Abstract
Substantial expert agreement was established for several occupational science concepts viewed as essential, providing a basis for future studies to refine the concepts for occupational therapy education and practice.
Since its formal introduction as an academic discipline (Yerxa et al., 1990), occupational science (OS), defined as “an emerging basic science which supports the practice of occupational therapy” (p. 1), has evolved in myriad ways. Yet, despite the evolution of an impressive infrastructure anchored by graduate programs, scientific societies, and publications, approaches to studying, teaching, and applying OS knowledge to occupational therapy have evolved unsystematically. The consequence is a lack of conceptual clarity that manifests in ambiguous terminology to describe interventions (Fisher, 2013) and disconnected research that impedes orderly theory development. As OS generates new knowledge about occupation, its most essential concepts should be identified and integrated into occupational therapy education and practice.
An ongoing study of educational process and content related to the concept of occupation among American occupational therapy education programs has been reported by Hooper and others (e.g., Hooper et al., 2018; Price et al., 2017). Through extensive interviews and artifact analysis, these studies have found that educators teach various OS concepts, such as occupational justice, more frequently than they explain the science itself (Hooper et al., 2016). Although evidence from the studies cited suggests that OS is influencing content in entry-level occupational therapy programs, systematic efforts to identify and organize concepts seen as essential for occupation-centered practice are missing. As a result, wide variation exists in the content, methods, and emphasis placed on occupation and the knowledge practitioners acquire.
Academic accreditation standards worldwide (e.g., Accreditation Council for Occupational Therapy Education, 2018; Canadian Association of Occupational Therapists, 2017; Occupational Therapy Council of Australia, 2018; World Federation of Occupational Therapists, 2016) describe requirements for occupation-related content in different and imprecise ways, contributing to variations in how graduates understand and are prepared to apply occupation-related concepts in practice. One approach to advancing occupation-based curricula and practice is to seek expert consensus on key concepts in OS that are viewed as essential for occupational therapists to know. This study applied that approach to a primary research question: “What OS concepts are essential to the occupation-based practice of occupational therapy?”
Method
Design
We used a complex convergent mixed-methods Delphi design (Figure 1). Complex mixed methods (Creswell & Plano Clark, 2017) combine qualitative and quantitative approaches within an established method, in this case the Delphi method. The Delphi process is well established as a consensus-gaining approach that relies on an expert panel to reach agreement on positions or opinions. It involves systematically sampling the panel over several rounds of inquiry until a predetermined level of consensus is reached. Originated by the Rand Corporation in the 1950s (Dalkey & Helmer, 1963), it has been widely used in various forms in many disciplines, including occupational therapy (e.g., Che Daud et al., 2015). The 14-mo Delphi process used in this study was unique in that it used two equivalent groups to generate comparative data and employed qualitative methods in the first round to generate potential consensus concepts. Surveys were distributed via Survey Monkey, and participants were given 4 to 6 wk to respond to each round with reminders sent 2 wk before closing dates. The a priori threshold for agreement was set at ≥70% .
Overview of study design depicting data collection and two analysis sequences.
We conducted four iterative rounds of data collection with two distinct analytical sequences—parallel-groups and full-group—to enable within-group and across-groups analyses. As illustrated in Figure 1, an initial qualitative round generated a set of concepts from two randomly assigned parallel groups of experts. In two subsequent rounds, we used numerical ratings consistent with Delphi methods to ascertain within-group agreement on concepts viewed as most essential. For the three rounds in the parallel-groups sequence, each group proceeded independently using identical methods. After analysis of the parallel-group results in Round 3, a fourth round was conducted with the combined full group of experts. This combined expert group was asked to select essential concepts after considering newly merged definitions derived from both groups’ third round. Consensus findings from this full-group round and results from each parallel group in Round 3 provided an enriched data set with three lists of essential concepts for comparative analysis and concept validation (Table 1).
Comparison of Parallel-Groups and Full-Group Consensus Concepts
Note. Boldface indicates concepts that achieved ≥70% agreement. Concept definitions for the parallel-groups analytical sequence can be found in Supplemental Appendix B; those for the full-group analytical sequence can be found in Table 2.
Parallel-groups results are ordered and grouped to show similarity of findings across analytical sequences.
All 52 experts from earlier rounds were invited to participate in Round 4.
Expert Recruitment and Selection
An international list of OS experts was created through an iterative, team-based process. Initial nominations by research team members were collaboratively ranked (as low, medium, or high) on the basis of publications, graduate degrees, and faculty roles in degree programs in OS. Experts ranked medium or high were invited to nominate other experts. These additional nominees were collaboratively ranked by the team using the same process, adding those ranked medium and high to the list of experts. Ultimately, 62 experts from 22 countries were invited to participate and randomized to one of two groups (A or B).
Study procedures were approved by the institutional review board at Colorado State University, and participants provided informed consent before starting Round 1.
Round 1: Identifying Concepts and Definitions
Data Collection
A design scenario for developing a hypothetical course titled “Occupational Science for Occupational Therapy” was used to prompt experts to identify and justify up to 10 concepts essential for practice to include in the course (Supplemental Appendix A, available online with this article at https://research.aota.org/ajot, contains the survey prompts and instructions to participants by round).
Analysis
Each group’s responses were analyzed by designated investigator dyads (Group A, Hooper and Pierce; Group B, Backman and Price) blinded to data from the other group. Only the study organizer–coordinator (Christiansen) had access to both data sets, which were maintained on separate secured institutional servers. Content analysis (Creswell & Plano Clark, 2017) was used to identify concepts and definitions. Each dyad read and reread proposed concepts, grouped similar concepts into clusters, and agreed on distinct concepts to represent the data as a whole. Concepts required identification by more than one panel expert to be included.
Results
Fifty-two (84%) of those invited consented and completed Round 1 (Group A, n = 24; Group B, n = 28). Proposed concepts ranged from 2 to 75 words in length. Group A provided 206 concept descriptions with rationales, totaling 26 single-spaced pages of raw text data. Group B’s 208 responses totaled 25 pages of raw text data. Content analyses reduced the raw data to 26 concepts in Group A and 25 concepts in Group B (Supplemental Appendix B).
Round 2: Scoring Concepts for Essentiality
Data Collection
Experts in each group were provided with concept labels and descriptors derived from their Round 1 data. Continuing with the course development scenario, experts were asked to rate each concept as absolutely essential, somewhat essential, less essential, or not essential.
Analysis
Each group’s responses were again reviewed by its assigned investigator dyad. An essentiality score (e-score) was calculated by summing the percentage of experts rating each concept as absolutely or somewhat essential. The resulting e-scores allowed equivalent and comparable ordinal distributions of concepts. Concepts achieving the a priori cutoff point (e-score ≥70) advanced to the next round for each group.
Results
Twenty (83%) experts from Group A and 25 (89%) experts from Group B responded to Round 2. After applying the cutoff, 18 concepts from each group were retained (Supplemental Appendix C, available online with this article).
Round 3: Prioritizing Concepts
Data Collection
In Round 3, participants were reminded of the course-building scenario and invited to consider their group’s top concepts from Round 2. Experts were given a binary choice to retain or eliminate each concept while considering the time constraints of a 16-wk course. No limits were placed on the number of concepts that could be included.
Analysis
The within-group percent agreement was calculated for each concept, indicating the proportion of experts who endorsed retaining the concept.
Results
Seventeen (71%) experts responded in Group A and 23 (82%) responded in Group B. For Group A, 10 of 18 concepts achieved ≥70% consensus; for Group B, 7 of 18 concepts achieved ≥70% consensus. Although the concept labels and descriptions varied, results from the two independent groups contained nine matching or overlapping concepts (see Table 1).
One group used broader and more inclusively defined concepts, and the other group expanded broad concepts into related subconcepts. Consequently, some subconcepts fell below the consensus threshold. For example, 88% of Group A experts retained “social and cultural context of occupation” for the hypothetical course, and 74% in Group B retained “culture” as an essential concept. However, in Group B, four additional concepts related to culture did not achieve the consensus threshold.
Round 4: Examining Consensus After Combining Groups
An extensive comparative review of data and results from the first three rounds indicated the potential usefulness of a fourth round that combined experts into one group. The purpose of this full-group round was to ask the combined experts to reconsider concepts from Round 1 using merged definitions.
To prepare the merged set of concepts and definitions from the parallel-groups data, the two dyads that had previously worked independently with their assigned group’s data now engaged in a systematic and collaborative process with the combined data set. Round 1 definitions included nine consensus concepts with a direct (one-to-one) match between groups. Definitions for these nine concepts were merged in a careful, qualitative iterative content analysis to capture all meanings contained in Round 1 raw data in order to maintain the authenticity of the original data and strengthen the trustworthiness of final results.
The dyads then used the same iterative content analysis to accomplish the more challenging goal of combining and producing definitions for concepts that had conceptual similarity but were not a one-to-one match. For example, analysis of expert input from Group A produced a broad definition of occupational justice, whereas Group B analysts distinguished occupational justice (and injustice), occupational deprivation/alienation, and political nature of occupation as distinct concepts. Yet, these overlapped with the broader definition of occupational justice generated in Group A. These concepts were merged into a single concept of occupational justice with a definition inclusive of all meanings offered in Round 1 data. This careful conceptual matching process generated six concepts, bringing the total set of merged concepts to be presented to experts in Round 4 to 15 (Table 2).
Names and Descriptors of Full-Group Final Round Concepts
Note. Consensus (≥70%) concepts are shown in boldface.
Data Collection
The experts were informed of the parallel-groups design used in earlier rounds, presented with the 15 merged concepts, and reminded of the course-design scenario. They were asked to make a binary choice to retain or exclude concepts for the hypothetical course after carefully considering the reconstructed definitions.
Analysis
The percentage of agreement that a concept was essential was calculated for each of the 15 concepts. Consensus results from the full group in Round 4 were then compared with the results from Groups A and B in Round 3 to identify similarities and discern conceptual nuances between parallel-groups and full-group results.
Results
Forty-two experts (81%) responded in Round 4. One concept achieved 100% consensus, and 10 others met or exceeded the 70% agreement threshold (see Table 2).
Discussion
In this study, we sought to identify OS concepts essential to the occupation-based practice of occupational therapy. “Occupation and wellness, health, and well-being” achieved 100% agreement as essential to a course on OS for occupational therapists, and it was the top ranked concept by both parallel groups in Round 3. Five additional concepts reached consensus in the full group and in the parallel groups: “social, cultural, and institutional context of occupations”; “occupation as core to occupational science and occupational therapy”; “occupational justice”; “occupational meaning”; and “occupation and identity.”
One concept, “spatial context of occupation,” achieved consensus in the full group and in one parallel group. Two concepts that achieved consensus in the full group achieved consensus in one of the parallel groups but achieved only ≥60% agreement in the other parallel group: “experience of doing occupation” and “habits, routines, and patterns.” Other concepts that reached consensus in the full group but not in the parallel groups included “temporal nature of occupation” and “occupational balance/imbalance.”
A complete understanding of these results also requires consideration of concepts that just missed the 70% criterion. “Occupational choice and agency,” “transactionalism in occupational science,” and “occupational change” were all above 60% consensus but below the criterion threshold. Replication studies may show different priorities with different samples or use an alternative scenario for eliciting the concepts and definitions from experts.
Although many of the concepts identified by the experts have appeared in the occupational therapy and OS literature (with varying definitions), additional research is necessary to determine whether the concepts identified in the current study are stable, mature, and important to the relationship between OS and occupational therapy.
Strengths and Limitations
Trustworthiness and credibility were addressed in several ways. The research team consisted of five experts in OS who collaboratively analyzed data. The team undertook extensive peer discussion throughout the study, from identifying and recruiting experts to decision making in each round. The team returned to raw data at key points to check and confirm the process and findings. Response rates were high, given the complexity, duration, and number of Delphi rounds. Two independently managed parallel groups helped validate consensus findings. Concepts were randomly presented in each round to control for order effects. A subsequent full-group analytical sequence checked for any shaping of the data by expert group composition and size or influence of the analyst dyads’ perspectives. Triangulation (convergence) of data demonstrated consistency of findings.
An untested, modified Delphi process was used. For some participants, English was an additional language, and variations in fluency between groups were assumed to be controlled through random assignment but may have affected concept definitions and rankings. The volume of data across two separate groups added complexity and time to analysis and planning of subsequent rounds. This lengthened the study’s duration and may have contributed to attrition as the rounds progressed. As previously noted, the participants were affiliated with academic settings, and their current involvement in practice was not determined. The participating experts represented 22 different countries, yet a majority were from Western nations, which may have biased the naming, framing, and defining of OS concepts reported here.
Implications for Occupational Therapy Practice and Education
These results are a beginning step toward identifying essential concepts in OS. Although these preliminary results may provide useful guidance for research and education, further work is necessary to determine whether the complex concepts identified here contain more discrete concepts that should be defined and studied individually. Because the inclusion criteria for experts required publications or advanced degrees in OS and faculty roles in degree programs, the findings may represent an academic rather than a practitioner perspective. Future research engaging practitioners as expert panelists would determine whether similar or different concepts are identified as essential to occupation-based practice.
The consensus concepts identified here could be used by practitioners to examine connections between the concepts and evaluations and interventions frame practice from an occupational justice perspective, considering opportunities and systemic barriers to clients’ occupational engagement, perhaps requiring a stronger emphasis on the practitioner’s role as advocate; and design continuing education offerings that examine how OS concepts can be used to inform occupational therapy practice.
The results may help organize and develop a more consistent approach toward teaching OS in occupational therapy, potentially enhancing curricula. Accordingly, educators may incorporate these concepts into courses, assignments, and learning activities; use the broad definitions from this study when teaching the concepts; justify the place of OS concepts in education and practice; and engage students in applying these essential concepts to professional reasoning and practice skills.
Conclusion
How, when, where, and why people spend time meaningfully in ways that influence their health are fundamental questions for OS and occupational therapy. International experts identified and agreed on 11 essential concepts for an introductory course on OS for occupational therapy. The study exemplifies the relationship initially envisioned between OS and occupational therapy, and the results offer a first step toward a more consistent approach in using OS knowledge for occupational therapy education and practice.
Supplemental Materials
Supplementary material for Occupational Science Concepts Essential to Occupation-Based Practice: Development of Expert Consensus
Supplementary material, sj-docx-1-aot-10.5014_ajot.2021.049090.docx for Occupational Science Concepts Essential to Occupation-Based Practice: Development of Expert Consensus by Catherine L. Backman, Charles H. Christiansen, Barbara R. Hooper, Doris Pierce and M. Pollie Price in The American Journal of Occupational Therapy
Supplementary material for Occupational Science Concepts Essential to Occupation-Based Practice: Development of Expert Consensus
Supplementary material, sj-docx-2-aot-10.5014_ajot.2021.049090.docx for Occupational Science Concepts Essential to Occupation-Based Practice: Development of Expert Consensus by Catherine L. Backman, Charles H. Christiansen, Barbara R. Hooper, Doris Pierce and M. Pollie Price in The American Journal of Occupational Therapy
Supplementary material for Occupational Science Concepts Essential to Occupation-Based Practice: Development of Expert Consensus
Supplementary material, sj-docx-3-aot-10.5014_ajot.2021.049090.docx for Occupational Science Concepts Essential to Occupation-Based Practice: Development of Expert Consensus by Catherine L. Backman, Charles H. Christiansen, Barbara R. Hooper, Doris Pierce and M. Pollie Price in The American Journal of Occupational Therapy
Footnotes
Acknowledgments
All authors contributed equally to this article. We sincerely thank the following participating experts, listed alphabetically and identified by countries of affiliation, who completed the study and agreed to be recognized: Sissel Alsaker, Norway; Eric Asaba, Sweden/Japan; Aaron Eakman, USA; Michelle Elliot, United Kingdom; Jean Gajardo, Chile; Roshaan Galvan, South Africa; Carita Håkansson, Sweden; Clare Hocking, New Zealand; Jeanne Jackson, Ireland; Hans Jonsson, Sweden; Staffan Josephsson, Sweden; Sarah Kantartzis, United Kingdom; Tomoko Kondo, Japan; Sheama Krishnagiri, USA; Nadine Lariviere, Canada; Elizabeth Larson, USA; May Lim, Singapore; Jacob Østergaard Madsen, Denmark; Kathleen Matuska, USA; Matthew Molineux, Australia; Chantel Morel-Bracq, France; Shoba Nayar, India; Etsuko Odawara, Japan; Melissa Park, Canada/USA; Ruth Kjærsti Raanaas, Norway; Elelwani Ramugondo, South Africa; Kathlyn Reed, USA; Valentin Ritschi, Austria; Charlotte Royeen, USA; Debbie Laliberte Rudman, Canada; Sandra Schiller, Germany; Lynn Shaw, Canada; Mandy Stanley, Australia; Elizabeth Townsend, Canada; Lisette Farias Vera, Sweden; Petra Wagman, Sweden; Gail Whiteford, Australia; Wendy Wood, USA; Valerie Wright-St Clair, New Zealand; and Ruth Zemke, USA.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
