Restraint and seclusion practices in American schools for students with disabilities

¶ … EBD Students

Time-Outs in the Classroom

Time-Outs for Students with Emotional and Behavioral Disorders

Time-Outs for Students with Emotional and Behavioral Disorders

On the second page of a 2010 report published by the National Disabilities Rights Network (NDRN) called School is Not Supposed to Hurt is a picture of a 7-year-old girl who died while being restrained and secluded in a Wisconsin school. This report went on to describe the wide-spread used of restraints and seclusion by schools in the United States and its publication triggered a congressional investigation. The Government Accountability Office (GAO, 2009) published its own report a few months later, which examined 10 court cases resulting in criminal convictions, civil adjudications, or settlements. These 10 cases formed the basis for judging the veracity of hundreds of allegations of mistreatment, injuries, and death resulting from children being restrained or secluded by school personnel. Even more troubling was the finding that no federal law regulates the use of restraints and seclusion on children, while state statutes vary widely in terms of existence, content, and effectiveness.

The GAO defined a restraint as "… any manual method, physical or mechanical device, material, or equipment that immobilizes or reduces the ability of an individual to move his or her arms, legs, body, or head freely" (2009, p. 1). The definition for seclusion was "… involuntary confinement of an individual alone in a room or area from which the individual is physically prevented from leaving" (GAO, 2009, p. 1). The 10th case examined by the GAO (2009) involved a 9-year-old boy with a learning disability who was secluded several times in a 'time-out' room by school staff in New York. The parent-approved individualized education plan (IEP) for this child included the use of a time-out room as a last resort, but school records revealed the room had been used 75 times during a six-month period. Some of the behaviors the teachers felt justified the use of the time-out room was whistling, slouching, and waving hands. Each time the child was escorted out of the classroom in from of his peers and when placed into the time-out room a staff member would physically hold the door closed. On at least one occasion the child's hands were blistered from trying repeatedly to open the 'unlocked' door. The size of the room was no more than a broom closet and lined with torn padding. Once the mother became aware of what was occurring she requested that her child be transferred to a different school and a jury awarded the parents $1,000 for each seclusion event, plus attorney fees.

The NDRN (2010) report described an IEP including time-outs in response to aggressive behavior for an autistic child. This child's parents eventually become aware of the fact that their child was spending about 75% of each school day in a time-out room (2010, p. 25, 38-39). A 9-year-old girl with ADHD, mild intellectual disability, and an anxiety disorder had tried to run away from school, so the teacher responded by placing the student in time-out under the teacher's desk. Often parents realize something is wrong when their child returns from school traumatized or injured and is unwilling to return to school voluntarily. Although the use of restraints, seclusions, and time-outs were originally intended to be therapeutic for children, the GAO (2009) discovered that teachers and staff untrained in the use of these methods were often relying on them to control non-aggressive behavioral problems exhibited by disabled children.

Unfortunately, the examples of maltreatment of disabled children by teachers and staff described in the GAO (2009) and NDRN (2010) reports are representative rather than exceptional. The response at the federal and state levels has been slow and limited, although many legislative bodies in the United States seem to agree that the use of restraints and seclusion should be banned outright or severely restricted. The use of time-outs, however, is still a widely accepted method of behavioral control. According to Kerr and Nelson (2009) there are three levels of time-outs used in a primary school setting: (1) planned ignoring, (2) contingent observation, and (3) exclusionary/seclusionary time-out (p. 82-83). These three levels represent a continuum from the least to most restrictive, with the latter triggering the most serious ethical and legal concerns. To better understand the use of time-outs as a classroom management tool for children suffering from emotional and behavioral disorders (EBD), this essay will examine what is known about this topic and review the findings of several relevant research studies.

The first topic to be discussed will be the theoretical justifications for the use of time-outs in the classroom, with a focus on EBD children. Textbooks dealing with managing student behavior will provide a foundation for this discussion. This is followed by an examination of empirical research studies that have investigated the efficacy and utility of time-outs in the school setting and which variables control outcomes. The theoretical foundations and empirical findings will then be discussed and conclusions drawn.

Literature Review

Variables Determining Time-Out Efficacy

The ideal classroom approach to time-outs, according to Grskovic and colleagues (2004), maximizes the amount of time a student remains engaged in classroom activities by minimizing time spent in time-out. The sequence of events surrounding a time-out are the following: (1) student engages in inappropriate behavior, (2) teacher issues the time-out, (3) delay between teacher mand and student beginning the time-out, (4) student performs time-out, (5) termination of the time-out, and (6) student reentry into classroom activities. Time-outs can also include contingent delay of time-out termination, teacher-student review of time-out justification, and alternative behavior suggestions, with the latter two occurring between termination of the time-out and before the student reenters classroom activities. Contingent delay represents time-out duration contingent upon a student's compliance with a time-out mand, such that it may be increased if the student is noncompliant and/or continues to engage in inappropriate behaviors (Mace, Page, Ivancic, & O'Brien, 1986).

Grskovic and colleagues (2004) attempted to reduce the frequency and length of time-outs by increasing the efficacy when administered. They implemented an Active Response Beads-Time Out (ARB-TO) intervention in a classroom populated by 11 students with ADHD, ranging in age from 7- to 12-years. When a student was given an ARB time-out they were required to retrieve a string with 10 wooden beads from the teacher's desk, return to their desk, and begin to count down from 10 to 1. With each breath, the child would move a single bead from one end to the other. All children were instructed in how to use the bead system during a time-out. Implementation of the ARB-TO intervention required teachers to praise the student for successful completion of a time-out task and verbally reengage the student in classroom activities.

What the authors of this study found was that both outcome variables, time-out frequency and duration, decreased dramatically regardless of whether the student was given a typical 10 second time-out with head down on the desk or an ARB-TO (Grskovic et al., 2004). The authors attributed this change to the implementation of ARB-TO, which then had the effect of reducing time-out outcome variables even when teachers administered a typical time-out. This possibility is supported by a dramatic increase in ARB-TOs in the first several classes, probably due to student interest in playing with the beaded string. Another important finding is that students remained engaged in a task during the time-out, which probably suppressed inappropriate behaviors and thus the prevalence of contingent delay. This may have also reduced the prevalence of escape behaviors, which would help explain the dramatic decrease in time-out frequency and duration compared to baseline. The main limitation of this study was the absence of an outcome variable quantifying the amount of time teachers and students were engaged in instructional activity. Despite this limitation, anecdotal teacher and student reports indicated general satisfaction with the ARB-TO strategy.

What is remarkable about the findings of Grskovic and colleagues (2004) is that a 10 second time out was sufficient to significantly reduce the incidence of inappropriate behaviors in children with ADHD. This is quite a bit shorter than the 5 and 15 minute time-outs examined in the study by Fabiano and colleagues (2004). More recently, researchers examined the effect of time-out duration, but from the perspective of rewarding quick compliance with a shorter time-out (Donaldson, Vollmer, Yakich, & Van Camp, 2013). Six children between the ages of 3 and 5 with behavioral problems were included in the study. The time-out settings used in the study were school recess and free-time at home. The settings choices would have eliminated the risk of escape behavior skewing the study's findings, because time-outs would be expected to be universally aversive in these settings. When the children engaged in an inappropriate behavior there were given a 4-minute fixed time-out or a 4-minute time-out reduced to 1 minute if the child complied with the mand within 10 seconds. Both time-outs reduced problem behaviors to almost zero, but the contingent time-out duration was associated with an improved compliance. The limitations of this study are the small sample size and the non-quantitative analysis, thereby precluding generalization to the wider community of EBD children. Despite this limitation, these findings do suggest that time-out duration is significantly aversive.

Shortening time-outs increases the amount of time a student can remain engaged in classroom activities. Another method that has been tried is the use of time-out ribbons (also known as time-in ribbons). Yeager and McLaughlin (1994) evaluated compliance rates for a 4-year-old child in pre-school special education before and after the ribbon intervention. The ribbon is awarded to a child for compliance with classroom activities, which confers eligibility for rewards. If a child engages in a problem behavior then the ribbon is removed during the time-out period, in addition to withholding reinforcement. An ABACBC single subject replication study design was used, with A, B, and C. representing baseline, time-out ribbon, and being read to by the teacher as a reward for compliance, respectively. Baseline compliance rates were never higher than 13%, but the first ribbon time-out phase increased compliance as high as 78%. Compliance rates during the second ribbon phase, however, never went above 20%. Using 'teacher reading' as a reward improved compliance to a high of 100%. The poor performance of the ribbon time-out method during the second exposure led researchers to conclude that additional parameters were controlling child responsiveness to the intervention. In support of this conclusion, the teacher informed the researchers that the child had begun to develop seizures during the latter phases of the study and underwent changes in medication and living situation. The use of a single child as a research subject therefore precluded drawing any conclusions about the effectiveness of time-out ribbons for increasing time spent engaged in classroom activities.

The suggested advantages of the time-out ribbon procedure are that the child remains present for classroom activities and is required to continue working (Kostewicz, 2010). The use of the ribbon method also tends to discourage teacher overreliance on more coercive strategies for classroom management, while encouraging improved time-in reinforcement quality. Kostewicz (2010) conducted a systematic review of the research literature on this topic and found 6 studies that met inclusion criteria. All studies reviewed revealed a reinforcement-rich time-in period, which was necessary for increasing the effectiveness of time-outs. The reinforcers used by the studies varied, but the most consistent variable was frequent use of reinforcers during time-in. The author concluded that improved student behavior also tended to reinforce teachers' willingness to provide frequent reinforcement. The other notable conclusion reached by the author is that problem behaviors can be reduced through a time-out punishment, including noncompliance, but the incidence of appropriate behaviors tends to be affected less. Time-out punishments therefore function primarily by reducing problem behaviors.

Although time-out from reinforcement is an effective behavioral intervention, addressing noncompliance after a child is instructed to engage in a time-out can be frustrating. Warzak and Floress (2009) examined the efficacy of a deferred time-out (DTO) strategy with two children between the ages of 4 and 5, who were referred for noncompliance, verbal aggression, and minor physical aggressive behaviors. DTO consisted of the parent informing the child that he or she would not receive parental assistance for any activity the child could not perform on its own until the time-out had been completed. Despite expectations that the delay between noncompliance and completing a delayed time-out would reduce its effectiveness, the mean delay between the mand and compliance was reduced substantially for both children. In addition, the improvements persisted over time. The obvious limitation of this study is the small sample size and its limited relevance to classroom management, but it does offer one more tool that could be used to address time-out mand noncompliance.

Extinguishing Escape Behaviors

As mentioned above, some children will engage in behaviors intended to trigger a time-out mand because classroom activities are perceived to be aversive for a variety of reasons. The student may feel the tasks at hand are too difficult or demanding, for example (Taylor & Miller, 2007). In these situations, time-outs become negative reinforcement for escape behaviors (Everett et al., 2007). Taylor and Miller (2007) described two interventions which could reduce or extinguish escape behaviors. The first involved allowing the student to 'escape' an unpleasant task after the student engages in prosocial behaviors, such as asking for help or a break. The second is avoiding the use of time-outs during the unpleasant task, thereby encouraging the student to 'work through' escape behaviors.

Taylor and Miller (2007) then conducted two studies into how effectively time-outs were being used and the best way to manage escape behaviors, respectively. The criteria that determined appropriate use of time-outs were the following: (1) accuracy in target behavior identification, (2) labeling misbehavior, (3) delivering the time-out mand, (4) allowing a 5-second delay before praising compliance or physically moving the child to time-out, (5) setting timer, (6) planned ignoring, and (7) instructing the student to reenter classroom activities. Four children with developmental disabilities between the ages of 9 and 12 were included in the study. Following researcher administered training on when and how to use time-outs, the teacher and aids involved in the study made significant improvements in all outcome variables mentioned above. The increased accuracy and consistency in administering time-outs had a significant impact on student misbehaviors as well, with some behaviors approaching zero frequency, while others increased in frequency. For example, the prevalence of tantrums for one student declined from 59 to 4%, while screaming increased from about 60 to over 90% for another. The latter outcome was attributed to escape behaviors.

Taylor and Miller (1997) then tested whether escape behaviors could be reduced using a variety of strategies. These strategies were social attention, demand, and toy play. The social attention condition involved ignoring the student during play time until a problem behavior occurred, then the teacher or aide would admonish the student for 30 seconds. Demand condition involved presenting a task to the student that had previously triggered escape behavior, but in a step-wise manner. If the student exhibited escape behavior during the task, the teacher or aide would turn away and ignore the student for 30 seconds. The teacher or aide would reengage the student once 30 seconds had passed since the last escape behavior. The toy play condition involved the student and teacher sitting together at a table and playing together, but if the student engaged in problem behaviors there were no consequences. All four students were well-behaved during the toy play condition; however, higher levels of problem behaviors occurred during the social attention condition for two students and during the demand condition for the other two students. These findings were interpreted as attention-seeking and escape behaviors, respectively. Although there are significant limitations to this study, including small sample size, the results are informative because they help define at three possible outcomes following a time-out mand: (1) compliance, (2) increased attention-seeking behaviors, and (3) increased escape behaviors.

The overall conclusion reached by Taylor and Miller (1997) is that time-outs should not be used when a child is engaging in attention- or escape-maintained behaviors, a conclusion supported by other researchers (Everett et al., 2007). Everett and colleagues (2007) took issue with this dogma because the potential effectiveness of extinction has been ignored by researchers. From their perspective, confronting the child with the same task that triggered escape behavior immediately after completion of the time-out task would tend to extinguish escape behaviors. They tested this theory on four children between the ages of 4 and 5 with noncompliance issues. Functional assessment revealed that all four children engaged in behavioral patterns consistent with escape behaviors, rather than attention-maintained behaviors. Noncompliance was then followed by either a typical time-out or time-out with escape extinction (TO-EE). The escape extinction protocol involved administering a time-out in response to noncompliance and then exposing the child to the same task immediately after time-out completion. Compared to baseline, both time-out and TO-EE dramatically increased compliance. The effect sizes were large for both interventions, but varied considerably between students. Overall, however, TO-EE was more effective for three of the four children. The main limitation of this study is the small sample size, which prevents generalizing the findings to all children with noncompliance issues; however, these findings suggest that a significant proportion of children engaging in escape-maintained behaviors could respond favorably to a TO-EE approach.

Extinction was also applied to four male children between the ages of 2 and 7 with escape- and attention-maintained noncompliance (Benshoof, 2012). The contingent attention condition provided parental verbal statements referring to noncompliance behavior or continued interactions to compliance. The contingent escape condition provided planned ignoring in response to noncompliance or continued interaction to compliance. An ABCB study design was utilized, representing baseline, TO-EE, contingency reversal, and TO-EE, in that order. Contingency reversal closely approximated baseline compliance rates for all four children, as would be expected. The TO-EE intervention, however, dramatically increased compliance rates for all four children even though two of the children were judged to be primarily attention-maintained noncompliant. Again, the small sample size is limiting, but these results reveal the generalizability of TO-EE to both escape- and attention-maintained noncompliance. The authors also mentioned that the maintenance of noncompliance depended on factors other than escape and attention, thereby providing empirical evidence of the complex interplay of multiple reinforcers contributing to noncompliance.

Importance of Time-In

As mentioned above, the quality of time-in can determine how effective a time-out will be (Kostewicz, 2010). Simply increasing the frequency of praise, for example, can increase or decrease the prevalence of appropriate and inappropriate behaviors, respectively; however, when time-in and time-out considerations are combined the impact on undesirable behaviors can be dramatic (Olmi, Sevier, & Nastasi, 1997). For example, when a mother touched her child as a reward for not pinching there was a modest reduction in this behavior, but when combined with a 5-minute time-out pinching was reduced to almost zero. To test the effectiveness of the time-in/time-out 'package,' Olmi and colleagues (1997) administered the intervention to two EBD children between the ages of 4 and 8. The time-in portion of the intervention consisted of gentle touching and verbal praise, while the time-out consisted of an exclusionary time-out imposed when teacher instructions were not obeyed within 5 seconds. For the younger child, compliance increased from 10 to 98% and the prevalence of verbal/physical aggression and elopement decreased to zero from 1 per 2.7 min. And 1 per 4.5 min., respectively. The older child frequently threw objects placed in her hand, but after the intervention object throwing decreased to almost zero. In addition, during follow-up evaluations at 2, 24, and 40 weeks the improvements had been retained by both subjects. The obvious limitations are the small sample size, but the dramatic improvements tend to negate this concern.

Leaf and colleagues (2012) evaluated the efficacy of a time-out ribbon intervention for a 16-year-old autistic boy who was engaging in aberrant behaviors interfering with classroom activities. The aberrant behaviors (stereotypic) were not the results of attention-seeking or escape, but were being reinforced automatically. The authors chose the time-out ribbon because it had considerable empirical support and produced good results in their clinic when administered to autistic children. The first phase involved linking positive reinforcement to the wearing of the ribbon, while the second phase helped the boy understand which behaviors would result in removal of the ribbon. The third phase involved the therapist asking the child to remove the ribbon as a consequence of engaging in aberrant behavior, which could be earned back by taking five deep breaths. The prevalence of aberrant behaviors decreased from 70 to 6% once the child began wearing the ribbon. The main limitation identified by the authors was the number of potential confounding factors, which will require further evaluation in future studies.

Discussion

Restraints, seclusion, and time-outs, as methods for classroom behavioral control, have come under increasing legal and ethical scrutiny in recent years, as claims of misuse, injury, psychological trauma, and even death have entered the public arena. The greatest concern surrounds the use of restraint and seclusion; however, time-outs are not immune to some of the same criticisms. Seclusionary and exclusionary time-outs, for example, can result in the same adverse outcomes as with the use of restraints, even when the letter of the law has been followed. Aside from the risk of harm to both student and staff, time-outs can also consume valuable class time and thus degrade the academic progress of all students. There is thus a need to evaluate empirically whether time-outs are effective and if so, under what circumstances, especially for EBD students.

A large number of studies have revealed that time-outs are clearly effective when administered consistently and judiciously (Turner & Watson, 1999; Fabiano et al., 2004; Vegas et al., 2007; Goldstein & Brooks, 2007; Kerr & Nelson, 2009). Efficacy, however, can be undermined by a lack of training, weak time-in reinforcers, student disengagement from classroom activities, noncompliance, continued disruptive behavior, and unintended reinforcement of attention-seeking and escape behaviors. Noncompliant, disruptive, attention-seeking, and escape behaviors were reduced significantly by the introduction of a brief task during time-out (Grskovic et al., 2004). Noncompliance with a time-out mand was improved when time-out duration was contingent upon quick compliance (Donaldson et al., 2013) or when a deferred time-out restricted positive reinforcement until the time-out mand had been obeyed (Warzak & Floress, 2009). Keeping students working on their assignments at their desk during time-outs was made possible through the use of a time-out ribbon (Yeager & McLaughlin, 1994); however, the efficacy of time-out ribbons depended on the quality of time-in reinforcers (Kostewicz, 2010).

Escape behaviors can be addressed in a number of ways. One method is to allow students to escape an unpleasant task after engaging in prosocial behavior, while another would be to avoid the use of time-outs during demanding classroom activities (Taylor & Miller, 2007). Remarkable results were obtained, however, simply by improving the accuracy and consistency in time-out administrations. The logical conclusion from this finding is that proper training in time-out administration tends to promote generalization across academic settings. Attention- and escape-maintained behaviors were not reduced sufficiently by this intervention, so Taylor and Miller (2007) recommended simply avoiding the use of time-outs during these behaviors. Everett and colleagues (2007) provided evidence that exposing the child again to the same demanding classroom activities after the time-out led to extinction of escape behaviors. Similar results were obtained by Benshoof (2012) regardless of whether the child was primarily engaging in attention- or escape-maintained behaviors.

None of the interventions tested by the above researchers, however, would have been very effective if the quality of time-in reinforcement was insufficient to represent an aversive condition when withdrawn during time-out (Olmi et al., 1997; Kostewicz, 2010). This is the basis for administering a time-out ribbon method, which Leaf and colleagues (2012) revealed was effective even for an autistic boy engaging in aberrant behaviors that were automatically reinforced.

The research findings presented here reveal that time-outs are effective, especially when administered accurately and consistently. Behavioral scientists have also begun to identify and test various modifications that can help increase the efficacy of time-outs, even for the most challenging students. Overall, there is an obvious shift toward less coercive and shorter time-outs, thereby reducing the risk of harm to everyone concerned. This shift tends to moderate the pitfalls that can accompany the use of time-outs, while at the same time increasing the amount of time students remain engaged in learning activities. Future studies will likely continue this trend by elucidating the many variables that influence time-out efficacy. These studies will likely emphasize planned ignoring and deemphasize exclusionary and seclusionary time-outs. In addition, the quality of time-in reinforcers will probably occupy the minds of researchers more and more in the coming years, so there is also a shift toward modifying the behavior of teachers and staff in response to problem behaviors.