Abstract
A single-case withdrawal design replicated across three families was used to evaluate the impact of the manualized Prevent-Teach-Reinforce for Families (PTR-F) process for addressing challenging behavior. Results across three families showed that (a) families reached high-fidelity implementation of the behavior support plans (BSPs) developed using the PTR-F process, (b) functional relations existed between parent implementation of the BSP strategies and child behavior, and (c) families perceived the PTR-F process and the BSPs as socially valid. Our results replicated findings from prior research on the general efficacy of the PTR process when extended to its use by adult family members in home settings. Implications for practice and future research are discussed.
Keywords
During their early childhood years, young children experience and demonstrate challenging behavior (Dunlap et al., 2006). It is not uncommon for young children to intermittently exhibit behaviors such as tantrums, noncompliance, withdrawal, and aggression, and families often question whether these behaviors are developmentally appropriate and momentary or whether they warrant more serious concern (Fox & Armstrong, 2004; Powell, Dunlap, & Fox, 2006). As children grow and develop social and emotional competence through nurturing and responsive caregiving and instruction, many children will become capable of independently regulating their behavior by the time they enter elementary school (Fox, Dunlap, & Cushing, 2002; Powell et al., 2006). For some young children, however, sustained behavioral difficulties that are more intense than what is considered developmentally appropriate persist despite the best efforts of families and early childhood professionals (Fox, Dunlap, & Cushing, 2002). Toddlers and young children who exhibit these severe challenging behaviors also experience limited learning opportunities and added family stress, which can be difficult to remediate if not addressed early with effectives supports (Buschbacher, Fox, & Clarke, 2004; Campbell, 1995; Didden et al., 2012; Fox, Dunlap, & Powell, 2002; Powell et al., 2006).
One approach that has shown great promise in teaching social and emotional skills and reducing the challenging behavior of young children is positive behavior support (PBS; Fox, Dunlap, & Powell, 2002; Lucyshyn et al., 2007). PBS promotes the likelihood that successful future outcomes will occur by focusing on the provision of supports that lead to meaningful outcomes (Fox, Dunlap, & Cushing, 2002; Fox, Dunlap, & Powell, 2002). When used with families during the early intervention and preschool years, PBS (referred to as family-centered PBS herein) can improve the family’s quality of life by reducing the challenging behaviors of young children in ways that are acceptable to the family and that lead to stable, long-lasting outcomes (Fox, Dunlap, & Philbrick, 1997).
Family-centered PBS has been effective for a variety of populations of children and families and categories of behaviors (e.g., communication, eating and food acceptance, self-regulatory skill difficulties) including the significant needs of children and families who have experienced neglect and abuse (Binnendyk & Lucyshyn, 2009; Buschbacher, 2002; Fettig, Schultz, & Sreckovic, 2015; Hardaway, Wilson, Shaw, & Dishion, 2012). Furthermore, it has consistently been found that families are able to successfully implement behavior support plans (BSPs) and that greater treatment gains and generalization of such gains have been noted with the participation of families in the PBS process (Buschbacher et al., 2004; Fettig & Barton, 2014; Lucyshyn, Albin, & Nixon, 1997). Research has also demonstrated that the maintenance of intervention effects is more likely when families are involved in the PBS process (Buschbacher et al., 2004).
Studies have shown that family-centered PBS creates positive immediate and long-term parental and family outcomes and has been shown to reduce stress levels and promote parental self-efficacy (Brookman-Frazee & Koegel, 2004; Buschbacher et al., 2004; Smith-Bird & Turnbull, 2005). In addition, anecdotal reports from this research have suggested that family perspectives and outlooks for their futures and for the futures of their children improve with family-centered PBS (Buschbacher et al., 2004).
Conceptual Framework and Key Components of Family-Centered PBS
PBS draws on the principles of applied behavior analysis, the inclusion movement, and person-centeredness as its main foundational sources (Carr et al., 2002). While extending the foundations of PBS, family-centered PBS also builds on the behavioral parent training literature (Buschbacher et al., 2004). In family-centered PBS, person-centeredness is expanded to include family-centered practices, and families are provided with comprehensive supports that are individualized to their needs and strengths to encourage optimal functioning within the family system (Dunlap & Fox, n.d.). Because each member of the family affects the family system, family-centered PBS focuses on the development of comprehensive and multicomponent intervention plans that consider the social context of the family and that promote the ability of family members to support young children by resolving behavioral difficulties (Dunlap & Fox, n.d.). Family-centered PBS also involves a multitude of key features that have been described in the literature and that further define its theoretical framework and delineate the practical implications of the approach (Lucyshyn, Horner, Dunlap, Albin, & Ben, 2002). Among the key features cited by Lucyshyn et al. (2002) are (a) working in the context of collaborative partnerships, (b) seeking meaningful lifestyle outcomes, (c) basing support plans of functional assessments, (d) developing and implementing multicomponent BSPs, (e) viewing challenging behaviors as learning problems, (f) insuring that BSPs have contextual fit, (g) adopting the activity setting (routine) as a unit of analysis and intervention, (h) providing support for families’ implementation of BSPs, (i) continuous evaluation, and (j) sincerity and humility of professionals. Prevent-Teach-Reinforce for Families (PTR-F) was as manualized approach to PBS including these characteristics.
PTR-F
The PTR-F intervention process is a revised version of Prevent-Teach-Reinforce (PTR), an evidence-based, manualized form of PBS for school-aged children (Dunlap et al., 2010; Iovannone et al., 2009) and Prevent-Teach-Reinforce for Young Children (PTR-YC), a manualized, evidence-based form of PBS for young children in early care and education settings (Dunlap, Lee, Wilson, & Strain, 2013; Dunlap, Strain, Lee, Joseph, & Leech, 2017). The PTR-F manual outlines the five-step PTR-F process by dedicating a chapter to each step that provides in-depth information in addition to all of the necessary documentation that the step requires. The process is scripted as much as possible for home visitors and families, and each step includes a team-and-self-check evaluation to ensure implementation fidelity.
The purpose of this study was to conduct an experimental analysis of PTR-F with three families of young children with challenging behavior. The aims of this study were to (a) determine whether families reached a high level of intervention fidelity by using the PTR-F process, (b) examine the relation between parent implementation of the BSP and child behavior, and (c) determine how families perceive the PTR-F process and the BSP developed through the PTR-F process. Research questions included the following:
Method
Facilitator
The primary author and researcher served as the PTR-F facilitator for all families who participated in this study. The researcher was White, had a master’s degree in social work, and was pursuing her doctorate in an early childhood special education related field. She was a Board-Certified Behavior Analyst and had worked in homes and schools for 8 years.
Participants
Staff at a childcare center initially nominated the children and families for study participation. Three young children (ages 3 years 1 month through 3 years 9 months) with challenging behavior and their families participated. All young children and families were White and living in suburban neighborhoods outside of a large, metropolitan city in the western United States. Each child lived in a two-parent family home with both parents participating in the PTR-F process. All parents had received a bachelor’s degrees or higher.
Henry and his family
Henry was 3 years 8 months old when his family enrolled in the study. He had no developmental delays or reported diagnoses. He attended preschool at a nonprofit childcare center 2 days per week in the morning for 3 hr. All meetings and sessions took place in the family’s home. Henry’s father, who was 36 years old, worked outside of the home as a software engineer, and his mother, 37 years old, referred to herself as a stay-at-home parent. Henry’s younger sibling also resided in the family home.
Olivia and her family
Olivia was 3 years 9 months old when her family began participating in the study. Residing in the family home was Olivia’s sister (aged 6), Olivia’s mother (aged 37), and Olivia’s father (aged 46). Olivia’s mother was employed full-time in restaurant marketing, and her father was employed full-time in restaurant operations. Olivia’s family had a nanny who cared for Olivia and her sister 5 days per week.
Olivia attended preschool three afternoons per week at a nonprofit childcare center. She was diagnosed with childhood apraxia of speech for which she was receiving speech and language therapy at the childcare center 3 times per week. She was also diagnosed with developmental delays related to her motor skills, and she received occupational therapy 1 time per week at the childcare center.
Nathan and his family
Nathan was 3 years 1 month old when his family entered the study. Nathan’s parents and his sibling also resided in the home. Nathan’s mother, a graphic designer who worked from home, was 36 years old, and his father was 45 years old and worked as a publisher. Nathan did not attend preschool or childcare at study entry.
Settings and Routines
The settings included the interior spaces of each of their families’ homes. Each family chose a particular target routine (i.e., the most difficult routine of the family’s day) on which to focus PTR-F facilitation, and the spaces associated with the routines (e.g., bathroom, bedroom) were the settings for all study and PTR-F steps. The settings and target routines for each family are described next.
Henry and his family
The setting for Henry’s family consisted of the upstairs, downstairs, and garage spaces of the family home. Henry’s family indicated that the most difficult routine of their day was in the afternoon directly before Henry and his mother left to pick up his sister from school. The family indicated that leaving the house was always somewhat difficult for Henry, but that this particular routine was the most difficult due to the time restrictions associated with his mother’s need to leave the house within a certain timeframe to ensure a timely pick up for his sister.
Olivia and her family
Olivia’s bedroom and shared bathroom made up the setting for Olivia’s family. Olivia’s family chose to target their second most difficult routine, Olivia’s morning routine. The family’s most difficult routine was bedtime. Following a discussion of the family’s upcoming schedule changes, and due to the research plan and the nature of a single-case withdrawal design, the researcher and family decided that targeting the family’s second most difficult routine would create less stress for the family. The researcher and family agreed that the researcher would provide support for the bedtime routine after the family had completed all of the experimental conditions for the single-case withdrawal design for the morning routine. Information regarding the bedtime support that was provided for Olivia can be found in the Monitoring Plan Implementation and Child Progress section specific to Olivia’s PTR-F process.
Nathan and his family
The setting for Nathan and his family consisted of the first floor and upstairs spaces of his family’s home. Nathan’s family reported that the most difficult routine of their day was Nathan’s bedtime routine. Nathan’s family indicated that if both his mother and father were home, Nathan would insist that his mother complete his bedtime routine with him. Nathan’s family also reported that if Nathan’s mother completed his routine with him, she would have to lie down with him until he fell asleep. When Nathan’s mother was not home and his father completed his routine with him, Nathan would fall asleep independently after his father left the room without having to lie down with Nathan.
Measurement
Videotaped observational data (i.e., percentage of intervals with challenging behaviors, families’ implementation of the BSP) were collected with a digital video camera. Video data were obtained during the target routine on days that the researcher met with the family for meetings, observations, or to provide coaching during intervention conditions. The primary dependent variable in the study was the percentage of intervals with challenging behaviors.
Percentage of intervals with challenging behaviors
Videos were scored for occurrence of challenging behaviors using modified operational definitions of the behaviors that were chosen by the family and researcher in Step 1 of the PTR-F process; modifications were made to more precisely operationalize each behavior. A 10-s partial-interval recording system was used to score video segments to estimate count for challenging and desirable behaviors. Because all challenging behaviors were defined as discrete behavioral responses, partial-interval recording was used. In addition, partial-interval recording allowed the primary observer to capture any occurrence of the challenging behavior that happened at any time during an interval. The binary coding system was mutually exclusive and exhaustive with only two possibilities for intervals (i.e., challenging behavior or appropriate routine behavior). That is, each interval was scored as a challenging behavior or desirable behavior interval. If challenging behavior occurred during the interval, it was scored as a challenging behavior interval. If challenging behavior did not occur during the interval, it was scored as a desirable behavior interval. The operational definitions used for video coding of challenging behaviors were the following:
Henry: refusal comments (e.g., never, no, I want to play), throwing self onto the floor, crying, whining, running away, and doing something other than what mother instructed to do.
Olivia: refusal comments (e.g., saying, “Olivia not ready,” “No buttons,” “I don’t want that shirt”), making comments to distract adult (e.g., asking for hugs; saying “my boo boo”), flopping to the floor, and running away.
Nathan: crying, yelling, screaming, saying “no mommy,” negotiating for mother to perform the routine, avoiding dad (e.g., hugging mom, grabbing for mom, running away from dad to other room), engaging in behaviors not related to the routine (e.g., running to brother’s room, jumping on bed, hiding in closet), and being held by an adult to perform particular routine steps (e.g., parent holds for teeth brushing).
Family perception of the PTR-F process and BSP (social validity)
To measure social validity, families completed a modified version of the social validity measures developed for the randomized controlled trial of PTR-YC (Dunlap, Strain, Lee, Joseph, & Leech, 2017). The social validity scale included 10 items that were answered based on the family’s agreement with a 5-point Likert-type scale. For most items, a score of one indicated low social validity and a score of five indicated high social validity depending on the nature of the question. Some items were reverse-scored (i.e., a score of one indicated high social validity with a score of 5 indicating low social validity). Items assessed areas such as the family’s satisfaction with changes in their child’s behavior, the family’s perception of the amount of effort and time required to implement the BSP, and whether the family believed that the BSP fit contextually with their familial norms.
Interobserver Agreement
A primary independent observer scored all observational measures. The observer was White, had a master’s degree in school psychology, and was pursuing her doctorate in early childhood special education. The primary independent observer’s scores were used for all condition change decisions. The first author was the secondary observer. Once a minimum interobserver agreement (IOA) of 85% was reached, both observers were permitted to score observational measures. To ensure that observations were independent, each observer scored all sessions at different times and videos that were not used for calculating IOA were used for training purposes. The primary observer noted on each scoring sheet the exact moment (i.e., the time stamp including the hundredths of the second) of the video when coding began. IOA was calculated and reported for all participants and families for a minimum of 33% of all sessions across all study conditions (i.e., 33% of baseline sessions, 33% of Intervention 1 and 2 sessions, and 33% of withdrawal sessions for each participant and family).
Procedural Fidelity
Two forms of procedural fidelity were recorded. First, the procedural fidelity of all experimental conditions was documented using experimenter self-checklists after each session for a total of 33% of sessions across baseline and intervention. These checklists were created for this study to monitor experimenter behavior across all conditions and families. Second, all videos were scored using the PTR-F Fidelity of Intervention Checklist to document parent implementation of each child’s BSP. The checklist included the items on each family’s BSP. Only the checklist components regarding the family’s behavior were scored because child behaviors were scored separately. An overall average fidelity intervention checklist percentage score was obtained for each observation by dividing the total number of obtained “Yes” scores by the total number of possible “Yes” scores and multiplying the answer by 100. IOA was also recorded for 33% of these checklists using video recording and the procedures described in the IOA section.
Experimental Design
A single-case withdrawal design replicated across three families was used to examine the relation between family use of a comprehensive BSP developed through the PTR-F process and child challenging behaviors (Kennedy, 2005). The single-case withdrawal research design has been established in the literature (e.g., Clarke, Dunlap, & Vaughn, 1999; Crozier & Tincani, 2005) as a feasible and efficient single-case research design for use with similar populations of children and their families. Unlike time-lagged single-case research designs (e.g., multiple baseline design) that can delay intervention for prolonged periods of time, the withdrawal design was considered superior for use in this study because an experimental effect was demonstrated through a brief removal of the intervention, causing minimal disruption for children and families.
Baseline
During baseline, families were encouraged to engage in their typical, existing routine without being given any instruction. Baseline observation sessions were videoed using a digital video camera and lasted for 10 min. Each baseline observation session occurred during the family’s target routine. All planning steps in the PTR-F process involved in developing each BSP occurred during baseline, but no intervention implementation occurred until the onset of the first intervention condition.
Intervention 1
Following the completion of Steps 1 through 3, the researcher engaged in coaching with the families during the target routine to establish fidelity of intervention.
Withdrawal
The researcher instructed the families not to implement the BSP during the withdrawal condition. As soon as a sufficient pattern of behavior (i.e., a return of child behavior to baseline rates) was noted, the second intervention condition was administered.
Intervention 2
The researcher again met during the target routine to provide coaching and feedback for families to reestablish fidelity of intervention.
Independent Variable: PTR-F BSPs
The independent variable in this study was the individualized intervention represented by the PTR-F BSPs. The process of PTR-F is outlined in Table 1 and described in detail in the PTR-F manual (Dunlap, Strain, Lee, Joseph, Vatland, & Fox, 2017). Table 1 lists the specific elements of the BSPs for each of the participating children. The following paragraphs describe the PTR-F procedures that applied to all participants and, then, the individualized strategies for each child.
Intervention Strategies on the Behavior Support Plans.
General Procedures
Following recruitment and signing informed consent, families met in individual sessions with the facilitator once per week for the three planning and assessment sessions (Steps 1–3). The first meeting was intended to define roles for the parents and facilitator, agree on short and long-term goals and target routines, develop an operational definition of the primary challenging behavior, and review the methods for parents to collect data for progress monitoring. In all cases, these evaluation data were collected in the home using a 5-point behavior rating scale as described in the PTR-F manual (Dunlap et al., 2017).
The second meeting was devoted to conducting a functional behavioral assessment (FBA). This was accomplished by having the parents complete three PTR-F Assessment Checklists pertaining to antecedent variables (Prevent), functions of and possible replacement behaviors for the challenging behavior (Teach) and consequences (Reinforce). When the checklists were completed, responses were summarized and hypotheses were developed using the PTR-F forms (available in Dunlap et al., 2017).
The assessment-based hypotheses were used in the third meeting to create individualized BSPs with intervention strategies selected from each of the PTR categories. Research-based intervention strategies for each category are identified in a menu in the PTR-F manual, and instructions for implementation of each strategy are also provided. The facilitator guided the parents to select strategies that were logically associated with the hypotheses and that seemed to have contextual fit with the family’s values and routines (Albin, Lucyshyn, Horner, & Flannery, 1996). The strategies on the BSPs were then listed on the PTR-F Fidelity on Intervention Checklist that the facilitator used in subsequent sessions for coaching and to evaluate fidelity.
Step 4 involved the initiation of intervention. The facilitator provided coaching to help the parents implement the BSP to at least an 80% level of fidelity. During coaching sessions, the facilitator reviewed the strategies on the BSP, described how to implement the strategies, asked whether the parents had questions and provided answers, gave verbal reinforcement and feedback on the parents’ implementation and, on occasion, gave brief modeling illustrations of the procedures (Fettig et al., 2015; Fox, Hemmeter, Snyder, Binder, & Clarke, 2011). Coaching continued throughout the intervention conditions of the study (Steps 4 and 5), but was withheld when the parents achieved high levels of fidelity and the child’s behavior showed clear improvement. BRS data collected by the parents was reviewed during each meeting to determine whether progress was satisfactory and whether any changes to the BRS needed to be made.
Individualized Strategies
Henry
The team comprised Henry’s parents and the facilitator. They met over a period of 15 weeks, with video data collection during the final 13 weeks. Due to Henry’s father’s travel schedule, the majority of the meetings and intervention implementation included only Henry’s mother. In Step 2 of the process, using the FBA data and the PTR-F Assessment Form, Henry’s parents and facilitator developed the following hypothesis statement: When Henry is directed to terminate a preferred activity or to transition to a nonpreferred activity, he will demonstrate refusal behavior and, in response, the transition will be delayed or terminated. Based on this understanding, the team selected intervention strategies to comprise the BSP.
A decision was made to use a visual schedule as an antecedent (Prevent) intervention because the FBA data suggested that Henry’s behavior might be improved with enhanced predictability. In addition, Henry’s mother wrote a brief story describing the steps of the routine and read the story to Henry prior to the onset of the routine. For a Teach intervention, the team opted to provide instruction to Henry on how to respond to the visual schedule. Prompts were provided for Henry to attend to the schedule and the steps were reviewed when the routine was completed. For the Reinforce intervention, it was decided to use a high rate of contingent praise, to ignore Henry’s refusals, and to provide access to a favored DVD following successful completion of the routine.
Olivia
Olivia’s team included her mother and father and the facilitator, and both parents participated in the majority of sessions, including all planning meetings, although one or the other parent had to miss an occasional coaching session due to work-related travel. Following the FBA data collection, the team developed the following hypothesis statements: When Olivia does not receive adult attention, she will demonstrate noncompliance and, in consequence, she will receive adult attention. When Olivia is directed to transition, she will demonstrate noncompliance and, as a result, the transition will be delayed or terminated.
The BSP for Olivia included three strategies from the Prevent menu of interventions. First, a visual schedule was created and was hung on the wall so that parents could direct Olivia’s attention to the pertinent steps of the routine. Second, Olivia was provided an opportunity to make a choice among clothing options that had been laid out by her parents and, third, she was given a 10-to-1 countdown during hair brushing so that Olivia could anticipate the approaching end of this nonpreferred step in the routine.
The Teach element of the BSP selected by the parents was to teach Olivia to follow the schedule as independently as possible by prompting her to attend to the schedule and verbally indicate that she knew the sequence. The Reinforce component included verbal praise, providing comments with a ratio of at least five positive statements to one correction, and providing a treat from a special snowman cookie jar if Olivia completed the full routine successfully.
Nathan
Nathan’s two parents and the facilitator were the team for Nathan’s participation. They met over a period of 20 weeks. Nathan’s father was unable to meet during the first planning meetings, but both parents were present during all of the coaching (intervention) sessions. The hypothesis statements developed following the FBA data collection were the following: When Nathan is given a nonpreferred directive or told that an activity that he wants will not occur, he engages in tantrum behavior. As a result, he gains access to what he wants. When Nathan is given a nonpreferred directive or told that an activity that he wants will not occur, he engages in tantrum behavior. As a result, he receives adult attention. And, when Nathan is given a nonpreferred directive or told that an activity that he wants will not occur, he engages in tantrum behavior. In consequence, he escapes the nonpreferred directive.
Three intervention strategies were selected for the Prevent component. First, distracting materials and events were removed during the routine. For example, efforts were made to allow Nathan to complete the routine without his brother or favored toys being present. Second, a visual schedule was created and hung of the wall. The schedule had pictorial representations of the steps that were attached with Velcro and that could be removed and placed in an envelope that was labeled “Done.” Third, Nathan’s parents conspicuously referred to the schedule after each step was completed. As with the other participants, Nathan’s team used the Teach strategy of instructing Nathan to use the visual schedule independently. The Reinforce component included the following: (a) ignoring challenging behavior and prompting engagement with the routine, (b) providing verbal praise for participation with the routine, and (c) allowing Nathan to lie on the bed with his mother (which had been identified as a reinforcer) following successful completion of the routine.
Results
IOA
For challenging behavior, the mean occurrence IOA for Henry across all study conditions was 95.2% (range = 87%–100%), and the mean nonoccurrence IOA across all study conditions was 94.7% (range = 82%–100%). The mean occurrence IOA for Olivia across all study conditions was 85.6% (range = 78%–97%), and the mean nonoccurrence IOA across all study conditions was 92.6% (range = 67%–100%). The mean occurrence IOA for Nathan across all study conditions was 98.9% (range = 94%–100%), and the mean nonoccurrence IOA across all study conditions was 100%.
Procedural Fidelity
First, across baseline and intervention conditions, procedural fidelity for all sessions was 100%. That is, the experimenter followed all planned procedures during baseline conditions and did not implement the intervention. Likewise, the experimenter implemented the intervention as planned during intervention conditions.
Second, regarding parent implementation of the BSP, each family achieved a preestablished criterion (i.e., 80% or more) level of fidelity of BSP implementation score during intervention conditions. Across families, elements of the BSPs were implemented during the baseline conditions indicating that the families were already using some of the strategies that were chosen as PTR-F strategies to include in the BSPs prior to their developing the BSPs. Some of the strategies were also used during the withdrawal condition indicating that a complete withdrawal of the BSPs did not occur during the withdrawal conditions. Table 2 shows the mean fidelity scores and ranges for each family across baseline, intervention, and withdrawal conditions.
Mean Fidelity Scores and Ranges Across Baseline, Intervention, and Withdrawal Phases.
The mean occurrence IOA for Henry’s family’s fidelity of intervention across all study conditions was 96.5% (range = 82%–100%), and the mean nonoccurrence IOA across all study conditions was 100%. Olivia’s family’s fidelity of intervention checklist mean occurrence IOA across all study conditions was 98.3% (range = 90%–100%), and the mean nonoccurrence IOA across all study conditions was 91.3% (range = 57%–100%). Nathan’s family’s fidelity of intervention checklist mean occurrence IOA across all study conditions was 100%, and the mean nonoccurrence IOA across all study conditions was 96% (range = 80%–100%).
Children’s Challenging Behaviors
Visual analysis (Kratochwill et al., 2013) was used to evaluate the effect of the families’ implementation of the BSP on children’s challenging behaviors. The analysis included within and across condition examination of level, trend, and variability and across condition evaluation of immediacy of effect, overlap, and consistency of data patterns. A functional relation was identified for all child participants between child challenging behavior and family implementation of the PTR-F generated BSP. There were three intraparticipant replications (i.e., between baseline, intervention, and withdrawal conditions) and three interparticipant replications (i.e., for Henry, Olivia, and Nathan). When parent implementation of the BSPs was withdrawn, all three participants demonstrated increases in challenging behavior that subsequently decreased with the reimplementation of their families’ BSPs. Figures 1 through 3 show the percentage of intervals with challenging behavior for each participant across sessions.
Percentage of intervals with challenging behavior for Henry across study phases.
Percentage of intervals with challenging behavior for Olivia across study phases.
Percentage of intervals with challenging behavior for Nathan across study phases.
Henry
During the initial baseline condition, Henry demonstrated a high level of challenging behavior. (range = 11%–84%). Variability in baseline data was primarily due to one occasion, on which he engaged in 11% challenging behavior, which was well below his other challenging behavior scores (52.7% lower than his next lowest challenging behavior score). This outlier was the only point of overlap with either intervention conditions. Excluding this outlier, Henry’s percentage of challenging behavior during baseline was stable and demonstrated a low, upward trend. Once intervention commenced, Henry’s challenging behavior rapidly decreased (range = 0%–33%). Intervention data showed low to medium variability with a low, downward trend. When the intervention was withdrawn, Henry’s challenging behavior scores immediately increased (range = 68%–72%). When the BSP was reimplemented again, Henry’s challenging behavior immediately decreased (range = 0%–25%). Intervention data demonstrated low to medium variability and a decreasing trend. Overall, Henry’s percentage of challenging behavior was consistent across similar conditions.
Olivia
During baseline, Olivia’s levels of challenging behavior were high (range = 50%–78%). Her baseline percentage of challenging behavior data had some variability and a moderate to high increasing trend. Once the intervention was implemented, Olivia’s challenging behavior immediately decreased to a mean of 30.1% (range = 13%–45%). Intervention data showed medium variability due to a break that occurred between the third and fourth intervention data points. The family went on a 5-day trip to visit Olivia’s grandparents, at which time the BSP was not implemented. The family began implementing the BSP upon their return home; however, a brief increase in Olivia’s percentage of challenging behavior scores occurred. With this increase in challenging behavior, Olivia’s intervention data demonstrated a medium upward trend. However, eventually, Olivia’s challenging behavior reached pretrip levels, and the final three data points in the intervention condition had a downward trend. During the withdrawal, Olivia’s challenging behavior data had an immediate increase (range = 67%–65%) to a level similar to baseline. With the reintroduction of the BSP, Olivia’s percentage of challenging behavior immediately decreased to levels similar to the first intervention condition (range = 27%–54%). On one occasion during this intervention condition, Olivia exhibited 54% challenging behavior, which was 23% higher than her next highest challenging behavior score. There were a number of changes occurring at this time for Olivia (e.g., her school schedule changed, her mother was out of town for work), but no particular explanation for this increase in challenging behavior was identified. This spike in challenging behavior was the only intervention point that overlapped with any baseline data, and Olivia’s percentage of challenging behavior during intervention was otherwise stable and demonstrated a low, upward trend through the third to last data point. Then, a decrease in challenging behavior was noted between the second and last data point indicating a rapid downward trend.
Nathan
Throughout the baseline condition, Nathan’s level of challenging behavior was high. During the final two baseline sessions, Nathan demonstrated challenging behavior for the entire duration of each videotaped observation period. His percentage of challenging behavior scores, therefore, demonstrated high rates of challenging behavior with a low, increasing trend. Once his family began implementing the BSP, Nathan’s mean percentage of challenging behavior immediately decreased by 25% between the final two baseline data points and the first intervention datum point, and then by 75% between the first and second intervention observation period. A meeting occurred between the primary researcher and Nathan’s family between the first and second videotaped intervention observation period during which the team thoroughly discussed the BSP and broke the steps of the BSP down into concrete tasks for Nathan’s father. Following this meeting, Nathan’s challenging behavior decreased to 0% and his percentage of challenging behavior remained low and stable throughout the remaining duration of the intervention (range = 0%–10%). When the intervention was withdrawn, Nathan’s percentage of challenging behavior rapidly increased to a level similar to baseline (mean of 85% and range of 75%–95%). Once the BSP was reimplemented, Nathan’s percentage of challenging behavior immediately decreased to 0%, which was similar in level to the first intervention condition, and it remained stable at 0% across all four videotaped intervention observations.
Family Goals
All families noted that the absence of challenging behavior was their goal for the routine. Therefore, any interval that did not contain challenging behavior contained desirable behavior. All families indicated that their children learned desirable behaviors through their implementation of the BSPs.
Family Perception of the PTR-F Process and BSP (Social Validity)
All three families favorably rated both the PTR-F process and the BSPs that were developed and implemented. On questionnaire items with a rating of 5 indicating the most favorable score, average social validity ratings for Henry and his family, Olivia and her family, and Nathan and his family were 4.83, 4.83, and 5, respectively. On questionnaire items with a rating of 1 representing the most favorable score and a rating of 5 the least favorable score, average social validity ratings for the families included a score of 2 for Henry and his family, a score of 2.25 for Olivia and her family, and a score of 1 for Nathan and his family.
The most highly rated questionnaire items indicated that families found the BSPs acceptable, the BSPs fit well into their existing routines, the children learned desirable behaviors through the families’ implementation of the BSPs, and the BSPs were congruent with each family’s goals for their child. Items related to the family’s willingness to carry out the BSPs and the amount that families liked the BSPs were also highly rated. Families rated items regarding BSP disadvantages, side effects, and child discomfort more variably. Henry’s family reported that they rated the questionnaire item addressing undesirable side effects a 3 out of 5 because their children now watched a DVD in the car (Henry’s reinforcing activity for completing his routine), which prevented their ability to listen to music while driving. Olivia’s family noted that they rated the questionnaire item addressing child discomfort that might occur through their BSP implementation a 3 out of 5 because the “first time [implementing the BSP] was rough.”
Discussion
The purpose of this study was to experimentally analyze the PTR-F process with three families who have young children with challenging behavior. All three families reached high-fidelity implementation of the BSPs developed using the PTR-F process. Also, functional relations were identified between parent implementation of the BSP strategies and child behavior. Importantly, families perceived the PTR-F process and the BSPs as socially valid. Families reported increased satisfaction with their target routines, increased self-confidence regarding their ability to implement the BSPs, and positive perceptions of the PTR-F process and the BSPs.
This study supports existing studies that have demonstrated that families are able to implement function-based BSPs with their young children with challenging behavior in ways that produce desired outcomes (Fettig & Barton, 2014). The results of this study are consistent with an ever-increasing literature base (e.g., Buschbacher et al., 2004; Clarke et al., 1999; Fettig & Barton, 2014; Fettig et al., 2015; Koegel, Steibel, & Koegel, 1998; Lucyshyn et al., 1997) that provides support for the use of family-centered PBS to effectively decrease child challenging behaviors and improve the quality of life of young children and their families in home settings.
The findings of this study also add support for the efficacy of the PTR model when used with families of young children with challenging behavior in home settings. Although some studies have used modifications of previous versions of PTR with families (Bailey & Cho Blair, 2015; Sears, Kwang-Sun Cho, Iovannone, & Crosland, 2013), we believe that the current investigation is the first peer-reviewed study to have evaluated the PTR-F model specifically. Thus, the results from this study extend current research by providing evidence of the effectiveness of the PTR-F model when used with families of children with challenging behavior. Additional replications are warranted to corroborate our findings.
In regard to the families’ fidelity of intervention scores across conditions, as previously noted, each family used some of the strategies that were included in the BSPs during baseline and withdrawal conditions. The inclusion of these strategies in the BSP was important for the family-centered nature of the PTR-F process because it ensured that the plan was built on the strengths of the family and on practices that were already in place in the home setting. Furthermore, all families indicated that including specific strategies in the BSP served as a reminder for them to use the strategies during the routine. For example, while Henry’s mother was using verbal praise during baseline, her use of this strategy increased during the intervention condition, and she used it more regularly across varying steps of the routine. It could be concluded that even without large changes in fidelity of intervention scores, behavior change was observed across families. It is more likely, though, that the comprehensive nature of the BSPs led to the inclusion of some strategies that were not necessarily critical for child behavior change to occur. No experimental manipulation occurred to determine which strategies from the BSPs led to changes in child behavior. However, overall, the strategies that were used across families during baseline and withdrawal conditions were related to the families’ use of verbal praise and positive attention during the routines and to the families’ following of the individual routine steps in a specific order. None of the families used visual schedules at baseline or during the withdrawal condition, and access to functional, individualized, tangible reinforcers (e.g., Henry’s DVD, Olivia’s prize jar, Nathan’s mother’s lying with him in bed when the routine was finished) did not occur during the baseline or withdrawal conditions.
The social validity findings of this study substantiate findings from previous research that suggest that families favorably rate family-centered PBS (Binnendyk & Lucyshyn, 2009; Buschbacher et al., 2004; Fettig et al., 2015). Overall, positive ratings were noted for the PTR-F process and for the BSPs that were developed by each team.
Limitations
There are several limitations to this research study. First, the generalizability of study findings is limited given the small sample size and limited participant diversity. Second, this single-case withdrawal design implemented only two withdrawal data points per family. Although clear and immediate-level changes were observed for each child’s percentage of challenging behavior between intervention and withdrawal conditions, results should be interpreted with caution due to the limited number of data points that were used to establish the functional relation between the withdrawal and reimplementation of the BSP. Third, although it might be extremely difficult to have a blind coder score the videos for these particular families given the number of changes to the routines that were made from baseline to intervention, the primary video observation scorer was not blind to study conditions. Therefore, the video observation scorer may have had some knowledge of the condition changes that occurred. It is unknown whether or not this affected scoring. However, IOA between observers was sufficient, which adds support for scoring validity. Along these lines, partial-interval recording, which lacks sensitivity when compared with other measures such as event recording, was used to estimate count of child behaviors. The use of this procedure may have resulted in an underestimation of behaviors (Cooper et al., 2007). In addition, no generalization, maintenance, or follow-up data were collected after the primary researcher’s final close-out meeting with the families. Therefore, no information is available regarding the families’ use of BSP strategies during other routines or the families’ continued implementation of the BSP once their regular visits with the primary researcher stopped. One might wonder, also, about the clinical significance of the effects of PTR-F on challenging behavior, at least for Olivia, whose challenging behaviors were never reduced to zero levels during intervention. Although the effects were rated as very favorable on the social validation scales, it might have been better if the intervention condition was extended or the BSP refined so that zero levels had been documented for all participants.
Recommendations for Future Research
Given the findings and also the limitations of this study, a number of recommendations can be made for future research studies. It is important that the efficacy of the PTR-F model continue to be studied with young children and families. Future studies should, for example, focus on the implementation of the model across children with a variety of disabilities and special needs, across families from varying culturally and linguistically diverse backgrounds, and with families of different socioeconomic statuses.
While all families noted relatively rapid changes in their children’s behaviors, future research should consider the average number of sessions that are required for families to report changes in child behaviors in practice-based settings in which home visitors and families are using the PTR-F model. It is likely that the PTR-F process might be even more efficient in these practice-based settings without the research study requirements that were dictated by the single-case withdrawal design that was employed in this study. In addition, it is recommended that future studies consider the generalization of the families’ learned skills to other routines and settings and also the changes in child challenging and desirable behaviors that might occur in routines that are not directly targeted through the PTR-F process. Finally, BSP implementation maintenance and follow-up data should also be collected in future studies.
Implications for Practice
The results of this study indicate that families can successfully participate in the PTR-F process and that their participation and implementation of BSPs can lead to desired changes in child behavior and increased family satisfaction and confidence implementing BSPs. All three families successfully participated in all steps (including PTR-F Assessment) of the PTR-F process implemented in this study. The results also indicate that family-centered PBS, and specifically the PTR-F model, may be an efficient and effective way to reduce challenging behavior in the home settings of families of young children with challenging behavior.
A primary implication for practice emerging from this study was the need for flexibility and availability from the researcher to support the families who participated in the most efficient and effective way possible. For example, the researcher was present for each family during the time that the actual routine occurred. This meant that the researcher arrived at the families’ homes during very early (e.g., 6:30 a.m.) and somewhat late (e.g., 8:30 p.m.) hours. Meeting times tended to fluctuate throughout the research process. Furthermore, families were not always available on the same day every week, so the primary researcher met with families when they were available as opposed to meeting with each family on the same day and at the same time each week. In addition, there were multiple occasions during which the researcher was en route to a family’s home when the family had to cancel the session at the last minute. The researcher’s ability to reschedule with families during the same week that sessions were canceled contributed greatly to the efficiency of the process.
Also contributing to the efficiency and effectiveness of the process was the researcher’s ability to work on the development and organization of materials (e.g., visual schedules, detailed written instructions) that the families would use while they were implementing the BSPs during indirect (i.e., nonsession) hours. The primary researcher’s schedule flexibility and use of indirect time reveal another implication for practice. That is, it is important to consider the number of families who home-visiting professionals work with at one time as it relates to the quality of the service that families receive. The efficiency of child behavior change might be affected by a home visitor’s inability to use indirect time, to reschedule during the same week, or to provide support and coaching for families during typically occurring routines.
Conclusion
The PTR-F process resulted in families’ high-fidelity implementation of the BSPs, and a functional relation existed between each family’s implementation of their child’s BSP and changes in his or her behaviors. Furthermore, this study has shown that family-centered PBS, and specifically PTR-F, can improve familial routine satisfaction with target, difficult routines and increase familial confidence with implementing BSPs. In addition, the findings of this study indicate that families favorably rate the PTR-F process and the comprehensive BSPs that are developed through their participation in the PTR-F process.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.