|Table of Contents|
Table of Contents
List of Tables
List of Figures
Introduction and review of the literature
Specific aims and hypotheses
Appendix A. Procedural outline for conducting SIT training
Appendix B. Demographics questionnaire
Appendix C. Consumer satisfaction questionnaire
Appendix D. Behavior definitions
ENHANCING PRODUCTIVE PHYSICAL THERAPY IN CHILDREN
USING STRATEGIC INTERACTION TRAINING
STACEY J. HOFFMAN
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
I would like to express my sincere gratitude to my chairperson, Dr. Stephen R. Boggs
for his guidance during my graduate school training and his support in completing this
project. I would also like to thank committee members Drs. Sheila Eyberg, Kathye Light,
Mary Jane Rapport and Trevor Stokes for their support and encouragement.
This research could not have been accomplished without the assistance of the
physical therapists and the children at Kids on the Move, Shriners Children's Hospital and
All Children's Therapy Center. The research would not have been completed without the
tremendous efforts and dedication of my research assistants, Stephanie Strobeck and Karen
Whitlock, who provided countless hours of coding and videotaping. I am also grateful to
the Center for Pediatric Psychology and Family Studies for providing partial funding for this
I am especially thankful for the ongoing support of my family and friends. In
particular, I am indebted to my husband and friend, John Barbour, for his never-ending
patience, support, and encouragement; to my best friend Murphy; to Ardis Hanson for her
friendship, generosity, and assistance in editing this manuscript; and to my parents Sandra S.
Hoffman and William R. Hoffman for their emphasis on education and faith in my abilities.
TABLE OF CONTENTS
A CK N O W LE D G E M E N TS .............................................. ........................................... ii
LIST O F TA BLE S ............................................................................................................... V
LIST O F FIG U RE S .......................................................................................................... v i
A BSTRA CT ................................................................................................................................ vii
INTRODUCTION AND REVIEW OF THE LITERATURE .......................................... 1
Behavior M odification for M otor Skills ....................................................................... 5
Behavior Problem s in PT ................................................................................................ 12
Training Program s in Behavior M odification ............................................................ 14
SPE CIFIC A IM S A N D H YPO TH E SE S .......................................................................... 21
Therapist Target Behaviors ........................................................................................... 21
Child Target Behaviors .................................................................................................. 22
Consum er Satisfaction Predictions ............................................................................... 22
M ETH O D ................................................................................................................................... 23
Participants ........................................................................................................................... 23
Setting ..................................................................................................................................... 25
M easures ................................................................................................................................ 26
Procedure D esign ............................................................................................................ 34
RE SU LTS ..................................................................................................................................... 38
Therapist Behaviors ......................................................................................................... 38
Child Behaviors .................................................................................................................... 47
Consum er Satisfaction ................................................................................................... 60
D ISCU SSIO N ............................................................................................................................. 62
D ifferences betw een settings ......................................................................................... 65
Strengths and w eaknesses of the study ........................................................................ 67
Im plications ........................................................................................................................... 68
Future directions ................................................................................................................... 69
A PROCEDURAL OUTLINE FOR CONDUCTING SIT TRAINING ............... 74
B DEMOGRAPHICS QUESTIONNAIRE ............................................................... 81
C CONSUMER SATISFACTION QUESTIONNAIRE ........................................... 82
D BEHAVIOR DEFINITIONS .................................................................................... 84
R E FE R E N C E S ........................................................................................................................... 89
BIOGRAPHICAL SKETCH ............................................................................................... 96
LIST OF TABLES
1 Child demographics, diagnosis and behavioral issues grouped by treating
th erap ist .......................................................................................................................... 2 4
2 Summary of reliability for behavior categories ..................................................... 33
3 Mean occurrence of target behaviors at baseline and intervention for all
th erap ists ......................................................................................................................... 39
4 Average change in the inpatient setting vs. outpatient setting ........................... 59
5 Average change in the inpatient setting vs. outpatient setting ........................... 60
6 Mean scores on Consumer Satisfaction Questionnaire ....................................... 61
LIST OF FIGURES
1 Percent occurrence of direct commands inpatient setting ........................ 40
2 Percent occurrence of direct commands outpatient setting ....................... 41
3 Percent occurrence of ignoring inappropriate behavior -
Inpatient setting .............................................. ............................................ 42
4 Percent occurrence of ignoring inappropriate behavior -
outpatient setting ................................... ........................................... 43
5 Percent occurrence of all attention to appropriate child behavior -
inpatient setting... ..................................................................................... 45
6 Percent occurrence of all attention to appropriate child behavior -
outpatient setting ....................................................................................... 46
7 Rate of praises inpatient setting ............................................................... 48
8 Rate of praises outpatient setting ............................................................... 49
9 Percent occurrence of compliance--inpatient setting .................................... 51
10 Percent occurrence of compliance--outpatient setting .................................. 52
11 Percent occurrence of noncompliance--inpatient setting ............................. 53
12 Percent occurrence of noncompliance--outpatient setting .......................... 54
13 Rate of inappropriate child behavior--inpatient setting ................................ 56
14 Rate of inappropriate child behavior--outpatient setting ............................. 57
Abstract of Dissertation Presented to the Graduate School of the University of Florida in
Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
ENHANCING PRODUCTIVE PHYSICAL THERAPY IN CHILDREN
USING STRATEGIC INTERACTION THERAPY
Stacey J. Hoffman
Chairperson: Stephen R. Boggs, Ph.D.
Major Department: Clinical and Health Psychology
Many children with developmental disabilities or orthopedic impairments require
physical therapy (PT) to help maximize function, independence, and mobility. Children with
developmental disabilities may also have impaired cognitive abilities and a higher incidence
of negative behavior that can interfere with PT. Negative behaviors such as noncompliance,
resistance, tantrums, and other delay tactics can often prevent the therapist from
implementing a treatment program and thus impede the process of PT. Yet without PT,
these children are at increased risk for progression of their disabilities and immobility as well
as further impairment of their social development.
The present study used a training intervention to teach physical therapists behavioral
techniques to deal more effectively with children who exhibit behavior problems in PT. A
multiple baseline design was used with two therapists and four children in an inpatient
children's hospital, and two therapists and three children in an outpatient setting. Results
indicated that physical therapists learned the behavior modification techniques and
incorporated these skills into their daily PT session. Results also indicated that therapists
increased their levels of direct commands, praise, attention to appropriate child behavior,
and ignoring inappropriate child behavior. Further, child compliance rates increased,
and inappropriate child behaviors decreased. The intervention was rated positively
by all therapists.
INTRODUCTION AND REVIEW OF THE LITERATURE
Many children with developmental disabilities or orthopedic impairments require
physical therapy (PI). Goals of PT may be to increase or maintain strength or range of
motion, prevent contractures or other deformities, provide appropriate seating to maximize
function and promote postural alignment, and improve functional mobility and
independence. However, many of these children may not receive adequate PT due to refusal
by private insurance or managed care companies to pay for PT or to authorize adequate
numbers of sessions. Other children may not receive adequate services due to time
constraints on the therapist to provide more than standard range of motion exercises. Still
other children may not receive the PT they need due to behavior problems that interfere
with the therapy and limit progress towards therapeutic goals.
Compared with typically developing children, children with developmental
disabilities often have impaired cognitive abilities, attention deficits, a higher need for
attention from adults, and a high incidence of negative behavior (Briener & Beck, 1984).
Conduct problems such as noncompliance, aggression, resistance, crying, tantrums,
withdrawing, and other delay tactics can often prevent the therapist from implementing a
treatment program and thus impede the process of PT. Studies using functional analyses to
identify influences on problem behaviors found that behavior problems occur more
frequently in high demand situations (Repp & Karsh, 1994; Carr, Newsom, & Binkoff,
1980). Carr and Durand (1985) found that inappropriate behaviors increased during high
demand situations and in the absence of adult attention. Physical therapy is a high demand
situation requiring the patient to perform multiple exercises and activities that are often
difficult and may be painful. Thus, the pediatric physical therapist is frequently faced with
children who are aggressive and noncompliant often without the skills to handle such
behavior. Yet without PT, these children are likely to be at increased risk for progression of
their disabilities and immobility as well as further impairment of their social development.
There is a plethora of literature focusing on interventions for aberrant behaviors in
persons with developmental disabilities (Repp & Karsh, 1994). Such research has typically
been conducted by psychologists and has been conducted in a variety of settings such as
schools (e.g., Repp & Karsh 1994), the dental office (Allen & Stokes, 1987), the home (Day,
Homer, & O'Neill, 1994), inpatient settings (Mace, Page, Ivancic, & O'Brien, 1986), and
during PT sessions (Rapport & Bailey, 1985).
According to the literature (Carr, 1977; Durand & Carr, 1985), four primary
motivating conditions for engaging in aberrant behaviors such as self-injurious and
aggressive behavior have been identified. These include attention seeking behaviors, escape
motivated behavior, sensory motivated behaviors (e.g. self-injurious behavior; SIB), and
obtaining tangible consequences such as food or a toy. Escape and attention have been
found to be the most common motivating conditions (Iwata et al., 1994).
The appearance of aggression and noncompliant behavior in PT is not uncommon.
Reynell (1965) found an interfering, negative behavioral response pattern in postoperative
children with cerebral palsy. One of the more notable behaviors reported during treatment
observations, described by both the therapists and the children's parents, was aggression
toward the therapist. Lalli, Mauk, Goh, and Merlino (1994) and Manella and Varni (1981)
noted that problem behaviors such as aggression and self-injurious behavior increased when
the patients were requested to participate in their PT sessions. Homer (1971) reported the
occurrence of aggression in a 5 year old child with spina bifida and mental retardation as the
requirements for walking were increased. Carr, Newsom, and Binkoff (1980) found that
aggression in children with retardation increased in demand situations. Furthermore, it was
shown that aggression was maintained by negative reinforcement in the form of escape from
demands. The general conclusion has been that aggression often functions as an escape
response in such high demand situations and when successful, is maintained through
negative reinforcement. Depending on the physical therapist's response to such behaviors,
the child may adapt and progress physically or the child may regress in his/her behavioral,
emotional or physical repertoire (Stokes, Mowrey, Dean, & Hoffman, 1997).
Not only might physical therapists have to overcome overt behaviors such as
aggression, tantrum behavior, resistance, and pain behaviors, but they might also deal with
other delay tactics that impede the flow of therapy. Such problems may include inattention,
self-stimulating behaviors, noncompliance, and dawdling such as stopping an exercise to ask
questions or to engage the therapist or others in conversation. These behaviors are equally
interfering with the process of PT and increase the amount of time spent in therapy while
Behavior therapy has been shown to be an effective approach to evaluating and
changing behavior. Behavior modification techniques have long been used with children and
adults with cerebral palsy, mental retardation, autism, and other developmental disabilities to
treat behaviors such as self-injurious behavior (Durand & Carr, 1985), aggression (Carr,
Newsom, & Binkoff, 1980), noncompliance (Rapport & Bailey, 1985), pain behaviors
(Fordyce, Shelton, & Dundore, 1982), and others. However, the use of behavior
modification techniques in PT by physical therapists has been slow to evolve.
The use of behavior modification in PT can serve to modify the environment to
reduce difficult behaviors that interrupt therapy. Yet, the majority of applications of
behavior modification in the PT setting reported in the literature have focused on using
behavior modification to increase motor skills (Martin, 1976). Additionally, such
interventions appear to have been conducted by psychologists and not the physical
therapists. However, the majority of children with physical disabilities are treated by physical
therapists, not behavior therapists. Few attempts have been made to train physical therapists
in behavioral techniques to modify children's behavior.
Lalli, Mauk, Goh, and Merlino (1994) successfully trained two physical therapists to
use a behavioral intervention to increase compliance to commands, decrease rates of
problem behaviors, and increase the distance ambulated in PT. The training of the physical
therapists occurred after the behavior therapist had implemented the intervention and had
determined its effectiveness. Unfortunately, the authors failed to describe how the therapists
were trained and over what period of time, nor was there any follow-up to determine if the
therapist's training carried over to children exhibiting aberrant behaviors who were not
participants in the study.
By incorporating behavioral techniques into the physical therapist's repertoire, the
therapist may be able to decrease negative behaviors that interfere with PT. Additionally,
behavioral techniques such as positive reinforcement could result in higher rates of success
and lead to greater treatment gains in fewer sessions.
The remainder of the chapter critically examines the literature regarding the use of
behavior modification to promote the development of motor skills and to improve behavior
and participation in PT. Early studies in the area (Fuller, 1949; Harris, Johnston, Kelley, &
Wolf, 1964; Rice, McDaniel, & Denney, 1962; and Trotter & Inman, 1968) focused on using
behavior modification to promote changes in motor skills. It was generally concluded that
such use of behavior modification in the treatment of persons with physical and mental
disabilities was effective. As Maneila and Varni (1981) pointed out, however, the use of
behavior modification in physical rehabilitation can be used not only to promote optimal
physical functioning, but also to promote modification of behaviors that often occur in
children with long-term disabilities and that may interfere with PT. Literature on training
behavior change agents is reviewed as well. The review concludes with a summary and
critique of existing literature followed by a discussion of the hypotheses to be examined
in this research.
Behavior Modification for Motor Skills
Perhaps the earliest reference to the use of behavior modification to increase
movement in a person with a disability was made by Fuller (1949). Fuller described the use
of operant conditioning with an 18-year-old male with profound impairment. By using a
squirt of sugar-milk solution each time the subject moved his arm, Fuller was able to
increase the rate of upper extremity movements from .67 per minute to 3 per minute.
Almost twenty years later, Foss (1966) described several studies that provided
evidence for the successful use of behavioral techniques to control movement in humans,
and suggested its use for children with cerebral palsy. Early literature addressing behavior
modification with children who have cerebral palsy (CP) continued to focus primarily on the
promotion of learning motor skills. Rice, McDaniel, and Denney (1962) described a single-
case study in which reinforcement of arm movements was provided to a six-year-old boy
with profound mental retardation. After they increased the operant rate of arm movements
by rewarding the subject for any arm movements, the authors used ice cream as a reinforcer
to shape the response of reaching for a ring placed over the subject's bed. With ice cream as
a reinforcer, the subject touched the ring over 700 times during a 45-minute period.
Early reports of the use of behavioral techniques in the PT environment were
provided by Meyerson, Kerr, and Michael (1967). These authors presented four case studies
demonstrating the use of behavior modification in a rehabilitation setting. The target
behaviors included developing independent ambulation with two subjects, promoting on-
task behavior during a typing task with one subject, and decreasing high rates of self-
injurious and self-stimulating behavior with another subject. In each case, a behavior
therapist designed and implemented specific interventions for each subject. Interventions
included shaping target behaviors through successive approximation, providing positive
reinforcement of target behaviors with attention from adults or edible rewards, and
providing alternative sensory stimulation for non-injurious behavior. Each intervention was
successful in achieving the target goal. However, the physical therapists and the
occupational therapist who had been treating the children prior to the research intervention,
were not incorporated into the study to learn the interventions themselves.
Trotter and Inman (1968) described the use of positive reinforcement to promote
the performance of progressive resistance exercises in patients with paraplegia and
quadriplegia involved in physical rehabilitation. By providing positive reinforcement in the
form of attention from the therapist, and increased attention to appropriate rehabilitation
goals by charting, the authors hypothesized that the group receiving positive reinforcement
would achieve greater gains in their strengthening program. Indeed, a significant difference
was found in the average increase in weights lifted for the experimental group versus the
control group. However, due to several limitations of the study, the reasons for these
differences could not be isolated. Differences between the control and experimental group
other than the use of positive reinforcement significantly limited the conclusions that could
be drawn from this study. For example, half of the control group (subjects with
quadriplegia) performed their exercises in the presence of and with the assistance of a
nonprofessional who also provided attention in the form of "usual conversation." The
experimental group received attention from a trained physical therapist. Status, knowledge
and experience of the physical therapist versus the nonprofessional aide could contribute to
the differences observed. Additionally, the control group not only failed to receive positive
reinforcement, but also failed to receive any prompting or advice to increase the weights
being lifted. Such differences make it difficult to delineate which component of the
treatment was most instrumental in creating change.
Trotter and Inman's (1968) assertion regarding the benefits of using behavior
therapy techniques in the PT setting seemingly did little to promote the practice. In fact,
when one considers the tremendous success in using behavior modification to effect motor
skills as well as behavior in general, it is quite unfortunate that the PT literature remains
limited in its presentation of the successful combination of behavior therapy and PT.
In the PT literature, Kolderie (1971) provided a brief literature review related to
behavior therapy techniques and then described how behavior modification could be used
with children with cerebral palsy to learn motor skills. A single-case study was described
involving a seven-year-old, nonverbal female of average intelligence with mild spastic
diplegia. At the time of the study, the child had a vocabulary of 10 words, could attain a
sitting position with assistance, was learning to crawl, and could ambulate with maximum
assistance. Using a poker chip reward system and immediate verbal reinforcement in
physical, occupational, and speech therapy, the subject progressed to independent
ambulation and established an age appropriate vocabulary. Over the course of treatment,
reinforcements were gradually shifted to natural contingencies. However, because there was
no baseline or reversal of contingencies, the authors could not be conclusive in stating that
behavior modification was the critical variable accounting for the changes in the child's
behavior. Additionally, no follow-up assessment was done to determine if the progress
continued. Furthermore, the author did not describe if the physical, speech and
occupational therapists were trained to implement the intervention or if a behavior therapist
conducted the treatment.
In a single-case design, Chandler and Adams (1972) used a behavior modification
program to promote independent ambulation in an eight-year-old boy with multiple
handicaps, who would not ambulate despite being physically capable of doing so. During
the baseline period, the highest mean number of steps obtained from nine trials became the
base rate. Implementation of the modification phase involved providing the child a reward
of music or chocolate, contingent on exceeding the base rate from each previous session.
Independent ambulation was achieved after twenty-eight sessions, and continued after the
reinforcers were withdrawn. Several problems were evident in this study. Primarily, the
target behavior being studied, number of steps, varied tremendously across time.
Additionally, the target behavior had not achieved a stable rate during the baseline phase and
before the intervention was begun. During the intervention, the number of steps could
range between 37 and 85 over a span of about five days. This inconsistent performance led
the investigators to change the reinforcement system three times. Another shortcoming of
the study was an inadequately described baseline condition. It was not clear if the baseline
condition was a true baseline based on a typical PT session, or if the setting, or the physical
therapists' behavior (verbal encouragement or physical assistance) were different.
Bragg, Houser, and Schumaker (1975) examined the effects of rewarding appropriate
sitting positions in children with cerebral palsy in a multiple baseline design across six
children. Baseline sessions consisted of engaging the child in structured play during which
no comments were made regarding the way the child was sitting. The percentage of
appropriate and inappropriate sitting and the length of time spent in each position were
recorded. The subjects were then randomly assigned to one group that received contingent
praise, affection and food, or a second group that received consistent priming to sit with
their legs in front of them in addition to the praise, affection, and food. Appropriate sitting
was increased in all children and it was also believed to generalize to the classroom setting.
However the authors failed to obtain a baseline rate of inappropriate sitting in the classroom
prior to the study. Additionally, inappropriate sitting was not completely eliminated for any
subject, and there was no follow-up phase to assess continued effects. The authors
hypothesized that because the investigator played with the child regardless of how they were
sitting they may have inadvertently reinforced inappropriate sitting positions.
Pierce and Garland (1977) used an extensive combination multiple baseline and
reversal design to compare the effects of four interventions on promoting motor skills in six
people with physical and mental disabilities. The four conditions were as follows: (1) prior
instruction only before each motor task, (2) instruction plus social reinforcement when
performing well, (3) instruction, social reinforcement, and material reinforcement, and (4)
instruction, social reinforcement, and goal setting. Although each condition was
differentially effective, social reinforcement and material reinforcement were found to be the
most effective combination for increasing performance levels. This led the authors to
conclude that the material reinforcement was the critical component in that intervention.
Despite the proven effectiveness of the intervention, all subjects performance levels returned
to baseline once a procedure was removed. It was recommended that future studies assess
how to maintain motor skills after reinforcement is removed.
According to Hester (1981) ambulatory deficiencies in children with profound
mental retardation may be the result of misplaced reinforcement contingencies. He
described how nonadaptive behaviors may be inadvertently reinforced, whereas, adaptive
behaviors may not be reinforced at all. If indeed deficient skills had been learned through
accidental conditioning, then Hester hypothesized that behavior modification using physical
guidance and positive reinforcement could significantly affect change in such behaviors.
In a multiple-baseline design, Hester (1981) described how he used positive
reinforcement consisting of physical, social, and edible rewards to increase the standing and
walking behaviors of an institutionalized, 14 year old female, with profound mental
retardation. During the course of training, as the child made progress, the physical and
edible rewards were slowly withdrawn. A strength of this study was the addition of the
generalization phase following the intervention, during which another staff member
observed the procedure once before duplicating it for two 30-minute sessions each day.
During the generalization phase, the behaviors initially decreased, but returned to
appropriate levels. An additional strength in this study was the apparent use of the physical
therapist in implementing the training. Unfortunately, despite the therapists attempt to train
other staff members, the high staff turnover rate resulted in inconsistent use of the
procedure. Thus, at two months follow-up, the child's behavior had returned to baseline.
Hill (1985) also presented a rationale for combining behavior modification
techniques with PT in the habilitation of patients with cerebral palsy. Similar to Hester
(1981), she described how a person with cerebral palsy can develop maladaptive motor
function as a result of learning and interactions with the environment. More specifically, an
individual with cerebral palsy is likely to be limited motorically and thus lack opportunities to
practice appropriate movements. Without an adequate foundation of basic motor skills,
more complex motor skills may never be developed. In addition, because an individual with
cerebral palsy may experience limited range of motion, weakness, and increased spasticity
that may also cause increased pain, certain movements may actually be punishing. Thus,
such an individual will adapt to these problems by avoiding those movements and
developing abnormal movement patterns that are easier and likely to be reinforced with
success and accommodate his/her abilities. Such behaviors are often maladaptive, and may
be resistant to treatment.
Rapport and Bailey (1985) used a multiple-baseline design across outcome measures
designed to assess fine and gross motor skills after implementing a home-based behavior
therapy program combined with an existing PT program. The goals of the program were to
increase compliance and improve fine and gross motor skills for an 8 year old with spina
bifida. A unique addition to this study was the involvement of the parents, who were trained
to use a variety of game-like tasks to practice the skills. In addition, a motivational chart was
used to indicate those activities that were to be worked on each day and to display a point
system for earning rewards for each task. Points could be exchanged for a variety of
reinforcers at the end of each week. Although the authors concluded that the combined
behavioral intervention with the existing PT program was successful, it was not possible to
determine the factor(s) primarily responsible for the change. The improvement in this study
compared to other studies reviewed was the use of a one-month follow-up period that
indicated the performance gains had increased or remained stable.
There are several shortcomings in the existing literature on behavior modification
and PT. First, the studies described only used behavior modification techniques to promote
the learning of motor skills. What has been lacking in the research conducted with children
with developmental disabilities who receive PT, is the use of behavior modification to
address behavioral problems that interfere with PT. A second shortcoming in several of the
studies reviewed was the inability to conclude that the intervention provided was the primary
factor affecting change. Third, several studies failed to provide follow-up data to verify that
such behavioral techniques can provide a lasting effect. Finally there is a lack of attention to
providing training to other people, such as parents and therapists, who are involved in the
care and treatment of the children with physical disabilities.
Behavior Problems in PT
Perhaps it is because physical therapists focus on the development of motor skills
and maximizing mobility that most studies have focused on the application of behavior
modification to the learning of motor skills. However, another aspect of PT in which
behavior modification techniques can be useful is in dealing with behavior problems.
Behavior problems such as aggression, noncompliance, and tantrum behaviors, are
frequently found in PT (Laffi, Mauk, Goh, & Merlino, 1994; Homer, 1971) and often
interfere with progress in PT.
Homer (1971) reported the occurrence of aggression in a 5-year-old child with spina
bifida and mental retardation, as the requirements for walking were increased. The subject
had received surgical procedures to correct bilateral hip dislocations at the age of four.
Following his surgery, he received PT on a regular basis that focused on strengthening
exercises in preparation for walking. Despite ongoing PT, and having the physical
capabilities to ambulate, attempts at enforced ambulation were met with increased negative
behaviors such as tantrums and resistance.
Initially, Horner used a reversal design with a six-step successive approximation
sequence to establish the use of parallel bars to assist with walking. Root beer had proven to
be a reinforcer during the baseline phase. When the child was ambulating independently in
the parallel bars, an extinction procedure was initiated demonstrating control of the
reinforcement contingency over the behavior. Following the extinction period, the subject
received a "free" reinforcer and returned to the final step in the initial treatment period.
After ambulation was obtained in the parallel bars, the use of crutches was
established using a ten-step successive approximation sequence. During the baseline phase
for this stage of the treatment, aggressive behaviors started to occur, such as throwing
crutches and tantrum behavior. The subject was prevented from throwing the crutches by
securing them to his hands with elastic bandages, and a 3-minute time out was enforced
contingent upon his aggressive behavior. Although systematic records of the subject's
resistive behavior were not kept, the author stated that these behaviors responded to the
time-out and use of restraint. Also during this phase of treatment, the reinforcer was
changed after session 54 from root beer to noise that the subject was allowed to make after a
trial was successfully completed. This was reportedly due to the difficulty in being able to
immediately reinforce correct responses. Independent ambulation was achieved, after which
a contingency management program was designed to ensure continued independent
ambulation. The sound methodology allows one to conclude that behavior modification
was indeed the critical variable in training this child to ambulate independently in the
In a single-case study, Manella and Vani (1981), described a four and one-half-year-
old girl with myelomeningocele who presented with a short attention span, tantrum
behaviors and refusal to walk during PT. Using social attention, affection, and game playing,
positive reinforcement was contingent upon the partial performance of certain motor skills.
As the child became more skilled, the amount of assistance needed was decreased, and
independence was reinforced. All inappropriate behaviors were ignored. After the child
exhibited appropriate behaviors in PT, the child's mother was trained in the behavioral
techniques for home use. Results indicated that the child reached the therapeutic goals set
by the physical therapist. Additionally, the mother was effective in implementing the
behavioral program in the home resulting in continued progress. As might be expected, the
child's ambulation deteriorated when the mother began implementing the program, but
quickly returned to and then exceeded the performance she had obtained with the physical
therapist. A limitation of this study was that the frequency of the problem behaviors that
had previously been exhibited before the onset of the study were not mentioned after the
initial description. The occurrence of the behaviors was not documented in the study other
than to state that they were ignored. Thus, it is not known if these behaviors were
eliminated before, during, or after the intervention, or were not eliminated at all. However, a
strength of the study was training the parent in the application of the behavioral technique.
Such training was an essential component in the treatment, and one that has rarely been
addressed in the literature.
Training Programs in Behavior Modification
Programs that train parents and healthcare professionals in behavioral procedures for
children have been shown to promote more positive parent-child interactions at home and
optimize health care provision in the hospital (Eyberg, Boggs, & Algina, 1995; Babbitt et al.,
1994; Singer, Nofer, Benson-Szekely, & Brooks, 1991). Parent training has probably
received the most attention in the literature. Various child-treatment procedures have been
taught in parent-training programs. Most frequently, parents have been trained to apply
differential reinforcement either as the only form of treatment or with other procedures
(Budd, Green & Baer, 1976). One of the key advantages of using a behavioral approach to
deal with the behavior problems of children is the relative ease with which an individual who
may have relatively little skill in the use of sophisticated therapy techniques can learn and
implement the basic principles of behavior change (Riley, Parrish, & Cataldo, 1989).
One of the most common parent training approaches involves the use of modeling,
practice and immediate feedback (e.g., Forehand and King, 1977). Parent-child interaction
therapy (PCIT; Eyberg & Matarazzo, 1980; Eyberg, Boggs, & Algina, 1995) for example, was
developed to treat children with conduct problem behaviors and their families. In PCIT, the
therapist models for the parents certain skills to be learned. The parents then practice the
skills with their child, while receiving feedback through a bug-in-the-ear device. Following
each session, the therapist provides feedback to the parent regarding their performance.
PCIT outcome research has demonstrated statistically and clinically significant
improvements in child problem behavior (Eisenstadt, Eyber, McNeil, Newcomb, &
Funderburk, 1993; Eyberg, Boggs, & Algina, 1995; Eyberg, Robinson, 1982).
Hudson (1982) conducted a component analysis of a group training program for
parents of children with developmental disabilities to teach their children new skills. Forty
mothers were randomly assigned to one of four treatment groups: verbal instruction, verbal
instruction plus the teaching of behavioral principles, verbal instruction plus the use of
modeling and role-playing with immediate feedback, and a wait-list control group. Results
indicated that the inclusion of modeling, role-play, and feedback led to significantly greater
improvement in the ability of parents to develop new skills in teaching their child.
Eyberg and Matarazzo (1980) compared two types of training to promote behavior
management skills in mothers. The didactic group received five, 90-minute sessions where
they were taught the principles and application of behavioral techniques. Mothers in the
individual interaction group were trained in the basic rules of two components of parent-
child interaction. The rules were taught to the mothers in five, 20-minute sessions, using
description, modeling, interaction and feedback. Results indicated that mothers in the
interaction group improved significantly on all targeted behaviors.
"Train the trainer" techniques are perhaps an inevitable extension of the research
demonstrating the effectiveness of behavioral approaches to change human behavior.
Trainers, or behavior change agents, that have been studied include teachers to increase
social interactions (Hendrickson, Gardner, Kaiser, & Riley, 1993), institutional attendants to
apply behavioral techniques (Gardner, 1972), graduate students to train parents @saacs,
Embry, & Baer, 1982), medical residents to increase communication skills (Branch, 1990),
and foster grandparents working with institutionalized children (Fabry & Reid, 1978).
Research on training the trainer has frequently focused on the most effective ways to
train others in the implementation of behavioral techniques (Bernstein, 1982). In a review of
consumer satisfaction in parent training programs, practice with the child was found to be
the most useful training tool, followed by performance-oriented teaching, and trainer
demonstration. The use of written materials was found to be least useful (McMahon &
Forehand, 1983). Gardner (1972) found role playing to be more effective than lecture in
teaching behavior modification techniques to nonprofessionals. Cunningham, Davis,
Bremner, Dunn, and Rzasa (1993) found modeling, role play, and homework most useful in
promoting mastery of new skills. Hosford and Johnson (1983) found that using the self-as-
model technique, with only appropriate behavior viewed, resulted in completely
extinguishing inappropriate interviewing behaviors in counselors.
It is generally agreed upon that instruction alone is insufficient to teach behavior
change skills (Bernstein, 1982; Delamater, Conners, & Wells, 1984; McMahon & Forehand,
1983). However, sufficient support exists for modeling (Bandura, 1969; Cunningham et al.,
1993), rehearsal, role play, and feedback (Delamater, Conners, & Wells, 1984).
Modeling has repeatedly been shown to be a powerful technique for behavior
change. Effective modeling has been used to teach speech and language skills (e.g. Charlop
& Milstein, 1989), improve interviewing skills (e.g. Miltenberger & Veltum, 1988), acquire
motor skills (e.g., Carroll & Bandura, 1985), reduce avoidant behavior (e.g., Meichenbaum,
1971), teach parenting skills (e.g., Webster-Stratton, 1981), and reduce child uncooperative
behavior (e.g., Stokes & Kennedy, 1980). Symbolic modeling in the form of videotapes was
the logical outgrowth of the documented effects of live modeling. Videotape modeling has
several advantages over live modeling, including the ability to create naturalistic modeling
sequences, greater control over creating the modeling scene, ability to present multiple
models or repeated observations of the same model, efficiency, and self-administered
treatment sessions (Thelen, Fry, Fehrenbach, & Frautschi, 1979).
For a videotape model to be effective, the modeling procedure should focus on the
skill to be learned, its context, and its consequences. Bandura (1969) identified four
components that mediate observational learning: attention to modeled events, retention of
what is observed, ability to replicate modeled behaviors, and motivation to reproduce those
behaviors. The characteristics of the model contribute to the effectiveness of the procedure.
Model characteristics such as competence (Baron, 1970), status level (McCullagh, 1986), age
(Bandura & Kupers, 1964), and similarity (Kazdin, 1974) have been shown to strengthen the
degree of observational learning.
The empirical evidence for the use of videotaped modeling allows it to stand alone as
an effective intervention in its own right. However, packaging it with other effective
interventions, especially with opportunities to practice, should be even more effective
(Dowrick, 1991). Bandura (1971) suggested that the effects of modeling could have a greater
impact if it were followed with guided practice or rehearsal in a natural environment.
In his review of the parent training literature, O'Dell (1985) argued that the best
parent training programs have been a combination of visual, verbal and interactive training
methods. Kazdin (1994) also promoted the use of multicomponent interventions for several
reasons: (1) combining individual techniques that have been shown to be effective should
strengthen their impact; (2) a combination of techniques may increase the rate of skill or
knowledge acquisition; and (3) a multicomponent intervention may be able to address a
range of problems that often occur together.
Forehand and King (1977) used cueing, feedback, instruction, role play, and practice
with feedback to train parents of noncompliant children. After observing the parent-child
interaction, the parent was instructed on ways to change his/her behavior to increase the
child's compliance, and the therapist modeled these behaviors. The parent was then able to
practice the skills with the therapist who provided feedback. Finally, the parent used the
skills in playing with his/her child while receiving further feedback from the therapist via a
bug-in-the-ear device. Results indicated that the behavioral criterion for the study were
achieved in an average of nine sessions. Additionally, follow-up assessments indicated that
the behavior changes were maintained at three months.
Matson and Stephens (1978), and Delamater, Conners, and Wells (1984) used
training interventions that included modeling, behavioral rehearsal and feedback to train
staff to use behavioral techniques in their interactions with children on an inpatient
psychiatric unit. Webster-Stratton (1990) found greater effectiveness in decreasing deviance
in children through the use of videotaped modeling plus therapist consultation for parents.
This chapter has described the literature on the use of behavior modification in the
field of PT, primarily with children. Behavior modification has been used most frequently to
facilitate the learning of new motor skills. Although few in number, research studies have
been conducted demonstrating the use of behavior modification to address behavior
problems of children during PT. However, the behavior change agents in such studies were
most often a behavior specialist and the physical therapists were rarely trained in the
behavior modification techniques. This type of approach continues to leave the physical
therapist without the skills necessary to deal effectively with the behavior problems that
frequently occur in the PT setting.
A review of the literature on effective components for training nonprofessionals in
behavior modification techniques was also described. It has been well established that
parents, teachers, psychiatric technicians, and therapists can successfully learn and
implement behavioral techniques (O'Dell, 1974; Eyberg & Matarazzo, 1980; Harris,
Johnston, Kelley, & Wolf, 1964; Rapport & Bailey, 1985; Gardner, 1972). Successful
training components include modeling (O'Dell, 1974), videotaped modeling with behavior
therapist feedback (Webster-Stratton, 1990), practice and immediate feedback
(Eyberg & Matarazzo. 1980).
Based on the repeated success of training behavior change agents through the use of
modeling, rehearsal and feedback, it seems likely that a training intervention combining these
practices would be an ideal training package to promote the acquisition of a new skill. The
present study used a training module, Strategic Interaction Training (SIT), that incorporates
videotaped modeling, individual interaction training, and feedback and reinforcement to
determine if physical therapists can learn to use behavior analysis and therapy reduce
behavior problems in children during PT. SIT was modeled in part, after Parent-Child
Interaction Therapy (Eisenstadt, Eyber, McNeil, Newcomb, & Funderburk, 1993; Eyberg,
Boggs, & Algina, 1995; Eyberg, Robinson, 1982). To date, no other work has trained
physical therapists to apply general behavioral techniques to the problem behaviors that
occur in PT.
Compared to previous studies involving the use of behavior therapy in PT, the
current study differs in several ways. First, in this study, the physical therapists were
directly trained to use behavioral modification techniques to effect change in the behavior
problems that children present with during PT. Second, multiple PTs were trained in these
skills, and the outcome is assessed across several children. Finally, two very different
settings, an outpatient and inpatient setting, were used to demonstrate the generalizability
of such training.
SPECIFIC AIMS AND HYPOTHESES
This study specifically focuses on training physical therapists in behavior analysis and
therapy to decrease resistant, noncompliant and aggressive behaviors in children during PT
and to increase child compliance rates. The training intervention, Strategic Interaction
Training (SIT), was the independent variable of interest. Dependent variables were the
frequency of target behaviors coded for both the physical therapist and the child across
sessions, and social validity measures. It was believed that the results of this study would: (1)
identify problem behaviors that physical therapists frequently encounter; (2) show that
physical therapists can learn to use behavior analysis and therapy; (3) demonstrate that the
use of behavioral techniques by the physical therapist will result in a decrease of negative
behaviors in the children receiving PT.
Specifically, the following hypotheses were tested regarding the
Therapists Target Behaviors
(1) Prior to SIT, it was expected that physical therapists would exhibit low
frequencies of the following behaviors: praise, direct command, ignore, attention to
(2) It was hypothesized that physical therapists could learn to implement behavioral
techniques to address problem child behavior. Specifically, it was expected that the physical
therapist would increase rates of the following behaviors to a pre-determined goal level:
praise, attention to appropriate child behaviors, ignore, and direct commands.
(3) Additionally, in follow-up sessions, it was hypothesized that the training would
promote the use of the skills with other children who present with negative behaviors in PT.
Child Target Behaviors
As a result of the physical therapists training, the following hypotheses were made
regarding the behavior of children being treated.
(4) Child compliance rates would increase.
(5) Child inappropriate behaviors (e.g., whining, crying, aggression, pain behaviors,
yelling, and smart talk) would decrease.
Consumer Satisfaction Predictions
(6) On a measure of consumer satisfaction, it was predicted that the physical
therapists receiving SIT would rate it as an acceptable, effective, and beneficial training
module to reduce behavior problems in children and increase compliance during PT.
Four currently practicing, licensed pediatric physical therapists provided informed
consent and participated in the training. There were two therapists from an outpatient
setting and two therapists from an inpatient setting. All physical therapists were Caucasian
females and had received a bachelors degree in PT. The average number of years practicing
PT was 15 years and 3 months (range 6 20 years). The average number of years practicing
PT with children was 11.5 years (range = 6 18 years). One therapist from each site was
certified in neurodevelopmental therapy (NDT). One therapist at Shriners treated three
children in the study, one therapist at All Children's Therapy Center treated two children
while the other two therapists each treated one child at each site.
Table 1 presents information on the children and the grouping of children with each
therapist. A total of 7 children were recruited for the study, 4 from Shriners Hospital and 3
from All Children's Therapy Center. The parent(s) of each child provided informed consent
to participate, and for those children over 7 year of age, assent was obtained from the child.
Six children were Caucasian, one child, Sherry, was African American. There were 3 males
and 4 females, and they ranged in ages from 3 years to 12 years old j = 7.9). Each child
was nominated by his/her physical therapist, who was also participating in the study, based
on the presence of behavior problems that interfered with the process of PT. Children in
the inpatient setting had been working with their therapist for 1-2 months prior to entering
the study. Children in the outpatient setting had been working with their therapists for
Child demographics, diagnosis and behavioral issues grouped by treating therapist.
Therapist Participants Gender Age Medical Behavioral Issues Number of
F 8 multiple congenital
tibia hemiamelia with
resultant leg length
discrepancy on left,
treated with an
F 8 spastic diplegic
cerebral palsy; status
M 12 Legg-Calve Perthes
disease of the left leg;
status post left
adductor release and
months prior to his
M 3 spastic quadriplegic
cerebral palsy; status
post bilateral adductor
releases and bilateral
F 8 spastic diplegic
cerebral palsy and a
M 9 spastic diplegic
cerebral palsy; status
F 7 Downs syndrome,
was disruptive to
the flow of PT,
refusing to walk,
limp and falling
over or collapsing,
talking back to the
back to the
and talking back to
approximately two years prior to their involvement in the study. Child behavior problems
included noncompliance, whining, screaming, aggression, becoming limp and falling over or
falling to the floor, talking back to the therapist, and dawdling.
Because of the need for the child to understand language in order to assess
compliance with therapist requests, children with a history of moderate to severe mental
retardation were not included and the ability to understand therapists requests was
determined by observing the child/therapist interactions prebaseline. If any therapists
achieved training criterion during the first two baseline sessions on the target behaviors to be
described they were excluded from the study. No therapist met all criterion during baseline.
Participants were recruited from and treated in two different settings. Two physical
therapists and four children were recruited from Shriners Hospital for Children in Tampa,
Florida. Shriners is a 60-bed pediatric orthopedic, inpatient hospital that serves children
from around the state of Florida, southern Georgia, and the Caribbean. For the purposes of
this research, PT was conducted in a smaller gym adjacent to the main treatment gym and
away from the noise and distractions of other people. This was done for several reasons.
First, by using a quieter room, we were able to obtain a better sound for coding therapist and
child behaviors. Second, it prevented the first therapist from influencing the second
therapist's behaviors after completing the intervention. Finally, baseline and treatment were
both conducted in the smaller gym in order to maintain a stable environment during baseline
and the intervention. Ambulation activities occurred in the main hall, which was typical for
that setting. Because Shriners Hospital is an inpatient rehabilitation setting, each child was
seen for PT every day, excluding weekends. The time at which each child was seen during
the day was relatively constant throughout the research, and each child was seen an average
of 60 minutes, 5 days a week.
Two therapists and three children were recruited from All Children's Therapy
Center, an outpatient therapy center affiliated with a large children's hospital located in St.
Petersburg, Florida. Physical therapy at All Children's Therapy Center was conducted in
one large room, often with other children and therapists present. Because of the physical
layout of the setting, it was not possible to separate the therapists. Two children were
scheduled for PT once a week, one for thirty minutes and one for one hour, and one child
received PT three times a week for thirty minutes each. Additionally, one child scheduled
for PT once a week, had a very high no-show rate resulting in the therapist having less
consistent opportunities to utilize her skills with this child and implement feedback she
received on a regular basis. Both PT departments were equipped with similar PT equipment,
which included bolsters, therapy balls, mat tables, and small staircases.
Each PT session was videotaped during all phases and coded for target behaviors to
be described below. The coding system consisted of portions of the Dyadic Parent-Child
Interaction Coding System II (DPICS-I; Eyberg, Bessmer, Newcomb, & Edwards, and
Robinson, 1994), and revised portions of OTIS (Observation, Training, and Interaction
System-revised; Stokes & Mowrey, 1999), which was adapted from the Systematic Carousel
Observation of Performance (SCOOP; Osnes & Stokes, 1987), as well as the addition of a
pain behavior category that was not covered under DPICS-II or OTIS. Two sequences of
behavior were also coded: 1) the child's response to the therapist's commands and 2) the
therapist's response to the child's inappropriate or pain behavior.
The DPICS-II is a revised and expanded version of the DPICS (Eyberg & Robinson,
1983), a behavioral coding system designed for the assessment of the quality of parent-child
social interactions. It was originally designed to be used as a pretreatment and posttreatment
observational measure, as well as a measure of treatment progress and outcome. DPICS-II
and its categories were designed for multiple purposes and can be modified depending on
the needs of the user. Its reliability and validity have been well supported in a recent study
by Bessmer (1996), who found that the reliability estimates of the DPICS-II codes used in
this study fell in the good (% agreement >70%; kappa = .60 .75) to excellent
(kappa >.75) range.
The SCOOP (Osnes & Stokes, 1987) was originally designed to be used in diverse
settings as a measure of the contingencies in effect in various situations. OTIS was adapted
from SCOOP and was designed for a program serving children with physical impairments.
There are no existing data on the reliability or validity of this observation and coding system.
The child behaviors that were coded included responses to commands, inappropriate
behavior, and pain behavior and were defined as follows:
Compliance (CO; DPICS-II). Compliance with instructions or commands was
coded when the child obeying or beginning to obey a command within 5 seconds after
it was given.
Noncompliance (NC; DPICS-II). Noncompliance is coded when the child does
not begin to complete an instruction, or demonstrates a behavior that is clearly incompatible
with compliance of a command within five seconds of the request.
No opportunity (NOC; DPICS-II). After a therapist issues a command, the child
has five seconds to respond. No opportunity is coded when the child is not given an
adequate chance to comply either because the therapist completes the behavior for him/her,
or the command is clearly out of the range of the child's ability.
Inappropriate behaviors (IB; OTIS). Inappropriate behavior includes behavior
that is distracting, off-task, or disruptive including crying, whining, yelling, smart talk,
aggression. Crying consists of inarticulate utterances of distress (audible weeping) at or
below the loudness of normal conversation. Whining consists of words uttered by the child
in a slurring, nasal, high-pitched, falsetto voice. Yelling consists of a loud screech, scream,
shout or loud crying. The sound must be loud enough so that it is clearly above the intensity
of normal indoor conversation. Smart talk consists of impudent or disrespectful speech.
Aggression is any physical touch that is intended to be antagonistic, aversive, hurtful, or
restrictive of the therapists activities, or harmful to self, or environment.
Pain behaviors (PB; SIT). Pain behaviors include obvious grimacing, vocalized
complaints of pain such as yelling out about pain, whimpering, or crying related to pain.
Inappropriate behaviors and pain behaviors are coded according to the therapist response
Therapist behaviors that were coded included:
Direct command (DC; DPICS-II). A direct command is a clearly stated order,
demand, or direction in declarative form. The statement must be sufficiently specific as to
indicate the behavior that is expected from the child.
Indirect command (IC; DPCIS-II). An indirect command is an order, demand, or
direction for a behavioral response that is implied, nonspecific, or stated in question form.
Praise (P). Praise includes any specific or nonspecific verbalization that indicates
liking, approval, or expresses a favorable judgment upon an activity, product, or attribute of
the child (DPICS-II; combination of labeled and unlabeled praises).
Ignore (I; OTIS). Ignore is coded when the therapist does not provide any
evidence of having heard or seen an inappropriate behavior that occurs (passive ignore); or
the therapist may take action by withdrawing attention, withdrawing physically from the
child, turning away, or removing an object from the child's reach without responding to the
inappropriate or pain behavior (active ignore).
Physical guide (PG; OTIS). If the child is noncompliant following a command,
the physical therapist physically guides the child to complete the command with minimal
physical contact. This could include pushing a child's walker forward so that the child
will continue walking or guiding the child's hand using a hand-over-hand motion to reach
for an object.
Inappropriate behavior or pain behavior followed by attention (A- or PBA;
SIT). Inappropriate behavior or pain behavior followed by attention is coded when a
therapist talks to the child or provides physical touch, other than is therapeutically necessary,
when the child is exhibiting inappropriate behavior or pain behavior. This includes
negotiating (offering a reward such as a desired toy, the presence of a parent, or an activity
contingent on performance), "threatening" (a specific verbalization indicating a negative
consequence will follow a behavior) the child, or commenting on or questioning the child's
inappropriate or pain behavior.
Inappropriate behavior or pain behavior followed by ignore (I- or PBI; SIT).
Inappropriate behavior or pain behavior followed by ignore occurs when a therapist
removes attention by withdrawing from the child, turning away from the child, ceasing
verbalization with the child, or not providing any evidence of having seen or heard the
inappropriate or pain behavior exhibited by the child.
Inappropriate behavior or pain behavior followed by escape (E- or PBE; SIT).
Inappropriate behavior or pain behavior followed by escape is coded when the therapist
discontinues an activity, demand situation or therapy for more than 30 seconds due to
inappropriate child behavior or pain behavior or when an activity is completely terminated
due to inappropriate or pain behavior.
Attention following appropriate child behavior (A+; SIT). Talk or physical
contact with the child occurring before, during, or after appropriate child behavior is coded
as attention following appropriate child behavior.
Two behavior categories, while initially coded separately, were combined in the final
summary of behaviors. Attention to pain behavior was subsumed within attention to
inappropriate behaviors. Although it is certainly recognized that pain often occurs during
physical therapy and can be an important indicator for physical therapists to assess pathology
or the extent to which a patient can be pushed in his/her exercises, depending on the
response following a pain behavior, pain behaviors can also serve as a nurturance trap for
attention and escape (Stokes, Mowery, Dean, & Hoffman, 1997). The criterion level for
ignoring inappropriate behaviors still allowed for therapists to acknowledge 25% of
The primary author served as the primary coder. A second coder who was blind to
the research hypotheses, served as a reliability coder. Before beginning to code videotapes
for the current study, each coder completed a minimum of 30 hours of training. Coder
training consisted of (1) reading and studying relevant portions of the DPICS-II Manual
(Eyberg, et al., 1994) and the OTIS behavioral definitions. (2) discussion of DPICS-II,
OTIS, and the other behaviors described; and (3) successfully completing quizzes that
pertain to each category of the DPICS-II system in The Workbook: A coder training manual
for the Dyadic Parent-Child Interaction Coding system II (Eyberg, Edwards, Bessmer, &
Litwins, 1994). Quizzes involving behavior codes not covered in the DPICS II training
manual were also provided by the primary author. After obtaining 90% on all quizzes of
relevant behaviors, the reliability coder practiced coding videotapes that had been coded and
transcribed by the primary coder. Training was considered complete when the reliability
coder met 80% agreement with the primary coder on two consecutive criterion tapes. The
reliability coder and the primary coder met on a weekly basis throughout the study to
practice coding and discuss differences during practice sessions.
Each session during baseline, intervention, and follow up was videotaped in its
entirety and coded for therapist and child behaviors. The outpatient, thirty-minute sessions,
were blocked into five-minute segments. The 2nd, 4th, and 6th five-minute segments were
coded for target behaviors. During the inpatient sessions that could be one hour or longer,
sessions were blocked into ten-minute intervals followed by five-minute intervals throughout
the session. Frequency coding for target behaviors was conducted during each five-minute
interval. This allowed for a sample of behavior to be coded from the beginning, middle, and
end of each PT session. Frequencies were summed for each behavior category and percent
occurrence or rate per minute was calculated. Percent occurrence was calculated by dividing
the frequency of a specific category, such as direct commands by the total frequency of the
behavior category, such as total commands, in each session. The observers and the video
camera were present for several sessions before initiating data collection to allow the
participants to adapt to the novelty of a new person and the camera. The PT session and
timing began when the patient entered the PT gym, and terminated when the patient exited
the gym at the end of therapy.
The primary author conducted the training intervention with each therapist using a
detailed procedural outline (see Appendix A). Undergraduate research assistants recorded
treatment integrity data from videotapes of each training session using the outline of the
intervention in checklist format. Each intervention session was videotaped and coded for
integrity. Percent agreement was calculated by dividing the number of agreements by the
number of agreements plus disagreements. Treatment integrity was 100%, indicating that
the intervention was implemented exactly as it was outlined and indicating a high degree of
fidelity with the SIT protocol.
Thirty percent of all intervention sessions were randomly selected to be coded
independently by an undergraduate research assistant to assess reliability. The
undergraduate research assistant was naive to the purpose of the investigation. Percent
agreement was calculated by dividing the number of agreements by the number of
agreements plus disagreements on occurrences only. Interrater reliability using Cohen's
kappa (k; Cohen, 1960), which takes into account chance agreement and is thus a more
stringent reliability measure was also assessed. Reliability was calculated on each behavioral
category and data are presented in Table 2.
The mean interrater reliability coefficient was obtained for each behavior category
and ranged from 77% 100% and kappa ratings ranged from .82 .96. Inappropriate or
pain behavior followed by escape was a very low frequency behavior resulting in an average
percent agreement of 50% and thus was removed as a dependent variable.
Summary of reliability for behavior categories
No Opportunity for Compliance
Inappropriate Behavior followed by
Attention to appropriate behavior
Demographics and Medical History
A demographics and medical history form (see Appendix B) documenting the child's
age, gender, medical history, and goals for PT was completed by each physical therapist. In
addition, the therapists provided information regarding their educational background,
number of years practicing as a physical therapist, number of years working with children,
and whether they were NDT certified.
Consumer Satisfaction Questionnaire
Therapist satisfaction with the treatment intervention was evaluated using the
Consumer Satisfaction Questionnaire (see Appendix C). This 13-item questionnaire was
administered to each therapist to evaluate various aspects of the training on a 5-point scale
(1 = negative, 5 positive), such as relevance of the training intervention, time
commitment needed for the intervention, and whether or not the therapist perceived
behavior changes in the children. The questionnaire also included four open-ended items
that requested the therapist to provide feedback on the degree to which the intervention was
acceptable to them, the most helpful and least helpful aspects of the training, along with any
other feedback for the improvement of Strategic Interaction Training.
Procedure and Design
A multiple-baseline design across children and therapists (Baer, Wolf, & Risley, 1968)
was used in each setting to evaluate the effect of Strategic Interaction Training on therapist
and child behaviors. A multiple baseline design allows for the visual inspection of the data,
allowing the reader to quickly see the relationship of one set of data to another set of data.
The visual analysis of data tends to be more conservative and is thus commonly used to
assess variables that may impact behavior rather substantially (Parsonson & Baer, 1992).
Additionally, the multiple baseline design is tailored for unique, applied settings and
circumstances when only small numbers of participants are available.
The study was first conducted in an inpatient setting with three children being
treated by Therapist 1 and one child being treated by Therapist 2. Baseline data were
gathered on the target behaviors for each Therapist/Child dyad. Because of the multitude of
target behaviors, it was difficult to achieve a point in the baseline where all target behaviors
were stable. The intervention occurred with Therapist 1 when a stable baseline occurred for
the majority of target behaviors and baseline observations continued for Therapist 2. Once
the behavior of Therapist 1 showed a clear change on the target behaviors, Therapist 2
received the intervention. This design was replicated in the outpatient setting, where
Therapist 3 who was treating one child, received the intervention first, followed by Therapist
4 who was treating two children. To assess generalization of the therapists skills across time
and with non-targeted children, follow-up data were also collected one month
postintervention. Thus, this study involved two multiple baseline designs with three phases:
baseline, intervention, and follow-up on therapist skills.
During the baseline phase, therapists were instructed to conduct their PT sessions as
usual. No other instructions or recommendations were provided. Each PT session varied
between children but was similar for each individual child across sessions, and was based on
each child's individual PT goals. This most often consisted of requests to perform multiple
repetitions of strengthening and range of motion exercises, and ambulation activities. Play
activities were often incorporated into each PT session.
The Strategic Interaction Training was conducted in three phases:
Phase one -Didactic training and modeling videotape
During this phase, the physical therapists viewed a videotape of another female
physical therapist providing PT to two different children who exhibited a variety of negative
behaviors such as crying, noncompliance, and resistance. The therapist in the videotape
modeled the use of the following behavioral techniques for which participating therapists
would receive training: direct commands, praises, attention contingent upon appropriate
behaviors, and ignoring inappropriate behaviors.
The physical therapist and the behavioral consultant, the experimenter, then
discussed the definitions of target behaviors and the rationale for the therapist's target
behaviors. The therapist was provided with a handout of behavioral definitions (see
Appendix D) in order to promote concurrent use and understanding of the language being
used to define behaviors.
The basic guidelines of a behavior analysis and therapy approach with an emphasis
on differential reinforcement were then reviewed. Therapists were instructed to ignore all
inappropriate behaviors, provide positive reinforcement in the form of praise and attention
contingent upon compliance and positive behaviors, use time-out or a physical guide for
noncompliance, use clear, concise and age appropriate commands to direct the child's
behavior, and be consistent in the use of the techniques. This didactic and modeling phase
typically took 90 minutes to complete.
Phase two -Interactive training
Following Phase 1, the therapists practiced their skills by providing PT to non-target
children. After a ten-minute warm-up period, the therapist's behavior was coded for the
next five minutes without receiving prompts or feedback. The primary author provided
prompting and feedback for the remainder of the session to help shape the therapist's
appropriate use of the behavioral techniques. Additionally, target behaviors of the child
were labeled to facilitate the identification of behavior problems.
Following each PT session, the therapist received prompt verbal feedback on her
overall performance. Criterion for discontinuing Phase 2 included the therapist providing
75% of all praise and attention contingent upon appropriate behavior, 75% of all requests
for action were direct commands, and 75% of negative behaviors were ignored during the
five minute segment that was coded. Therapist 2 had four practice sessions during this
phase, and Therapists 1, 3, and 4 all had five practice sessions. Each practice session
averaged one hour.
Phase three -Feedback
When the therapist reached criterion, she resumed PT with the target children from
the baseline period. All sessions continued to be videotaped. Each session began with a
brief reminder of the behavioral goals for that session based on the observations from the
previous session. The therapist then received feedback only at the end of each session.
Because of the time constraints on the therapists at All Children's Therapy Center, the
therapists frequently received feedback from the author via telephone, or occasionally by fax.
This phase lasted until the completion of the research. However two children were
discharged prior to the end of this study and thus have limited data intervention.
A follow-up observation period was attempted with each therapist approximately
one month after completing the training to test the hypothesis that the intervention allowed
the therapists to generalize their skills to other non-target children. Each therapist was
observed during 2-5 PT sessions with different children. Each session was videotaped and
coded for therapist target behaviors only for comparison across baseline and intervention
data. Because of difficulty in scheduling these sessions, and in recruiting appropriate
participants, some follow-up sessions were not conducted until 8 months postintervention.
Percent occurrence or rate per minute was calculated for each of the child and
therapist target behaviors. Percent occurrence was calculated by dividing the total frequency
of a behavior by the total number of occurrences in the general behavior category for each
session. For example, to calculate the percent occurrence of direct commands for each
session, the frequency of direct commands was divided by the total number of commands
(direct and indirect). Results are summarized first across all therapists and children in relation
to the original hypotheses and then compared between the two settings in which the study
Direct commands. The percent occurrence of direct commands for all dyads is
shown in Figures 1 and 2. During the intervention phase, all four therapists exceeded the
goal level for 75 percent of all commands to be direct. Prior to the intervention the average
percent occurrence for direct commands was 62 percent (range = 42% 80/6) (see Table 3).
During the intervention the average percent of direct commands increased to 83 percent
(range = 79% 87/6). In the inpatient setting, therapists increased their direct commands an
average of 27 percent, while therapists in the outpatient setting increased their direct
commands by 15 percent. Therapist 1 (with Ron) and Therapist 4 (with Bryan) had already
attained the goal level for direct commands prior to the intervention. Additionally, Therapist
4 demonstrated an upward trend during baseline. As can be seen in the multiple baseline
Mean occurrence of target behaviors at baseline and intervention for
Behavior Category Baseline Intervention
Direct Commands 62% 83%
Compliance 82% 88%
Noncompliance 16% 12%
No opportunity for 55%/
Inappropriate child 1.4/n 1/mm
Praise 1/min 2.7/min
All attention to
appropriate child behavior96
behavior followed by 20% 78%
design, the increase in direct commands for Therapist 2 did not occur until after the
intervention, and follows the increase in direct commands attained by Therapist 1. The upward
trend demonstrated by Therapist 4 with Bryan began after the intervention had already been
implemented with Therapist 3. However, Therapist 4 did not reach the goal level with Mary
during baseline and, in fact, had a downward trend with Mary.
Inappropriate behavior followed by ignore. The percent occurrence for ignoring
inappropriate child behaviors was calculated by dividing the number of inappropriate child
behaviors that were ignored by the total number of inappropriate behaviors followed by ignore
plus attention. The percent occurrence of ignoring inappropriate child behavior for each
child/therapist dyad in the inpatient setting is shown in Figures 3 and 4. If there was no
inappropriate behavior during a session to ignore, then no data point is present in the graph. All
four therapists demonstrated significant increases in ignoring inappropriate behavior
-I t i
. 0 0
1 i 1 1 1 t 1 1 I
NN~~A AlA A
Figure 1: Percent occurrence of therapist direct commands--inpatient setting
- 0n- -0 --------
I i I I I I
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4 5 6
0 / \ 0-0-* / _* --,.
A A A-~ A
N _a -d-+--
0 0 +
Figure 2: Percent occurrence of therapist direct commands--outpatient setting
- -a-- -- --- --l -
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4 5
I[ I n I q
I I I I t
-- I I
._ ___ I_ _ _ __ __t_ _
N-- ar -----en
41 h I I I I I I III I i I I I I I
. 9 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4
Figie 3: Percent occurrence of ignoring inappropriate child behavior--outpatient setting
I I I
1 i t,
I I I I I I I I I
II I I I
80 Therapist 4/Bryan
_-_-- -4- ---+-
-~~~ I I I I I --.r, F I I I I I
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4
Figure 4: Percent occurrence of ignoring inappropriate child behavior--
during the intervention. During baseline, very little inappropriate behavior was ignored (L4
20%; range = 9% 30%). The goal for this category was to ignore 75 percent of all
inappropriate child behavior.
During the intervention, there was a dramatic increase by all therapists to an average
rate of 78 percent (range = 65% 91%). Therapists in the inpatient setting attained an average
increase of 62 percent while the therapists in the outpatient setting attained an average increase
of 53 percent. The average percent occurrence for this behavioral category at baseline and
during the intervention for all therapists is shown in Table 3. Although all therapists
demonstrated significant increases in ignoring inappropriate behaviors, the average percentage
of two of the therapists from the outpatient setting did not meet the goal level with two of the
children (Therapist 3 (T3)/Katy = 65%, & T4/Mary = 65%). Therapist 1 exhibited an obvious
upward trend during baseline with Ron, which continued during the intervention and was stable
during the last two sessions at 100 percent. Therapist I also exhibited a subtle upward trend
with Sherry and Sierra. However, following the intervention, the mean shifts between baseline
and intervention were more obvious for Sherry and Sierra.
Attention to appropriate child behavior. Figures 5 and 6 show the total attention to
appropriate child behavior, a combination of both attention and praise. The percent occurrence
of attention to appropriate child behavior was calculated by dividing the total amount of
attention to appropriate behavior (A+ and Praise) by the total amount of attention to both
appropriate and inappropriate behaviors. During baseline, all therapists demonstrated high
levels of attention to appropriate child behavior. Attention to appropriate child behavior was
more variable during baseline, whereas during the intervention, the rates of attention were
higher and more stable.
Non-0- - -
I I I I I t 1 I
. 9 0
I' I I I I I I I
1 i i I i I i i i i i f 1 i I
. . 0 *
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4 5 6
Figure 5: Percent occurrence of therapist attention to appropriate child behavior--
a 0 0
I I $ I I
I I I I I I I
-4- -4- cd
5~S 0 0
A a 0 +
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4 5
Figure 6: Percent occurrence of therapist attention to appropriate child behavior--
Therapists in the inpatient setting had an average increase of 17 percent during the
intervention, while therapists in the outpatient setting had an average increase of 12 percent The
mean occurrence of attention to appropriate behavior across all child/therapist dyads at
baseline and intervention are provided in Table 3. Therapist 1 demonstrated an obvious upward
trend during baseline with Ron that increased further during the intervention. However,
Therapist 1 did not exhibit the same trend with Sherry or Sierra. Therapist 4 also demonstrated
a slight increase in this target behavior during baseline with Mary that increased further
during the intervention.
Praise. Praise was defined as a separate and specific form of attention to appropriate
child behavior. Praise was directed at child conduct not motor learning. The rates of praise per
minute for each child/therapist dyad are depicted in Figures 7 and 8. As predicted, all four
therapists demonstrated low levels of praise during baseline. Only after the implementation of
the intervention did all therapists showed substantial improvements in providing praise. The
mean rates of praises per minute for all therapists at preintervention and intervention are shown
in Table 3. Before the intervention, therapists provided an average of 1 praise per minute
(range = .42 2.1). During the intervention, there was an increase by all therapists to an average
rate of 2.4 praises per minute (range = 1.8 3.8). The decline in the rate of praise for Therapist
1, with Ron, is likely the result of the increased independence Ron exhibited in completing his
exercises prior to discharge. Ron's increasing independence with his treatment program
resulted in decreased interactions with his therapist..
Compliance. As a result of the therapists' training, it was hypothesized that child
compliance rates would increase and inappropriate child behaviors would decrease. The percent
occurrence of compliance across each therapist/child dyad is shown in Figures 9 and 10.
/ / \ 4
- arg- -et- --
Figure 7: Rate of therapist praise to child--inpatient setting
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4 5 6
I I I I
I I I I I I I I I I I I
I I I I I I
. -- .*
A _A X _"_,A /% --
Figure 8: Rate of therapist praise to child--outpatient setting
I t I I I I
f _I I I I I I I I I I i I I I I I
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4 5
The percent of commands complied with was calculated by dividing the number of
commands complied with by the total number of commands followed by compliance plus
noncompliance. Thus commands with no opportunity to comply were not included. There
was variability for all of the children during baseline and the intervention, ranging as much as
35 percent between sessions. Although the rates of compliance were relatively high to begin
with, during the intervention, 5 of the 7 children improved their rates of compliance further.
However, of the remaining two, one child (Sierra) had a small decrease in compliance and
another child (Sherry) had no mean change overall. Two children (Mary & Sherry) had an
upward trend during baseline that confounded the improvements seen during the
intervention. The average percent occurrence of compliance for both baseline and
intervention across all children is shown in Table 3.
Noncompliance. Figures 11 and 12 show the percent occurrence for child
noncompliance with therapist commands. Rates of noncompliance were calculated by
dividing the frequency of noncompliance by the total number of commands given with
opportunities to comply. During the intervention, five of the children (Ron, Alan, Katy,
Bryan and Mary), exhibited declines in their rates of noncompliance. Although one of the
graphs (Sherry) indicated a downward trend in noncompliance, the two spikes in
noncompliance during sessions 6 and 9 resulted in no change in the mean rate of
noncompliance overall. Another child's rate of noncompliance (Sierra) declined throughout
the intervention, however her mean rate of noncompliance across all sessions increased
from a baseline rate of 7 percent to 15 percent during the intervention.
Inappropriate child behaviors. Figures 13 and 14 show the rates of inappropriate
child behavior per minute across all therapist/child dyads. There was tremendous variability
in the frequency of inappropriate child behaviors across sessions during both baseline and
A A A A
-- --- -
Figure 9: Percent occurrence of child compliance--inpatient setting
-6 -0 -
I t I I I I I I I
A A -,_A
0 A 0-,
1111111 -I 1 -I 1 1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4 5 6
!. I . I t I i I I .! I I I I I I
J . .
I I I I I I I I I I I I
7 9 11
13 15 17
Non- e ---+--
Figure 10: Percent occurrence of child compliance--outpatient setting
19 21 23
___________ 4 +
I I I I I I I I I I I I I I I I I I I I I I I
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4 5
-- tar-et h------dr
0 --- --0
0 o i
t _I 1 I --I
-0- -- --
0 0 \
F FF 1T T
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4 5 6
Figure 11: Percent occurrence of child noncompliance--inpatient setting
I-- ---0 --0- -+--
1 2 3 4 5 6
7 8 9 10 11 12 13 14 15 16 17 18
1 2 3 4 5
Figure 12: Percent occurrence of child noncompliance--outpatient setting
I I I i I
I I I I I I
I I I I
the intervention for six of the seven children. On average, the frequency of inappropriate
behaviors across all children during the intervention decreased from 1.4 per minute (range = 1 -
1.7) to 1 per minute (range = .2 1.7) (see Table 3). However, for two children (Katy, and Mary),
the average rate of inappropriate behaviors during the intervention increased. The average
decrease in inappropriate behavior was greater than the average increase. Specifically, for the two
children whose rates of inappropriate behavior increased, the average increase in frequency was .2
inappropriate behaviors per minute (range = .1 .3) while the average decrease in the frequency
of inappropriate behaviors per session for the four other children was .8 (range = .5 1.1). For
one of the children (Alan), there was a very large and stable reduction in the occurrence of
inappropriate behavior. Katy, on the other hand, demonstrated an increase in inappropriate
behavior. The increases in her inappropriate behavior appear to be an extinction burst directly
related to increases in the frequency of ignoring inappropriate behaviors by her therapist. Mary
also demonstrated a tendency towards increasing inappropriate behaviors; however, the limited
data points for Mary during the intervention preclude making strong conclusions about her
behavior. Sierra's inappropriate behavior showed an increase during baseline with an upward
trend during the intervention, although the mean rate of inappropriate behaviors did not change.
Ron demonstrated a downward trend during baseline that continued during the intervention;
however, his awareness of his impending discharge may have contributed to the decrease in his
Follow-up. Follow-up was conducted to determine if the behavior management skills
the therapists learned in the intervention would generalize to non-target children. Follow-up data
for therapist target behavior are shown in Figures 1-8. Mean occurrence of target behavior is
found in Table 3. Follow-up data for therapist target behaviors indicate that the majority of
+8 2.5 .
= 2 1 Non-target ch
3 Therapist I/S
2.5 Non-target ch
0.5 *----. F
0 4- F
3 Therapist 1/
.5 Non-target ch
0 F F F I F F F F F F F F F F F F F F
2.5 ---Therapist 2/,
2 Non-target ch
1 2 3 4 5 6 7 8 9 101112131415161718 1 2 3 4 5
Figure 13: Rate of inappropriate child behavior--inpatient setting
I I -
Figure 14: Rate of inappropriate child behavior--outpatient setting
I I I I I I I I I I I
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4 5
the skills learned from the intervention were generalized to non-targeted children
postintervention. Specifically, direct commands, praise and attention to appropriate
behaviors remained at or above the criterion level. During the follow-up sessions, the levels
of praise decreased somewhat for most therapists but still remained well above the baseline
level. The frequency of praises per minute decreased during follow-up for Therapist 4 and
Therapist 2 but increased further for Therapist 1 and Therapist 3. All therapists' follow-up
levels of praise remained well above their baseline levels. The percent occurrence for direct
commands fell during follow-up sessions for Therapist 1 when compared with the rates of
direct commands with all three target children. However, the average rate of direct
commands remained well above Therapist 1's baseline level achieved with Sherry and Sierra.
The rate of direct commands also fell during follow-up sessions with Therapist 2, but again
the follow-up level remained well above the average achieved at baseline.
Attention to appropriate child behavior remained high and stable for all therapists.
The ability to ignore inappropriate child behaviors fluctuated among the therapists.
Although the average rate of ignoring remained higher than the baseline levels, it fell for all
therapists. This may have been due to the very low rate of inappropriate behaviors exhibited
by the follow-up participants (1-2 inappropriate behaviors per session).
Comparisons between settings. Table 4 shows the average change on several
behavior categories for therapists and children in both settings. Overall, the therapists made
positive changes in the target behaviors in both settings. As a result of the therapists'
changes, the majority of child behaviors also improved. However, the degree of change of
behavior varied in each setting. Therapists in the outpatient setting had lower rates of
ignoring inappropriate child behavior compared to therapists in the inpatient setting. On
other target behaviors, they attained very similar rates of behavior although they may have
Average change in the inpatient setting vs. outpatient setting.
Behavior Category Average change Average change -
inpatient setting outpatient setting
Direct commands 27% increase 15% increase
Child compliance 3% increase 11% increase
Child noncompliance 0 10% decrease
Commands with no
opportunity for 17% decrease 15% decrease
Praise 1.5/minute increase 1.9/minute increase
Attention to appropriate .3/minute decrease A/per minute increase
Inappropriate child .7/minute decrease 0
behavior followed by 62% increase 52% increase
started at different levels at baseline. For example, therapists in the inpatient setting made a
27 percent increase in their direct commands across all children; however, this group of
therapists started with a rate of direct commands 14 percent lower than that of the
outpatient therapists. Overall, both groups attained a similar average rate of direct
commands (see Table 5).
Children in the outpatient setting had a larger increase in compliance (11%)
compared to the children in the inpatient setting (3%). However, children in the outpatient
setting started with a 10 percent lower rate of compliance in the inpatient setting. Yet both
groups reached a similar level of compliance during the intervention phase. Additionally,
Table 5. Average rate of occurrence for target behavior at pre- and post-intervention and
No opportunity for
All attention to
behavior followed by
children in the outpatient setting had a larger decrease in noncompliance (10%), whereas
setting experienced no change, on average, in rates of noncompliance. However, once again,
children in the outpatient setting had higher rates of noncompliance during baseline (24%)
than children in the inpatient setting (11%).
On the measure of consumer satisfaction (see Appendix B), it was predicted that the
therapists receiving SIT would rate the training as acceptable, effective and beneficial in
reducing behaviors problems and increasing compliance in children during physical therapy.
Table 6 shows the mean scores on the satisfaction questionnaire. Overall, the items on the
questionnaire were responded to positively except the time commitment needed for the
study was on average rated as "a little long". Additionally, one therapist rated the behavior
of one child as "somewhat worse" (Therapist 1/Sierra). Two therapists commented that the
most helpful aspects of the training included "learning to reinforce positive behavior and to
ignore negative behavior", another therapist reported that "definite guidelines to follow
when there was a compliance problem" was most helpful. When asked to describe the least
helpful aspects of the training, one therapist reported "nothing really". Another therapist
reported that "time out just did not fit in this setting, nor did time out or a physical guide
work with all kids or all ages." Two therapists made no suggestions for improvement.
Mean scores on Consumer Satisfaction Questionnaire.
Item Mean Range
1. Rationale for the training 4.75 (4-5)
2. Relevance of training to work 4.25 (4-5)
3. Learning techniques for dealing with behavior 4.25 (4-5)
4. Time commitment needed for the training 2.00 (1-3)
5. Acceptability of techniques 4.25 (3-5)
6. Child's behavior following training 4.14 (2-5)
7. Child's compliance following training 4.43 (4-5)
8. Changes observed in non-target children 4.50 (4-5)
9. Intervention facilitated more productive PT 4.50 (4-5)
10. Overall rating of training (10-point scale) 8.25 (7-9)
This study sought to determine if physical therapists could learn general behavior
analysis and therapy skills using an intervention program that incorporated a modeling
videotape, individual interaction training, feedback and reinforcement. An additional
purpose of this study was to determine if the physical therapist's use of behavior analysis and
therapy would result in decreased levels of inappropriate child behaviors and increased levels
of compliance during PT. A multiple baseline design was used with two therapists and four
children in an inpatient children's hospital, and two therapists and three children in an
outpatient PT setting.
Results indicated that after participating in Strategic Interaction Training (SIT), the
physical therapists learned the behavior analysis and therapy skills and incorporated them
into their daily PT sessions and, that the changes in their behavior were the result of the
intervention. Specifically, following the intervention, therapists increased their levels of
direct commands, ignoring inappropriate child behavior, attention to appropriate child
behavior, and praise. As a result, compliance increased, and noncompliance and
inappropriate behaviors decreased for most children. Results also indicated differences in the
amount of change in target behaviors between participants in the inpatient setting and the
outpatient setting. The intervention was rated positively by the therapists.
Ignoring inappropriate child behavior seemed to be the most difficult behavior
change for therapists to make. Two therapists did not reach the goal level during the
intervention, and the rates for this behavior fell during follow-up for all therapists. One
possibility for this finding is that the goal level was too high. However, it is worth noting
that the average percent occurrence of inappropriate child behaviors increased for the two
children whose therapists did not reach the goal level set for ignoring inappropriate child
behaviors. It seems likely that the intermittent reinforcement in the form of therapist
attention to the inappropriate behaviors contributed to the increase in inappropriate child
behaviors. The children whose therapist's average percent of change in ignoring
inappropriate behavior was highest, and thus had a more consistent rate of ignoring, had the
greatest decrease in inappropriate behaviors. Katy appeared to experience an extinction
burst for her inappropriate behaviors. During sessions when the rate of therapist ignoring
was high, the rate of child inappropriate behaviors was also high. When the rate of therapist
ignoring was low, the rates of inappropriate child behavior were also low. Thus, it may be
that for this dyad, the therapist was punished with an increase in inappropriate child
behaviors when she ignored more consistently.
During the intervention, therapists were coached and given feedback to provide a
high level of target behaviors. Following the intervention, the rates of some therapist target
behaviors, such as direct commands, ignoring inappropriate child behavior, and praise, fell
for most therapists. Follow-up sessions were conducted at least one month post-
intervention. During that time, it is likely that therapists assumed a more natural rate
of target behaviors. Whereas a more consistent rate of ignoring would be best for
extinguishing inappropriate child behaviors, an intermittent rate of positive reinforcement
in the form of praise would have a better effect on the maintenance of appropriate
Five of the seven children demonstrated an increase in compliance to therapist
commands following the intervention. Only one child had a decrease in compliance and one
child's rate of compliance stayed the same. It is difficult to determine cause and effect for
changes in a particular behavior when an intervention targets multiple behaviors, and several
factors may have influenced the positive changes in compliance rates. First, all therapists
made increases in the rates of direct commands. Direct commands tell the child specifically
and clearly what to do versus indirect commands that may inadvertently present a command
as if the child had a choice about complying. Second, the increase in compliance to
commands may be the result of a decrease in rates of commands given with no opportunity
to comply. It seems probable that eliminating multiple commands for the same behavior,
and allowing the child time to comply, would result in the increased compliance rates.
On the consumer satisfaction questionnaire or in comments made to the
experimenter, all therapists commented on the use of time out being difficult to utilize due
to time constraints or the inability to enforce time out through the use of a restraint process.
Although time out has been used extensively and effectively with children with
developmental disabilities and behavior problems (Olmi, Sevier, & Nastasi, 1997), the use of
time out may not have been appropriate for these settings because of the issues of time and
restraint. Handen, Parrish, McClung, Kerwin, and Evans (1992) found time out to be more
effective in promoting compliance in children with mild developmental delays when
compared to a physical guide. A physical guide is often perceived as less aversive, easier to
administer, and does not negatively reinforce the child by allowing the child to escape from a
difficult situation (Handen et al., 1992). The physical guide was viewed differently by
different therapists in this study. One therapist, who only used physical guides and not time
out, commented that she liked having definite guidelines to follow when there was a
compliance problem. Another therapist reported that the physical guide did not seem to
work with older children.
Differences Between Settings
Differences between the settings included the outpatient setting being a more
structured setting where therapists had sequential appointments throughout the day with
very little flexibility in their schedules. This had implications for the amount of feedback
they received as well as the format in which they received their feedback. Because the
therapist usually had another child waiting for the next appointment, these therapists
frequently did not have time immediately following a session with a target child to receive
and discuss the feedback. As a result, the researcher had to provide feedback in different
formats such as via telephone and written feedback via fax. Therapists in the inpatient
setting had a great deal more flexibility in their schedules and were available to receive and
discuss verbal feedback following each session with the target children.
Another aspect of the research that was different between the settings was the length
of the relationship each therapist/child dyad had prior to their involvement in the study.
Children in the outpatient setting had been working with their therapists, for over 1-2 years.
Their behavioral repertoires were strongly engrained and perhaps more resistant to change.
These lengthy relationships may account for the higher rates of noncompliance and lower
rates of compliance seen during baseline. In the inpatient setting, the children had been
working with their therapists for only 1-2 months. These relationships were still forming and
the therapists may have been more lenient, using more indirect commands versus direct
commands, and the children may have been more eager to please, possibly contributing to
the higher rates of compliance seen during baseline.
Another aspect of the research that was different between settings was the presence
of a research assistant or the primary investigator. During the inpatient intervention, the
primary investigator (PI) provided all coaching and feedback during the training and
intervention, with a research assistant observing. Due to significant time constraints, the PI
was not always able to provide the coaching and feedback directly to the outpatient
therapists. Instead, the P1 frequently informed the research assistant as to what behaviors to
coach a therapist in during Phase 2 of the intervention and what feedback to give during
Phase 3. The PI did conduct Phase 1 of the intervention with all therapists.
Additionally, the environment the children were in when not participating in PT and
the frequency of PT with the target children differed between the two PT settings. In the
inpatient setting, children were patients in the hospital for many weeks, away from their
families, schools, and homes. The hospital was the environment in which they were
residing. Additionally, these children were hospitalized for rehabilitative purposes and they
received PT every day. On the other hand, children being seen in the outpatient setting were
living in their own homes with their families, going to their own school every day and
coming to the PT setting for their therapy. These children had different needs for therapy
and were seen once, twice, or at most, three times a week. These children could also miss
their therapy and frequently did. These variations between the settings may have impacted
the rates of change in various behaviors. In spite of the fact that there were several key
differences between the settings, all therapists involved in the study demonstrated positive
changes on target behaviors. This lends support for the generalizability of the intervention
across different settings and therapy formats.
Strengths and Weaknesses of the Study
Strengths of this study include conducting the study in multiple settings.
Additionally, the variety of children, with varying diagnoses and behavior problems indicate
the generalizability of this intervention in promoting positive changes in both children and
therapist's behaviors. Incorporating follow-up sessions that occurred at least one month
postintervention with primarily different, non-target children, further demonstrated the
generalization of the intervention as well as the maintenance of the skills learned.
An additional strength of the study is the involvement of several therapists. Few
studies have involved therapists in the intervention, and then it has only been one therapist
that has been involved. This study is the first study to train multiple therapists in behavior
analysis and therapy.
Weaknesses of the study include differences in the implementation of various
components of the study. The greater use of the research assistant in the outpatient setting
may have accounted for some of the differences in the outcome. However, this also speaks
to the likelihood that such training could be conducted by less well-trained personnel, on a
less frequent basis, perhaps resulting in a more cost-efficient intervention.
Additionally, although this study had more therapists involved in the intervention
that any previous study, 4 therapists and 7 children is still a small sample size. A further
weakness is that two children left the study prematurely during the intervention. In the
inpatient setting, Ron was told he would be going home just prior to implementing the
intervention with his therapist. This information may have had a positive impact on his
behavior above and beyond that of the intervention. In the outpatient setting, Mary left
for a six-week vacation with her family and thus we were only able to obtain three
intervention data points.
The use of a multiple baseline design has both strengths and weaknesses. For the
purposes of this study, a new intervention, it allows for a detailed analysis of both PT and
child behavior. It also allows for ongoing assessment of the intervention as well as
refinement and change in the intervention depending on the data. However, while there are
certain guidelines used with graphic analysis, the reader's interpretations of the findings may
be influenced by their own experiences and views regarding the procedures, meaning, and
importance of the research (Parsonson & Baer, 1992). Additionally, there are concerns
regarding external validity, and the generalizability of the findings outside the sample.
Replication of the results in future studies can increase the confidence of these findings.
This study is an important contribution to both fields of physical therapy and
psychology. It describes an effective training intervention that psychologists can use to help
physical therapists learn behavior analysis and therapy skills to enhance productive physical
therapy. It is the first study to train multiple physical therapists on the use of general
behavior management skills that can be applied with many children with a variety of
behavior problems. Such techniques will allow the therapist to work more autonomously
and deal more effectively with children who present with difficult behavior.
Few studies have incorporated training the physical therapists to use a behavioral
intervention in the PT setting. The majority of studies have typically involved a behavior
specialist to conduct the intervention, leaving the therapist dependent on a behavior
specialist each time they encounter a difficult child whose behaviors interfere with PT. This
study extends the current literature by demonstrating that physical therapists can learn
behavior management skills and effectively implement these skills into their PT sessions.
Additionally, this study demonstrated the effectiveness of the intervention in two very
diverse PT settings, inpatient and outpatient, which demonstrates the generalizability
of the findings.
This study has further implications for the future of collaborative efforts between
physical therapy and psychology. Overt, disruptive behavior in children is but one aspect in
which psychologists could help facilitate improved physical therapy. Other problem
behaviors can consist of depression, anxiety, fear, chronic pain, noncompliance, coping with
loss (of limb or function), and death and dying issues. Through better integration of mind
and body functioning through increased collaborations between PTs and psychologists, it is
likely that PTs can further enhance the effectiveness of their interventions with both
children and adults facing a myriad of health problems.
This study clearly demonstrates that Strategic Interaction Training resulted in
positive changes in the therapist's behaviors, and positive changes in the majority of target
child behaviors. However, noncompliance and inappropriate child behaviors that were
disruptive to the treatment process still occurred during PT. Having the ability to identify
specific variables maintaining noncompliance and problem behaviors would likely enhance a
therapist's ability to further impact the occurrence of these problems. Recently, researchers
have been training other professionals such as teachers, and parents to conduct their own
functional analyses in the classroom or at home and to generate hypotheses of the functional
relationships of the aberrant behaviors being exhibited (Foster-Johnson & Dunlap, 1993;
Frea, Koegel, & Koegel, 1993).
Boggs, Danforth, & Stokes (1986) demonstrated that mothers of children with
attention deficit hyperactivity disorder could learn to provide contingent rewarding, clear
instructions, and positive attention to their child and that as a result, child compliance rates
increased and inappropriate child behaviors decreased. However, untreated child behaviors
were not affected until the parents were trained to analyze functionally their child's behavior.
Therapists in this study were not trained in a manner that would promote
generalization (Stokes & Baer, 1977). Having learned basic behavioral techniques during SIT,
the ability to analyze functionally child problem behavior would have allowed therapists to
apply more readily the principles and techniques learned during SIT to new behaviors
exhibited by other non-target children during therapy. Such training would be more
consistent with generalization programming promoted by Stokes and Osnes, 1989. The
better equipped the therapist is to assess the functional contingencies of behavior, and utilize
that information, the more likely that treatment techniques will be effective.
This seems a logical next step in training other professionals in behavior analysis and
therapy. In this study, although positive changes were evident for the majority of target
behaviors, individualized approaches for some of the children would likely have resulted in
greater behavior changes. For example, Sherry had both legs immobilized for six weeks in
long leg casts following her surgery. During her admission for rehabilitation, she
demonstrated extreme anxiety about bending her knees. Her inappropriate behaviors,
including aggression, noncompliance and screaming, escalated during activities that focused
on bending her knees. An individualized approach could incorporate relaxation training
exercises such as deep breathing, imagery, and positive self-talk, outside of the PT setting to
provide her with better coping skills during PT. However, not only have physical therapists
not received training in behavioral techniques to deal with problem behavior, they also rarely
receive training in integrating cognitive-behavioral techniques such as relaxation training into
Had the therapist been able to conduct a functional analysis for Sierra, it likely would
have demonstrated that many of Sierra's most difficult behavior problems escalated when
she was asked to perform a particular exercise that was more painful than the others. In this
situation, if the therapist followed Sierra's noncompliance with a time out, her behavior was
reinforced with escape from the difficult and painful exercise. Again, an individualized
approach could have incorporated additional techniques such as relaxation training and
teaching Sierra different ways to perform the exercise independently, thus providing her
more self-control during difficult exercises.
Physical therapists frequently conduct exercises that are painful for their patients.
Many patients may experience fear or anxiety during PT because of the discomfort they
experience. Physical therapist would benefit from knowing how to utilize cognitive-
behavioral techniques such as relaxation training, deep breathing, and imagery. The
integration of psychological interventions with physical interventions can only serve to
further enhance the effectiveness of therapy, decrease some of the aversiveness of therapy,
and empower patients by encouraging them to take more control over their pain, anxiety and
fears. Health and pediatric psychologists could be instrumental in promoting this training
and providing ongoing consultation for such issues.
Parent training is the most commonly implemented behavior management program.
It has been well-established that a group format can be equally if not more effective than
individual training (Eyberg & Matarazzo, 1980; Webster-Stratton, 1990). The format used
for Strategic Interaction Training could easily be adapted for use in a group format for future
use. This format would have the advantages of sharing problems and intervention ideas
among therapists and be a more efficient use of both the psychologist's and therapist's time.
Additionally, child behavior problems are certainly not limited to the PT setting.
Occupational therapists, speech therapist, nurses, and doctors, as well as the child's parents,
could all potentially benefit from similar training in behavior management techniques.
Training multiple exemplars outside of the PT setting would allow for generalization of
child's improved behavior outside of the PT setting as well (Stokes & Baer, 1977). Because
consistency is so important in effecting change in children's behavior, it seems likely that if a
child's behavior is treated consistently by a variety of people he/she comes into contact with,
it would result in further reductions in inappropriate and noncompliance behaviors.
Finally, now that it has been established that physical therapists can learn behavior
therapy skills and effectively implement these skills, it will be meaningful to assess if PT was
in fact more productive in terms of attaining PT goals in a more timely manner following a
training intervention in behavior analysis and therapy. The benefit of such an intervention
to decrease the number of PT sessions needed for some children and increase the
effectiveness of PT should be assessed in future studies. Additionally, the cost effectiveness
and cost offset of such an intervention needs to be evaluated. For children who do not
receive PT due to severe behavior problems, or whose PT does not progress due to their
problem behaviors, a behavioral intervention such as SIT would be clearly indicated, and yet
is infrequently employed due to lack of knowledge by the therapist.
A future study to measure progress made towards PT goals is warranted. Such a
study could take the form of an experimental design involving two groups of children
receiving PT, who are matched with similar disabilities and PT goals, and exhibit significant
disruptive behaviors that interfere with PT. The experimental group would receive PT from
therapists who have received SIT, while the control group would continue to receive
standard PT from untrained therapists. The number of sessions to reach the therapeutic
goals would be the dependent variable.
Several single-subject designs have been conducted with children who were not
progressing in therapy until a behavioral intervention was implemented (Homer, 1971;
Chandler & Adams, 1972). These studies lend support to the efficacy of a behavioral
intervention for children with behavior problems that impede progress in physical therapy.
However, more controlled studies, with larger samples may need to be conducted to further
promote the benefits of incorporating behavior analysis and therapy into the PT setting.
PROCEDURAL OUTLINE SIT TEACHING SESSION
1. Set up room;
a) Put modeling videotape in VCR and set up television.
b) Have handouts on Behavioral Definitions, Key points about Ignoring, and time
c) Have video camera set up with tape and microphone turned on.
1. Provide overview of SIT.
PHASE 1: Today we will talk about certain behaviors and behavioral definitions
and watch a videotape of a physical therapist working with a child who has some
PHASE 2: Starting tomorrow, your PT sessions will be observed, coded, and
videotaped with non-participating children. I will be coding different intervals in
each session. I will also be helping to label target behaviors of the child, and provide
prompting for you to use target the behaviors we discuss today. At the end of each
session, we will review the data on your performance. Your goal is to make 75% of
all praise and attention contingent upon appropriate behavior, ignore at least 75% of
all negative behavior, and to make at least 75% of all requests for action direct
PHASE 3: When the goals as stated above are met, PT will resume therapy with
children from the baseline phase. Each session will be observed, coded, and
videotaped. After each session, you will receive feedback on the target behaviors.
2. Provide handout of behavioral definitions to therapist. Review each definition
and the examples provided before reviewing the tape. Attention will focus on the
therapist's target behaviors noted from baseline.
3.- Commands should be direct rather than indirect.
a) A direct command is a clearly stated order, demand, or direction in declarative
b) The statement must be sufficiently specific as to indicate the behavior that is
expected from the child
c) Give examples of "let's", "how about", and "why don't you" commands.
Examples: "Lets do your leg lifts now." "Can you walk now?" "Lets transfer
to the mat, OK?" "How about ten more?"
Many indirect commands are expressed in question form.
A direct command should leave no question in the child's mind that he/she
is being told to do something, no illusion of choice.
You can still be polite and preface the command with "please". Give
examples of direct commands: "Transfer to the mat please." "Please take
five steps forward."
4. Commands should be positively stated.
a) Tell the child what "to do" instead of what "not to do".
b) Try to avoid "don't" commands. In many situations, it is possible to give a
positively stated command that is incompatible with the negative behavior you
are trying to eliminate (e.g., "Straighten your arm" instead of "Don't bend your
elbow." Or "Put the toy down" instead of "Don't throw that."
5. Commands should be simple.
a) Commands should be things that the child is capable of doing as much as
possible. However, I realize that is difficult to always know what a child in PT is
physically capable of doing, and that as a PT, you are frequently working to
increase the child's ability.
6. Commands should be given one at a time.
a) Children have a hard time remembering more than one thing at a time. Avoid
stringing together commands (e.g., "Go transfer onto that mat, do 10 straight-
leg-raises, 10 heel slides, and 10 short-arc-quads." That's a lot to ask a child to
do all at once.
b) Instead break that big command down into its smaller parts.
c) Another problem in giving a big command is that the child has an awful lot of
work to do before he/she gets positive feedback from you for complying. We
will talk about compliance and praise in detail in a few minutes.
7. Commands should not be vague.
a) Make sure that your commands tell the child specifically what to do.
b) Commands like "be careful", "relax", "be good", "look" are so nonspecific that
the child does not know exactly what to do in order to comply. Also, commands
such as "Do your bridges" does not tell the child how many to do. Instead use
commands like "Put both crutches on the floor please." "Walk to the door
please", "Do ten bridges".
8. -Indirect Commands are orders, demands, or directions for a behavioral response
that is implied, nonspecific, or stated in question form. This type of command
implies that the child has a choice when often they do not. Provide examples. "Lets
walk now", "You might want to sit down now", "How about we do your leg lifts
first?", "Can you reach for the bean bag?"
9. Ask the therapist what he/she would do if the child complied with the
a) Instruct them to give the child a labeled praise contingent upon child
b) A praise is any specific or nonspecific verbalization that indicates liking,
approval, or expresses a favorable judgment upon an activity, product, or
attribute of the child.
c) Distinguish b/w labeled and unlabeled praise.
d) Give examples of praise in terms of compliance:
"You did a great job sitting up all by yourself" (labeled praise)
"I like how well you are walking today" (labeled praise)
"Thank you for transferring to the mat so quickly!" (labeled praise)
"Good job." "Way to go!" (unlabeled praise)
10. -Ask the therapist to provide examples of appropriate behaviors that occur in
PT. List the examples provided by the physical therapist:
a) Discuss how the therapist's attention (verbal, as in talking, or physical touch)
should only be delivered as long as the child is behaving appropriately (e.g., as
long as the child is doing his/her exercises, or the child is calm while being
stretched, the PT can provide attention to him/her).
b) Emphasize how this is different from praise.
c) Encourage therapist not to talk following a command, until the command is
d) Discuss how this might interfere with learning new motor patterns, as well as just
being distracting and interrupting, which may interfere with the child completing
11... Ask the therapist to provide examples of negative behavior that occur in PT.
List the examples the PT provides:
a) Reinterpret these as compliance, noncompliance, inappropriate behaviors, and
b) List your response/reinterpretation next to the example of negative behavior
the PT provides.
c) The therapist will probably list many behaviors that are attention seeking
behaviors or escape motivated behaviors. Identify these.
d) Inform therapist that all inappropriate behaviors are to be IGNORED.
e) Provide Ignoring handout.
a) -When you ignore negative behavior you give no verbal or nonverbal reaction, and
continue what you are doing, except for providing further talk other than to give
b) At first, when you ignore negative behavior, it may get worse.
c) But with CONSISTENCY on your part the inappropriate behavior will go
d) Be sure to be consistent in your ignoring or it will not work. Give example of
child wanting candy in the grocery store. If you do not think you can
consistently ignore a behavior, do not begin to ignore, or stop ignoring early
rather than later.
e) As soon as the negative behavior ceases, provide attention and praise.
13. Discuss Compliance and Noncompliance.
a) If you give a command and the child is noncompliant, meaning he/she does not
initiate the command in 5 seconds and complete the behavior without further
prompting, you are to physically guide the child through the command or give
the warning discussed below.
b) As long as the child is continuing to follow through with the command,
regardless of inappropriate accompanying behavior such as crying, even after 5
seconds, it is compliance.
14. Discuss Time out vs. Physical Guide.
a) If the child is noncompliant after five seconds, you may physically guide the child
through the motion, or command, showing no emotion and without comment.
Then, immediately provide another simple direct command and praise the child
if he/she complies or pause at least five seconds before providing attention to
the child, and then provide attention only to appropriate behavior.
b) If the therapist opts for time out, instruct the therapist to give the following
warning if the child disobeys: "Ifyou don't .......you're going to sit in time out."
Do not repeat your command or ask if he/she heard you. You need to use
these exact words each time. Again, after the warning, you need to decide if the
child is being noncompliant.
15. Ask the therapist what he/she would do if the child obeyed the warning.
Instruct the therapist to give a labeled praise.
16. Instruct the therapist in the time out procedure using the time out diagram.
a) Place the child on the mat and say only 'You didn't do what I toldyou to do soyou
have to stay here on the mat." Ask the therapist what he/she thinks the child will do
at this point.
b) Discuss issues about time out, safety, mobility of the child, etc.
c) Discuss how to get the child to the mat quickly and safely, possibly carrying the
child from behind.
d) Once the child is on the mat, the therapist is to say 'Now stay here until I tellyou to
get off." The child has to stay in the room for exactly three minutes with five
seconds of silence before the therapist can return.
17. Discuss the importance of using those exact words and nothing extra. The
words were chosen to be the shortest simplest way to tell the child the reason he/she
is in time out and what he/she has to do now. The child gets no extra attention.
Time out is effective only if the therapist is in control of when the child can leave
18. -When the child is in time out, what kinds of things could he/she do? Discuss
these issues and how to handle them. Basically, anything the child does is
ignored unless the child's safety is at risk.
19... Define the following:
a) Inappropriate Behaviors includes crying, whining, or yelling, and smart talk.
Crying consists of inarticulate utterances of distress (audible weeping) at or
below the loudness of normal conversation. Whining consists of words uttered
by the child in a slurring, nasal, high-pitched, falsetto voice. Yelling consists of a
loud screech, scream, shout or loud crying. The sound must be loud enough so
that it is clearly above the intensity of normal indoor conversation. Smart talk
consists of is impudent or disrespectful speech.
b) Aggression verbalization of threats, even if said in a joking tone of voice, and
physical threats or behavior directed at (1) therapist, (2) self, or (3) environment.
c) Pain behaviors grimacing, vocalized complaints of pain, such as yelling out,
whining about pain.
20. Discuss Attention to Inappropriate Behavior.
a) Attention to inappropriate behavior includes talking to the child, reacting by
pulling away, or stopping an activity, providing physical touch, other than is
therapeutically necessary, when the child is being noncompliant, crying/whining,
dawdling, or being aggressive. This also includes negotiating, "threatening" the
child, or allowing the child to escape from a demand situation, activity, or
Child is purposefully falling in the walker, and therapist says "Stand up!"
Child is having a tantrum and the therapist tries to calm him/her by talking
or touching, or allows the child a break. As soon as the child calms down,
and complies, he/she can then have a break.
Child is refusing to scoot back on the mat, and therapist continues to say
b.)Describe how attention to inappropriate behaviors increases the likelihood of
increasing inappropriate behavior.
21. Describe Information Descriptions and Behavioral Descriptions.
a) Information descriptions introduce information about people, objects, events, or
activities, but do not clearly describe the child's current or immediately
Examples of Information Descriptions:
"I can tell you have been doing your exercises".
"Tomorrow is Saturday."
"I'm going to work on your legs first."
"Your legs feel tight today".
"This is the last set".
b) Behavioral Descriptions are statements used to describe the child and a verb
describes the child's ongoing or immediately completed verbal or nonverbal
Examples of Behavioral Descriptions:
"You'retransferring to the mat."
"I see you're tLi_ hard to lift that leg."
"You're pushing your wheelchair."
22. Review videotape.
a) Point out examples of praise, direct commands, ignoring negative or
inappropriate behaviors, and attention to appropriate behaviors.
23. Answer any questions. Write down what questions are asked.
24- Review basic rules of behavior modification.
a) Ignore all negative behaviors.
b) Review "Key Points About Ignoring" handout.
c) Provide positive reinforcement contingent upon positive behavior.
d) Use direct commands to increase compliance with requests for action.
e) Use a lot of praise following compliance for commands.
f) Consistency is key!!!
Participants Initials: Subject # Date:
Brief medical history:
Goals for physical therapy:
Physical Therapists initials:
Number of years practicing PT:
Number of years working with children:_
Number of years at Shriner's:
Highest degree obtained:
NDT Training? Yes No
CONSUMER SATISFACTION QUESTIONNAIRE
Please circle the response for each question that best expresses how you
1. Regarding the goals/rationale for the training, I feel that they were:
1. not stated 2. poorly stated 3. stated 4. stated well 5. stated very
2. Regarding the relevance of the training to my work, I feel that the goals for the training
1. not at all 2. a little relevant 3. somewhat 4. very relevant 5. extremely
relevant relevant relevant
3. Regarding the techniques for dealing with negative behavior of children in PT, I feel I
1. nothing 2. very little 3. a few 4. several useful 5. many usefu
techniques techniques techniques
4. Regarding the time commitment needed for the training, I feel it was:
1. too long 2. a little long 3. just right 4. short
5. How acceptable are the techniques you learned in the training to you?
1. not at all 2. somewhat 3. neutral 4. acceptable
6. After the training, how would you describe the patient's behavior
1. considerably 2. somewhat worse 3. the same 4. somewhat
5. very short
7. After the training, I felt that my patient's compliance to my commands or requests was:
1. considerably 2. somewhat worse 3. the same 4. somewhat 5. greatly
worse improved improved
8. How would you describe any behavior changes in the children you work with as a result of
the techniques you learned?
1. much worse 2. somewhat worse 3. the same 4. somewhat 5. greatly
than before than before improved improved
9. How much do you feel that the techniques you learned helped facilitate more productive
physical therapy with the children you worked with?
1. not at all 2. very little 3. no change 4. somewhat 5. very
10. Overall, how would you rate the training you have received to deal with negative behavior
in physical therapy?
1 2 3 4 5 6 7 8 9 10
not at all helpful
11. What aspect of the training do you feel was most helpful?
12. What aspect of the training was least helpful?
13. Please provide any feedback you feel would be helpful enhance the Strategic Interaction
Direct command is a clearly stated order, demand, or direction in declarative form. The
statement must be sufficiently specific as to indicate the behavior that is expected from the
child. Commands should be direct vs. indirect, positively stated, simple, concise, and given
one at a time.
"Lift your leg up."
"Transfer to the mat."
"Do ten more leg lifts please."
"Take five steps forward please."
"Walk to the door please."
Indirect command is an order, demand, or direction for a behavioral response that is
implied, nonspecific, or stated in question form. This type of command implies that the
child has a choice when often they do not.
Lets do your leg lifts now.
Do you want to walk now?
Lets transfer to the mat, OK?
How about ten more?
Compliance Compliance with instructions or commands. Child begins to comply
following a direct command, or an indirect command within 5 seconds after it is given and
completes the behavior without further prompting. Any negative behavior occurring while
the child is complying does not affect compliance and should be ignored. Ex: PT says "Lift
your leg up" and child says "NO!" but lifts leg anyway.
Noncompliance Child does not begin to complete an instruction, or starts but fails to
complete the command, direct or indirect, within ten seconds of the request. This includes
delay tactics that interfere with therapy behaviors that allow a delay in activity.
self-stimulating behaviors (rocking, touching self, etc.)
engaging others in interaction, such as saying "hi" to people in the room/hall
arguing with therapist, parents, or others
distractibility, not paying attention,
commenting/conversing about events or objects in environment
making an excuse, or just saying "NO".
any diversion from the task at hand that interferes with the child initiating the
command within ten seconds and completing the command without further
No Opportunity for Compliance After a command is given, the child is given 5 seconds
to respond. No opportunity for compliance occurs when the child has not been given an
adequate chance to comply.
Commands that request a behavior to be performed in the not-immediate future
(> 5 sec.).
"You can ride the bike when we finish."
*Child dumps out all the bean bags and PT says "You have to put them away
e When therapist completes the action requested in the therapists command, thus
preventing the child from complying. Ex: PT says "Lift up your leg." And then
picks up the child's leg for him/her.
Verb phrases in a command that do not provide enough information for the
child to perform the expected behavior.
"Be careful" can be replaced with a direct command "Put both crutches
on the floor."
"Wait a minute." Can be replaced with a direct command "Stop at the
Praise any specific or nonspecific verbalization that indicates liking, approval, or expresses
a favorable judgment upon an activity, product, or attribute of the child.
You did a great job sitting up all by yourself (labeled praise)
I like how well you are walking today. (labeled praise)
Thank you for transferring to the mat so quickly! (labeled praise)
Good job. (unlabeled praise)
Wow! (unlabeled praise)
Way to go! (unlabeled praise)
Attention following appropriate child behavior Talk or physical contact with the child
contingent upon appropriate child behavior.
Child is ambulating well, and therapist talks to him/her.
Child completes a set of exercises well, and therapists pats his/her arm
and praises his/her efforts
Child is whining, therapist ignores (see below), child stops crying, and
then therapist provides attention after child stops crying.
Child is tolerating stretching and is not whining, and therapist uses
distraction by telling child a story.
If the child is whining but complying, the therapist can continue to
interact and praise the child for the positive behavior that is occurring.
Ex: PT says "Walk to the door." Child complies with the command, but
whines, therapist can praise the compliance and ignore the whining.
Inappropriate Behavior includes crying, whining, or yelling, and smart talk. Crying
consists of inarticulate utterances of distress (audible weeping) at or below the loudness of
normal conversation. Whining consists of words uttered by the child in a slurring, nasal,
high-pitched, falsetto voice. Yelling consists of a loud screech, scream, shout or loud crying.
The sound must be loud enough so that it is clearly above the intensity of normal indoor
conversation. Smart talk consists of is impudent or disrespectful speech.
Do I haaavvvvveeee to? NO! (very loud; yelling)
This is too hard. STOP IT! (very loud; yelling)
I don't want to do this anymore. Your stupid. (smart talk)
Why should I? (smart talk) You can't make me! (smart talk)
I don't like you anymore! (smart talk) Inappropriate grunting/noise
I'm gonna get you back (smart talk) Stop or I'll hit you!
Aggression any physical touch that is intended to be antagonistic, aversive, hurtful, or
restrictive of the therapists activity or aggressive behavior directed at (2) self, or (3)
Ripping up a book
Breaking or throwing a toy
hitting or attempting to hit someone or something (pounding fist in mat), swats
at therapist or parent
Refusing walker and kicking it over or grabbing the walker out of PT's hand.
child pinches therapist
child pushes therapists hand away
Pain behaviors grimacing, vocalized complaints of pain, such as yelling out.
"That's far enough!!!! That's far enough!!!"
Attention following inappropriate child behavior Talking to the child, looking at the
child or providing physical touch, other than is therapeutically necessary, when the child is
exhibiting inappropriate behavior, is noncompliant, or is being aggressive. This would
include negotiating, such as offering a reward such as a desired toy, the presence of a parent,
or an activity contingent on performance, "threatening" (a specific verbalization indicating a
negative consequence will follow a behavior) the child, or allowing the child to escape from
a demand situation, activity, or therapy.
Child is yelling, and therapist continues to talk and try to calm him/her.
Child has stopped ambulating to ask a question, and therapist answers.
Child is screaming inappropriately, and therapist is providing comfort.
Child is being noncompliant and therapist says "If you walk to the door
then we will stop." (negotiating)
Child is not standing appropriately and therapist says "You can play this
game only if you stand up." (negotiating)
If you don't stop crying mommy will have to leave. (threat)
If you throw that toy you will never play with it again. (threat)
Child starts whining during an exercise and therapist responds "When
you stop whining, we will continue."
Child is noncompliant with therapists requests and is sent back to room.
Ignore inappropriate behavior. There is active ignoring and passive ignoring. During active
ignoring, the therapist may take action by withdrawing physically from the child, restrain the
child or remove an object from the child. During passive ignoring, the therapist does not
provide any evidence of having heard or seen the inappropriate behavior that occurs, and
works through the behavior. Often times, the child is allowed to escape from an activity by
using aggression such as trying to prevent you from stretching by pushing your hands away.
Always try to resume the activity, even if for a short duration, so that the child is not
reinforced for aggression or other negative behavior.
Examples of Passive Ignoring:
Child whines, and therapist does not react in any way, or acknowledge the child's
whining. Does not ask child to stop crying or ask what questions. Therapist
continues with the exercise.
Child acts out by purposefully not doing SLR's correctly; therapist may not pay
any attention to this, does not attempt to correct.
Child throws toy; therapist does not scold or acknowledge that behavior
happened; leaves the toy on the floor (passive).
Child tries to prevent therapist from bending his/her knee further by pushing
therapists hand away and PT continues to bend knee, not
acknowledging/addressing child's behavior. When child stops resisting for 2
seconds, therapist can stop bending the knee. This way, the child is not
reinforced by getting the PT to stop when he/she wants.
Child stops walking and begins crying. Therapist can physically guide the child
and push the walker forward briskly to complete the task. The crying is not an
issue and is not addressed. When the task is completed, the therapist can still
PT "Walk to the door."
child is walking to the door, and is almost there when he/she begins to
cry and stops walking
PT briskly and firmly guides the child to the door with no emotion,
comments, or extraneous physical touch.
Once at the door, PT can say "You walked to the door" Does not
comment on crying.
Child is complying with a command but is crying or screaming. PT ignores the
crying and praises the child's compliance.
PT: "Do ten bridges."
child: "No!" but begins doing them
PT: "One, good job doing your bridges. Two..."
Examples of Active Ignoring:
Child acts out by purposefully not doing SLR's correctly; PT physically holds the
child's leg preventing the exercise, but does not address the child. Once child
stops the negative behavior, physical therapist can provide a direct command
such as "Tighten this muscle first, keep your knee straight and lift your leg. OR,
therapist can firmly guide the child through the SLR's, but not address the child's
Child acts like he/she will throw a toy, and therapist removes the toy from the
child's hand and move it away, without speaking or other acknowledgment
Information Descriptions introduce information about people, objects, events, or
activities, but do not clearly describe the child's current or immediately completed behavior.
Examples of Information Descriptions:
"I can tell you have been doing your exercises".
"Tomorrow is Saturday."
"I'm going to work on your legs first."
"Your legs feel tight today".
"This is the last set".
Behavioral Descriptions are statements used to describe the child and a verb describes
the child's ongoing or immediately completed verbal or nonverbal observable behavior.
Examples of Behavioral Descriptions:
"You're transferring to the mat."
"I see you're trying hard to lift that leg."
"You're pushing your wheelchair."
Allen, K. D., & Stokes, T. F. (1987). Use of escape and reward in the management of
young children during dental treatment. journal of Applied Behavior Analysis 20. 381-390.
Babbitt, R. L., Hoch, T. A., Coe, D. A., Cataldo, M. F., Kelly, K. J., Stackhouse, C., &
Perman, J. A. (1994). Behavioral assessment and treatment of pediatric feeding disorders.
Developmental and Behavioral Pediatrics, 15, 278-291.
Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Some current dimensions of applied
behavior analysis. Journal of Applied Behavior Analysis, 20, 313-328.
Bandura, A. (1969). Principles of behavior modification. New York: Holt, Rinehart and
Bandura, A. (1971). Psychotherapy based on modeling principles. In A. E. Bergin & S.
L. Garfield (Eds.), Handbook of psychotherapy and behavior change (pp. 653-708). New
Bandura, A., & Kupers, C. J. (1964). Transmission of patterns of self-reinforcement
through modeling. Journal of Abnormal and Social Psychology, 69, 1-9.
Baron, R. A. (1970). Attraction toward the model and model's competence as
determinants of adult imitative behavior. Journal of Personality and Social Psychology, 14,
Bernstein, G. S. (1982). Training behavior change agents: A conceptual review.
Behavior Therapy, 13, 1-23.
Bessmer, J. L. (1996). The dyadic parent-child interaction coding system II (DPICS-II):
reliability and validity (Doctoral dissertation, University of Florida, 1996). Dissertation
Abstracts International, 58-07B, p. 3961.
Boggs, S. R., Danforth, J. S., & Stokes, T. (1986). Functional problem-solving skills:
Increasing the generality of parent training. Paper presented at the meeting of the American
Psychological Association, Washington, DC.
Bragg, J. H., Houser, C., & Schumaker, J. (1975). Behavior modification: Effects of
reverse tailor sitting in children with cerebral palsy. Physical Therapy, 55, 860-868.
Branch,, W. T. (1990). Teaching models in an ambulatory training program. Journal of
General Internal Medicine, 5, S15-26.
Briener, J., & Beck, S. (1984). Parents as change agents in the management of their
developmentally delayed children's noncompliant behaviors: A critical review. Appied
Research in Mental Retardation, 5, 259-278.
Budd, K. S., Green, D. R., & Baer, D. M. (1976). An analysis of multiple misplaced
parental social contingencies. Joumal of Applied Behavior Analysis, 9, 459-470.
Carr, E. G. (1977). The motivation of self-injurious behavior: A review of some
hypotheses. Psychological Bulletin, 84, 800-816.
Carr, E. G., & Durand, V. M. (1985). Reducing behavior problems through functional
communication training. journal of Applied Behavior Analysis, 18, 111-126.
Carr, E. G., Newsom, C. D., & Binkoff, J. A. (1980). Escape as a factor in the
aggressive behavior of two retarded children. Journal of Applied Behavior Analysis,
Carroll, W. R., & Bandura, A. (1985). Role of timing of visual monitoring and motor
rehearsal in observing learning of action patterns. Journal of Motor Behavior,
Chandler, L. S., & Adams, M. A. (1972). Multiply handicapped child motivated for
ambulation through behavior modification. Physical Therapy, 52, 399-401.
Charlop, M. H., & Milstein, J. P. (1989). Teaching autistic children conversational
speech using video modeling. Journal of Applied Behavior Analysis, 22, 275-285.
Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and
Psychological Measurement, 20, 37-46.
Cunningham, C. E., Davis, J. R., Bremner, R., Dunn, K. W., & Rzasa, T. (1993). Coping
modeling problem solving versus mastery modeling: effects on adherence, in-session
process, and skill acquisition in a residential parent-training program. Journal of Consulting
and Clinical Psychology, 61, 871-877.
Day, H. M., Homer, R. H., & O'Neill, E. 0. (1994). Multiple functions of problem
behaviors: Assessment and intervention. Journal of Applied Behavior Analysis, 27, 279-289.
Delamater, A. M., Conners, C. K., & Wells, K. C. (1984). A comparison of staff training
procedures. Behavioral applications in the child psychiatric inpatient setting. Behavior
Modification, 8, 39-58.
Dowrick, P. W. (1991). Practical guide to using video in the behavioral sciences. New
York: Wiley Interscience.
Durand, V. M., & Carr, E. G. (1985). Self-injurious behavior: motivating conditions
and guidelines for treatment. School Psychology Review, 14, 171-176.
Eisenstadt, T. H., Eyberg, S. M., McNeil, C. B., Newcomb, K., & Funderburk, B.
(1993). Parent-child interaction therapy with behavior problem children: Relative
effectiveness of two stages and overall treatment outcome. Journal of Clinical and Child
Psychology, 22, 42-51.
Eyberg, S., Bessmer, J., Newcomb, K., Edwards, D., & Robinson, E. (1994). Dyadi
Parent-Child Interaction Coding System-II: A Manual. Unpublished Manuscript, University
of Florida, Gainesville, FL.
Eyberg, S. M., Boggs, S. R., & Algina, J. (1995). New developments in psychosocial,
pharmacological, and combined treatments of conduct disorders in aggressive children.
Psychopharmacology Bulletin 31, 83-91.
Eyberg, S., Edwards, D., Bessmer, J., & Litwins, N. (1994). The workbook: A coder
training manual for the Dyadic Parent-Child Interaction Coding System II. Unpublished
manuscript, University of Florida, Gainesville, FL.
Eyberg, S. M., & Matarazzo, R. G. (1980). Training parents as therapists: a comparison
between individual parent-child interaction training and parent group didactic training.
Journal of Clinical Psychology, 36, 492-499.
Eyberg, S. M., & Robinson, E. A. (1982). Parent-child interaction training: Effects on
family functioning. Journal of Clinical and Child PsycholoM, 11, 130-137.
Eyberg, S. M., & Robinson, E. A. (1983). Dyadic Parent-Child Interaction Coding
System: A manual. Psychological Documents, 13, Ms. #2582.
Fabry, P. L., & Reid, D. H. (1978). Teaching foster grandparents to train severely
handicapped persons. Journal of Applied Behavior Analysis, 11, 111-123.
Fordyce, W. E., Shelton, J. L., & Dundore, D. E. (1982). The modification of avoidance
learning pain behaviors. Journal of Behavioral Medicine, 5, 450-414.
Forehand, R. L., & King, H. E. (1977). Noncompliant children: Effects of parent
training on behavior and attitude change. Behavior Modification, 1, 93-108.
Foss, B. M. (1966). Operant conditioning in the control of movements. Developmental
Medicine and Child Neurology, 8, 339-340.
Foster-Johnson, L., & Dunlap, G. (1993). Using functional assessment to develop
effective, individualized interventions for challenging behaviors. Teaching Exceptional
Children, 25, 44-50.
Frea, W. D., Koegel, L. K., & Koegel, R. L. (1994). Understanding why problem
behaviors occur: A guide for assisting parents in assessing causes of behavior and designing
treatment plans. Santa Barbara: University of California.
Fuller, P. R. (1949). Operant conditioning of a vegetative human organism. American
Journal of Psychology, 62, 587-591.
Gardner, J. M. (1972). Teaching behavior modification to nonprofessionals. Journal of
Applied Behavior Analysis, 5, 517-521.
Handen, B. L., Parrish, J. M., McClung, T. J., Kerwin, M. E., & Evans, L. (1992). Using
guided compliance versus time out to promote child compliance: a preliminary comparative
analysis in an analogue context. Research in Developmental Disabilities, 13, 157-170.
Harris, F. R., Johnston, M. K., Kelley, C. S., & Wolf, M. M. (1964). Effects of positive
social reinforcement of regressed crawling of a nursery school child. Journal of Educational
Psychology., 55, 35-41.
Hendrickson, J. M., Gardner, N., Kaiser, P. & Riley, A. (1993). Evaluation of a social
interaction coaching program in an integrated day-care setting. Journal of Applied Behavior
Analysis, 26, 213-225.
Hester, S. B. (1981). Effects of behavioral modification on the standing and walking
deficiencies of a profoundly retarded child. Physical Therapy, 61, 907-911.
Hill, L. D. (1985). Contributions of behavior modification to cerebral palsy habilitation.
Physical Therapy, 65, 341-345.
Homer, R. D., (1971). Establishing the use of crutches by a mentally retarded spina
bifida child. Journal of Applied Behavior Analysis, 4, 183-189.
Hosford, R. E., & Johnson, M. E. (1983). A comparison of self-observation, self-
modeling, and practice without video feedback for improving counselor interviewing
behaviors. Counselor Education and Supervision, 23, 62-70.
Hudson, A. M. (1982). Training parents of developmental handicapped children: A
component analysis. Behavior Therapy, 13, 325-333.