Table 1. 
Image not available
Demographic Features of Responding Patients*
Table 2. 
Image not available
Provider Characteristics Deemed Significant, With Totals From Survey, by Sex of Provider*
Gjerdingen  DKSimpson  DETitus  SL Patients' and physicians' attitudes regarding the physician's professional appearance. Arch Intern Med. 1987;1471209- 1212Article
Molloy  JT Dress for Success.  New York, NY Warner Books1975;30- 50210- 212
Molloy  JT The Woman's Dress for Success Book.  New York, NY Warner Books1977;112- 114
Taylor  PG Does the way housestaff physicians dress influence the way parents initially perceive their competence? Pediatr Notes. 1985;91
April 2002

Patients' Attitudes Regarding Physical Characteristics of Medical Care Providers in Dermatologic Practices

Author Affiliations

From the Departments of Dermatology, Stanford University School of Medicine, Stanford, Calif (Drs Kanzler and Gorsulowsky), and University of California at San Francisco (Dr Gorsulowsky); and Division of Dermatology, Santa Clara Valley Medical Center, San Jose, Calif (Dr Kanzler).


Copyright 2002 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2002

Arch Dermatol. 2002;138(4):463-466. doi:10.1001/archderm.138.4.463

Objective  To assess the present attitudes of patients toward physicians' physical attributes.

Design  Written survey offered to all patients seen during a 1-week period.

Setting  Two outpatient dermatologic clinical practices (a county hospital and a private practice).

Participants  Of 315 patients offered the survey, 275 agreed to complete it.

Main Outcome Measures  Opinions regarding physicians' demographic characteristics and opinions regarding desirability of 19 and 18 appearance-related characteristics in male and female physicians, respectively.

Results  Analysis of the responses revealed 25 characteristics that were significantly desirable or undesirable (defined as being selected desirable or undesirable by at least 25% of respondents). Further analysis revealed that patients in a private practice setting typically had more polar opinions about providers' appearances than did patients from a large county hospital. Most patients had no preference with regard to the sex, age, or race of their medical care providers. Age and sex of the patient did not independently contribute significantly to patient preferences, as determined by cross-tabulation analysis. Clinic site (private practice vs county hospital clinic) alone was the sole or most important predictor of preferences in 13 of the 25 significant characteristics.

Conclusions  Several characteristics of providers' dress and grooming were important to patients. There seemed to be little attitudinal change from similar studies performed 2 decades ago. Cognizance of these preferences may facilitate better interactions between medical care providers and patients.

THE PRIMARY goals of the medical care provider are to prevent and treat disease. This is most commonly accomplished through direct interactions with patients. The information obtained from patients during this interaction must be accurate and complete if we are to provide optimal care. Nonverbal cues such as physical appearance might influence these short interactions, especially for patients who have never met the provider previously. Increasing patient comfort level during patient-provider encounters may facilitate the exchange of information and hence improve patient care.

The medical care provider's appearance has been found to be a key symbol that not only identifies the individual as a professional, but also defines certain characteristics of the provider.1 A carefully dressed provider might convey the image that he or she is meticulous and careful. Alternatively, an unkempt appearance might convey impressions of uncaring or disorganized behavior.1

Several studies from 2 decades ago addressed the subject of physicians' appearance. Molloy2 found that the most desired outfit for male physicians was a white coat with slacks, shirt, and tie. Female physicians' desired characteristics were a white jacket with a male-tailored shirt and dark wool skirt.3 Taylor4 found a strong association between physician dress and parents' initial perceptions of physician competence in a pediatric setting of a university-affiliated hospital. A large study by Gjerdingen et al1 confirmed the importance of the physician's appearance in physician-patient communication. Patients gave positive responses to physicians with traditional appearances. Desirable items in the presentation of male physicians included dress shoes, groomed facial hair, shirt and tie, and dress pants. Traditionally feminine items found desirable for female physicians included nylon stockings, lipstick, makeup, skirts, and blouses. A white coat and name tag identifying the individual as a physician were very desirable in both sexes.

The public image of how physicians should look may have changed in recent years because of the informal appearance of physicians on popular television programs such as ER. The image of physicians portrayed on ER is far different from that of past television physicians such as Marcus Welby, MD, and Dr Kildare.

The purpose of the present study was to see to what extent, if any, patients' views have changed over the past generation with regard to the importance of physical appearance of physicians. We chose to use the term "medical care provider" instead of "physician" in our survey since other professionals such as physician assistants and nurse practitioners now render many services. In addition, we queried how various demographic characteristics such as race, education, and income level affected patients' responses. We hoped that information gleaned from this study would help future providers to better meet patients' needs.


Questionnaires were distributed to all patients seen in 2 dermatologic practice settings during 1 calendar week of February 2000: (1) a part-time private practice office in Fremont, Calif; and (2) an outpatient clinic in a large county teaching hospital in San Jose, Calif (Santa Clara Valley Medical Center). Both of these practices were primarily medical dermatology practices and did not specialize in cosmetic procedures. This type of practice was chosen to minimize the effect of attitudes that might be present in patients with greater-than-average cosmetic concerns.

In addition, these 2 settings were compared because, while both practices were staffed by the same physicians (thus lessening selection bias with regard to the caregivers), one group of patients (private practice) in theory "selected" its caregiver, while the other group (county hospital) typically had little choice in selecting the provider. Questionnaires were available in English, Spanish, and Vietnamese (the 3 major languages spoken in the geographic areas).

The questionnaires asked for demographic data of the patients as well as their opinions about various general visual characteristics of providers in the office/outpatient setting. Requested demographic traits of the respondents included their age, race, sex, level of education, and type of insurance coverage.

Patients were asked which characteristics they preferred in a medical care provider with regard to sex (male, female, or no preference), age group (20-40, 40-60, >60 years, or no preference), and racial background (white, black, Hispanic, Asian, or no preference). In addition, several physical characteristics (eg, items of dress and grooming) were evaluated as desirable, neutral, or undesirable. These characteristics were selected because of their having been used in previous research. For both sexes, the items were name tag, white coat, sport coat/blazer, dress pants, blue jeans, surgical scrubs, dress shoes, tennis shoes, clogs, sandals, traditional hairstyle, and nontraditional hairstyle. For male providers, patients were queried about open shirt, long hair/ponytail, mustache, beard, cologne, and earrings. Finally, for female providers, items were perfume, nylon stockings, lipstick, and makeup.


Data were initially analyzed for all patient responses. Cross-tabulation analyses were performed between patient age and preference for provider of a certain age, patient race and preference for provider race, and patient sex and preference for provider sex. Frequency analyses were performed for all items of dress or grooming characteristics on the questionnaire.

There exists no predetermined "level of significance" for frequency analyses. Intuitively, a physical characteristic identified by most respondents as "neutral" would not be deemed a significant (desirable or undesirable) physical characteristic of medical care providers. To determine at which point a physical characteristic would be considered significant, 20 physicians at one of the 2 clinics (Santa Clara Valley Medical Center) were asked the following question: "What is the percent of respondents who would have to view a characteristic as desirable or undesirable before you would feel obliged to display or refrain from displaying that characteristic?" All responses were either 1 (25%) in 4 patients, or 1 (33%) in 3 patients. Therefore, these percentages were arbitrarily chosen to determine which items of dress or grooming were deemed significant.

Physical characteristics found to be significantly important by these criteria were then further analyzed to determine if differences in opinion regarding their importance were present between various subsets of the patient population. The following bivariate demographic groups were analyzed: college/graduate school vs elementary/high school education; white vs other race; private clinic vs county hospital clinic setting; age younger vs older than 40 years; and insurance coverage via private vs indigent programs. The last grouping was used to approximate patients' level of income.

A condition of qualifying for public assistance insurance at the time of this study was a maximum yearly income of $8352 for an individual, or $17 052 for a family of 4. Private insurance, when present, was provided almost exclusively through employers, and the average annual salary for workers in the immediate geographic area (colloquially known as Silicon Valley) at the time of this survey was $53 700.

Cross-tabulation analysis was performed between each dependent variable (eg, blue jeans) and each demographic feature (eg, level of education). Those demographic features determined by cross-tabulation analysis to contribute individually to patient preferences were further subjected to stepwise linear regression analysis against the various dependent factors to determine which demographic features were most important in predicting patient preferences. All analyses were performed using the SPSS statistical program (SPSS Inc, Chicago, Ill).


A total of 275 of the 315 patients seen during the week agreed to complete questionnaires: 84 from the private practice setting (25 refusals; 77% response rate), and 191 from the county hospital clinic (16 refusals; 92% response rate). Reasons for refusal were not identified. Demographic data of the patients are presented in Table 1. Significant differences between the 2 patient populations (private practice vs county hospital clinic) were noted for class of insurance (Pearson χ21 = 81.6; P<.001), level of education (Pearson χ21 = 9.0; P = .002), and race (Pearson χ23 = 25.7; P<.001). No significant differences were noted between the groups with regard to age (Pearson χ22 = 1.5; P = .48) or sex (Pearson χ21 = 0.5; P = .48).

The questionnaire responses revealed that most patients had no preference with regard to the sex, age, or race of their medical care provider. Fifty-seven percent of all respondents had no preference regarding the age of their provider. Of the patients who did show a preference for a specific age in a provider, approximately half wanted a provider in his or her own age category, while half wanted a provider 40 to 60 years old.

Sixty-six percent of both male and female patients had no preference regarding the sex of their provider. Twenty-nine percent of male patients preferred a male provider, while 31% of female patients preferred a female provider.

Seventy-four percent of white and 80% of Asian patients had no preference in the race of their provider, while 58% of Hispanics voiced no preference in this characteristic. Those patients with a preference wanted providers of their own race. An interesting finding was that while the number of Hispanic patients requesting Hispanic providers seemed high, this preference disappeared when these patients were analyzed according to the language in which their questionnaire was printed. Only 17% of English-speaking Hispanic patients requested a Hispanic provider, while 65% of Spanish-speaking Hispanic patients requested someone of that group. Therefore, the ability to speak the same language seemed more important to patients than the race of the provider.

Twenty-five physical characteristics of male and female medical care providers were deemed either desirable or undesirable by at least 1 (25%) in 4 respondents (Table 2). Many characteristics actually surpassed 1 (33%) in 3 levels, and these highly desirable or undesirable characteristics are also noted.

Stepwise linear regression analysis was performed using each of the 25 desirable/undesirable characteristics as dependent variables to determine whether patients' demographic characteristics predicted preferences for providers' appearance. Age and sex of the patient did not independently contribute significantly to patient preferences using cross-tabulation analysis. Therefore, only clinic site, level of education, race, and type of insurance were chosen as independent variables. Clinic site (private practice vs county hospital clinic) alone was the sole or most important predictor of preferences in 13 of the 25 significant physical characteristics. Level of education was a significant predictor for 4 dependent variables, level of income (insurance class) for 4 dependent variables, and race for 3 dependent variables.


We have confirmed findings from studies conducted in other settings 2 decades ago.14 Despite a general trend in Western society toward more casual attire in public, very little change has occurred in patients' preferences regarding the preferred attire of their medical care providers. Patients in the current settings still find very traditional attire (eg, name badge, white coat, dress shoes, etc) preferable for their providers, while finding casual attire (eg, blue jeans, clogs, sandals, etc) undesirable.

One might expect younger patients to show less preference about the appearance of their medical care providers. However, interestingly, the age or sex of the respondents did not affect this preference. Responses from different racial groups were also extremely uniform with regard to this characteristic.

The most significant factor determining patients' preferences was the setting at which they were seen. However, the differences were merely in degree of preference. No factor was found to be desirable at one clinical setting and undesirable at the other. Patients from the private practice setting found desirable characteristics more desirable than their counterparts at the county hospital clinic, and undesirable characteristics more undesirable. Significant differences were noted between the 2 populations with regard to race, level of education, and income (insurance classification). Stepwise linear regression analysis revealed that the clinic setting was more important in determining views than any other item analyzed individually. Therefore, the interaction of individual demographic factors was more important than any 1 factor itself.

Most patients had no preference with regard to the sex, age, or race of the provider. Not surprisingly, the minority of patients with a preference preferred a medical care provider with demographic features similar to their own. The ability to speak the same language seemed more important to patients than the actual race of a provider.

This study was carried out in an outpatient medical specialty setting (dermatology). However, both clinical settings catered to patients referred by primary care providers for medical dermatologic problems rather than patients seeking cosmetic consultation. While our findings are similar to findings from a similar study carried out in a primary care setting,1 these results may not be applicable to all clinical settings. For example, preference for female providers might be found in a gynecology practice, and surgical scrubs may not be considered undesirable in a surgical clinic.

Based on the results of this study, to best serve their patients, medical care providers in similar settings should wear a name badge, white coat, and dress shoes and should avoid wearing blue jeans, clogs, and sandals while on duty. Male providers should also wear dress pants and avoid open shirts, long hair/ponytails, and earrings. These characteristics were preferred or frowned on, respectively, by more than 1 in 3 patients, and thus may be considered minimum requirements in standard dress code policies.

Other suggested characteristics found desirable by at least 1 in 4 patients include traditional hairstyles for both sexes and skirts/dresses or dress pants for female providers. Tennis shoes were found to be undesirable for both sexes by at least 1 in 4 patients, as were surgical scrubs, cologne, and nontraditional hairstyles for male providers.

The results of this study show that some physical characteristics of medical care providers are important to patients. Respecting patients' preferences with regard to physical appearance might help put the patient at ease during the history and physical examination.

Back to top
Article Information

Accepted for publication September 6, 2001.

Corresponding author: Matthew H. Kanzler, MD, Division of Dermatology, Santa Clara Valley Medical Center, 751 S Bascom Ave, San Jose, CA 95128 (e-mail:

Gjerdingen  DKSimpson  DETitus  SL Patients' and physicians' attitudes regarding the physician's professional appearance. Arch Intern Med. 1987;1471209- 1212Article
Molloy  JT Dress for Success.  New York, NY Warner Books1975;30- 50210- 212
Molloy  JT The Woman's Dress for Success Book.  New York, NY Warner Books1977;112- 114
Taylor  PG Does the way housestaff physicians dress influence the way parents initially perceive their competence? Pediatr Notes. 1985;91