The Influence of Background and External Exposures on Non-Cognitive Traits in Doctor of Physical Therapy Applicants

The Influence of Background and External Exposures on

Non-Cognitive Traits in Doctor of Physical Therapy Applicants

by Conor J. McClure

Background

I have long been intrigued by what is often referred to as psychologically-informed practice, a healthcare model which incorporates cognitive-behavioral techniques alongside traditional medical intervention to address psychosocial factors (e.g., beliefs, emotions, coping strategies) in patients limited by debilitating pain or disability.¹ Critical aspects of successful psychologically-informed physical therapy intervention include patient-centered language, clinical equipoise, sound interpersonal skills, and the development of a strong therapeutic alliance.² Additional research has identified a wide variety of personal attributes that contribute to a positive therapeutic alliance and improved patient outcomes, including personal flexibility, confidence, conscientiousness, extraversion, tolerance for ambiguity, grit, and more.^3-5 These are referred to as non-cognitive traits. These encompass aspects of one’s personality, such as behavior, personality, emotion, motivation, and others, the majority of which are not measurable by existing cognitive metrics. Increasingly, these traits are known to associate with more skillful, empathetic, and successful healthcare clinicians.

While my professional interests are myriad, the common denominator between them is that they all involve the human experience; “treat the person, not the tissue,” I have been known to say. The human experience is the reason I enjoy teaching—I have aspirations of professorship in academia down the road, and I get a great deal of satisfaction from helping others learn and better understand novel information. It is also the reason I enjoy research—most of my peers would probably vote me the “nerdiest” member of the class, and I’ve established a reputation of being the go-to PubMed scholar of my cohort. I frequently entertain notions of pursuing a PhD and have been longing to participate in real research projects. So, when presented with the opportunity to participate in a research project examining non-cognitive traits in DPT students, I could not pass up this opportunity to further explore my passions.

Purpose and Statement of Need

Applicants to Doctor of Physical Therapy (DPT) programs are typically evaluated by a number of metrics; however, these are mostly cognitive in nature. Grade point average, GRE scores, physical therapy shadowing hours, and other objective measures remain the standard for evaluating applicants’ abilities and attempting to predict future occupational success. Research suggests that the relationship between cognitive intelligence and emotional intelligence is weak at best.^6-8 And with the relative paucity of validated non-cognitive assessment metrics, precise and reliable identification of these traits remains a difficult task.⁹ Most graduate programs incorporate an interview process in an attempt to better identify non-cognitive traits in applicants—but is this enough? I have participated as an interviewer of applicants to the UNC DPT program for two years now and can attest to its limited utility in distinguishing and identifying those who will excel both academically and professionally.

In an attempt to solve this problem, researchers at Duke University, the University of Colorado, and University of Indianapolis have recently developed a novel tool for evaluating non-cognitive traits of DPT applicants¹⁰ at multiple institutions. This survey was demonstrated to have strong construct, content, and face validities; however, to increase its future utility and value, this tool requires additional research to assess its concurrent validity with more objective metrics, such as academic performance, National Physical Therapy Examination pass rates, and other academic and professional milestones. Furthermore, it would be useful in identifying characteristics that might associate with or predict higher scores on these preferred traits.

Not only do those who possess these types of non-cognitive traits tend to develop into quality clinicians, they also perform better in academic settings.¹¹ It is therefore prudent for graduate admissions committees to select applicants who possess both desirable cognitive and non-cognitive attributes.

How can we better identify aspiring healthcare professionals who possess these desirable traits?

As part of an annual collaboration between Duke University and the University of North Carolina, I joined a team of faculty researchers and DPT students to undertake a study to help further establish validity and value in this tool to further enhance DPT programs’ abilities to evaluate and identify ideal DPT student candidates. More specifically, this project examined the influence of background and prior experiential exposures on the non-cognitive traits evaluated by this tool. These variables included baseline demographic characteristics such as age, gender, or minority status, in addition to other characteristics, such as military experience, prior application to a DPT program, high school graduation rates, undergraduate program ranking, socioeconomic status, and many others. Any and all information provided to the Physical Therapy Centralized Application Service (PTCAS) application service was open for analysis. Our hope was that these findings would prove useful to future DPT admissions committees, who would be able to potentially predict non-cognitive traits in applicants based on their previous exposures and personal characteristics.

Project Overview

This research team consisted of six members. Dr. Deborah (“Debby”) Givens represented the lone UNC faculty member. Drs. Marcus Roll and Chad Cook were the participating Duke University physical therapy researchers. Cannon Hanebuth and Daniela Ortiz, two Duke DPT students in their second year, joined me as the student co-authors.

This project began at the beginning of the Fall semester of 2019. I registered for an independent study elective with Dr. Givens, in which I spent upwards of 20 hours researching and drafting a literature review. This paper examined available research on topics including the relationship between non-cognitive traits and academic and professional performance in healthcare fields, the relative contributions of genetics and environment to the personality development, associations between various individual characteristics (e.g., age, socioeconomic status) and non-cognitive traits, and more. The purpose of this was twofold: to establish and justify our reasoning behind this project, and to later be integrated into the final manuscript for publication.

Beginning in the early Fall 2019 semester of 2019 and continuing through early Spring 2020, I began to meet both virtually and in-person with my colleagues and co-authors at both the UNC and Duke campuses. In these meetings, we discussed initial findings and directions to take our research project. With a variable deadline between late April and early Summer, the end goal was determined to be a manuscript to be submitted for publication, with workload distributed between all participants. We determined that I would be nominated as lead author of the final paper alongside Dr. Roll. In writing the manuscript, I took the lead on the Introduction and Discussion sections, my Duke DPT student colleagues writing the Results section, and Dr. Roll wrote the Methods section. Drs. Roll and Cook conducted data analysis with IBM’s SPSS software.

This process continued through mid-March 2020. Virtual collaboration was facilitated through services such as Box and Google Docs, in which regular updates of statistical analyses, tables, and manuscript edits were shared. Our ability to maintain constant communication through text messaging, email, and application-provided chat features reduced the need for continued in-person meetings.

The individual characteristics (de-identified information obtained via PTCAS) that were analyzed include:

• Age
• Gender
• First generation in one’s family to attend college
• Physical therapy shadowing hours (paid, volunteer, and total)
• Prior applicant to DPT program
• Graduated from a high school from which a low percentage of seniors received a high school diploma
• Family receives public assistance or lives in a medically underserved area
• From a school district where 50% of less of graduates go to college
• English as a second language
• Military experience
• Undergraduate school ranking by U.S. News
• And more

The non-cognitive domains assessed and the validated measurements from which they were derived include:

• Adaptability
  • Intolerance of Uncertainty Scale (IUS)
  • Psychological Flexibility Questionnaire
• Intuitiveness
  • Schutte Self Report Emotional Intelligence Test
  • Measuring Social Intelligence (MSI) scale
  • Short Grit Scale (Grit-S)
• Engagement
  • Interpersonal Reactivity Index (IRI)
  • Psychological Flexibility Questionnaire

The construction of these latent non-cognitive domains is outlined in greater detail in our previous study, cited above.¹⁰ As our study is not yet published, I will not be reviewing the precise results of our data analysis on this page. However, a broad overview of our findings can be found in the presentation mentioned in the Products section below. There were numerous statistically significant (p < .05) associations between individual characteristics and non-cognitive traits, which can be viewed below:

Non-Cog Domains	Extreme Non-Cog Domains
Adaptability	Adaptability
Age (p < .01)**	Total shadowing hours (p < .05)*
Prior Applicant (p < .05)*	Family receives public assistance (p < .05)*
Military Experience (p < .05)*	Intuitiveness
Intuitiveness	50% or less high school graduates go to college (p < .01)**
Prior Applicant (p < .05)*	English as a second language (p < .01)**
Engagement	Military experience (p < .01)**
Undergraduate school ranking (p < .05)*	Underrepresented minority status (p < .05)*
Total Non-Cognitive Scores	Socioeconomically disadvantaged area (p < .05)*
Prior Applicant (p < .01)**	Low graduation rate high school (p < .05)*
Military Experience (p < .05)*	Family receives public assistance (p < .05)*
	Engagement
	N/A
	Total Non-Cognitive Scores
	Low graduation rate high school (p < .05)*
	Socioeconomically disadvantaged area (p < .05)*
	Family receives public assistance (p < .01)**

---

The vast majority of these positive correlations, while statistically significant, were generally weak. What might explain this?

The concept of “nature versus nurture” has existence since Shakespearean times and beyond. The relative contributions of environmental factors and genetics have formed a debate that continues to this day in some circles. However, with continued advances in psychological and genomic research, the modern scientific consensus is that such a stark dichotomy is both unnecessary and unsupported. For almost everything of importance, both genetics and our environment seem to play nearly equally important roles.

This is perhaps no more easily demonstrated than through studies examining siblings, identical twins, and adoptees.^12,13 On average, these studies show a large heritability component to non-cognitive traits; however, they also suggest that one’s personality traits are more malleable than one’s cognitive skills, the degree of which seems to vary throughout one’s life. Relevant environmental factors vary from basic familial influences to life-changing events. This subject and others is investigated in greater detail in my literature review, which can be found in the Products section below.

The results of this study leave us with many more questions to be answered. Are we truly able to predict non-cognitive traits on background influences alone? And if so, are the characteristics used in our analysis enough? Do certain environmental factors develop desirable traits or simply select for desirable traits? Is extreme response style, in which participants select “1” or “5” on a 5-point Likert Scale, a desirable or undesirable trait? There are countless avenues for continued exploration in this field, and with continued evolution of this tool and the accumulation of larger and more diverse cohorts over the coming years, the possibilities seem endless.

Limitations of our study include: the length of the survey (39 items), the fact that it was distributed during interview day (a particularly stressful time), that we were limited to only analyzing PTCAS data, and the ever-present question of if applicants answer all questions honestly.

Future directions of this research include establishing concurrent validity by tracking these cohorts over time and investigating non-cognitive correlations with profession-specific metrics such as academic performance, NPTE pass rates, if the student becomes a clinical instructor, if the student joins a residency program, and more. This survey will continue to be administered to incoming classes, and our analyses will benefit from larger and more diverse cohorts over the coming years.

Products

The initially proposed final products for this project included a manuscript to be submitted for publication in the Journal of Educational Evaluation for Health Professions (JEEHP) and an abstract and poster for submission to a national conference such as the annual Combined Sections Meeting (CSM). Unfortunately, the COVID-19 pandemic and the abrupt nationwide transition to virtual learning resulted in the timelines for these products to be extended beyond the deadline of this Capstone project.

• The literature review (described above) produced as part of my independent study served as a valuable springboard to initiating this research project. This review, totaling 14 pages, nearly 4,000 words, and 57 citations, can be downloaded and read in PDF format here.
• Additionally, a PowerPoint presentation was created to help summarize our methods, results, and additional topics of discussion. This document resembles a presentation that might be presented at a national conference in the future and provides an overview of our findings and future directions of research. It can be downloaded and viewed in either PowerPoint (.pptx) or PDF format.

Lastly, it is worth emphasizing that while the timeline of the abstract, manuscript, and additional materials has been delayed, the project is still quite alive and ongoing. I look forward to continuing to contribute to this project in the future.

Evaluation

Feedback from my advisor, Dr. Givens, and the lead author, Dr. Roll, was provided consistently throughout this process. Dr. Givens in particular provided multiple edits and suggestions for my literature review and was integral in helping me refine my scientific writing skills. I also collaborated frequently with Cannon and Daniela, my co-authors from Duke University, during the writing process.

My Capstone committee members, Dr. Marcus Roll and Dr. Chad Cook, provided feedback through a form sent to them upon completion of the Capstone project. This feedback was returned to my advisor, who then relayed it me.

Health Literacy

Health literacy is a key component of UNC’s Capstone curriculum. It can defined as one’s capacity to obtain, communicate, process, and understand health information and apply it effectively. This project sought to address the health literacy of physical therapy faculty, clinicians, students, and applicants, as opposed to patient populations or the general public. Research cited in my literature review demonstrates the strong relationship between non-cognitive skills and improved academic and professional performance and positive healthcare outcomes. Furthermore, our study’s findings reveal notable background characteristics of applicants who exhibit these desirable traits. In this sense, the health literacy of our profession is addressed by empowering admissions committees, faculty, and current or potential students to maximize their clinical potential.

Self-Reflection

While I am ultimately disappointed that this research project was not completed by the end of the school year, I am able to reflect on advice given to me by Dr. Givens and Dr. Thorpe: “Don’t set expectations too high and set yourself up to fail.” This is a tendency of mine that I am still trying to tame. With that said, I am eager and excited to continue to work on this project (and more) even after graduation from this program.

This entire process has revealed a great deal to me. I’ve always known that I am both a fast writer and an efficient integrator of research, and these skills leant themselves well to the construction of both the literature review and manuscript draft. I have become best friends with my reference manager, Microsoft Word, and PubMed. For a couple of years now, I have become profoundly interested in the field of clinical research and have been increasingly considering PhD options and academic positions further down the path of my career. The process of collaborating with experienced research faculty has been a valuable formative step towards progressing these future goals. I’ve also learned a great deal through trial by fire; research is often a long and arduous undertaking—more than anticipated—and collaborating with others requires unique skills that I must continue to develop. Time management, prompt communication, efficient workflows, and knowing when to ask for help are all skills that I have been able to cultivate throughout this experience.

Acknowledgements

This project would not have been possible without Dr. Deborah (“Debby”) Givens, PT, DPT, PhD. She has served as the advisor for many joint projects between UNC and Duke in addition to master’s and PhD students, most of whom were most certainly more talented and qualified than me. This project initially required two DPT students from UNC—I am grateful and humbled by her faith in me to take on this project without the assistance of a colleague from my own program. As the director of our program during a once-in-a-century pandemic, Dr. Givens had more than enough on her plate throughout this semester, but somehow still managed to find the time to provide helpful feedback and guidance. She has proven to be an irreplaceable resource for me both professional and personally.

The conception of this project is credited to Dr. Marcus Roll, PT, DPT, NCS, CBIS. As a participant in the Duke DPT Faculty Development residency, his passion and enthusiasm for academia is palpable. This study is a continuation of previous research of his and will no doubt be part of a much longer line of research in the future. I am grateful for his trust in me and allowing me to take on a significant role in this project, in addition to his frequent feedback and direction.

I am grateful and honored to have Dr. Chad Cook, PT, PhD, MBA, FAPTA as a co-author and Capstone committee member. A prolific research and eminent clinician and academician in our field, his skill and knowledge of data extraction and analysis were critical in determining our findings and outcomes.

Additionally, thanks and appreciation for my Duke DPT colleagues Cannon Hanebuth and Daniela Ortiz, who served as my co-authors and collaborators. They took on some of the more challenging aspects of this manuscript and worked closely with Duke faculty when I was not able to. They were always available for communication and were always on top of their game, sometimes more than I was. It was a pleasure working with these two—even if they represent the wrong shade of blue.

I would also like to acknowledge Dr. Vicki Mercer, PT, PhD, for discussing this project with me during our service-learning trip over Spring Break, offering advice, suggestions, and thoughts from an outside researcher’s perspective. Dr. Deborah Thorpe, PT, PhD deserves thanks for managing the overarching Capstone coursework and providing with me with suggestions as well. Additional thanks to Jon Hacke, PT, DPT, and Michael McMorris, PT, DPT, for allowing me to serve as a teaching assistant for a first year class, which helped me refine my teaching skills in hopes of one day becoming a professor like them.

♥

And lastly, I’d like to acknowledge and thank every single one of my classmates in the UNC DPT Class of 2020. Over the past three years, I’ve developed some of the best friendships of my life—the type of friendships I hope to maintain throughout the rest of my life. We’ve stuck with each other through the best and worst of times, and I’ve had the great pleasure of watching all of us succeed, struggle, and ultimately grow into the outstanding students and soon-to-be clinicians we are today. I couldn’t have asked for a better group to accompany me on this wild ride.

♥

Thank you,

Conor McClure

---

Bibliography

 

1. Archer KR, Coronado RA, Wegener ST. The role of psychologically informed physical therapy for musculoskeletal pain. Curr. Phys. Med. Rehabil. Rep. 2018;6(1):15-25. doi:10.1007/s40141-018-0169-x.
2. Keefe FJ, Main CJ, George SZ. Advancing psychologically informed practice for patients with persistent musculoskeletal pain: promise, pitfalls, and solutions. Phys. Ther. 2018;98(5):398-407. doi:10.1093/ptj/pzy024.
3. Ackerman SJ, Hilsenroth MJ. A review of therapist characteristics and techniques positively impacting the therapeutic alliance. Clin. Psychol. Rev. 2003;23(1):1-33. doi:10.1016/s0272-7358(02)00146-0.
4. Buining EM, Kooijman MK, Swinkels ICS, Pisters MF, Veenhof C. Exploring physiotherapists’ personality traits that may influence treatment outcome in patients with chronic diseases: a cohort study. BMC Health Serv. Res. 2015;15:558. doi:10.1186/s12913-015-1225-1.
5. Kooijman MK, Buining EM, Swinkels ICS, Koes BW, Veenhof C. Do therapist effects determine outcome in patients with shoulder pain in a primary care physiotherapy setting? Physiotherapy 2019;107:111-117. doi:10.1016/j.physio.2019.08.009.
6. Stankov L. Low Correlations between Intelligence and Big Five Personality Traits: Need to Broaden the Domain of Personality. J. Intell. 2018;6(2). doi:10.3390/jintelligence6020026.
7. Kretzschmar A, Spengler M, Schubert A-L, Steinmayr R, Ziegler M. The Relation of Personality and Intelligence-What Can the Brunswik Symmetry Principle Tell Us? J. Intell. 2018;6(3). doi:10.3390/jintelligence6030030.
8. Saggino A, Balsamo M. Relationship between WAIS-R intelligence and the five-factor model of personality in a normal elderly sample. Psychol. Rep. 2003;92(3 Pt 2):1151-1161. doi:10.2466/pr0.2003.92.3c.1151.
9. Guffey JS, Farris JW, Aldridge R, Thomas T. An evaluation of the usefulness of noncognitive variables as predictors of scores on the national physical therapy licensing examination. J Allied Health 2002;31(2):78-86.
10. Roll M, Canham L, Salamh P, Covington K, Simon C, Cook C. A novel tool for evaluating non-cognitive traits of doctor of physical therapy learners in the United States. J. Educ. Eval. Health Prof. 2018;15:19. doi:10.3352/jeehp.2018.15.19.
11. Cook CJ, Cook CE, Hilton TN. Does emotional intelligence influence success during medical school admissions and program matriculation?: a systematic review. J. Educ. Eval. Health Prof. 2016;13:40. doi:10.3352/jeehp.2016.13.40.
12. McGue M, Rustichini A, Iacono WG. Cognitive, noncognitive, and family background contributions to college attainment: A behavioral genetic perspective. J Pers 2017;85(1):65-78. doi:10.1111/jopy.12230.
13. Weiss A, Bates TC, Luciano M. Happiness is a personal(ity) thing: the genetics of personality and well-being in a representative sample. Psychol. Sci. 2008;19(3):205-210. doi:10.1111/j.1467-9280.2008.02068.x.