Uncovering the theoretical underpinnings of an interprofessional preschool intervention program: Tumbling Together

Tumbling Together

Tumbling Together was established in 2003 and was the brainchild of two clinicians (an occupational therapist and a speech-language pathologist). It uses recreational gymnastics as a therapeutic medium for the provision of play-based interprofessional rehabilitation. The program takes place in a gymnastics facility and comprises ten 45-min weekly sessions. It uses a small-group model for 3- to 5-year-olds based on play, movement, and social interaction to deliver occupational therapy and speech therapy. Its objectives include developing early group participation skills, providing sensorimotor experiences, working on language goals, and educating parents (Loiselle & Chamberlain, 2010). Children who participate in the program form a heterogeneous group, including children with an assortment of complex communication needs, developmental delays, physical challenges, and self-regulation challenges. They participate in activities such as circle times, gross motor circuits, and group activities in which appropriate language is integrated. Throughout the program, the Goal Attainment Scale (King, McDougall, Palisano, Gritzan, & Tucker, 2000) is used to carefully track progress in various areas (e.g., consistently attending to activities, initiating peer interactions). Parents are consulted during a parent session at the beginning of the program, throughout the program, and at the end of the program, when they are asked to complete a written feedback form.

Literature Review

Some important characteristics drawn from the program description are interprofessional collaboration, situational learning in a gymnastics facility, and use of a group model. A review of the literature by the first author looked at studies examining characteristics mentioned in the Tumbling Together program description for paediatric populations for a more complete definition.

The concept of interprofessional collaboration has been the subject of many studies. Interprofessional collaboration involves professionals from various disciplines, working together in an environment of trust and respect to address the complex needs of clients (D’Amour, Ferrada-Videla, San Martin Rodriguez, & Beaulieu, 2005). Symon (2001) indicates that there have been increased collaborative efforts in designing interventions for children with special needs. Collaboration between professions is a recommended practice (Rapport, McWilliam, & Smith, 2004), and interprofessional teams have proven to be an effective approach for paediatric populations (Bell, Corfield, Davies, & Richardson, 2010; Golos, Sarid, Weill, & Weintraub, 2011).

Another recommended practice for paediatric rehabilitation is to ensure that interventions include situational learning that incorporates the concept of naturalness (Owens, 2010). Children must learn skills in an environment in which the skills are required, thus making their use natural. Play is natural for children and allows them to develop important skills (Burriss & Tsao, 2002; Kasari, Gulsrud, Freeman, Paparella, & Hellemann, 2012). Use of groups in paediatric populations is also supported by literature. Groups provide opportunities for children to monitor their performance, learn from their peers, socialize with others, and increase their social participation and cooperation (Camden, Tétreault, & Swaine, 2012; Dunford, 2011; Quigg, 2003).

Research by Pennington and Thomson (2007) showed that qualitative methods are helpful in identifying key concepts of intervention programs and in studying implementation issues prior to evaluating their general effectiveness. To support an in-depth investigation into the theoretical model guiding the Tumbling Together approach, a transition from descriptive to analytical was required to identify all important concepts and their organization without being limited to the program description alone. Concepts are considered the building blocks of a theory and include central ideas and their characteristics (Strauss & Corbin, 1998). The aims of this study were therefore to take an in-depth look at the program to answer the following research questions: (a) What are the major concepts of the program? and (b) What is the underlying theoretical model that illustrates the relationships between concepts?

Study Design

A qualitative design, grounded theorizing using situational analysis, was chosen as it is well suited for analyzing situations that have not yet been widely researched. Grounded theory involves creating theory from data that have been systematically collected and analyzed (Strauss & Corbin, 1998). Situational analysis considers both human and nonhuman factors while exploring the relationships between them to gain a better understanding of the social world (Clarke & Friese, 2007). The social world as considered in the context of this study involved the individuals implementing and participating in the Tumbling Together program.

Participants and Recruitment

Ethics approval was obtained by the Health Sciences and Science Research Ethics Board at the University of Ottawa. Participants were recruited by way of advertisement posters displayed in the office building of rehabilitation professionals and at the gymnastics club where the program takes place.

Four sampling methods were used in this study: convenience sampling, purposeful sampling, theoretical sampling, and theoretical interviews (see Morse, 2007). Archival historical materials, documents, and participants were selected in sampling as allowed by grounded theorizing using situational analysis. Convenience sampling consisted of using the program manual and transcription of the program DVD as these provided a good overview, helped establish the scope of the study, and were accessible to the authors. Purposeful sampling was used to recruit program participants to gather data exploring their perspectives as they related to the initial findings. Theoretical sampling involved anonymous parent feedback forms and Goal Attainment Scale forms that were provided to the first author to aid in theory development and supplementation. These forms were administered outside the context of this study; therefore, no exchange between researcher and participants was possible when compared to the traditional process of selecting additional participants to interview. However, the forms were selected to help supplement information about linkages between emerging concepts as they provided information from another perspective, thus contributing to theory development. Theoretical interviews included the recall of two participants to “expand on and verify the emerging theoretical model” (Morse, 2007, p. 235).

Five key informants with involvement in the Tumbling Together program were recruited: one occupational therapist, one speech-language pathologist, one gymnastics coach, and two university students in rehabilitation science programs. All participants were retained until the end of the study. The inclusion criteria for the participants were recent and direct involvement in the implementation of the program.

All participants were women. The occupational therapist and speech-language pathologist each had 35 years of experience, were the co-creators of the program, and had been involved in 50 Tumbling Together series. The gymnastics coach had 27 years experience and had been involved with 39 Tumbling Together series. The rehabilitation science students were in their final years of their programs. Having three professionals participate provided a good sampling of staff involved when compared to the total number of staff who have been involved with running the program since its inception. Participants were aware of research questions prior to interviews. The first author is related to one of the co-creators of the Tumbling Together program and was compelled to research it subsequent to witnessing anecdotal success stories. Bias was managed by having the third author review coding and question conclusions.

Data Collection and Analysis

The grounded theory approach used was consistent with the methodology described by Strauss and Corbin (1998). The iterative process of data collection and analysis was followed, and one complete cycle of levels of coding was accomplished.

Data were collected through document review of program materials, transcript analysis, and interviews. All interviews were semi-structured and flexible to ensure appropriate potential concepts could be explored as dictated by analysis of data. Interviews were held in a quiet room and were audio recorded. Interviews were on average 30 min. The first author (female graduate student) conducted all interviews and made field notes. Audio recordings of the interviews were transcribed prior to coding.

The coding paradigm described by Strauss and Corbin (1998) was used for data analysis. Data analysis involved three types of coding (open, axial, and selective) as well as memos, situational analysis, mini-frameworks, and theoretical model drafts. The resulting theoretical model was then validated during validation interviews with 2 participants. Figure 1 illustrates the cyclical process of data collection enabled by sampling methods and types of analysis used.

OPEN IN VIEWER

Figure 1.

Visual representation of data collection and analysis methods.

Memos were created and continued to be generated throughout the study to assist with theory development. Memos helped ensure the emerging theory had conceptual density and were divided into three types of notes: code notes, theoretical notes, and operational notes (Strauss & Corbin, 1998). Code notes yielded a total of 430 significant units of meaning and were classified under 51 codes during open coding.

Constant comparison between concepts and memos helped to ensure the analysis was valid. Axial coding defined codes and identified patterns. Codes were then organized into concepts according to their relationships. Selective coding, which integrated and refined concepts, was done in conjunction with situational maps—the first type of map involved in situational analysis as described by Clarke and Friese (2007) and allows for the inclusion of all major concepts instead of restricting it to a central concept. Table 1 outlines the contributory codes for each concept identified and their definitions. The 14 main concepts were divided into three categories to facilitate explanation. Concepts, related directly to the types of skills targeted in the program and their generalization, were grouped into one category. Concepts surrounding the circumstances of the setting of the program were grouped into a second category related to context. Finally, concepts used as intervention strategies were grouped into a third category. These categories should not be viewed as part of the theoretical underpinnings of the program as certain concepts fit into more than one category as described in the Findings section.

OPEN IN VIEWER

Table 1

Axial and Selective Coding Concepts and the Contributing Codes From Data Analysis

Concept	Definition	Contributory code
Related to skills
Optimal state (being calm, alert, and focused)	A state in which individuals are ready for learning because they have appropriate levels of arousal, attend to the task at hand, and achieve adequate self-regulation	Calm, alert, focus
Integrated skill development	Simultaneously targeting and developing functional skills from a variety of disciplines	Skill development, integrated skills, therapeutic/learning foci of activities, parental input, sensory motor/gross motor development, communication and language skills, self-regulation, co-regulation, socioemotional skills, confidence
Generalization	The ability to transfer skills to other contexts and settings	Generalization
Related to context
Community-based program	A program that involves community partners, includes professional training informed by the community’s needs, and improves community competency	Partnerships, professional training, increasing community capacity
Interprofessional collaboration	Professionals from various disciplines working together in an environment of trust and respect to address the complex needs of clients (D’Amour Ferrada-Videla, San Martin Rodriguez, & Beaulieu, 2005)	Staff training, highly trained staff, interprofessional learning, student education, interprofessional teamwork, SLP role, OT role, coach role, mental health role, positive focus
Family-centred approach	An approach that engages the family as a member of the interprofessional team and includes family education, input, support, feedback, and advocacy.	Parent support, parent input, parent education, strategies for dealing with parental frustration, parent feedback, advocacy, motivation, parental outcomes
Related to intervention strategies
Movement	Activities in which the body is physically active in order to change positions or locations	Movement
Situational learning	Learning that occurs in a natural context in which the learning can be directly applied	Learning in context, learning through experience, gains experienced by the child, measuring changes
Individualization	Tailoring aspects of a program to each individual, which includes components of both assessment and intervention	Therapist program, high adult-to-child ratio, expecting responses, parental input, individualized goals, assessing children
Peer interactions	Exchanges between peers in which they are engaged with each other	Peer interactions
Expert/novice pairing	A combination of children who have complementary strengths and weaknesses in various skills, which enable them to both teach others at times and learn from them at other times	Expert/novice pairing
Explicit teaching	Teaching that involves breaking down skills into smaller steps and providing specific instruction on how to accomplish each step to achieve the desired result; involves conscious learning guided by modeling, explanation, and practice	Explicit teaching
Scaffolding	Graduated assistance provided to novice learners to help them achieve higher levels of competence (Kaderavek, 2011)	Scaffolding
Additional supports and strategies	All additional supports and strategies that facilitate the development of integrated skills and the ability to achieve an optimal state, such as use of a structured natural environment, routines, sensory processing strategies, and visuals	Supportive environments, non-human factors, routine

Social worlds/arenas maps and positional maps as described by Clarke and Friese (2007) were created after document review and analysis of feedback forms and Goal Attainment Scale forms. These included social and organizational aspects and concepts viewed from differing perspectives.

Drafts of a theoretical model were created based on the above-mentioned concepts, memos, and maps. A table of code frequencies was made and questions were created to check the relative importance of concepts. An interview guide with questions stemming from these analyses was used for the final semi-structured interviews in the data collection phase.

Validation interviews were then conducted with each of the two therapists who co-created Tumbling Together at the end of the study to ensure the analysis concurred with the reality of the program. A printout of the final theoretical model draft was presented. No major inconsistencies were found. Two small graphic changes were made following these interviews, but both therapists agreed that the proposed theoretical model was a good representation of the program.

Concepts Related to Skills

Concepts Related to Context

Concepts Related to Intervention Strategies

Negative Case

One comment questioned the generalization of skills with this program. This participant had not yet seen the program to completion. Participants who had seen the full program commented that the skills transfer to other contexts and provided specific examples. Studying whether generalization always occurs and how soon it occurs during the program would require further investigation.

Limitations

Due to time constraints, participants were selected for purposeful sampling prior to initial analysis. Thoughtful consideration was given to recruitment strategies and appropriate numbers of participants to maintain as much rigour as possible. However, this process deviates from requirements of grounded theory methodology. Grounded theory calls for continuation until a point of saturation has been reached. Saturation is defined as the point where new information no longer contributes to theory development (Strauss & Corbin, 1998). The process used appeared to facilitate saturation of main concepts within the sample of participants as no new information for theory development emerged in the later interviews as evidenced through coding. Although more in-depth analysis of secondary concepts may have been possible by interviewing parents of participants as a source of triangulation, the use of historical written parent feedback forms allowed for an examination of the thoughts of a greater number of parents. Twenty written parent feedback forms were available for analysis, which surpassed the limited number of parents involved in the program at the time of the study. While only one researcher coded the data, a second researcher challenged the data, reviewed coding, and suggested changes as appropriate to improve reliability.

Study Implications and Future Research

This study highlights important theoretical concepts of an intervention program that has had anecdotal success. Other program developers may be inspired to follow this model without necessarily reduplicating the Tumbling Together components exactly to build other intervention programs for preschoolers. Tumbling Together’s co-creators state they believe it is not the exact execution of the program that makes it successful but the combination of its concepts.

The theoretical model is also the first step toward meeting the criteria for evidence-based practice. It facilitates future testing of the program’s efficacy and generalizability. One or multiple concepts of the program could be studied at a time. Long-term effects and changes to participant diagnoses could also be studied. The theoretical model might also be used to explain the program more thoroughly to stakeholders to obtain grants that would ensure the continuation of the program and expand the program to other sites. Continuation and expansion would benefit a population in need.

As the methods used in this study were able to successfully address the research questions, they can provide a framework for studying other programs that combine various concepts. Grounded theory using situational analysis can be considered valuable in rehabilitation science program evaluations. A combination of qualitative and quantitative methods could be considered for future research (e.g., Glogowska, Campbell, Peters, Roulstone, & Enderby, 2002).