Close this window to return to the Glenholme School Site

Intensive Behavior Support: A Protocol to Address the Needs of Treatment Non-Responders

Patrick L. Queenan, Krista L. Kulpa, Scott D. Hewitt, Karen D. Pugh
Devereux Glenholme

A protocol for children who did not respond to universal and group level interventions in a structured school setting was established, and measured.  All children in the study had been referred to this school due to their having previously not responded to universal, group or individualized interventions in more traditional school settings.  The goal of the protocol was to quickly recognize children who required more intensive supports, and to utilize positive behavior supports specific to each child that would reduce the need for reactive and/ or deceleration techniques.  The results indicated that the protocol was very effective, and created no additional strain on the school as a whole.

Intensive Behavior Support: A Protocol to Address the Needs of Treatment Non-Responders
Managing disruptive behavior in school settings appears to be an ever-increasing concern.  Educators are faced with the challenge of delivering a curriculum while simultaneously managing a variety of complex learning and behavioral problems (Lewis and Sugai, 1999).  Without a plan for handling the behavioral problems of individual children, the learning of all children within a classroom can be negatively affected.  Additionally, the identified child with emotional and behavioral disorders in the public school setting is at increased risk of academic failure, dropping out, unemployment, illicit drug use, teen pregnancy, and other negative outcomes (Eber, Sugai, Smith, and Scott, 2002).

Disruptive behaviors create much discord in the classroom.  Children who display severe emotional and behavior problems in schools represent a relatively small proportion of a school’s total enrollment; however, they require significant amounts of expertise, time, and resources (Eber, et al., 2002).  Although most educators are dedicated to meeting the academic and behavioral needs of these children, they become frustrated by their inability to establish and sustain effective programming options (Eber et al.). According to the National Commission for Teaching and America’s Future (2002), 50% of urban teachers leave the profession within the first five years of their career, citing behavior problems and management as factors influencing their decision to leave.

High rates of disruptive behavior in school are associated with punitive disciplinary strategies, lack of clarity about rules, expectations, and consequences, lack of staff support, and failure to consider and accommodate individual differences (Mayer, 1995). Many schools lack the capacity to identify, adopt and sustain policies, practices and systems that effectively and efficiently meet the needs of all children. Schools often must rely on outside behavioral expertise because local personnel lack specialized skills to educate children with significant problem behaviors (Mayer, 1995; Sugai & Horner, 1994, 1999; Taylor-Greene et al., 1997; Walker et al.,1996). This reliance on external consultants results in limited knowledge and implementation of behavioral techniques within the school setting. In sum, educators face relentless and significant challenges that if not addressed, can dramatically impact students, school personnel, families, and community members (Sugai, Horner, et al., 2000).

Fortunately, the Positive Behavior Support (PBS) approach gives direction for educators and educational institutions in how to address problem behaviors on school-wide, smaller group, and individual levels.  PBS procedures emphasize assessment prior to intervention, manipulation of antecedent conditions to reduce or prevent the likelihood that a problem behavior will occur, development of new social and communication skills that make problem behaviors irrelevant, and careful redesign of consequences to eliminate factors that maintain problem behaviors and to encourage more acceptable replacement social skills and behaviors (Sugai, Horner, et al., 2000).

McCurdy, Mannella, and Eldridge (2003) examined how the application of a school-wide positive behavioral support model affected the number of office discipline referrals utilized in an urban elementary school. The model was applied in two phases over the course of two years. Universal (primary) supports that included behavioral expectations that operationalized school-wide rules, a motivation system to teach and strengthen desired behaviors, and correction procedures to reduce problematic behaviors were applied in the first year.  Group (secondary) and individual (tertiary) supports based on Functional Behavioral Assessments and data trends were implemented in the second year. The results of the study indicated a 46% reduction in the number of office discipline referrals per child at the end of the second year.   Significant results were also found after the 1st year of implementation during which only universal supports were applied, suggesting an immediate impact. The results were found despite the unique challenges that often face urban schools and contribute to antisocial behavior such as the children’s increased rates of aggression and lack of school readiness skills, as well as the schools’ lack of qualified staff and necessary resources.

Scott and Barrett (2004) investigated the effectiveness of school-wide PBS (universal) in reducing office referrals and suspensions in an urban elementary school.  A cost benefit analysis was also performed, assessing the monetary value of the administrator time associated with office referrals and suspensions and resulting student instruction time lost.  The universal supports in this study included specific behavioral expectations, clear and consistent routines, physical manipulation of the environment, and a reinforcement system. The PBS had an immediate impact in significantly reducing office referrals and suspensions, which continued to decrease through out the course of the study. It was determined that the reduction in referrals and suspensions led to significant savings in time and money despite the costs of implementation.  Thus, PBS has been demonstrated to be both effective and efficient.

While many studies have now demonstrated the efficacy of PBS, no treatment is 100% effective.  In traditional schools, a small percentage of children are referred for more intensive services, as they are not able to be maintained within the setting.  When these children are placed in a more intensive setting, a well functioning milieu will attend to the needs of approximately 90% of those children who were non-responsive to interventions in less intensive settings.  This leaves 10% of the original non-responders to treatment requiring even more intensive support.  The current study sought to establish a protocol for identifying when the PBS interventions are not working (e.g., how to identify the non-responders within the population), why they are not working, and how to address the treatment needs of these non-responders to treatment. The current study will describe the methodology used to attend to children requiring more intensive behavioral supports, and will present the response rates of individual children, as well as analyze the efficacy of the program collectively. Because there are limited resources in any program, the current study also sought to monitor the effects on the overall milieu when resources are focused on a small portion of the population.


The Devereux Glenholme School, a boarding school for special needs children, heightened its focus on children that require more intensive individualization to their treatment planning. For some of the highest functioning children, this meant identifying ways for them to gain exposure to the stimuli that typically triggered maladaptive behavior in the past, in order to help prepare the children for less structured environments. For others who were less responsive, or non-responsive to the treatment milieu, this meant minimizing exposure to these triggers while positive coping skills or replacement behaviors were developed. The purpose of this study was to monitor the impact of this individualization of treatment, specifically for the latter group of children, those who required the most intensive behavioral support (henceforth referred to as the Intensive Behavior Support or IBS group).

The milieu therapy component of treatment in this boarding school is viewed as the most powerful treatment agent, given that this treatment modality encompasses the child’s life. This universal level treatment approach includes a positive-based token economy wherein children earn tokens approximately every 15 minutes for displaying positive and prosocial behaviors. Tokens have been operationally defined in order to reinforce positive social skills with adults, positive skills with peers and one specific behavior that is prioritized and individualized for each child by the treatment team. Children gain access to reinforcing activities, tangible items, and specific privileges through their token earnings.

Another critical component to the program in this boarding school is engagement in creative and preferred activities that involve the artistic, athletic, technological, and intellectual strengths and interests of the children that we serve such as equestrian, art, drama, music, sports, technology, and photography. Outcome studies conducted at the school over the past 3 years have demonstrated that while children are engaged in such activities, disruptive or problematic behaviors are almost non-existent. At the same time, many more opportunities for reinforcement of positive behaviors and developing skills emerge, allowing for greater ability to implement positive programming. It is noteworthy that while this structured boarding school does not allow for inclusion, the school does stress normalization through exposure to these activities for two reasons. First, the children in this school have not succeeded or been maintained within a mainstream or inclusion environment. Most importantly, these children were not able to establish social connections within or outside of their mainstream school. For many of the children, the activities within this structured boarding school provide their first opportunity to participate on a school team sport, within the school play, or to go to the school dance.  Second, through exposure to these engaging and preferred activities, the children gain exposure, interest, and skill in activities that may provide them with a link to connect socially when they return to an inclusion environment.

It has been speculated that a well-functioning milieu (universal interventions) should address approximately 90% of the individuals in treatment (Sugai, Horner, et al., 2000). The remaining 10% require modifications or more intensive treatment approaches (specialized group and individual interventions). Over the course of this study, the descriptive statistics compiled on our milieu were consistent with the predicted statistics given in the model of positive behavior support (Sugai, Horner, et. al., 2000, p. 136). Averaging over the 6 months that data was collected, 88.5% of the children in treatment made positive progress through the universal level treatment techniques, and the structure, consistency and learning opportunities of the program offerings. The percentages of children not responding positively to the program varied between months, with a range of 5 to 17% of children requiring IBS monthly, and overall, 21% of our population utilized IBS at some point over the 6 month period.

Participants were all children who were enrolled at the school from September 2005 through June 2006.  The school is a boarding school for special needs children in the northeastern United States.  The children are predominantly Caucasian and middle to upper middle class.

Inclusion in Intensive Behavior Support (IBS).  The decision to be placed in this group is based on index scores, (composite scores which are defined below). Based on their monthly index score, children are either placed in the IBS group, or remain in the Milieu group. Once in the IBS group, the child’s index score needs to fall below the inclusion threshold for two consecutive months in order to return to the Milieu group.

Once a child is placed in the IBS group, behavior analysts observe the child in each setting utilizing Functional Behavior Analysis (FBA). Based on the observations, the child’s treatment plan is modified to include behavioral interventions and contingency contracting derived from the FBA. Once per week, the program supervisors, the behavior analysts, and behavioral psychologists meet for an hour to review the progress of the children in the IBS group, and make modifications to interventions as necessary based on the observations of faculty, supervisors, and behavior office personnel. Any modifications to the specialized interventions of children in the IBS group are posted to an “Intervention List” during this meeting which is automatically synced to PDAs that all supervisors and boarding school faculty carry in order to enhance the fidelity of treatment. The treatment flowchart presented in Figure 1 visually depicts the process utilized for the IBS group.

Index scores.  Index scores were computed for each child by summing their use of the following four treatment interventions which indicate behaviors of heightened concern.  These treatment interventions are non-preferred, as they indicate use of reactive strategies and/ or deceleration techniques:

  1. Supplemental Observation- children viewed as being at heightened risk for self-harm or elopement are placed on a heightened level of observation, this level of observation is designed so that the child is not aware of the heightened monitoring.
  2. Physical Control- reactive techniques to aggressive behavior.
  3. Supervisory Intervention- brief removal from activity by program supervisor.
  4. Programmatic Consequence- any deceleration techniques reactive to aggressive or destructive behavior.

(See Figure 1 below)


Overall, 85% of children who were placed in the IBS group at any point during the course of the study responded favorably enough that they were back in the Milieu group at the time that the study was concluded, indicating that they were able to function without additional modifications. It is noteworthy that 100% of the children who were placed in the IBS group improved to the point where they were able to return to the Milieu group. However, four (or 25%) of the children that returned to the Milieu group were placed back in the IBS group at some later point. Two of these four children who returned to the IBS group then improved to the point where they were able to return to the Milieu group a second time, and remained there at the time of this analysis.

Before presenting the data, it is important to note that the current study poses the typical threats to internal validity presented any time that assignment to grouping is not random. In this case, assignment to the IBS or Milieu treatment groups was done on the basis of clinical need, and therefore randomization would not have been possible or advisable. Of further concern in this study, assignment to the IBS group was done on the basis of extreme scores on the dependent variable (the Index Score) at the time of assignment. Regression to the mean is always a possible explanation when extreme scores move closer to the middle. Thus, due solely to the statistical principal that outliers tend to score closer to the population average on repeated measures, it is expected that without any intervention some improvement should be noted.

Despite these obvious threats, it is difficult to think of how the study could have been performed differently, as intervention was required for these children specifically because of their extreme score. Because these children were struggling in spite of the positive programming described above, their progress was tracked and monitored. The data will be presented both at an individual level, as well as an analysis of the IBS program as a whole. It is important to keep the threats to internal validity described above in mind when interpreting this data, however.

Graphs of the children who were placed in the IBS group at any point during the course of the study are presented in Figure 2. In all, there are 20 graphs, 17 of which display trend lines demonstrating improvement. Following this, Figure 3 shows a positive trend for the IBS group as a whole. It is also noteworthy that the children in the Milieu group did not trend negatively, thus demonstrating that focusing more intensive services on the needier children did not negatively affect the rest of the population.

T-tests were performed on the change in index scores from month to month to assess for statistical significance. Significant differences were found for the IBS group in 2 out of 5 months. Specifically, the change between months 2 and 3 (t(9)=2.919, p<.05) and between months 5 and 6 (t(15)=3.152, p<.01) were found to be significant. None of the differences in change scores for the Milieu group reached statistical significance.

(See Figures 2 and 3 below)


The IBS protocol for the children in this study appears to have been largely effective. These 20 children accounted for the portion of the population over a six month period that were the least responsive to the universal level treatment program. With the intensive behavior supports added to the milieu program, 100% of the children placed in the IBS group improved to the point where they were returned to the Milieu group. As described above, 25% of them required the IBS group again when the supports were reduced, however 50% of the children who returned to the IBS group were able to quickly return to the Milieu group again. This may indicate that this heightened level of intervention was a better match for their level of need. Alternatively, it might indicate that implementation of the universal level of treatment needs enhancement, which is a question of the reliability and validity of behavioral interventions, or treatment fidelity (Bellg, Borrelli, Resnick, Hecht, Minicucci, Ory, Ogedegbe, Orwig, Ernst, and Czajkowski, 2004).

These results have given confidence that the methodology of the IBS program is associated with positive responses in the children that are having the most behavioral difficulty in the program. This IBS program methodology included data analysis of the entire treatment population every 30 treatment days, and weekly meetings between program supervisors, behavior analysts, and behavioral psychologists resulting in expedient modifications to treatment plans and interventions. With the confidence gained through this data analysis, the IBS methodology will be maintained in this school setting. In order to address possible concerns regarding the fidelity of treatment at the universal level, the school will also focus efforts on enhancing the efficacy and fidelity of interventions for the entire population with the goal of allowing the milieu program to maintain more children after they have received IBS treatment.

Training which is already in progress toward this end includes the modeling of behavioral techniques for school faculty by behavior analysts and program supervisors. Program supervisors and behavior analysts also provide school faculty with observations and immediate feedback paired with positive praise and opportunities for enhancement. The next phase of treatment fidelity enhancement, which is already in the planning stages, includes unobtrusive in-vivo coaching of the milieu to the school faculty. Fortunately, with the implementation of the IBS protocol, more time is now available to plan these proactive universal level intervention enhancements.


Figure 1: Model for Intensive Behavior Support


Figure 3: Efficacy of Intensive Behavior Support

Text Box: Index   Scores



Bellg, A.J., Borrelli, B., Resnick, B., Hecht, J., Minicucci, D.S., Ory, M., Ogedegbe, G., Orwig, D., Ernst, D., Czajkowski, S. (2004). Enhancing treatment fidelity in health behavior change studies: Best practices and recommendations from the NIH Behavior Change Consortium. Health Psychology, 22, 443-451.

Carr, E.G., Dunlap, G., Horner, R.H., Koegel, R.L., Turnbull, A.P., Sailor, W., Anderson, J.L., Albin, R.W., Koegel, L.K., Fox, L. (2002). Positive behavior support: Evolution of an applied science. Journal of Positive Behavior Interventions, 4, 4-16.

Eber, L. Sugai, G., Smith, C., Scott, T. (2002).  Wraparound and positive behavioral interventions and supports in the schools.  Journal of Emotional & Behavioral Disorders, 10(3), 171-186.

Gardner, R. (2003).  Intensive positive behavior supports for adolescents with acquired brain injury.  Journal of Head Trauma and Rehabilitation, 18(1), 52-74.

Lewis T., Powers L., Kelk M., Newcomer L. (2002).  Reducing problem behaviors on the playground: An investigation of the application of school wide positive behavior supports.  Psychology in the Schools, 39(2), 181-190.

Lewis, T., Sugai, G. (1999). Effective behavior support: A systems approach to proactive schoolwide management.  Focus on Exceptional Children, 31(6), 1-25.

Mayer, G. (1995).  Preventing antisocial behavior in the schools.  Journal of Applied Behavior Analysis, 28, 467-478.

McCurdy B., Mannella M., Eldridge N. (2003). Positive behavior support in urban schools: Can we prevent the escalation of antisocial behavior? Journal of Positive Behavior Interventions, 5(3), 158-170.

Scott M., Barrett S. (2004).  Using staff and student time engaged in disciplinary procedures to evaluate the impact of school-wide PBS. Journal of Positive Behavior Interventions, 6 (1), 21-27

Sugai, G., & Horner, RH (1994).  Including students with severe behavior problems in general education settings: Assumptions, challenges, and solutions.  The Oregon Conference Monograph, 6, 10-120.

Sugai, G., & Horner, RH (1999).  Discipline and behavioral support: Practices, pitfalls, & promises.  Effective School Practices, 17(4), 10-22.

Sugai, G., Horner, R.H., Dunlap, G., Hieneman, M., Lewis, T.J., Nelson, C.M., Scott, T., Liaupsin, C., Sailor, W., Turnbull, A.P., Turnbull, H.R., Wickhan, D., Wilcox, B., Ruef, M. (2000). Applying positive behavior support and functional behavioral assessment in schools. Journal of Positive Behavior Interventions, 2, 131-143.

Sugai, G., Sprague, J., Horner, RH, & Walker, H. (2000). Preventing school violence: The use of office discipline referrals to assess and monitor school-wide discipline interventions.  Journal of Emotional and Behavioral Disorders, 8(2), 94-101.

Taylor-Greene, S., Brown, D., Nelson, L., Longron, J., Gassman, T., Cohen, J., Swartz, J., Horner, RH, Sugai, G., & Hall, S. (1997).  School-wide behavioral support: Starting the year off right.  Journal of Behavioral Education, 7, 99-112.

Walker, H., Horner, RH, Sugai, G., Bullis, M., Sprague, J., Bricker, D., & Kaufman, M. (1996).  Integrated approaches to preventing antisocial behavior patterns among school-age children and youth.  Journal of Emotional and Behavioral Disorders, 4, 193-256.

Wright, J., & Dusek, J.(1998).  Compiling school base rates for disruptive behaviors from student disciplinary referral data. School Psychology Review, 27(1), 138-147.