
IDEA has two ways to resolve the complaints: impartial due process procedures and the state complaint resolution system. Due process claims may be made by parents, state complaints may be made by an organization or individual on behalf of a student or a group of students (Apling & Jones, 2007). Due to the fact that legal complaints cannot be made against educational systems, many parents have turned to IDEA to make complaints regarding the use of seclusion (Kaplan, 2010). As a neutral party interpretation of a dispute between parents and schools, due process hearings present an opportunity to analyze the intersection of special education law and practice, helping to inform and improve practice (Rock & Bateman, 2009).
The purpose of this study is to discover the nature of seclusion of students with disabilities in schools using a grounded theory analysis of due process hearings involving a complaint about seclusion. It seeks to answer the following question: Under what conditions are students with disabilities subject to seclusion in schools?

METHOD
Due to the exploratory nature of my research question, a grounded theory analysis of due process hearings containing evidence of seclusion was conducted. Grounded theory was first identified by Glaser and Strauss (1967) as the discovery of theory from data through comparative analysis. The objective of grounded theory is to explain phenomenon in new theoretical terms, explain the properties of the theoretical categories, and to delineate the causes, conditions, and consequences of the phenomenon (Charmaz, 2006).
Grounded theory is appropriate for a study in which there is not adequate existing theory or research to form a hypothesis. In addition, it enables the researcher to optimize the sample by extracting more information from it, leading to enhanced understanding of a phenomenon (Collins, Onwuegbuzie, and Sutton, 2006;Creswell, 1994).
Initially, this study was planned as a mixed methods two phase study. The first phase involved a qualitative grounded theory analysis answering the question: Under what condition are students with disabilities subject to seclusion? The next phase was planned as a quantitative content analysis answering the question: Is there a relation between seclusion and the Individuals with Disabilities Act (IDEA)? The content analysis was not possible due to the small and scattered sample of due process hearings that contained a specific IDEA complaint about seclusion. However, there was a rich amount of narrative in the due process hearings, including hearing officer decisions and opinion, which was included and analyzed using grounded theory in order to address this question.

Characteristics of the Study Population
Due process hearings are part of the procedural safeguards for parents embedded within IDEA. IDEA has generated more litigation than any other education legislation (Russo & Osborne, 2007) and as a neutral party interpretation of a dispute between parents and schools, due process hearings present an opportunity to analyze the intersection of special education law and practice, helping to inform and improve practice (Rock & Bateman, 2009). Due process hearings are usually initiated by parents, although they may be initiated by schools (Getty & Summy, 2004). There are no specifications in IDEA regarding the qualifications of hearing officers, other than they must be neutral parties and have no involvement with the student. Due process hearing officers are usually lawyers or educators. Once a hearing officer has been appointed and heard the case from each party, the hearing officer renders a report and a decision on the merits of each complaint, including a remedy if warranted. The law does not detail the contents or length of the report, and reports differ across hearing officers, as well as across states. Typically, there are witnesses called to testify regarding the complaints of the hearings, and documents such as Individualized Education Plans (IEPs), reports, and correspondence are reviewed. Expert witnesses may be provided by the parents or the school. Most hearings list the witnesses that testified in the hearing, and include "findings of fact" based on the hearing officers' interpretation of the information provided at the hearing. Usually identifying information about the student is redacted, and in some cases names and identities of school staff are redacted as well.
The due process hearings in this study were downloaded from the Individuals with Disabilities Education Law Reporter, which is published by LRP Publications through the website Special Ed Connection (www.specialedconnection.com), a comprehensive nationwide data base containing special education case law (Zirkel, 2010). It has been in publication since 1979 and is commonly used in special education law research (Rose & Zirkel, 2007;Zirkel, 2008). The database compiles the text generated by the local education agency of due process hearings that take place nationwide. The due process hearings are considered to be public information and not subject to confidentiality issues or copyright violations.

Objectivist Framework
This study, while acknowledging the construction of due process hearings, takes an objectivist approach towards the data. This entails a positivist stance that the data are real in and of themselves, and are found and interpreted by the researcher (Charmaz, 2006). This is due to the goal of the study as well as the construction of due process hearings.
Due process hearings are an example of extant texts, documents that the researcher did not elicit or help create, and were intended for another purpose (Charmaz, 2006). As such, there are advantages in availability, unobtrusive method of data collection, and seeming objectivity (Charmaz). The relative objectivity of the due process hearings guides the decision to treat these data with an objectivist stance. The goal of this exploratory study is to look at conditions under which students with disabilities are subject to seclusion, and will focus on descriptions of these events. The focus will be on the facts of the case, as determined by the due process hearing officer.
Due to the construction of the due process hearings and the goal of this study, a constructivist approach would not be appropriate. Charmaz defines a constructivist approach as placing a "priority on the phenomena of study and sees both data and analysis as created from shared experiences and relationships with participants" (p. 130). In a constructivist approach, the focus is on how, and why, participants construct meanings and actions in specific situations (p. 130). The goal of this study is to uncover underlying conditions, not meanings or interpretations around seclusion.
There are distinct parameters to due process hearings. There is no way of knowing how and why information was included in the hearing narratives, and what information may have been left out. There is no way of knowing about seclusion events that never resulted in due process hearings, as there are several other ways to resolve complaints (Getty & Summy, 2004). However, the advantage of using due process hearings as data include the opportunity to gain access to practices and events, nationwide and over time, descriptions that are not easily accessed through other qualitative research methods. Additionally, these hearings present the opportunity to examine events that have occurred, not perspectives or opinions on what may be happening. As such, these hearings are approached with a positivist stance and will be analyzed as objective findings of fact, without attending to the precision of their production, the social context in which they were written, or the influence of the researcher (Charmaz). The focus of objectivist grounded theorists is as a "conduit for the research process rather than the creator of it", and the goal is theoretical understanding as the result of careful applications of methods (Charmaz,p. 132). It is not a verification method, but a method of generating theory, or plausible accounts (Glaser & Strauss, 1967).

Data Collection
A search of the Special Ed Connection due process hearing database was conducted in February 2013. Due process hearings taking place from 1997 to February 2013 were reviewed. The year 1997 was selected for the cut date due to the reauthorization of IDEA in 1997, when disciplinary provisions were added . These provisions included the consideration of functional behavior assessment and behavior intervention planning and limits on suspension of students with disabilities (Osborne, 2001).
This study used a criterion sample in order to ensure that all due process hearings involving the practice of seclusion were included, as well as to validate that events being described were in fact seclusion events. Due to the many different terms that may include seclusion, a search was done using the keywords (seclusion, isolation, timeout) and room. There were 297 cases that met initial criteria. These hearings were downloaded from the website for further analysis.
Each hearing was then read in order to determine if the key word in context met the following criteria: Is it apparent that the event describing the key word in context contains evidence of the 1. Involuntary confinement of a child or youth 2.
Alone in a room 3. From which the child is physically prevented from leaving (CCBD, 2009, p. 1). There were 26 hearings that met this criterion, of these, three were appeals, resulting in 23 hearings involving individual students.

Sample
Of the 297 due process hearings that met initial criteria from over 1000 reviewed in the database, 26 cases met all three qualifying criteria. This number may seem small considering the large amount of due process hearings that were reviewed, however, two things should be noted. First, there were many more cases that appeared to meet criteria but could not be confirmed. Additionally, there were many IDEA complaints involving the use of seclusion rooms. Although these complaints did not meet the criteria of being a due process hearing, they may prove to be a valuable data base for future study as they appeared to more specifically target the use of seclusion under IDEA, and contained a great deal of descriptive information regarding the physical structure of the rooms.
Three of the due process hearings were appeals, and were therefore coded with the initial case, resulting in 23 total due process hearings involving an individual student. Of those, 20 due process hearings involved the systematic use of seclusion, defined as multiple events over a period of time or as part of a behavior intervention plan. Three due process hearings involved a single seclusion incident that occurred. One case was an outlier (58 IDELR 267/7 GASLD 35/112 LRP 18857 Georgia, 2/1/12) in that it involved a chronically abusive teacher. However, it was included in the final analysis as it met all qualifying criteria. Throughout the text, the due process hearings will be identified by the reference numbers listed in the database, in the tables they will be referred to by number (listed below). The hearings that involve an appeal will be identified by the reference numbers from the initial due process hearing. One note-due to the fact that all of the hearings in this study met strict criteria of evidence of a seclusion event, the term "seclusion" is used to describe all such events, despite the term that may be used in the due process hearing narrative.
For reference, the cases that met criteria are listed below, numbered in chronological order.

Data Analysis
The hearings were analyzed through a reiterative grounded theory process of initial coding, focused coding, constant comparison, sorting, and memo writing.
Initial coding: First, the sample was read in full. Due to the highly individualized nature of each due process hearing, it was important to gain an understanding of its unique nature and to identify issues that are relevant for this study. Typically there are multiple issues that are brought forward during a due process hearing, and many issues may not be related to the seclusion event. After reading through each hearing, all text related to seclusion was isolated for data analysis using the following guiding question: Does this sentence contain information relating to the seclusion event or events? Information that was irrelevant was discarded. The hearings were then double spaced, printed out, and stored in binders for further coding. Due to the many issues that could impact the use of seclusion, a wide view was taken, and only text that clearly had no relation to the practice of seclusion was discarded. Through an open-minded examination and comparison of the action and process of seclusion, the question was asked: Which theoretical categories might these statements indicate? The goal of initial coding is to weave together two things: generalizable theoretical statements that transcend specific times and places and contextual analyses of actions and events. Constant comparison (Charmaz, 2006) of codes was used both across and within hearings. Due to the limited nature of the sample, every hearing that met criteria was coded, rather than stopping at saturation. Theoretical saturation was reached about two-thirds of the way through the hearings. During this phase in data analysis, demographic information on the student and descriptive information on the seclusion event were recorded, and the placement trajectory of students in the hearings was tracked for comparison.
Focused coding. After initial coding, focused coding was done, with the goal of looking for major themes and categories. Memos were used to record thoughts, observations, questions and patterns as they emerged from the data. This stage involved a reiterative process of returning to previously coded hearings to code a pattern that had become evident and to test codes in the sample through constant comparison of every theme and category.
At the same time, event maps were created for each hearing, recording conditions, interactions, and consequences that were present in the sample (Strauss & Corbin, 1990). Timelines were created to track the placement trajectory of students in the hearings. This sequence was followed for all 26 hearings, testing focused codes against the entire sample.
The final step involved writing advanced memos refining conceptual categories. Through advanced memo writing, data were sorted, conceptual categories were defined and described, and ideas, events, or processes were uncovered in the data. Comparisons were made between categories, and gaps in the data were considered. Strauss and Corbin's (1990) conditional matrix was used as an analytic framework, helping to define and link together the conditions that created the seclusion interaction. The conditional matrix is defined by Strauss and Corbin as "an analytic aid, a diagram, useful for considering the wide range of conditions and consequences relating to the phenomenon under study. The matrix enables the analyst to both distinguish and link levels of conditions and consequences" (p. 158).

Validity
In the qualitative tradition, validity and reliability are addressed through verification steps which typically involve feedback from informants (Creswell, 1994).
The grounded theory phase of this study was designed to provide verification from the data source by returning to the data in an iterative process to ensure internal validity, as well as to look for gaps in the data and to check codes through constant comparison.
Validity is the result of precise application of method. Dey (1999) defines validity as "well-grounded conceptually and empirically", and that the valid account of a phenomena generated by grounded theory is valid to the extent that the theories generated are grounded in the data from which it is produced (p. 268). Strauss and Corbin (1990) discuss grounded theory as a transactional system, "a method of analysis that allows one to examine the interactive nature of events" (p.

159). The process of seclusion fits Strass and Corbin's transactional systems as it has
the following properties; it is made up of interactive and interrelated levels of conditions, and central to the transactional system is the action/interaction that is processual in nature, which in turn results in various consequences. The transactional system can be captured in a conditional matrix, in which the outside rings specify the conditional features most distant to the action, with each ring specifying conditions closer to the event. The rings distinguish each level of conditions, but are also linked to each other-each set of conditions that emerge from grounded theory analysis is relevant to the event being studied. The conditional matrix enables the researcher to be theoretically sensitive to the range of conditions that effect the phenomenon under study, the range of potential consequences that result from the action/interaction, and aids analysis in that it provides a framework to systematically relate both (Strauss & Corbin).
This exploratory and interpretative phase of the study, like many qualitative studies, seeks interpretation over replication, and therefore has limited generalizability and reliability (Creswell, 1994).

RESULTS
The process of seclusion presents a clear path of conditions, interactions, and consequences. The conditions are illustrated by Figure 1. Each level of condition will be discussed in order of least to most effect and proximity to the event, followed by the findings regarding the seclusion interaction, and concluding with the short and long term consequences.

Loose Legal Boundary
The most exterior condition that is linked to the seclusion interaction is the loose boundary of law, primarily through the Individuals with Disabilities Education Act (IDEA), but also state laws on seclusion.
The relationship between IDEA and seclusion is unclear. There were three categories of complaints that were present in the hearings-denial of a free and appropriate public education (FAPE), placement, and other. These complaints resulted in a wide spectrum of hearing officer decisions, with decisions made in favor and against the use of seclusion.
The use, misuse, and overuse of seclusion was not a clear denial of FAPE. In two hearings the extensive use of seclusion was found to be a denial of FAPE (103 LRP 8586, 107 LRP 8925). However, the decision in 103 LRP 8586 was overturned in an appeal due to a lack of evidence (103 LRP 8590). In another instance, in which parents were seeking reimbursement for a private placement, the hearing officer found the practice of seclusion was a denial of FAPE, but the decision was made in favor of the school because the parents did not follow IDEA procedure when they immediately removed her from school upon discovering the use of seclusion (48 IDELR 266).
Three hearings found a denial of FAPE, not for the use of seclusion, but for failing to inform parents about it (107 LRP 63423, 58 IDELR 267, 112 LRP 51016). In another hearing, the use of seclusion was subsumed under the broader category of behavioral interventions, and the hearing officer found a denial of FAPE due to the lack of peer reviewed teaching methods (112 LRP 4046).
Placement decisions were dependent on the existence, or nonexistence, of seclusion rooms. In 43 IDELR 151 the hearing officer found that the student required a more restrictive setting, including the use of seclusion. This same finding was present in 109 LRP 23975, however, this decision was made against the parent's wishes that the student be placed in a setting without a seclusion room. In this case the hearing officer opined that being alone in the sensory room was an essential element in providing the student FAPE. Parents requested the use of seclusion rooms as an accommodation in 109 LRP 57374, and the hearing officer denied this request because "there is no unconditional legal requirement that a district maintain a safe room under state or federal law". Placement decisions were also made in favor of settings because the placement did not have a seclusion room (109 LRP 7122, 102 LRP 8176).
The broadest category of IDEA complaints involving seclusion fell under other, and all of these complaints centered on the use of seclusion. Of these, most hearing officer decisions were found in favor of the school and in support of the use of seclusion. The findings include-it is necessary and appropriate (108 LRP 607, 106 LRP 19250), the resources needed to administer seclusion are an asset for the student, the parent failed to complain prior to the hearing (106 LRP 19250), and the school does not need to provide a staff member near the student while administering seclusion (109 LRP 23975). One hearing, 48 IDELR 26, contained all three of the decisions found in favor of the student and against the use of seclusion, one was procedural: the school did not provide prior written notice when implementing seclusion, and two were substantive: the practice of seclusion found to not conform to the requirements of IDEA, and it was harmful to the student.
State laws, when apparent in the hearings, had varying effects and little overall impact. One hearing had two contrasting results of state law: in one school staff were changing practice due to recent legislation on seclusion, another school in the same hearing subject to the same law had no awareness of the law and was not planning to change practice (112 LRP 4046). In another hearing, the school was aware of the law and structured the practice of seclusion to avoid its impact. In this instance, the district retrofitted its seclusion rooms in order to make them too large to qualify as seclusion rooms under the legal definition provided by the Texas Education Agency (112 LRP 47218). A single hearing found the use of seclusion to be a violation of the use of conditional procedures under Minnesota law (102 LRP 7834).
In conclusion, the practice of seclusion in these hearings is loosely bounded by IDEA and state law. There is no consensus on whether seclusion is or is not permitted under IDEA. When present, state laws appear to have little impact. The lack of consensus on the process of seclusion is also present in the next level of the conditional matrix, expert recommendation.

Expert Recommendation
Similar to the inconclusive legal finding regarding the practice of seclusion, there was a lack of consensus in expert recommendation on the use of seclusion.
However, most experts either endorsed the process of seclusion by making recommendations around it or explicitly recommended it. One expert within the hearings was firmly and unequivocally against the use of seclusion, and his recommendation was disregarded. This independent evaluator warned the school that there was a serious problem in how the school was handling the student's behavior, as "the use of restraint and seclusion could result in aggression due to fearfulness or perceived or misperceived threat. These aversive measures typically induce greater fearfulness in the child and produce a worse outcome", especially given that the student could bang his head and injure himself. Despite this, the school team placed the student in a therapeutic day school that used a closed door timeout and documented in the student's IEP that student requires a behavior management plan that includes timeout (109 LRP 7122).
The majority of expert opinion throughout the due process hearings in this study endorsed or recommended the use of seclusion, in some cases for one category of disability over another. One behavior analyst testified against the use of a timeout room for a student with autism, because the student was falling asleep in the room and "the time-out room is not having the effect of having the student control his behavior and return to work". This was not due to the ineffectiveness of a timeout room, but due to the student's diagnosis, as "a student with autism is not going to remember why he is in timeout room and it would not be effective" (107 LRP 63423). This sentiment was echoed by a psychiatrist in 106 LRP 10179, who stated "removing a child from the classroom to the bathroom until he quieted down has some use with normally developing children with minor behavior disturbances, however, it is counterproductive with autistic children". A child psychologist in 103 LRP 8586, who when asked why a behavioral intervention plan that included the intensive use of seclusion for up to 60 minutes was not effective, concluded that the student did not truly have the diagnosis of oppositional defiant disorder (ODD), because "the behavioral intervention plan, properly administered by the district, did not provide improvement as it would have if ODD was the correct diagnosis". Some experts gave recommendations or testimony that did not explicitly endorse seclusion, but did not condemn it either, and could be misunderstood or misconstrued. One expert criticized the management of the time-out room as overused, poorly documented, with unclear expectations that resulted in the effective punishment of the student, but did not criticize the procedure itself (102 LRP 12751). Another expert testified in a due process hearing that if a seclusion room were to be used, there should be clearly defined protocol as to the conditions that mandate using the room, exit criteria for the room, and that the room should be "as devoid of material as possible" (107 LRP 8925). His opinion that the student should show 15-30 seconds of quiet behavior prior to removal from the room was countered by the district's behavior specialist, who advocated for placing a child "in some sort of isolation room or timeout room for a period exceeding thirty minutes" if the student was unsafe or had disrobed.
Experts made recommendations that were unclear or could be misconstrued: during one student's initial evaluation, a psychologist report recommended a clear behavior plan, along with "timeout for aggressive behavior" (48 IDELR 26). Some opinions were contradictory, for example a neuropsychologist/psychiatric evaluation requested by the district recommended the following, "systematic ignoring or use of a time-out room can be effective in combination with a positive reinforcement system".
The examiner acknowledges that this approach is "complicated" by the student's claustrophobia, and further states "timeout does not have to necessitate use of a timeout room. Time-out means removal from reinforcement and can involve sitting in a chair in an area in the classroom, sitting in a hallway, being sent to another classroom, etc., as long as these are perceived as nonreinforcing" (103 LRP 8586). It is noted in the appeal (103 LRP 8590) that the practice of timeout for this school turned into "directing student to a 12' by 14' carpeted, office-sized room without windows, equipped with a clock and camera for observation and recording, despite the student's claustrophobia. The hearing officer noted that the same expert noted the student's attachment issues, yet specifically recommended "when implementing timeout, there is very minimal if not any verbal communication until after the time-out is implemented" (103 LRP 8590). In this same case, the district contacted a clinical child psychologist, who found that the restrictive timeout regimen used by the district resulted in extreme behaviors that were likely due to "heightened anxiety related to fears of rejection, abandonment, and failure". Even then, the child psychologist did not remove the time-out plan, instead, the plan was modified and the time-out periods shortened to periods of up to 40-60 minutes (103 LRP 8586). The parent's own advocate declined to ban the use of seclusion, suggesting soundproofing the time-out room to muffle the student's screams. Even when timeout was found to be "not an effective way to deal with the student's behavior" a school disregarded this recommendation and the student was subject to repeated physical escorts to isolated timeouts where he was kept up to four hours (37 IDELR 209).
The majority of experts throughout the hearings recommended the use of seclusion: "isolation with immediate timeouts would be appropriate" (48 IDELR 26) including explicitly recommending a timeout room (102 LRP 8176,37 IDELR 209) and endorsing it as an accepted practice (102 LRP 12751). One expert, with a PhD in psychology, recommended that the district "build a separate room in order to keep staff safe. It is difficult for staff to educate a student when they do not feel safe around the student. The separate room would give the student an opportunity to be observed but would decrease the opportunities for the student to injure the staff" (107 LRP 63423), which the district then constructed over winter break.
Expert recommendations in favor of seclusion creates conditions for the seclusion interactions. These recommendations were made as part of independent evaluation in the special education context, which leads to the next level of conditions: special education settings.

Special Education Settings
The seclusion interactions in this study took place in special education settings and were administered by special education staff. Every single incident of seclusion across the hearings occurred in a special education setting. The majority of cases, 17, took place in special education classrooms located in general education schools. Six hearings took place in special education schools. The placement of students in the special education setting is further evidenced by the significant disabilities the students presented with across the hearings-none of the students presented with mild or moderate disabilities that might by easily accommodated in general education classrooms.

Manifestation of Disability
The students subjected to seclusion had significant and complex disabilities compounded by serious behavioral manifestations, as seen in Table 1. In fact, the term "complex student" was often used in the narratives of the due process hearings: "student is complex, presenting multiple medical, behavioral, education needs and profiles (102 LRP 12751); student has a "history of mysterious, vexing, and severe physical, mental, behavioral, health and educational problems (102 LRP 8176); "student has a complex history of diagnosis of disabling conditions" (106 LRP 19250) and "student is a complex and unusual individual with numerous medical, cognitive, behavioral, social, and educational issues" (106 LRP 10179). Fourteen of the 23 students in the sample were given multiple diagnoses, each diagnosis with significant potential impact, such as a student diagnosed with pervasive developmental disorder, ADHD, hyperkinesis, bilateral cerebral impairment, right hemisphere dysfunction, mild mental retardation, seizure disorder, depressive disorder NOS, epilepsy, oppositional defiant disorder, obsessive compulsive disorder, conduct disorder, autism, delusional disorder, paranoia, psychosis, enuresis, bipolar disorder, posttraumatic stress disorder, and early life trauma (in 102 LRP 8176) and another student diagnosed with ADHD, oppositional defiant disorder, agenesis of the corpus callosum (a rare birth defect in which the structure that connects the two hemispheres of the brain is partially or completely absent), mild mental retardation, hearing loss, processing issues (in 102 LRP 8176).
As a whole, students subject to seclusion in this study presented as vulnerable students requiring a high level of care and supervision. Terms such as "severely impacted" " (109 LRP 23975, 107 LRP 63423), "low cognitive" (106 LRP 19250) "low to no communication" (107 LRP 63423, 108 LRP 33822) and "fragile" (109 LRP 7122) were present in the narratives. Other students were challenging in that they were "twice exceptional" presenting as disabled and gifted (103 LRP 8586, 112 LRP 51016). None of the students in this study presented with what may be considered a mild or moderate disability requiring a low level of special education support.
This was the case in this hearing as the student had just transitioned to middle school when then the behavior that resulted in her being secluded began. The other exception was the negative case of 58 IDELR 267, in which a student with significant physical and mental disabilities was subject to seclusion by an abusive teacher. This student presented with no aggressive or disruptive behaviors, in fact he had minimal verbal ability and difficulty walking due to the effects of his multiple and severe disabilities.
Several students presented with highly unusual and significantly concerning behavioral manifestation of disabilities. To illustrate this point, I will look into 2 cases in depth, 102 LRP 8176 and 109 LRP 7122. In the first case, some of the behaviors displayed by the student include (as listed in the due process hearing): killing cats, growling at people, abnormal threats to painful stimulation, paranoia, delusions, auditory hallucinations, hoarding objects, extraordinary eating habits, ritualistic behaviors, inappropriate sexual behaviors, and using kitchen utensils as weapons.
However, he was also described as polite, wanting to be successful, resilient, streetwise, handsome, neat, social, enjoying music and displaying a sense of humor with unusual memory skills, and good ambulation and communication skills. Despite the unusual behavioral profile of this student, including both his strength and challenges, there was no evidence of a functional behavior assessment, behavior intervention plan, or school safety plan in the due process hearing. Similarly, in 109 LRP 7122, a student diagnosed with an emotional and behavioral disorder (bipolar) and other health impaired (ADHD) as well as significant learning disabilities experienced ongoing psychosis, including auditory hallucinations, extreme anxiety, and rapid cycling of moods that resulted in behavior dysregulation, poor working memory, extremely slow processing speed, slow work completion, difficulty transitioning and low frustration tolerance. As with 102 LRP 8176, notwithstanding the significance of these behavioral manifestations and the fact that he was educated in a therapeutic day school, there was no evidence in the due process hearing of a functional behavior assessment and minimal evidence of a behavior intervention plan. Table 1 provides demographic information on the 23 students, taken directly from the due process hearing and listed in chronological order. When possible, the IDEA eligibility category is listed under disability category. If that information was not provided in the due process hearing, the primary disability is listed. The related disability factors column provides comorbid disabilities, mental and physical health issues, and other manifestations of the student's disability that may be relevant to the process of seclusion. The next level of condition, lack of effective behavior planning, is closely linked to the significant disabilities and behavior manifestations of the students in the hearings.

Ineffective Behavioral Management
There was very little evidence in the sample of clearly articulated, effective behavior management procedures, particularly given the significant disabilities of the students involved. When present, behavior management was characterized by inaction, ineffectiveness and restriction.

Inaction
A great deal of the hearings show inaction. As evidenced by the demographic information on the students represented in the hearings, both in the students' disabilities and their behavioral manifestations, the students represented in these hearings are complicated with significant issues. Despite this, and despite the many problems that schools ran into, a frequent response to these behavior issues was not acting and/or seeking external control.
Not acting. Several school districts clearly identified maladaptive behavior, but took no steps to address it. For example, when transitioning to a new school, despite a history of aggression and a behavior plan used at the previous school, a school team decided that a behavior plan was not needed due to the student's current success at school, which rapidly deteriorated, leading to the use of seclusion, and the student's removal from school four months later (48 IDELR 26). In a second example, the school identified transitions, recess, and the student's impulsivity as behavioral challenges, but provided no evidence of any action taken to support the student through these issues (108 LRP 607). A third example can be found in 46 IDELR 239, in which a high school student with autism exhibited a pattern of predictable behavior, with no evidence of interventions taken by the school, eventually culminating in the student being secluded, staff being injured, the student suspended and the school seeking a more restrictive setting. In yet another case the school team decided not to conduct a functional behavior assessment (FBA) despite documenting that the student was struggling behaviorally and academically, and despite the parent's request for an FBA (109 LRP 7122).

Ineffectiveness
When present, behavior intervention plans appeared to be a poorly understood process. Special education staff members did not seem familiar with the procedures of first conducting a functional behavior assessment (FBA) to determine what is causing the behavior, then the development of a positive, proactive behavior intervention plan (BIP) to decrease the likelihood of the behavior occurring. One step was done without the other: an FBA was done but not a BIP (43 IDELR 151), or a BIP was done without an FBA (46 IDELR 239). At times the procedures were followed but not implemented in a timely manner (48 IDELR 266). Behavior intervention plans evidenced in the sample were also found to be ineffective because they were standard plans that were not individualized to meet the needs of the student (102 LRP 8176, 107 LRP 8925, 37 IDELR 209).
Some were not truly interventions intended to prevent the behavior from occurring, rather, they were a list of procedures to follow after the behavior had already occurred. For example, in 103 LRP 23975, staff members would take the student for walks and use a heating pad as behavior calming methods. In 43 IDELR 151 a behavior intervention plan was developed to help the student transition back into the district, yet, the text of the plan appears to be more of a response to unsafe behavior than a proactive behavior intervention: "BIP stated that staff would supervise student at all times and require she stay in the classroom unless she had permission to leave. When leaving, student would be supervised with line of sight observation. If student were to become a threat to safety of others or herself, a therapeutic hold would be used by trained personnel".
In 112 LRP 47218, the hearing officer found that the student's behavioral goals were not based on a current or comprehensive FBA, and that the response cost level system was not evidence or data based, not individualized, and not understood by the student. As a result, the "failure to reduce problem behaviors often results in the increased use of ever more intense punishers and inadvertently reinforce inappropriate behaviors".

Restriction
This leads to the next category of behavior interventions plans: restriction. As evidenced by the negative attributes given to student behavior in the hearings, schools often took the position that students were capable of controlling behavior and were willfully disobedient. This led to a pattern of restriction and exclusion.
Restriction. When present, behavioral programming was in the form of increased restrictions and punishments. In 108 LRP 33822, the student's IEP was amended a few weeks after he began preschool began to create an "aversive behavior plan". This pattern is evident across several more hearings: in response to several behavior issues exhibited by the student in 46 IDELR 239, rather than addressing the student's behaviors directly, the school restricted the classroom by removing the fish tank and bolting the furniture. In 107 LRP 63423, the student was given a 1:1 aide, then 2 aides to help control his behavioral issues. This student, too, was given an aversive behavior plan. , without consideration of providing related services and supports, as mandated by IDEA. In fact, the student in 107 LRP 8925 was placed in a self-contained classroom at her initial Individualized Education Plan (IEP) meeting, after one month of kindergarten. This trend was noted by the hearing officer in 112 LRP 4728, who stated that the school "failed to consider the alternatives mandated by IDEA such as increasing related services or supplemental aides and services." Two extreme examples of this pattern of exclusion can be found in 37 IDELR 209 and 58 IDELR 267. In the first instance, a student with behavioral issues was subject to efforts by the superintendent to "rid the student from the district", by documenting and photographing behavioral incidents, which were provided to the police along with police reports. In 58 IDELR 267, in which a teacher repeatedly abused her students with significant disabilities, the principal disregarded several complaints made by teachers. She had stated in the past that she didn't understand why the students were in school because "they can't really do anything". Her disregard and the fact that the students with disabilities were confined to a separate wing in the school created a situation in which the teacher was able to commit deplorable acts of abuse over the course of a school year.
In several hearings there was stronger evidence of a focus on restriction and seclusion than positive behavior interventions. For instance, in 48 IDELR 26, the student was subject to a "hand over hand" intervention to increase compliance to tasks, that evolved into a restraint done by up to three people (at the student's legs, waist, and hands) in order to make her color. As the student's behavior and the ensuing intervention escalated, the staff engaged in a discussion of these "holds", some expressing concern and some wanting to try some different holds that weren't authorized by the school. The school psychologist proposed, as an alternative to holding the student, "restricting her integration and making her earn it back" by isolating the student in the calming room, now to be called her "office". This in turn was documented in the student's IEP as a rationale for not including her: "student There were two hearings that showed evidence of behavioral intervention: one contained seclusion as part of a behavior intervention plan (106 LRP 19250). The other began with a behavior support plan developed by a team overseen by the Education Department at the University of Oregon, but then as the student's behavior continued to present a challenge, the district stopped revising the behavior intervention plan and relied on seclusion as it appeared to run out of other ideas (112 LRP 51016).
Although there was little evidence of effective behavior management, there were some references to interventions such as coping strategies ( Many schools confronted with a student with significant disabilities simply didn't know what to do. In 48 IDELR 26, when a parent complained after watching a video tape of her child in seclusion for three hours and 20 minutes, the teacher responded that she was "very upset" and "frustrated" and had "done everything I know how to do" to respond to behavior in "proactive and sensitive ways". Schools were unclear about the legal framework of IDEA. In 37 IDELR 209, the school psychologist testified that a behavioral intervention plan was not required in practice or in the IEP because "positive behavior supports were already being used". In 48 IDELR 26, a teacher testified that she decreased a student's time in general education because the student was too far behind, and "to be considered integration you need to have at least two general education peers working in that same group with the student". This was in sharp contrast to the parents of the students in the hearings, who by and large advocated for positive, proactive behavioral interventions for their children.
The significant manifestations of the students' disabilities in the hearings, compounded by the ineffective behavior management seemed to result in widespread feelings of frustration on behalf of the school staff as evidenced by negative connotations of disability, the next level on the conditional matrix.

Negative Connotation of Disability
From the due process hearings, a theme was clearly apparent: descriptions of student behavior and disability were widely negative. As previously mentioned, the students subject to seclusion in this study were significantly impacted by manifestations of their disability, but were seldom regarded this way. Often, rather than providing factual descriptions of behavior, actions were labeled with a negative attribute, which made it appear that the student was willfully engaging in inappropriate behavior. When discussing behavioral challenges, few hearings provided factual descriptions of student behavior. A factual description of student behavior is, "student became frustrated because s/he missed part of recess while working with the district's occupational therapist. Student threw chairs into desks, pounded her/his desk, and broke a pencil. S/he refused to follow staff directions, and kicked a chair toward a group of other children" (112 LRP 51016). Contrast this with the following description "student was loud in class, the student started acting out because the student didn't want to sit down and do the work" (109 LRP 680980 in which the student's actions are clouded with a negative connotation (loud in class, acting out, didn't want to).
Negative attributes used to describe behavior include: refusing (the most common by far)as in "student refused to do work and follow directions" (102 LRP 7834), "refusing to choose work or break"; defying: "student became angry and defiant" (102 LRP 7834); misbehaving "student misbehaved and ran out of the classroom" (112 LRP 51016); manipulating: student "was very manipulative…saying that his head hurt, or he didn't want to hold the pencil or anything, you know, always everyday it seemed to be something different-that student would just continue to try to get out of doing whatever he didn't want to do" (102 LRP 7834); blaming "parent looking for excuses for his behavior" (102 LRP 7834); annoying: "school psychologist …noted…components of the student's behavior…doing things that were annoying" (102 LRP 7834); complaining: "student complained about expectations of doing anything" (102 LRP 7834); lacking: "he lacks personal motivation" (107 LRP 63423); failing: student "suspended for failure to follow directions", acting out "student really began acting out" (48 IDELR 26) raging: "school psychologist…never observed the student during a period in which he was enraged" (102 LRP 7834); avoiding: "staff opined that student was sleeping to avoid work" (109 LRP 7122); and unacceptable: "district suspended student because his aggression is unacceptable" (107 LRP 63423).
These negative attributions were noted by at least one parent, who complained to an administrator about "the persistent and pervasive undertone that the student is responsible for his failure to learn or make meaningful progress" (109 LRP 7122).
There was one description of a student that did not list negative attributes, "the student did have days where he tried very hard and days when he made up his mind to have a good productive day". However, the next sentence shows that this good behavior is further proof of the willingness of the student's misbehavior "The student was able to control his behavior because he did control his behavior when he wanted to" (102 LRP 7834).
In one hearing, despite the diagnoses of bipolar affective disorder, oppositional defiant disorder, ADHD, major depressive disorder and brain injury, the fact that the student even had a disability was denied by the staff working with him. The paraprofessional working with the student testified at the hearing that he did not believe the student was disabled. This was substantiated by the special education teacher, who stated that student "is not disabled in the school environment". In a phone call with the student's parent she expressed her opinion that "the parent's claim of frontal lobe and brain injury problems was an excuse for allowing him to do whatever he wanted to do" (102 LRP 7834).

Rationale
The negative connotations for disability leads to the final condition for the seclusion interaction: the rationale for the use of seclusion. Seclusion occurred in this sample for a variety of reasons, and frequently in the absence of an imminent safety concern. Seclusion protocols often contradicted the rationale given, and the behavior exhibited by the student when subject to seclusion not only does not justify the rationale, but often made the situation worse.
The rationales for seclusion given in the hearings were disaggregated into the following broad categories, listed below in order of frequency. Each rationale provided in the hearing was counted. Rationale: is presenting a serious, probable imminent threat of bodily harm to the student and/or others". The other instance is less descriptive: "not used for minor infractions". There was a single instance of the term "only in emergency circumstance" used, as part of a behavior intervention plan. One hearing (108 LRP 607) contained information on the district procedure for the use of seclusion as an emergency procedure, however, in practice the student was secluded for disruptive and inappropriate behaviors. Several hearings contained the use of seclusion as part of a behavior intervention plan (107 8925, 106 LRP 19250, 103 LRP 8586, 112 LRP 51016, 108 LRP 607, 112 LRP 4046) and as part of an IEP (108 LRP 33822, 106 LRP 19250, 103 LRP 8586).
Of further concern are the many other reasons given for secluding students, such as noncompliance, not in area/classroom, defiance, verbal aggression, name calling, as a consequence or punishment, being off task, for fun, and as a form of abuse. In two instances seclusion was used because other (aversive) interventions weren't working. Further analysis found that students were physically escorted into seclusion (103 LRP 8586, 102 LRP 7834, 112 LRP 51016, 108 LRP 607, 108 LRP 33822, 106 LRP 19250). There were two instances of students voluntarily entering seclusion (112 LRP 51016, 102 LRP 7834).
Another finding was that in some cases students are secluded as a result of the manifestations of their disabilities. One illustration involves a due process hearing regarding a single incident of seclusion. In this case a high school student with autism was noted to have "diminished motor, social, coping and verbal skills". Due to his autism he had difficulty dealing with changes in routine, leading to frustration which the student would show by throwing small objects. On the day he was secluded in the "quiet room" in his classroom, he was upset because he had missed a hockey game over the weekend and he knew his mother was home from work that day visiting with relatives. He told his father that he did not want to go to school, then told his bus driver he was sick. As he entered his special education classroom that morning he told his teacher he was sick and gave her his mother's phone number. As his teacher called his mother, the student threw a pencil. His teacher directed him to go to the quiet room. Once in the quiet room his behavior escalated and he began to engage in property destruction, which resulted in a staff member holding the door closed. The student then bolted towards the door, was able to partially get out, at which time the staff member let go of the door. The student continued to escalate, eventually striking his teacher and received a 10 day suspension. In this situation, the behavior that resulted in the student's seclusion and eventual suspension was a clear and predictable manifestation of his communication, social, and coping deficits due to his autism. The behavior he engaged in, throwing a pencil, was a known response to his handling of frustration. It was not dangerous, yet led to a cascade of events that had serious ramifications for the student, his family, and the school.
Another strong example of this involved the systematic use of seclusion in 109 LRP 68098. This student, an adolescent of unknown gender with an emotional and behavior disorder ( including ADHD, oppositional defiant disorder, major depressive disorder, PTSD, a learning disorder and a visual impairment), was subject to seclusion for refusing to do assignments. The student's parents testified that they were not aware of any incidents of the student being secluded (and restrained) due to aggression.
Work refusal is a typical and expected behavior for students with learning disorders and oppositional defiant disorder, however this case went even further. It was known that this student had a 75% vision loss in the right eye. The student reportedly refused to complete any writing activities, and informed the classroom teacher that the student's "hand did not connect to student's brain, and the student could not write the information from the student's brain". Despite this, the student was secluded in the isolation room for failing to complete assignments then kept in the isolation room until assignments were completed. Later assessment showed that the student was reading at a post high school level yet writing at a 4 th to 5 th grade level. Further evaluation revealed that the student had a visual processing disorder that impacted the student's ability to write, requiring 12 months of visual information processing therapy. As with the others, in this case there was no evidence of a functional behavior assessment or behavior intervention plan, nor was there evidence of consideration of assistive technology as an accommodation or any classroom modifications to alleviate the student's challenge with writing. Instead, the student was secluded for behavior that was a manifestation of not only his learning and emotional behavioral disability, but a visual impairment.

Interactions
The seclusion interaction is characterized by the room in which students are being secluded, the term given for the process of seclusion, the frequency and duration of the seclusion event, exit criteria, and mental and physical health contraindications.

Room
The rooms being used to seclude students are built for this purpose, with features such as locks, viewing windows, padding, carpeting (including on sides of walls), and cameras. Five rooms in the sample were intended for another purpose and then used for seclusion (former coatroom, bathroom, and in the negative case several rooms were used to seclude the student: the adaptive art room, the handicapped bathroom, and the adaptive PE room).

Term for Seclusion
The practice of seclusion takes place under many different terms. Within the hearings, there were a total of 30 different terms used 55 separate times. Of those, four terms referred to a practice, (timeout, timeout intervention, isolated timeouts, and isolation) and 26 terms referenced the use of a room or area. Only three terms clearly referenced the practice of seclusion (seclusion, seclusion booth, and locked confines).
Often multiple terms were used throughout the hearings, although the practice being described clearly fit the definition of seclusion, there was no consensus on what to call it. It appears that there is a desire to couch the practice of seclusion within more acceptable terms. The terms found with the hearings can be divided into the following three categories: neutral terms, positive terms, and misleading terms.

Neutral Terms
Most of the references involved terms that could be considered neutral, such as: timeout, timeout rooms, (the two most commonly occurring) and variations of the term timeout: isolated timeouts, timeout interventions, timeout facility.

Positive Terms
Many of the terms used put a positive connotation on the practice of seclusion, such as quiet room and safe room (the two most commonly occurring), followed by recovery room, focus room, calming room, calming area, individual instruction room, sensory room, and safe room area. Again, although the process being described clearly fit the criteria of seclusion there seemed to be an effort on the part of the school staff to not only obfuscate the process that was occurring, but to put a positive spin on the process itself.

Misleading Terms
The remaining were terms that appeared to be deliberately misleading such as: work room, student's office and student's den. Other terms, despite the fact that the rooms used were clearly built for the practice of seclusion, instead referenced a vague type of room: separate room, room adjacent to the classroom. Acronyms were used in place or terms, adding increased ambiguity: "QR" for quiet room, and "IIR" for individual instruction room. In one hearing the school used the deliberately misleading term "office referral" for the practice of seclusion (112 LRP 47218). Table 2 contains the terms used to define seclusion, as compared to descriptions of the rooms from within the hearings. For hearings in which the student was secluded in more than one setting, both settings are listed. just two hearings contained a single seclusion incident. One trend noted was that after an initial incident of seclusion, there was an increase in the frequency of seclusion-in only 1 case was there was a claim that there was a decrease in incidents over time (107 LRP 63423).

Duration
There were two ways in which these data were reported in the hearings; average length of time and total length of time. In the three instances in which duration was reported as an average, all three were reported as an average of close to 30 minutes per day. However, in terms of total durations most were reported as an hour or more: two instances were reported as less than an hour, five instances of 1-3 hours, four instances of 3-6 hours, and four instances of 6 + hours, often longer than the school day. There was evidence of students being kept afterschool and into the evening (as late as 8:00 p.m.) to "finish" the seclusion time, (103 LRP 8586, 112 LRP 47218) as well as students being placed in seclusion the following morning for the same reason (48 IDELR 26,112 LRP 47218). Two of these cases involved strict exit criteria combined with significant levels of disability, specifically multiple disabilities (103 LRP 8586,48 IDELR 26). In the other, (112 LRP 47218) school staff implemented a "no limit time stealing" strategy, in which the student could be placed in seclusion for extended periods, even afterschool, with the intent of increasing the "severity of the punitive isolation for misbehavior with the desired effect of encouraging the student to make better behavioral choices". Table 3 contains information on the frequency and duration of seclusion. Many of the hearings did not include information regarding exit criteria, it is unclear if there were no exit criteria or if it was not included in the due process hearings.

Contraindications
The process of seclusion takes place despite evidence of significant mental and physical health contraindications. Out of the 23 total students within the hearings, 16 of the students in this study presented with a wide range of mental or physical conditions that should have contraindicated seclusion, (see Table 4).  The seclusion interaction resulted in several negative short and long term outcomes for students, including a placement trajectory that moved students into more restrictive settings.

Short Term Effects of Seclusion
Regardless of the rationale given for using seclusion, it appears that when students are placed in seclusion there is a significant escalation in behavior. Many students attempt to get out of the room, by banging themselves against the door, hitting or kicking the door or viewing window, trying to escape, crying and begging staff to let them out. Several students engaged in what may best be described as physical decompensations (listed in order from most to least frequently occurring): urinating, spitting, disrobing, vomiting, defecating, feces smearing, and masturbating.
One hearing showed evidence that the student was given water and bathroom breaks (107 LRP 8925). Some students engaged in extremely unsafe behaviors while in seclusion, such as head banging, attempting to scratch face, pulling hair out, and attempting to choke self. This extreme unsafe behavior led several schools to call in an outside resource such as police (102 LRP 7834), an emergency mental health evaluation, (102 LRP 7834,109 LRP 57374), and in one case the extreme behaviors exhibited in seclusion by the student prompted the school to file a child abuse complaint against the parents, which was later shown to be unfounded (103 LRP 8586). Three hearings showed evidence of schools contacting parents regarding the seclusion event (112 LRP 63423, 107 LRP 8925, 103 LRP 8586).
There was no evidence of the act of seclusion resulting in an immediate deescalation of student behavior or evidence of an immediate decline in unsafe behavior.
The only evidence that seclusion may have a calming effect on students was when students fell asleep, although at times this occurred after what appears to be a lengthy period of behavioral upset. In fact, there was evidence in six cases of the opposite: when placed in seclusion overall behavior patterns worsened and the process of seclusion became iatrogenic. An independent education evaluator found that the use of seclusion escalated the student's behavior (102 LRP 7834). In 103 LRP 8586, the hearing officer found "as student was placed in timeout almost daily her reactions became more severe, including hitting, kicking, or acting out at school personnel who were putting her in the timeout room", which was further exacerbated by her hypoglycemia and the delay or denial of lunch. Similarly, in 112 LRP 47218, "the student seems to be learning how to be more defiant and aggressive in response to longer office referrals (seclusion) and 'no limit stealing' disciplinary actions".

Long Term Effects of Seclusion
There was evidence of some negative long term effects of seclusion: most frequently post-traumatic stress disorder or related behaviors (such as nightmares and bed wetting), followed by fear of school, physical injuries, and suicidal gestures.
There was one instance of the student reportedly unaffected by the use of seclusion (109 LRP 68098). Table 5 contains information on the rationale given for seclusion, the student's behavior during seclusion, the criteria for exiting, and the effects of seclusion. arguing, and turning around pleaded with staff to let student out, to go home, to see student's mother, to turn down music, and stated student was injured, subject to ants crawling on him

Placement Trajectory
Across 13 of the hearings, half of the total sample, the process and iatrogenic effects of seclusion resulted in a placement trajectory that moved students into more restrictive settings, either into self-contained schools or removal from school, due to a temporary arrangement or homebound placements.

DISCUSSION
From this analysis, the conditions under which students with disabilities are subject to seclusion emerged as several layers of poor special education practice.
These are, in order of proximity and relevance to the seclusion event: a loose legal boundary, expert recommendations, special education settings, significant manifestations of disability, ineffective behavior plans, negative attributions of disability, and multiple rationales (see Figure 1).
The practice of seclusion is not bounded by IDEA or state laws regarding seclusion. The relationship between IDEA and the practice of seclusion is undecided, with due process hearing officer decisions widely spread along a continuum of requiring it for some students in order to receive a free and appropriate public education to banning it entirely. Placement decisions were just as divergent, with hearing officer decisions' made on placements in favor of the availability of seclusion rooms to placements that do not use seclusion rooms. Several hearings contained decisions regarding the practice of seclusion as part of a behavior support plan and were equally undecided along a continuum from being a necessary component of a behavior intervention plan to being harmful to students. There was minimal evidence of the impact of state law regarding the practice of seclusion, with school evidencing little awareness of the law or efforts to circumvent it. The use of seclusion was often recommended by outside experts consulting with schools. If not recommended, it was endorsed in that suggestions were made to alter, but continue, the practice of seclusion. In one instance seclusion was stated as contraindicated for a student, however, the school disregarded this recommendation. Seclusion events occurred exclusively in special education settings and were administered by special education staff, most often within a general education school. The demographic information provided by the due process hearings present some consistent and concerning findings: vulnerable students with a high level of diagnostic complexity are being subjected to seclusion for behaviors that are in many cases behavioral manifestations of their disabilities. There was little evidence of effective behavior management throughout the due process hearings. Schools met the behavioral challenges that manifested from the significant disabilities present in the students through: inaction, exclusion, and restrictive practices. Additionally, descriptions of student behavior were characterized by negative connotations. Behavior was regarded as willful disobedience as opposed to evidence of a skill deficit that should be addressed through positive behavioral intervention. There were 21 different rationales for the use of seclusion in the sample, the majority of which did not contain a safety concern.
In these hearings, the seclusion interaction is characterized by the room in which students are being secluded, the term given for the process of seclusion, the frequency and duration of the seclusion event, exit criteria, and mental and physical health contraindications.
Students were secluded in rooms built for the purpose of seclusion. Despite the clear intent of these rooms, the practice of seclusion is referred to by multiple terms, including neutral, positive, and misleading terms. Students in the sample were subject to high rates of seclusion for multiple hours. Moreover, in several cases students were kept afterschool or placed back into seclusion the following day to "finish" the seclusion event. This was often due to strict exit criteria. Exit criteria that were present in the hearings often contained a compliance task, quiet behavior for an extended period of time, and did not always relate back to the rationale given for seclusion.
Additionally, many students with contraindicating mental and physical health issues were subject to seclusion, and in some cases, locked door seclusion with what may be low levels of supervision.
There were significant negative short and long term effects of seclusion present in the sample. Once students were placed in seclusion rooms, there was an escalation of student behavior, at times resulting in physical injuries. Behavior escalated to the point at which schools called outside authorities such as police and emergency mental health centers. There were accounts of post-traumatic stress disorder as the result of seclusion. After an initial seclusion event, the practice of seclusion became iatrogenic, placing students on a trajectory out of educational settings that might have been successful, had procedural safeguards been followed, effective behavior management been used, and better working relationships with parents been present.

Interpretation
Several of the findings support existing literature on the practice of seclusion.
First, the practice of seclusion is referred to by many different terms. There were 30 different terms used for the practice of seclusion in the sample, consistent with many different terms present in the literature (Busch & Shore, 2000;CCBD, 2009;Day, 2002;Endres & Goke, 1973;TASH, 2011;Yell, 1990;Yell, 1994). The practice of seclusion was rarely referred to as seclusion, most terms seem to neutralize or obfuscate the practice of seclusion, or make it appear to be a positive experience. This is an important finding for several reasons. First, the use of many terms for seclusion confuses the practice. This may be deliberate. In a recent report issued by the United States Senate Health, Education, Labor, and Pensions Committee, investigating ten cases involving the use of seclusion and restraints in schools, the investigation found evidence of "code of silence" in which school deny or downplay the actual circumstances under which students are subject to seclusion and restraint. The terminology found in the sample stood in direct contrast to the seclusion interaction (see Table 2). The majority of the rooms used for seclusion in the sample were built for the purpose of seclusion, and outfitted with padding, viewing windows, cameras, and locking mechanisms, but the practice of putting students in these rooms was referred to by a wide variety of misleading, positive, or neutral terms. Second, the seclusion experience was neither neutral nor positive in the sample. Upon being placed in seclusion rooms, students experienced behavior escalations and physical decompensations, as evidenced by: banging themselves against the door, hitting or kicking the door or viewing window, trying to escape, crying and begging staff to let them out, head banging, attempting to scratch face, pulling hair out, and attempting to choke self, urinating, spitting, disrobing, vomiting, defecating, feces smearing, and masturbating.
The escalation in behavior makes it had to justify the use of seclusion, even with the rationale of an emergency situation. However, in the sample there were 21 different rationales given for the use of seclusion, the majority of which did not include a safety component. This too, is consistent with literature showing several different rationales for the use of seclusion (Busch & Shore, 200;CCBD, 2009;Ferleger, 2008;Gast & Nelson, 1977;Piersma, 1980;Ryan et al., 2008;Ryan, Sanders, et al., 2007;Yell, 1994).
Another finding consistent with the literature was the overall negative impact of the use of seclusion; there was strong evidence of the iatrogenic effects of seclusion, supported by Arthur (2008) and Ryan, Sanders et al., (2007). There was evidence of the diagnosis of post-traumatic stress disorder as a result of the seclusion interaction, supported by Finke (2001), Gutheil & Tardiff (1984), and Westling et al., (2010), as well as physical harm to students, supported by CCBD (2009). The practice of seclusion in schools may have its genesis in special education (Arthur, 2008;, and it continues to be a special education practice. All of the seclusion interactions in the sample took place in special education classrooms and were administered by special education staff. This is consistent with the surveys done by COPAA (2009) and Westling et al. (2010). The students subject to seclusion presented with significant, complex, and co-existing disabilities. This, too, is consistent with a recent U.S. Senate report (2014), which also found that the presence of significant student disability combined with a lack of parent notification impeded the parents' ability to advocate on behalf of their child. Another impediment identified by the report and unique to special education was what was called "halo" effect'the perceptions that teachers, particularly special education teachers are experts in the field and are overwhelmed by behavior issues, and should not be challenged in their behavior management strategies.
The goal of this study was to discover the conditions under which students with disabilities are subject to seclusion in schools, and the widely negative findings call for some immediate changes in practice. First, the finding that students with significant physical and mental health contraindications are being secluded, including locked seclusion, should be addressed. Students with contraindications such as: asthma, seizures, hypoglycemia, enuresis, trauma histories, claustrophobia, selfinjurious and suicidal behaviors should not be subject to routine and/or unsupervised seclusion (see Table 4).
Second, the only rationale that may justify the use of seclusion is a situation with an imminent safety threat for the students or others. Even this justification, though, must be carefully considered in that it should be expected that upon being placed in a seclusion room students will suffer an escalation and/or decompensation in behavior. Put another way, the process of seclusion will make the behavior worse in the short term, not better. Furthermore, the characteristics of the student must be taken into consideration, such as age, size, and manifestation of disability.
Third, the seclusion event, when used in case of an emergency, should end as soon as the safety concern has passed. Staff should be able to judge this in flexible ways-students should not be required to engage in strict compliance tasks in order to be allowed to leave the seclusion room.
Fourth, as with short term effects, it may be expected that the use of seclusion will become iatrogenic and lead to an increase in maladaptive behaviors. Incidents should be documented and monitored by school staff, IEP teams, and potentially a human rights committee composed of school staff as well as outside members.
Positive behavior interventions should be ongoing, and outside resources should be considered in the event there is an increase in the use of seclusion for an individual student.
Fifth, the process of the involuntary confinement of a student alone in a room or area from which the student is physically prevented from leaving must be referred to as seclusion. In order to effectively discuss and monitor the process of seclusion its intent must be clear, and it must be clearly communicated to parents. It is imperative that consistent terminology is used for a procedure with so many potential detrimental effects. Alternatively, the process of seclusion should not be confused or confounded with other practices, such as timeout and the use of sensory rooms. Clarity and consistency in the use of terminology about the use of seclusion may decrease the likelihood of schools misunderstanding or misconstruing expert recommendations regarding practices such as timeout and the use of sensory rooms.
The findings in this study extend the recommendations made for reducing the use of seclusion. Most of the recommendations focus on surface level interventions, such as staff training (Greene et al., 2006), intervention models (Martin et al., 2008) and technical assistance (CWLA, 2004). This study provides evidence that a deeper level of training may be needed to address the conditions that lead to the process of seclusion. This study found staff expressing negative attitudes around manifestations of disability, and often describing behavior in negative terms. In at least one case the actual presence of a disability was denied by the staff, including the special education teacher, working with the student. The students in the sample presented with complex and significant disability, and often were diagnosed with more than one disability.
These students present as needing intensive levels of support, which was not evidenced in the hearings. Instead, school staff seemed to interpret behaviors as willful disobedience, leading to punitive overtones in their disciplinary approaches. This occurred across many levels in the sample, from paraprofessionals working with students, to teachers, principals, all the way up to the superintendent level. Staff understanding of and attitude toward disability and its impact on the use of seclusion may need to be addressed as part of reduction efforts.
Additionally, there was little to no evidence of effective behavioral intervention in the sample, as well as some misunderstandings around behavior intervention, inclusion, and IDEA regulations. The finding that seclusion was used in situations with ineffective behavior plans was consistent with the surveys done by COPAA (2009) and Westling et al. (2010). Positive Behavior Interventions and Supports (PBIS), has been suggested as a proactive intervention to prevent the use of seclusion and restraint (Rozalski, Peterson, Ryan & Losinski, 2013), however, more attention may need to be paid at the highest and most individualized level of support needed for the most intensive tier of students.
A new finding was the fact that experts are recommending and/or endorsing the use of seclusion. This is surprising, given the fact that there is no therapeutic value to the practice of seclusion (Ferleger, 2008), there is minimal research done in this area in schools, and the fact that it is not an evidence based practice. This is a finding that needs further exploration.
An argument has been made that seclusion and restraints are not just a special education issue, but evidence is mounting that special education owns a large piece of it, according to OCR data (http://www2.ed.gov/about/offices/list /ocr/docs/crdc-2012data-summary.pdf). As such, it is a logical idea that IDEA should contain some specific regulations regarding the use of seclusion. This may be an area where the definition of seclusion is determined, in order to promote consistency in terminology and understanding of practice. The relation of seclusion and the legal guarantees of IDEA should be addressed in regulation, for the following reasons. Within this study, there were widely divergent IDEA decisions made in relation to complaints made about the practice of seclusion, with no discernable pattern. This finding is supported by the U.S. Senate report, which found inconsistent legal results, depending on the state in which the event occurred and the underlying legal claim. This report discusses the limited effect of current law about seclusion in detail, including the challenges parents face when advocating on behalf of their children. These include the requirement that IDEA remedies be exhausted in order to pursue legal action at the court level, often necessitating the removal of students from their current setting into a private placement. This was consistent with the placement trajectory of students within this sample, who were often removed to more restrictive settings by the parents or the schools. Often the seclusion issue is resolved when the student leaves-there is no consequence for the school and no building of capacity to handle the next student that comes through the door with significant behavior challenges. As the U.S. Senate report points out, this leaves schools with very little incentive to change practice, and furthermore, "the fact that the parents were unable to halt the use of these practices directly contradicts IDEA's mandate to provide children with disabilities a free and appropriate public education" (U.S. Senate, 2014, p. 16).
Some stronger recommendations include banning the use of seclusion as a behavior intervention in an IEP, and amending IDEA to allow families to file civil actions to stop the use of seclusion before exhausting remedies under IDEA.
Consistent with the findings in this study, the U.S. Senate report found that state laws on seclusion and restraint "have not had a discernable impact on resolutions of these cases to date. Few state laws or regulations provide parents with an effective enforcement mechanism regarding the use of seclusion and restraints" (p. 27). The U.S. Senate report recommends legislation to remedy this finding, but it still may not be enough. First and foremost, restricting restraints to emergency situations and banning the use of "all unsupervised and unmonitored use of" seclusion still leaves a gap, which school are likely to circumvent, in that if students are supervised and monitored they may still be involuntarily confined in a room or area from which they are physically prevented from leaving. The legislative recommendations also include: data collection on the frequency, durations and intensity of the use of seclusion at the local, state, and federal level with the ability to disaggregate data at the school level. It does not mandate who is ultimately responsible for this data, nor does it provide any oversight or enforcement mechanisms. It contains a recommendation that training programs focus on prevention and positive programming, as evidenced by these findings training may have to go deeper. It recommends parental notification within 24 hours, but provides no enforcement or monitoring mechanism to guarantee compliance with this mandate.
Seclusion has a lengthy and controversial history in the mental health field.
The negative short and long term effects of the seclusion event on students are significant and far outweigh any temporary benefits to the use of seclusion, and may undermine the safety rationale for its use. The practice of seclusion has clearly been established in school across the United States. It is imperative, for the psychological and physical health of children that we establish limits to this potentially harmful practice.

Limitations
This is an exploratory study in an area with very little research, and as such presents with limitations, most of which center around the data used for analysis.
The goal for this study was to gather information on what practices are actually occurring in schools nationwide, within the context of an IDEA framework. Moreover, a practice as controversial as seclusion is not easy to gain access to, and presents with challenges in passing the Internal Review Board process. In that regard, due process hearings presented an opportunity to cast a wide net, and to even go back in time to see if practices have changed over time. However, due process hearings are not examples of best special education practice, rather, they are evidence of a disagreement regarding educational practice that has taken place between parents and schools. As such, they are not a representative sample and likely to present with practices that are flawed. Due process hearings are generated by each states' education agency, therefore there was no control over how the narratives were written and what types of information was included, or excluded. It should be pointed out, that, despite the limitations in sample, there was consistency in the seclusion interactions that took place across the hearings, which spanned a decade and took place nationwide.
Additionally, this was a small sample of hearings that fit criteria. A different data source may show different findings.
In considering the nature of due process hearings, however, there is another aspect to consider. Not every disputed issue in schools becomes a due process hearing.
In order to be brought to the level of a due process hearing, parents must be aware of the issue, have the desire and ability to oppose the school about the issues, and pass through the mediation process in order to result in a due process hearing. A criticism of this process has been: "many parents of exceptional children do not know or understand their rights and thus do not exercise them; others do know their rights, but for a variety of reasons are unable to exercise them. Children of such parents remain vulnerable, as if the protections did not exist" (Zettel & Ballard, 1982, p. 7). The small sample in this study did not capture the extent of the use of seclusion in schools, and given the overall negative findings, this is an area that should be explored in further research.

Suggestions for further research
As an initial study, this area opens up several lines of research. First, the findings should be compared against a different data source, such as incident reports, surveys, or, when possible, observation. There is also a large number of IDEA complaints in the IDELR database, often with detailed descriptions of seclusion rooms and the seclusion process that may be used for a similar study to confirm, extend, or contrast these findings. Given the negative impact of the seclusion event in the sample, Another area for further investigation is the impact of seclusion on students who witness it and staff who administer it. It is my belief that the misuse of seclusion is not evidence of bad people, but bad practices that become the norm, and have lasting impact on the people doing it. I never planned on being a special education teacher that placed a student in the seclusion room and held the door closed. I inherited a seclusion room that had already been built, and when I did, my first thought was, can you do that? I never liked using the seclusion room and I never became comfortable with it. And yet I did use it, more times than I can count and many more times than I'd like to admit. My unease led me to this dissertation and my findings disturb me now. Might the impact on staff be another area to explore in reduction studies? Some research shows that staff who use seclusion more often