Patient Preference in Dermatologist Attire in the Medical, Surgical, and Wound Care Settings

Importance  Patients’ perceptions of their physician can affect subjective and objective outcomes. Physician attire influences patients’ perceptions of their physician and consequently may affect patient outcomes.

Objective  To determine patient preferences for different types of dermatologist attire in dermatology medical, surgical, and wound care clinics. We hypothesized that patients in the dermatology medical setting would prefer professional attire, while patients in the dermatology surgical and wound care setting would prefer surgical scrubs.

Design, Setting, and Participants  This study analyzed responses to a cross-sectional, anonymous survey by English-speaking dermatology patients (aged 18 years or older) at general, surgical, and wound care clinics in an academic center in Miami, Florida. Patients who could not read and understand the survey were excluded. Participants received pictures of a physician wearing business attire, professional attire, surgical attire, and casual attire, and responded by indicating which physician they preferred for each of 19 questions. Frequencies of responses were recorded, and χ² and regression tests were performed.

Main Outcomes and Measures  Response frequencies.

Results  Surveys were administered to 261 persons, and 255 participated and completed enough of the questions to be included in the outcome analyses (118 men, 121 women, 22 unknown [did not answer sex question]), mean (SD) age, 56.3 (18.6) years; about 49% of those who reported their sex were men; 56% were Hispanic; and 85% were white. Approximately 72% of respondents held a college degree or higher. About 63%, 24%, and 13% of respondents were medical, surgical, and wound care dermatology patients, respectively. Roughly 73%, 19%, 6%, and 2% of cumulative responses were for professional, surgical, business, and casual attire, respectively. Respondents who received a picture of a black male or black female physician were more likely to exclusively prefer professional attire: unadjusted odds ratios (ORs) 3.21 (95% CI, 1.39-7.42) and 2.78 (95% CI, 1.18-6.51), respectively, compared with respondents who received a picture of a white male physician. Nonwhite and unemployed respondents were less likely to prefer professional attire exclusively: ORs, 0.28 (95% CI, 0.1-0.83) and 0.28 (95% CI, 0.08-0.99), respectively. Respondents preferred professional attire in all clinic settings, though respondents in the dermatology surgery clinic were less likely to prefer professional attire compared with respondents in the medical dermatology clinic: race-adjusted OR, 0.74 (95% CI, 0.56-0.98). Wound care and medical dermatology respondents preferred professional attire comparably.

Conclusions and Relevance  In this study, most patients preferred professional attire for their dermatologists in most settings. It is possible that patients’ perceptions of their physicians’ knowledge and skill is influenced by the physicians’ appearance, and these perceptions may affect outcomes.

Physicians and patients quickly develop an opinion about each other during their initial encounter. Patients (subconsciously or consciously) evaluate whether they like and trust the physician, and this serves in the development of the physician-patient relationship.¹ Patients’ perceptions of their physicians affect outcome and have been linked to improvement (or lack thereof) in quality of life, depression, hemoglobin A_1c levels, blood pressure, and cholesterol levels.^2-4 Part of this perception is based on physician attire,^5,6 and consequently, it is possible that physician attire may affect patient outcomes.

In other medical fields, patients prefer their physicians to wear formal attire (button-down shirt, tie, and slacks for male physicians and blouse, skirt, or suit pants for female physicians), with or without a white coat, but this seems to vary geographically, by patient age, and by specific fields within medicine.⁶ Three prior studies have investigated patient preference in dermatologist attire. Thomas et al,⁷ used a written survey without pictures to inquire about the importance of dermatologists’ wearing a white coat with or without a tie and found that adult respondents preferred white coats, albeit only slightly. Similarly, most respondents in a study by Kanzler and Gorsulowsky⁸ reported the white coat as a “desirable” physical characteristic of their dermatologist. Maruani et al⁹ evaluated patients’ attire preference in a French inpatient and private-practice setting and found patients preferred professional attire (though surgical scrubs was not one of the investigated attire options). We sought to elucidate patient preference in dermatologist attire in the outpatient dermatology medical, surgical, and wound care settings to determine what additional factors may affect these preferences and how dermatologist attire affects patients’ perceptions of their dermatologist. We hypothesized that patients in the dermatology medical setting would prefer the white coat, while patients in the dermatology surgical and wound care settings would prefer surgical scrubs.

Box Section Ref ID

After obtaining University of Miami (UM) institutional review board approval, we prepared a validated,⁵ anonymous, 1-time survey to be administered to UM Miller School of Medicine Dermatology Department patients in outpatient general dermatology, Mohs surgery, and wound care clinics. A copy of the survey questionnaire is provided in the eAppendix in the Supplement. Eligible respondents were patients (both new and follow-up patients) in a UM dermatology clinic, at least 18 years old, and able to complete the English-language survey unassisted. We excluded respondents who were unable to visualize the pictures or comprehend the survey questions. All participants provided their written informed consent.

To make the survey better suited to dermatology, modifications were made to a prior survey⁵ as follows: 6 questions were added, 1 deleted, and 5 were slightly modified. These modified questions were administered to volunteers who affirmed their clarity. Respondents in the clinic waiting rooms were approached by investigators and invited to complete the survey without any inducements. If respondents were called to enter their examination rooms, they completed the survey in their examination rooms. Respondents were assured of the anonymity of their responses. Each respondent was administered a survey with 1 picture series of the 4 different attire types (Figure) selected in a consecutive fashion. Each series featured 1 physician (white male, white female, black male, or black female) wearing business attire (suit and tie), professional attire (white coat and tie), surgical attire (scrubs), and casual attire (t-shirt and jeans)⁵ on a single page, so the respondent was able to view all 4 attire types simultaneously. Each image depicted the physician using identical lighting, background, facial expression, hairstyle, and accessories. Respondents were asked, of the 4 physicians (same physician wearing the 4 different attires), which would they prefer “to be their dermatologist,” “to cut out skin cancer,” “for a dermatologic emergency,” and other questions.

Using Statistical Package for the Social Sciences, version 22, we performed χ² tests for each question to compare responses across respondent groups such as education level, race, and sex as well as the pictured physician’s race and sex. A multivariate logistic regression model was used to determine if respondent characteristics or pictured physician characteristics influenced the respondents’ exclusive preference for the white coat for all questions. Finally, a negative binomial regression analysis was performed including a Wald χ² test along with likelihood ratios to determine significant linear associations between all respondents’ clinic location and likelihood of choosing professional attire vs the other attire options for the sum of all questions. Statistical significance was set at P < .05 with 95% confidence intervals (CIs) for all tests. Surveys with missing responses were included in the analysis without imputing missing values.

Surveys were administered to 261 persons (mean [SD] age, 56.3 [18.6] years), and 255 participated and completed enough of the questions to be included in the outcome analyses; demographic data were reported for all 261 when available. Nearly half of respondents who reported their sex (49.4%, n = 118) were men; 55.5% (n = 142) were Hispanic; and 85.4% (n = 211) were white. Nearly three-quarters of respondents (72.1%, n = 186) had a college degree or higher; 86.2% (n = 219) were employed full-time or part-time or retired; and 55.6% (n = 140) were married. About three-fifths of respondents (62.6%, n = 154) were medical dermatology patients, 24% (n = 59) were surgical dermatology patients; and 13.4% (n = 33) were dermatology wound care patients (Table 1).

The rate of participation was greater than 95%. Only 3 surveys (1.1%) were returned nearly entirely incomplete and were excluded. In addition, 3 respondents refused to participate, 1 because he believed the project was not important, and 2 because they were “not interested.” We did not quantify the number of respondents who were excluded owing to inadequate proficiency with the English language because they were excluded based on inclusion and exclusion criteria.

Respondents overwhelmingly preferred professional attire across all clinic settings (Table 2). Of all responses, 73% preferred professional attire, 19% preferred surgical attire, 6% preferred business attire, and 2% preferred casual attire. The query “which would you prefer for a dermatologic emergency?” was the only case for which respondents preferred surgical attire more often than professional attire: 49% vs 47% (Table 3).

We also found that respondents who received a picture of a black male or black female physician were significantly more likely to exclusively prefer professional attire: unadjusted odds ratios (ORs) 3.21 (95% CI, 1.39-7.42) and 2.78 (95% CI, 1.18-6.51), respectively, compared with respondents who received a picture of a white male physician. Non-white and unemployed respondents, were less likely to prefer professional attire exclusively: OR, 0.28 (95% CI, 0.10-0.83) and OR, 0.28 (95% CI, 0.08-0.99), respectively. Associations were not significantly altered following adjustment for respondent’s picture series, race, and level of employment (Table 4).

Clinic setting did not affect the likelihood of selecting professional attire: wound care OR, 1.03 (95% CI, 0.72-1.47) and dermatologic surgery OR, 0.78 (95% CI, 0.59 to 1.02) compared with the medical dermatology setting (reference). However, after controlling for respondent race, we found that respondents in the clinic for dermatologic surgery still preferred, but were significantly less likely to choose, professional attire compared with respondents in the clinic for medical dermatology (OR, 0.74 [95% CI, 0.56-0.98]). Race did not affect preference in wound care respondents (OR, 0.99 [95% CI, 0.69-1.42]).

Across wound care, surgical, and medical dermatology clinics, respondents largely preferred professional attire. Unemployed respondents and nonwhites were less likely to prefer professional attire for all questions, while respondents who received a picture of black male or black female physician were more likely to prefer professional attire for all questions. Furthermore, surgical dermatology clinic respondents still preferred professional attire but were significantly less likely to prefer professional attire compared with medical dermatology respondents. Wound care respondents demonstrated preferences comparable to those of medical dermatology respondents.

Our finding of patients’ marked overall preference for professional attire is similar to findings in other attire studies.^5,6 Interestingly, in the study by Rehman et al,⁵ the study participants were less well educated than in the present study; most had received less than a high school education. In contrast, most of the respondents in our study were college graduates or beyond. Nonetheless, Rehman et al⁵ found a similar overwhelming patient preference for professional attire. This is consistent with our findings that respondent education was not associated with a particular attire preference. In contrast, we found that nonwhite respondents were significantly less likely to choose professional attire for all questions, while Rehman et al found that black respondents were more likely to prefer professional attire for all questions. Rehman et al⁵ conducted their study in a Veteran’s Affairs internal medicine clinic in South Carolina, and the respondents’ military background and/or differing geographic, cultural, and social influences may explain this discrepancy in attire preference. Similarly, in our study, geographic, cultural, and social influences may have contributed to perception of appropriate attire for black physicians.

In the dermatologic setting, Thomas et al⁷ also found that respondent age was not significantly associated with attire preference, and most of the respondents in their adult dermatology clinics (55%) preferred their dermatologist to wear a white coat. Likewise, Marauni et al⁹ found that their adult respondents favored the white coat and most highly ranked their confidence in physicians wearing the white coat, consistent with our findings that respondents were most confident in physicians wearing professional attire. Kanzler and Gorsulowsky⁸ studied a variety of physical characteristics and found that most of their respondents reported the white coat as a desirable physical characteristic. Other desirable characteristics included displaying a name badge and wearing dress shoes, while wearing sandals, jeans, and earrings (male physicians only) were all undesirable characteristics. Although patients seem to prefer their dermatologist to wear a white coat, 77% of dermatologists do not wear white coats when pictured on their practice website.¹⁰

Professional attire was favored in all 3 clinic settings. We found that respondents in the clinic for surgical dermatology were less likely to choose professional attire compared with respondents in the clinic for medical dermatology. Moreover, respondents’ preferences for surgical attire and professional attire were comparable for the question investigating their preferences for their dermatologist’s attire in an emergency. Respondents even preferred the physician to be wearing professional attire “to cut out a skin cancer,” though the proportion of respondents who chose surgical attire for this question was notably higher than that of nearly all other questions. This suggests in surgical and emergency dermatologic settings, patients prefer either professional or surgical attire. Perhaps patients associate physicians wearing scrubs with emergency situations based on their own observations in emergency departments or what they have seen on television.

Contrary to our hypothesis, respondents preferred professional attire over scrubs in the surgical clinic. It is possible that patients do not view cutaneous surgery as they do general surgery, at least pertaining to the need for scrubs (and possible sterility). In fact, this perception may be accurate: in a randomized clinical trial, it was found that “newly cleaned physician uniforms and infrequently washed white coats have similar rates of bacterial contamination after an 8-hour workday.”^11(p177)

Another unexpected finding was that respondents in the wound care clinic exhibited attire preferences more similar to medical dermatology respondents than to surgical dermatology respondents, even though wound care patients often undergo surgical procedures such as debridement, grafting, and dressing changes. In regard to attire, patients seemed to view their wound care dermatologist as more similar to a medical dermatologist than to a surgical dermatologist.

Notably, a common sentiment that respondents expressed to the study investigators was that they did not judge their physicians by their attire, but rather by their knowledge and skill. This may be the case consciously, though it is possible that patients’ perceptions of their physicians’ knowledge and skill is influenced by the physicians’ appearance: we found that respondents consistently believed that the physician wearing professional attire would be more knowledgeable and competent that the physician wearing the other types of attire, and would have superior diagnostic and treatment abilities.

Our study is limited in that survey questions modified or added to make the survey relevant to dermatology were not validated. More surveys were distributed to medical dermatology patients than to surgical and wound care dermatology patients, though the data was sufficiently robust to allow for our analysis. While multiple investigators distributed the surveys, survey instructions were simple and clearly printed on the front page of each survey, which allowed for respondents to complete the survey without asking the investigators any questions. By limiting the survey to English language only, there was some selection bias, as monolingual Spanish-speaking patients were excluded. This contributed to the majority of the respondents being white and college educated, which may limit the generalizability of the study.

With any survey study, there is a possibility of response bias, such as social desirability bias possibly resulting from the investigator’s attire or the respondent’s actual physician’s attire, which may influence the respondent’s choices, despite the fact that it was emphasized to each respondent that the survey was completely anonymous.

Our study was conducted in 3 clinics at 1 university academic medical center in Miami, and it is possible that patients’ attire preferences vary geographically, culturally, and socioeconomically.⁶ Future studies could enroll more respondents of varied ethnic, education and socioeconomic strata in multiple medical centers to more broadly evaluate dermatology patients’ attire preference. Moreover, other attire permutations could be assessed (eg, polo shirt and slacks vs white coat with scrubs vs white coat with shirt and tie vs white coat with shirt without a tie). Finally, future investigations could control for whether the respondent is a new or prior patient and whether the respondent completed the survey prior to, or after seeing the dermatologist.

In sum, we found that patient and physician characteristics as well as clinic setting all affect patient preference in dermatologist attire. Nonetheless, professional attire is the predominantly preferred option for dermatologists in the medical, surgical, and wound care setting. Future research is needed to define regional, geographic, social, and cultural perceptions of professional attire in other populations.

Accepted for Publication: March 29, 2016.

Corresponding Author: Robert S. Kirsner, MD, PhD, Department of Dermatology and Cutaneous Surgery, University of Miami Miller School of Medicine, 1600 NW 10th Ave, Rosenstiel Medical Science Bldg, Room 2023A, Miami, FL 33136 (rkirsner@miami.edu).

Published Online: June 1, 2016. doi:10.1001/jamadermatol.2016.1186.

Author Contributions: Drs Fox and Kirsner had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Fox, Kirsner.

Acquisition, analysis, or interpretation of data: Fox, Prado, Baquerizo Nole, Macquahe, Herskovitz, Rosa, Akhtar, Aldahan, Shah, Mlacker, Cardenas.

Drafting of the manuscript: Fox, Prado, Akhtar, Aldahan, Mlacker.

Critical revision of the manuscript for important intellectual content: Fox, Prado, Baquerizo Nole, Macquahe, Herskovitz, Rosa, Shah, Cardenas, Kirsner.

Statistical analysis: Fox, Prado, Baquerizo Nole, Macquahe, Akhtar, Cardenas.

Administrative, technical, or material support: Aldahan, Mlacker.

Study supervision: Kirsner.

Conflict of Interest Disclosures: None reported.

Additional Contributions: We are indebted to Carol B. Kittles, BA, for her regulatory work with the institutional review board, Aliette Espinosa, CCRP, our clinical research coordinator, and Gabrielle Benesh, BS, for her assistance in distributing surveys. They received no compensation for their contributions beyond that received in the normal course of their employment.