Does physician attire influence patient satisfaction in an outpatient obstetrics and gynecology setting?

Article Outline

• Abstract
• Materials and Methods
• Results
• Comment
• Acknowledgments
• Appendix
• References
• Copyright

Objective

The purpose of our study was to determine whether physician attire played a role in patient satisfaction after a new obstetrician/gynecologist encounter.

Study design

Over a 3-month period, 20 physicians were randomly assigned to dress in business attire, casual clothing, or scrub suit each week. One thousand one hundred sixteen patients who had an office visit with a new obstetrician/gynecologist for at least 10 minutes completed a satisfaction survey, which assessed patient comfort as well as perception of the competency and professionalism of the physician. Patients were blinded to the physician attire manipulation.

Results

There was no difference in the mean overall satisfaction score among the 3 physician attire groups. No differences existed in satisfaction scores when analyzed by individual survey item or by demographic factors, after controlling for attire.

Conclusion

Patients are equally satisfied with physicians who dress in business attire, casual clothing, or scrub suit.

Key words: patient satisfaction, physician attire, physician-patient relations

The physician-to-patient relationship has been characterized by a delicate balance between compassion and objectivity, leading to the description of the physician as the “intimate stranger.”¹ Nowhere is this term more applicable than in the field of obstetrics and gynecology, where the physician is allowed to share in some of the most personal and memorable events in the life of a woman. Paramount to a successful physician–patient interaction is the patient’s confidence in her physician’s competency and professionalism. The role of physician attire in formulating these initial impressions has been widely debated.² The public image of appropriate physician attire has varied, depending on time, cultural expectations, and generational preferences. Hippocrates recommended that physicians should “dress decorous and simple, not over-elaborated, but aiming rather at good repute, and adapted for contemplation, introspection and walking.”³ Although some physicians prefer the more traditional white coat attire to convey an air of cleanliness, professionalism, and authority, others choose a more informal attire in the hope of breaking down barriers, improving communications, and creating a more equal physician–patient relationship.

Surveys on patient preferences for physician attire have been conducted in both the hospital and office settings, involving many different medical specialties. In most studies, patients expressed preferences for specific physician items, such as white coat and visible stethoscope, as well as shirt and tie for men, and dress or skirt for women.⁴, ⁵, ⁶, ⁷, ⁸, ⁹ However, few studies have focused exclusively on patients in obstetrics and gynecology. In 1 such study, Cha et al¹⁰ distributed a questionnaire that asked patients in an obstetrics and gynecology clinic about their preferences for the attire of resident physicians, and displayed pictures of male and female residents in various styles of dress. Though more than 60% responded that physician attire did not influence their perceptions, responses to the pictures showed a significant trend toward greater comfort and confidence in the more formally attired physicians. This investigation, as with the majority of other attire studies, surveyed the patient’s idealized public image of a physician. Although this approach may assess preconceived preferences of physician appearance, it may not reflect the actual patient satisfaction levels with physicians in various attires.

The purpose of our study was to determine whether, in a new office encounter with an obstetrician or gynecologist, physician attire played a role in patient comfort, as well as the perception of physician competency and professionalism. We hypothesized that patient satisfaction would be unrelated to physician attire.

Back to Article Outline

Materials and Methods 

This observational trial was judged exempt by the Institutional Review Committee, as it was a patient satisfaction survey. The study involved outpatients seen by 20 full-time faculty members of the department of obstetrics and gynecology at Cooper University Hospital in Camden, NJ, from September through November, 2005. Nine of the 20 physicians were from the general division of obstetrics and gynecology, 6 from the division of maternal-fetal medicine, 3 from the division of female pelvic medicine and reconstructive surgery, and 2 from the division of gynecologic oncology.

Over the 3-month period, physicians were assigned to wear business attire, casual clothing, or scrub suit on a weekly basis. The 3 modes of attire described in Table 1 were clearly delineated, and conformed to the institution’s dress code. A buttoned white coat was required for business attire, prohibited in scrub attire, and though discouraged, was permitted (unbuttoned only) for casual dress. The schedule of attire for each physician was determined by the principal investigator, who placed 3 different colored sets of 4 identical-size plastic disks, each color representing 1 of the dress styles, into an opaque container. He then sequentially selected without replacement each of the 12 disks and recorded the attire sequence for each physician.

TABLE 1. Physician attire requirements

A copy of the randomized 12-week schedule was sent to each physician. Subsequently, the physicians were reminded each weekend by telephone and e-mail of their assigned dress for the upcoming week. If a physician was scheduled to be away for a week during the study period, the assigned attire sequence was maintained and the study was extended by 1 week for that individual so that every physician dressed in each of the 3 attires for a total of 4 weeks.

The front office staff at each of 12 outpatient sites maintained an attire log sheet to monitor the physicians’ adherence to the clothing assignments at each office session. The logs were faxed on a weekly basis to the principal investigator, who spoke with any physician who did not dress according to the assigned schedule.

Eligible patients for the study included women with an office visit of at least 10 minutes with a new obstetrician or gynecologist, and no previous participation in the study. Immediately after the physician encounter, the office staff asked eligible patients to participate in an anonymous survey. The 1-page questionnaire asked for the patient’s age, ethnicity, pregnancy status, and name of the physician. The patient was asked to confirm that the office visit lasted at least 10 minutes, that it was her first time seeing the physician, and that it was her first time completing the questionnaire. The patient satisfaction rating scale included 10 affirmative statements about the physician’s competence and professionalism, the patient’s sense of comfort and confidence in her physician, and whether she would return to or recommend the doctor to others (see Appendix). All statements were written at the 5th to 6th grade level as measured by the Flesch-Kincaid readability scale. Responses were marked on a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). There were no questions about, or references to, the physician’s clothing.

To assure patient anonymity, the date of the physician visit was not recorded on the questionnaire. As each physician’s attire was determined on a weekly basis, the surveys were color-coded by week so that the attire of each physician could be determined from a master schedule after study completion. At the conclusion of the study, participating physicians were also surveyed regarding their demographics and preferences for each of the 3 types of attire.

Statistical tests included 1-way analysis of variance (ANOVA) to compare means among the 3 modes of attire, and 2-way ANOVA to test the effects of patient characteristics and their interaction with attire. The χ² test was used to compare categorical variables. Cronbach’s alpha was calculated to estimate the internal consistency reliability of the questionnaire and product-moment correlations to estimate its validity. All statistical analyses were performed with either SPSS version 13.0 (SPSS Inc, Chicago, IL) or Stata (StataCorp LP, College Station, TX). Statistical significance was defined as a P value less than .05. Analysis was based on intent-to-treat.

Back to Article Outline

Results 

A total of 1116 patients completed the patient satisfaction survey over the 3-month period. No completed surveys of eligible patients were excluded. The demographics of the women and 20 participating physicians are shown in Table 2. The majority (64%) of respondents were white, with an average age of 37 years. Forty-one percent were pregnant at the time of their office visit. Among participating physicians, the average age was 42 years, with an equal number of men and women. Compliance with the assigned modes of attire was excellent. Of 670 separate office sessions held by the 20 physicians during the study period, there were only 11 instances of incorrect attire, for a compliance rate of 98.4%. No physicians dropped out of the study.

TABLE 2. Patient and physician demographics

Characteristic	Patients (n = 1116)	Physicians (n = 20)
Age (y)	37.3 ± 15.6	41.9 ± 1.7
Ethnicity
White	64%	65%
Black	20%	20%
Hispanic	9%	0%
Asian	4%	10%
Other	3%	5%
Pregnancy status
Pregnant	41%
Nonpregnant	59%
Physician gender
Female		50%
Male		50%

Results reported as percentage or mean ± SD of valid responses.

Overall patient satisfaction was measured by the mean Likert scale score of the first 8 affirmative statements of the survey, with a value of 5 representing the highest possible score. Among the 1116 respondents, patient satisfaction was very high, with an overall mean score of 4.8 and a SD of 0.4. The responses were analyzed to estimate the reliability and validity of the scale. Internal consistency reliability was extremely high, with a Cronbach’s alpha of 0.97. The correlation between the mean response to the first 8 questions and responses to each of the last 2 questions, which related to the patient’s willingness to return to or recommend the physician, was 0.87 for each.

Of the 1116 respondents, 375 were exposed to physicians in business attire, 373 to casual attire, and 368 to scrub suit attire. Patient age and pregnancy rates were similar among the 3 groups. There was an unequal distribution of ethnicities, (0.008 by χ²), with a greater percentage of Hispanic patients exposed to business attire and fewer white patients exposed to scrub attire. There was no statistically significant difference in overall satisfaction scores among the 3 groups (Table 3). In addition, there were no significant differences among the 3 groups with respect to any individual survey question or when grouped by patient comfort or perceived physician competency and professionalism.

TABLE 3. Patient satisfaction results

Variable	Business (n = 375)	Casual (n = 373)	Scrubs (n = 368)	P value⁎
Overall satisfaction score†	4.8±0.3	4.8±0.4	4.8±0.4	.80
Would return to physician	4.9±0.4	4.8±0.4	4.8±0.5	.85
Would recommend physician	4.8±0.4	4.8±0.5	4.9±0.5	.44

Data presented as the mean ± SD of valid responses.

An exploratory 2-way ANOVA was performed to investigate any potential influence of demographics on satisfaction scores, either independently or interacting with mode of attire. The analysis revealed no significant effect of patient age, physician gender, physician specialty, or study week on patient satisfaction scores. Although there were significant differences in mean satisfaction scores based on patient ethnicity, pregnancy status, physician, and outpatient office site, the analysis showed no effect of any of these factors on patient satisfaction scores in combination with different types of physician attire.

Among the 20 participating physicians, 8 preferred casual dress, 7 preferred business attire, and 5 preferred to wear scrub suits. There was no association between attire preference and physician age, race, gender, specialty, or years in practice.

Back to Article Outline

Comment 

Our study showed that women experiencing new encounters with obstetricians and gynecologists were equally satisfied with their physicians regardless of their attire. When analyzed in conjunction with physician attire, there were no differences in satisfaction for any subgroup of patient or physician characteristics.

These findings may at first glance seem to contradict earlier published reports.¹, ⁴, ⁵, ⁶, ⁷, ⁸, ⁹, ¹⁰ In virtually all previous studies in which patients were asked specifically about physician attire, preferences were expressed for items such as white coats, ties, dress pants, and dresses. However, these observations might be biased because of perceived physician stereotypes gleaned from television medical shows and physician advertisements. We were interested in how women, who were unaware that clothing was being evaluated, would respond to physicians in various attires. Therefore, our surveys deliberately avoided asking about physician attire.

The results of our study are consistent with 4 others in which patient satisfaction surveys were administered after encounters with physicians in different attires. Pronchik et al¹¹ randomized male attendings, residents, and students in an emergency department to wear a necktie or no necktie over a 6-week period. Patients completing a postencounter survey had similar ratings between the 2 groups with regard to medical care, interaction, and general physician appearance. In a study by Hennessy et al,¹² a single anesthesiologist visited inpatients in either formal or casual attire. Patient satisfaction with the anesthesiologist’s professionalism and approachability was not influenced by his attire. Ikusaka et al¹³ surveyed 611 outpatients at an urban Japanese hospital, with physicians wearing a white coat during the first week and no white coat during the second week. Overall satisfaction with the physician consultation was equivalent in the 2 groups. Finally, Baevsky et al¹⁴ assigned emergency medicine physicians and physician assistants to wear either formal (even days) or scrub suit attire (odd days), both with white coats. As in our study, patients were blinded to the nature of the trial, and physician attire was not directly mentioned. Though the patient response rate was only 45% and the physician noncompliance rate was approximately 20%, there was no difference in patient satisfaction by physician attire when analyzed by either intent-to-treat or by post-hoc analysis.

These 4 studies, along with our present study, highlight the importance of study design in assessing patient satisfaction. When patients were asked their opinions of physicians rather than apparel, physician approval scores did not differ by attire. Although patients may express a predilection for specific items of clothing or personal appearance, it is likely that other factors, such as medical knowledge, personal demeanor, and interpersonal skills play a much more important role in patient satisfaction.

Not only can the white coat and tie, part of the traditional male physician attire, have little influence on patient satisfaction, but it can be potentially deleterious as well. “White coat hypertension” is a long-recognized condition affecting 21% of individuals with borderline hypertension.¹⁵ Though it is unlikely that this entity is due solely to the white coat, more formal physician attire may lead to a more intimidating atmosphere, patient anxiety, and resulting hypertension. In addition, studies have suggested that the white coat and tie may harbor potentially virulent bacterial organisms. Wong et al¹⁶ cultured 100 physician white coats from the sleeve cuff, front pocket, and back. They found an average colony count of 23.8 colony forming units from the 300 samples taken, with the highest colony counts originating from the cuff and pockets. Although most of the isolates were skin flora, Staphylococcus aureus was isolated from 29% of white coats. Similarly, in another study, neckties from 20 of 42 health care providers harbored bacterial pathogens, compared with only 1 of 10 neckties from nonhealth care hospital personnel.¹⁷

The greatest strength of our study was the large sample size. With over 350 patients in each of the 3 groups, our statistical power to detect a 10% difference in overall satisfaction scores was greater than 0.95 at an alpha of 0.05. Our study also encompassed a mix of pregnant and nonpregnant women from a wide variety of different ages and ethnic groups. It should therefore be generalizable to women in various obstetrics and gynecology outpatient practices in our region. We only included women having a new physician encounter to avoid bias from earlier interactions. Blinding of patients to the nature of the study was ensured by not including any questions about physician attire on the survey, and by not having the physicians or office staff discuss attire with patients. The participating physicians were remarkably compliant during the study period, aided by the weekly reminders from the investigators.

Our study had a number of limitations. We used onsite handout evaluation forms to ensure higher completion rates and for cost containment. However, this may have led to an artificially inflated satisfaction score, as studies have shown that handout surveys tend to score higher than mailed surveys. Reasons include perceived lack of anonymity, less time for thoughtful physician evaluation, and immediate postencounter optimism of the treatment plans.¹⁸ Another limitation is that our results may not be generalizable outside of our geographic region, as patient attitudes may vary in other areas of the country. Finally, the overwhelmingly positive patient satisfaction responses may reflect a Hawthorne effect,¹⁹ in which physicians’ interpersonal conduct may have been improved because of their participation in this research trial. In fact, our results are similar to those in other comparably designed studies.¹¹, ¹⁴ The relative impact of a Hawthorne effect is difficult to quantify, as our institution did not distribute outpatient satisfaction surveys before this trial for comparison with our current study results.

Although patient satisfaction may be artificially elevated in this research setting, it does not diminish the important finding that physicians, even those on their best behavior, are seemingly valued for their medical and interpersonal skills, not their work attire. Given that medicine, like many other industries, is in large part a consumer-driven business, it underscores the importance of good physician-patient interaction to maintain customer satisfaction and loyalty. Our study results suggest that patient satisfaction is not affected by a physician’s decision to dress in business attire, casual clothing, or scrub suit.

Back to Article Outline

Acknowledgments 

The authors are indebted to all the participating Cooper faculty physicians and office staff for their cooperation during the study. We also thank Autumn Gill and Kimberly McCaffrey from Press Ganey Associates, Inc, for their assistance in constructing the patient satisfaction survey.

Back to Article Outline

Appendix 

Please blacken the circle below that best describes your level of agreement with the following statements about the doctor that treated you today:

Back to Article Outline

References 

1. Gooden BR, Smith MJ, Tattersall SJN, Stockler MR. Hospitalised patients’ views on doctors and white coats. Med J Aust. 2001;175:219–222
   • View In Article
2. Brandt LJ. On the value of an old dress code in the new millennium. Arch Intern Med. 2003;163:1277–1281
   • View In Article
   • MEDLINE
   • CrossRef
3. Bishop WJ. Notes on the history of medical costume. Ann Med Hist. 1934;193–218
   • View In Article
4. Menahem S, Shvartzman P. Is our appearance important to our patients?. Fam Pract. 1998;15:391–397
   • View In Article
   • MEDLINE
   • CrossRef
5. McKinstry B, Wang J. Putting on the style: what patients think of the way their doctor dresses. Br J Gen Pract. 1991;41:275–278
   • View In Article
6. Gjerdingen DK, Simpson DE, Titus SL. Patients’ and physicians’ attitudes regarding the physician’s professional appearance. Arch Intern Med. 1987;147:1209–1212
   • View In Article
   • MEDLINE
7. Keenum AJ, Wallace LS, Stevens ARB. Patients’ attitudes regarding physical characteristics of family practice physicians. South Med J. 2003;96:1190–1194
   • View In Article
   • MEDLINE
   • CrossRef
8. Kanzler MH, Gorsulowsky DC. Patients’ attitudes regarding physical characteristics of medical care providers in dermatologic practices. Arch Dermatol. 2002;138:463–466
   • View In Article
   • CrossRef
9. Dunn JJ, Lee TH, Percelay JM, Fitz JG, Goldman L. Patient and house officer attitudes on physician attire and etiquette. JAMA. 1987;257:65–68
   • View In Article
   • MEDLINE
10. Cha A, Hecht BR, Nelson K, Hopkins MP. Resident physician attire: does it make a difference to our patients?. Am J Obstet Gynecol. 2004;190;:1482–1486
    • View In Article
    • CrossRef
11. Pronchik DJ, Sexton JD, Melanson SW, Patterson JW, Heller MB. Does wearing a necktie influence patient perceptions of emergency department care?. J Emerg Med. 1998;16:541–543
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (102 KB)
    • CrossRef
12. Hennessy N, Harrison DA, Aitkenhead AR. The effect of the anaesthetist’s attire on patient attitudes. Anaesthesia. 1993;48:219–222
    • View In Article
    • MEDLINE
    • CrossRef
13. Ikusaka M, Kamegai M, Sunaga T, et al. Patients’ attitude toward consultations by a physician without a white coat in Japan. Intern Med. 1999;38:533–536
    • View In Article
    • MEDLINE
    • CrossRef
14. Baevsky RH, Fisher AL, Smithline HA, Salzberg MR. The influence of physician attire on patient satisfaction (letter). Acad Emerg Med. 1998;5:82–84
    • View In Article
    • MEDLINE
    • CrossRef
15. Pickering TG, James GD, Boddie C, Harshfield GA, Blank S, Laragh JH. How common is white coat hypertension?. JAMA. 1988;259:225–228
    • View In Article
    • MEDLINE
16. Wong D, Nye K, Hollis P. Microbial flora on doctors’ white coats. BMJ. 1991;303:1602–1604
    • View In Article
17. Nurkin S, Urban C, Mangini E, et al. Is the clinicians’ necktie a potential fomite for hospital acquired infections? [astract]. American Society for Microbiology; 2004;May
    • View In Article
18. Gribble RK, Haupt C. Quantitative and qualitative differences between handout and mailed patient satisfaction surveys. Med Care. 2005;43:276–281
    • View In Article
    • MEDLINE
    • CrossRef
19. Roethlisberger FJ, Dickenson WJ. Management and the worker: an account of a research program conducted by the Western Electric Company, Hawthorne works, Chicago. Cambridge, MA: Harvard University Press; 1939;
    • View In Article