Rehabilitative Health Care Professionals’ Perceptions of Appearance-Based Professionalism

We used a cross-sectional survey design to explore the perceptions of appearance-based professionalism by ATs, PTs, and their respective students. This study was approved by the Temple University Institutional Review Board.

Athletic trainers, PTs, professional-level athletic training students (ATSs), and student physical therapists (SPTs) were included. These professions were selected because both have roles in injury rehabilitation and often perform similar tasks with their patients. No additional exclusionary factors existed. A total of 344 participants consented to the study; 2 participants were removed due to self-reported inaccurate reporting in the open-ended response to the occupation variable (ie, they reported being physical therapy assistants). Of the 295 participants who began the survey, 88.8% (n = 262) completed the survey. Most participants (n = 133, 45.4%) were ATs. Athletic training students (n = 55, 18.8%), SPTs (n = 50, 17.1%), PTs (n = 32, 10.9%), and dual-credentialed providers and students (n = 23, 7.8%) also completed the survey. Participants were an average age of 30 ± 9 years (range = 19–68 years) and were predominantly White (86.8%, n = 236), non-Hispanic (91.2%, n = 248), and female (77.6%, n = 211). The ATs had an average of 10 ± 8 years (range = 0–33 years) of professional experience, whereas PTs averaged 14 ± 12 years (range = 0–40 years). Additional participant demographics can be found in Table 1.

We conducted this survey, which was created by 3 members of the research team (S.M., E.R.N., J.L.M.), using the web-based survey platform Qualtrics. The instrument was validated via an external Delphi panel of 4 experts from various universities: 1 clinician, 3 educators, and professionals in athletic training (n = 3) and physical therapy (n = 1) with various levels of clinical practice, educational practice, and research expertise. The experts were asked if each question was appropriate or if revisions were needed. Two rounds of revisions were conducted until the panel came to complete agreement and consensus about the questions, question delivery, and formatting. Photo participants were volunteers from the primary investigator’s institution. Students and faculty were sent a demographic survey and were asked to disclose different types of attire (eg, leggings) and self-expression (eg, visible tattoos) in which they could be photographed. Figures 1 to 8 illustrate the 8 photos used for the survey instrument.

Demographic questions appeared at the beginning of the survey. The second section consisted of 8 randomized mock pictures of rehabilitative health care professionals providing care. After viewing each picture, participants were prompted, “Do you believe this rehabilitative health care professional looks professional?” The answer choices were yes, no, or unsure/undecided. Once the participant made decisions for all pictures in the survey, the pictures selected as no or unsure/undecided were redisplayed one at a time with a prompt to write a specific explanation regarding their thoughts on why they chose that answer. The survey additionally asked for contact information from participants who wanted to volunteer for the semistructured interview protocol, which was analyzed separately.

Participants were recruited via listservs created by publicly available data of educational program administrators and through social media. For the listservs, program administration contact information was collected for Commission on Accreditation of Athletic Training Education–accredited athletic training programs and Commission on Accreditation in Physical Therapy Education–accredited physical therapy programs as of September 2022. Social media messages were posted on various platforms for participant recruitment throughout the time in which the survey was available. A secondary push for data collection occurred via email from the National Athletic Trainers’ Association Survey distribution list of 1000 ATs and ATSs. The survey opened in November 2022 and closed in April 2023.

We analyzed the data using SPSS (version 28; IBM Corp). The α level was set to P < .05, and the data were analyzed using descriptive and inferential statistics. We performed χ² and Fisher exact tests to identify any differences in responses of the perceptions of professionalism in the photos and various demographic information. Descriptive statistics were means, frequencies, and SDs calculated to examine differences in perceptions among groups.

The short-answer responses were downloaded and sorted by photo. For the analysis, we followed the consensual qualitative research–modified (CQR–M) approach, which has been used in previous athletic training–based research.^16,17  This method, which was adapted from the original CQR method first described by Hill et al, allowed the review of a large sample with brief responses.¹⁶  For new phenomena and unexpected ideas, CQR–M has been described as useful; therefore, we chose this method to ensure that controversial responses and ones that might not align with the team’s personal views were checked.¹⁶  By calling on 3 reviewers and an auditor, this process allowed for interrater agreement, which decreased the expectations, biases, and disagreements in the coding. Although auditing is not a required component of CQR–M, we decided to complete this optional step to ensure quality of coding and eliminate any possible biases from the coding team.

Trustworthiness was established through (1) multianalyst triangulation, (2) an auditor, and (3) reflexivity checks. For the first phase in the coding process, 3 researchers (S.M., E.R.N., J.L.M.) reviewed 50 open-ended responses. These responses spanned the 8 photos, with a minimum of 5 responses or 10% of the initial recruitment responses. From this initial phase, a code book was created with 6 domains. The 3-person coding team (S.M., E.R.N., J.L.M.) coded the 50 responses individually and then met for discussion. Codes were verified by a two-thirds vote from the team. For the second phase, 1 researcher (S.M.) coded all the responses, and the other members (E.R.N., J.L.M.) verified the coding. Any discrepancies were brought to a group decision with a two-thirds vote being needed to verify the coding. An auditor (D.M.M.) was given the instrument, code book, and coded responses and reviewed the codes for consistency and accuracy. The auditor determined that no changes needed to occur. After the second round of recruitment, the same process was followed in phase 2. This process also created dependability in the research. Using CQR–M, we were additionally able to perform reflexivity checks to examine our personal beliefs and influences to make sure they did not affect the coding.

Athletic trainers and PTs were primarily employed in the Northeast region (31.6%, n = 50). Athletic training students and SPTs predominantly studied in the South region (33.0%, n = 36; see Table 2 for full results). Athletic trainers mostly worked in the college and university setting (52 [37.7%]), whereas PTs typically worked in outpatient clinics (14 [40.0%]). The full summary by AT job setting is shown in Table 3, with the PT job settings shown in Table 4. The numbers of yes, no, and unsure/undecided responses by profession for the 8 photos are found in Table 5. Photo 7 had the greatest yes response rate, with 97.8% (n = 263) of participants perceiving this clinician as looking professional; photo 3 was thought to look the least professional, with only 85 (31.6%) participants choosing yes.

We identified significance when comparing students (ATSs, SPTs) and professionals for photo 1 (⁠ = 5.359, P = .021), photo 3 (⁠ = 17.593, P = .000), photo 4 (⁠ = 6.444, P = .011), photo 6 (⁠ = 4.826, P = .028), and photo 8 (⁠ = 4.080, P = .043) but not for photo 2 (⁠ = 1.017, P = .313) and photo 5 (⁠ = 0.42, P = .838). The proportion of participants who reported that the photos were unprofessional differed among professions for photo 6 (⁠ = 11.407, P = .001) and photo 8 (⁠ = 16.031, P = .000) but not for photo 1 (⁠ = 1.166, P = .280), photo 2 (⁠ = 0.244, P = .621), photo 3 (⁠ = 2.387, P = .122), photo 4 (⁠ = 0.983, P = .321), photo 5 (⁠ = 0.1754, P = .185), and photo 7 (⁠ = 2.091, P = .148). Regarding ethnicity, photo 8 differed among participants (⁠ = 4.080, P = .043) but not photo 1 (⁠ = 0.001, P = .980), photo 2 (⁠ = 0.606, P = .436), photo 3 (⁠ = 1.181, P = .277), photo 4 (⁠ = 0.606, P = .436), photo 5 (⁠ = 0.059, P = .809), photo 6 (⁠ = 0.019, P = .892), or photo 7 (⁠ = 0.399, P = .527). The proportion of participants who reported that all of the photos were unprofessional did not differ by gender (photo 1 [ = 0.403, P = .526]; photo 2 [ = 0.281, P = .596]; photo 3 [ = 2.380, P = .123]; photo 4 [ = 0.029, P = .866]; photo 5 [ = 0.002, P = .966]; photo 6 [ = 0.019, P = .891], photo 7 [ = 0.014, P = .906]; photo 8 [ = 0.152, P = .696]). The full χ² analyses are presented in Table 6 (photos 1–4) and Table 7 (photos 5–8).

Based on our qualitative analysis of the open-ended responses, we identified 6 domains: (1) unprofessional attire, hair, or both; (2) situation-dependent attire; (3) nonappearance related; (4) role confusion and health care employer or employee identification; (5) tattoo-related bias; and (6) rethinking after question is displayed. Responses that did not address the prompt given were coded as NA and not included in the analysis. A total of 361 open-ended responses for the 8 photos were coded using the CQR–M method for qualitative data analysis. A total of 195 photos had at least 1 response of no or unsure/undecided that allowed for an open response. The breakdown of open-ended responses by profession is outlined in Table 8. The final consensus codebook with frequency counts is also provided in Table 8. The following themes were supported by associated statements.

Comments regarding attire accounted for the most responses of unprofessionalism, addressed by 65.4% (n = 235) of participants. The unprofessional attire, hair, or both category encompassed any comments relating to, regarding, or questioning clinician appearance. Concerns ranged from how pieces of clothing matched to how hair was styled. In the open-ended response section, some participants commented on the fit of the clothing being worn in the photo. Kate, an SPT, noted in photo 1: “Her shirt appears to be very tight-fitting, which makes it unprofessional. Professionals should be conservative in their dress.”

Zoey, an SPT, remarked similarly about photo 3, calling into question the look of some articles of clothing from different angles.

In addition to clothing style, participants also mentioned hair presentation. Some responses were gender specific; regarding photo 6, Carol, a PT, shared: “I do not feel a professional should wear shorts or have a ‘man bun.’” Clothing and hair concerns were sometimes related to weather conditions. Still, Oliver, an SPT, thought the clinician’s weather-appropriate attire was unprofessional in the same photo (6):

This category included participants’ responses that the attire worn by the clinician in the photo might have been better suited for a different situation than pictured. A total of 82 participants (22.7%) described the attire as possibly situationally professional, with an emphasis on functionality and weather. Amelia, an AT, observed in photo 1:

Because rehabilitative health care professionals can work in many different locations and terrains, multiple participants pointed out the need for attire to be functional enough for the practice setting. Kim, an AT who was currently in school as an SPT, offered her thoughts on photo 3:

Participants had a similar point of view on the need for clothing to adhere to the active role some health care providers perform when working outside of the traditional office setting. Aya, an SPT, shared her thoughts about the situation-dependent clothing in photo 3:

One participant took a different approach and brought up the potential for weight bias, calling attention to the fact that some articles of clothing may be perceived differently, whether professional or unprofessional, on larger body types. Abbey, an AT, gave her perspective on photo 3:

The health care employer or employee identification theme includes the need for clinicians to wear more identifiable articles of clothing. Health care employer or employee identification was thought necessary to be professional by 21.1% (n = 76) of participants. Josh, an ATS, explained his opinion of photo 1:

Some participants felt that wearing health care employee or employer identification showcased authority and expertise, whereas nonbranded casual attire was seen as unprofessional. For photo 3, Helen, who was studying as an SPT, remarked:

A common thread in the responses was role confusion and the need to change attire or appearance to more clearly project the role of a health care professional. Regarding photo 8, Marguerite suggested more structured clothing for the clinician to help outside sources distinguish patient from provider in her work as an ATS:

This domain also encompassed responses that reflected a need to eliminate role ambiguity regarding the health care provider. Participants believed that being able to distinguish between patient and clinician on sight was necessary to be deemed professional. For photo 8, Louise, an AT, noted: “I know this is what ATs wear for practice all the time. I know some view this as professional attire, but nothing here suggests this person is anything more than a coach.”

Eleanor, who was an AT, expressed similar concerns with photo 3, stating that the clinician “look(s) sloppy, not like a sports medicine provider,” based solely on the visual presentation of the person.

The nonappearance-related theme encompasses perceptions of professionalism based on other items like the positioning of the clinician when stretching a patient (14.1%, n = 51). Donna, an AT, shared her thoughts on photo 8: “Poor positioning for stretching male athlete—AT is crouched over the athlete rather than on the side or standing, helping to stretch.”

Emerging from comments on improper positioning was the notion that some stances could be seen as sexual in nature. Similar to the previous comment, Abbey said that females, specifically with respect to photo 8, should be careful when stretching males as to not draw unwanted attention:

Regarding photo 8, Nancy, a PT, remarked: “Using her knee to stabilize the thigh could be construed as sexual in nature.” Alternatively, Courtney, an ATS, took a nonappearance-related approach in her comments on photo 3 in a different direction, focusing on the functionality of the clothing for the profession:

Leo, an ATS, also commented on the body language in photo 3 and how it could affect the patient encounter:

Tattoo-related concerns were identified in 5.5% (n = 20) of responses, with the common suggestion being they should be covered. Sarah, a PT, expressed her perceptions of the tattoos in photo 4: “The tattoos are not covered, and there [are] multiple ones.” Similarly, Alex, an AT, stated “In the clinical setting, I think tattoos should be covered.”

Health-related concerns were brought into question. Observing photo 5, Charlie, an AT, answered “The tattoos make me question the hygiene.”

Even when the style of clothing was thought to be appropriate, participants still questioned professionalism when the clinician had large or multiple tattoos. Kathryn, a PT, shared her view of how tattoos affected her perceptions of the provider’s professionalism in photo 5: “The amount of tattoos makes me question the professional appearance. The sleeve is an appropriate length, but the tattoos extend all the way down the arm.”

This theme included responses (5.3%, n = 19) that participants rethought or changed after being prompted to describe why they felt the clinician in the photo was unprofessional. Josh, an ATS, responded to photo 8:

Participants also rethought their motives and acknowledged gender biases. While admitting the bias in her response, Emily, an AT, offered her perspective of photo 8: “The length of the shorts is inappropriate. However, I fully acknowledge that most dress codes that have to do with pants/short length are gender biased toward females.”

Additionally, Amelia, an AT, remarked on the nonnormative expressions of self and culture in photo 5: “This was tough. I respect expression and culture. I think, if it wasn’t a T-shirt but a collared shirt, long or short, that could influence my decision about professionalism.”

Professionalism can be defined in diverse ways. Our results showed that professionalism is often influenced by various factors, and the perception of professionalism can grow, change, and adapt over time. First impressions are notable but not always the best indicators of a person’s professionalism. Additionally, what one believes to be professional may change over time, even superseding external influences. When we examined demographic differences, no significant findings were present for gender, the region in which the health care professionals lived, the region in which the students attended school, or ethnicity; also, no significant correlations were evident between the yes, no, or unsure/undecided responses and the photos.

Because no current standards for professionalism exist, it would be arrogant to think that everyone would have the same professional beliefs about each photo. Professionalism can be witnessed and learned in an unofficial, unwritten, or unintended manner, which can lead to contradictions, confusion, and misunderstanding.¹⁸  No differences were found among genders, indicating that similar views were held regardless of the gender of the participant. The 1 area in which responses to 5 of the 8 photos were different was in comparing professionals with students in the disciplines. This may have been because students were either socialized into following a set of professional norms or more willing to go against these traditions.

Throughout our study, participants noted the need for an easily identifiable distinction between providers and patients for several reasons, including easy identification for the public and patient recognition. Currently, no set dress code specifically identifies rehabilitative professionals, although we have seen attire such as khakis and a polo shirt used to identify ATs.¹⁹  Just as previous researchers have observed that physicians’ white coats were preferred by the majority of patients within hospital settings, our results were consistent: rehabilitative professionals expressed a need for clinicians to wear some form of employee or employer identification. To make the health care provider more easily recognizable to patients, athletes, and coaches, participants recommended name tags and school-branded attire. Some state laws require ATs to wear their identifying badges.^20,21  In addition to branded clothing, other ways to identify health care providers include medical bags or packs or medical station signage, increasing visibility in an emergency.

Generational workplace preferences for dress code differed, with Gen X, millennials, and even boomers favoring business casual as a more loosely enforced code.²²  Many participants in our study also discussed whether it was better to wear a collared shirt in a professional setting and identify when athleisure (ie, casual clothing designed to be worn both for exercise and for general use)²³  was appropriate based on the situation (eg, when functionality was important). One participant commented on a clinician’s athletic-appearing attire of biker shorts, supporting the need to be functional and able to demonstrate proper mechanics or engage in activity; however, other participants believed the athletic clothing was too tight and unprofessional and overall took away from the clinician’s professional appearance. Rehabilitative health care professionals work in a multitude of settings and may need to be ready for vastly different activities from 1 day to the next, which may support the need for flexibility in dress.

Gender expression is defined as how a person publicly presents gender via various possible combinations of behavioral and appearance-based attributes, including but not limited to dress, hair, body language, or voice.²⁴  Investigators have documented both negative and positive views as connected to freedom of gender expression, self-expression, or both, with appearance being a factor in patients’ perceptions of health care providers.⁵  Nontraditional forms of self-expression (eg, visible tattoos) are becoming more prevalent among physician and nonphysician health care providers.⁵  This will likely continue, with about 40% of millennials having a tattoo.²⁵  Values have changed; though tattoos were once tied to aggression and delinquency, the prevalence of tattoos in society and positive portrayals by the media have led to tattoos being viewed as creative status symbols.²⁵  Although communicated by only a small percentage of our participants, stereotypes were expressed when tattoos were visible on a clinician. Strong negative tattoo stigmas have given way to discrimination of persons with tattoos in both professional and personal life settings.²⁶ 

We would be remiss not to mention how often gender was cited, even though the question was never asked. This aligns with the notion that professionalism is a gendered construct that the participants subconsciously identified in their open-ended responses. For example, a photo depicting 2 female-identifying persons received multiple comments about how a woman should be cautious of being in this same stance if the patient was a man. In health care, gender perceptions continue to dictate what is appropriate attire and behavior. Gender biases favoring men as more professional than women persist in health care professions.²⁷  Our participants reflected a higher representation of women (77%) than in the 2 professions (athletic training = 55%, physical therapy = 65%).

Historically, professionalism in health care settings has been defined by and for heterosexual White males and is not always inclusive or accepting of diversity in the workplace or its patient population.²⁴  This sentiment held true in the current study, in which a White, cisgendered male had the highest professional rating, even though he was pictured in the same position as 2 other female-presenting clinicians. The models presenting as females in this survey received more criticisms on their clothing choices than the males, with functional athletic shorts being deemed as too tight or too casual and dress shirts and shawls being seen as too dressed up for professional demands.

Our purpose was to identify ideas related to appearance-based professionalism. Surprisingly, participants also noted characteristics of the persons in the photographs that dealt with nonappearance-related critiques. These included patient positioning for stretches and taking blood pressure. All patient and clinician positioning was in accordance with patient care techniques and not meant to be deemed sexual in nature. One example was when a more petite female used her body positioning to assist with leverage on the nonstretched leg during an assisted hamstring stretch. Although this is a generally accepted functional practice for smaller clinicians, outside observers may question techniques and intent. Clinicians should take note with respect to more public-facing athletic training duties (eg, sideline care).

Our work had several limitations. We believe that true diversity includes more than what we can see on the outside; however, for this appearance-based study, depictions of diversity were limited (eg, skin color, gender identity, self-expression). Another limitation was that selection bias could have been present in the participant pool. Participants who had a vested interest in the topic may have been more likely to take the survey, which perhaps affected the perceptions reported.

Future researchers should expand beyond identifying bias against minoritized groups and begin to look at debiasing techniques. Creating and implementing more diversity, equity, and inclusion workshops with a specific focus on bias recognition and management could aid in redefining professionalism to reflect knowledge, skills, and abilities rather than subjective personal attributes. Additionally, with the increasing number of women in the rehabilitation workforce, they should have more of a voice in the dress they wear. Body shapes differ for individuals but especially between women and men. Better choices for women (eg, pants with pockets) may not always be available if following the workforce practices dictated by men.

On review of rehabilitative health care professionals’ perceptions, what is considered appropriate and professional is not concrete. Rehabilitative health care professionals should be cognizant of their perceptions of others to make certain bias is accounted for and minimized. The term professionalism has been suggested as being outdated and favoring White, cisgender men. As health care professionals, we must ensure that, with our implicit biases, we do not judge those who do not fit the mold of what is currently being accepted as professional, which this study demonstrated is not an exact science.