Patient Perceptions and Preferences when Choosing an Orthopaedic Surgeon

Abstract

Purpose

Information regarding patient preferences is important to develop more diversity in healthcare providers. To our knowledge, no information exists regarding how patients choose their orthopaedic surgeon. The purpose of this study is to determine which demographic factors, if any, affect patient preferences when choosing an orthopaedic surgeon.

Methods

Five hundred new patients presenting to a large, urban, academic orthopaedic clinic from May 2011 to May 2013 were prospectively asked to participate in this study. Patients were asked to complete a survey designed with the help of the Division of Population Health that focused on demographic, professional and physical attributes of theoretical surgeons. Specifically, patient preference of surgeon age, gender, race, religion, importance of education prestige, training program prestige and number of medical publications were evaluated. Patients were then stratified by age, gender, race, religion, educational level and income level to assess whether their own demographics were related to their preferences. The data was then analyzed to determine whether correlations existed between patient preferences and their own demographics.

Results

Five hundred patients agreed to participate in the study. There were 195 (39.0%) males and 281 (56.2%) females with an average age of 40.8 years (SD=20.5), 24 patients (4.8%) did not respond to the question. Two hundred and twelve (42.4%) patients were Caucasian, 116 (23.2%) were Hispanic, 53 (10.6%) were African American, 44 (8.8%) were Asian, 32 (6.4%) were listed as other and 43 (8.6%) did not answer. 78.0% of patients had no preference for their surgeon's gender, but for those who did, both men and woman preferred male surgeons (weak positive correlation, not statistically significant, r=0.096, p=0.373). The majority of patients (84.8%) had no preference for the race of their surgeon, but those that had a preference tended to prefer surgeons of their own ethnicity (p<0.001). With increasing patient education level, medical school, residency and fellowship training prestige had more importance as a selection criterion. Increasing patient education level also demonstrated a corresponding importance given to physician education and training as categorized by the perception of residency training program prestige (p=0.04). A majority of patients (84.0%) had no preference for their surgeon's religion, but for those who did there was a strong correlation (r=0.65), between the patients' own religion and that of the physician (p<0.001). There was universal agreement in perception that neither physician age nor years in practice made any difference as selection criteria when choosing an orthopaedic surgeon (p>0.05). Finally patient income level had no effect on specific criteria when choosing a surgeon.

Conclusion

The vast majority of patients surveyed had no preference in age, gender, race, or religion of their potential surgeon. However, patients who had preferences in these categories tended to choose surgeons of the same age, race and religion. These findings neither support or refute the need for diverse health care providers in the field of orthopaedics.

Introduction

Diversity within Orthopaedic programs has recently become an increasingly focal concern. According to Gebhardt et al., programs all over the United States consist mainly of Caucasian men¹. In some cases, this may impair aspects of the quality of orthopaedic care and hinder the development of positive physician-patient relationships. It has been shown that physicians from different backgrounds enrich other physicians' understanding of patients whose cultures are different from their own, and therefore improve their ability to successfully serve heterogeneous populations².

The purpose of this investigation was to identify preferred qualities of an orthopaedic surgeon that patients seek when given an opportunity to select their orthopaedic surgeon. We explored the value placed by patients on physician characteristics: racial background, religious affiliation, spoken language, gender, age, socioeconomic status, and educational status.

Methods

Patients presenting to a large urban, academic medical center general orthopaedic clinic over a 24-month period were prospectively asked to participate in this study. They were given a survey designed with the help of the Division of Population Health from our university. Photographs of ethnically and racially diverse male and female orthopaedic surgeons within our department were used within the survey. With each surgeon, all educational and practice information was provided as a caption beneath the picture. Patients both read the professional information about the provider and were able to visualize that specific provider. The relative age, gender and ethnic background of the provider was inferred from the picture. Within this survey, patient demographic data and information pertaining to educational status and income level was self-reported. For our analysis, participants were given questions regarding specific preferences for a theoretical orthopaedic surgeon. The survey was divided into two sections. The first section focused on the importance of educational prestige (medical school), training program prestige (residency, fellowship) and medical publication amount. The second section focused on how strongly or mildly participants felt about having a surgeon of a certain religious background, race or age group. Once decisions regarding these parameters were made, surgeon religion, race, age, and number of years in practice was specifically asked. Participants' responses were anonymous, and surveys were labeled only with numbers. Not all questions were answered on every survey. About 44.0% of the surveys were incomplete, however, questions that were not answered were included in the final analysis and categorized as “no answer noted”.

Descriptive statistics were used to report the outcomes of the survey. Subgroup analysis was performed to see if the patient's own demographic information was associated with their surgeon preference. Spearman's correlation tests and ANOVA were used to evaluate these relationships in this subgroup of patients. Statistical analysis was performed using Spearman's rho along with ANOVA to find correlations for preferences among subgroups of patients. Patients who did not report preferences for specific surgeon attributes were excluded from statistical analysis of those attributes. All statistical analyses were performed with SPSS version 19.0 software (SPSS, Inc., Chicago, Illinois) and significance level was set at p < 0.05.

Results

Five hundred patients agreed to be enrolled into this study. The demographics of the cohort are described in Table 1. Anticipating that patient preferences might vary as much within, as between, age, gender, and race groups, we decided to focus on relationships between seven different patient and surgeon attributes.

Table 1.

Preference of surgeon age based on patient age

Surgeon Age Preference (Years)	Mean value for Patient's Age (Years)	Number of Patients Participating in this Question	Standard Deviation
30-40	38.0	35	14.2
40-50	47.3	133	16.0
50-60	51.9	42	17.0
>60	55.3	3	15.3
No preference	43.6	261	17.3

Race

The racial distribution in our cohort was Caucasian (42.4%) followed by Hispanic (23.2%), African American (10.6%), Asian (8.8%) and other (6.4%). The majority of patients 424 (84.8%) had no preference for the race of their surgeon. However, a moderate positive correlation (r=0.627, p<0.001) was seen among patients who indicated a racial preference, preferring surgeons of their own race.

Gender

The gender distribution in our cohort was 281 (56.2%) females, 195 (39.0%) males, and 24 (4.8%) who did not answer. The majority of participants, 390 (78.0%), had no preference for the gender of their surgeon. Amongst those with a preference, 14.2% versus 4.0% preferred male or female surgeons, however this was not statistically significant (p=0.373).

Age

The average age in our cohort was 40.8 (SD=20.5). The majority of participants, 261 (52.0%), had no preference for the age of their surgeon. However, of those patients with an age preference, a trend was seen demonstrating preference to surgeons who were similar in age (Table 1).

Religion

The religious distribution in our cohort was Christian (48.6%) followed by Jewish (14.6%), Agnostic (5.6%),Atheist (4.4%), Muslim (3.2%) and other (3.0%); 20.8% did not answer. The majority of participants, 420 (84.0%), had no preference for the religious background of their surgeon. A strong, statistically significant, positive correlation (r=0.65, p<0.001) was seen for those indicating a religious preference, demonstrating preference of surgeons of their own religious background.

Educational Background

Participants were asked to rate on a scale of 1 to 10 (1 being not important at all and 10 being very important) the importance of a given item on choosing their surgeon. The educational background distribution of our cohort was less than high school (8.8%), high school (24.4%), technical school (11.8%), 4-year college (26.2%), and graduate/professional level schooling (20.2%). A positive trend demonstrated that as patient education level increases, more emphasis is placed on the prestige of a surgeon's medical school, residency, and fellowship (Table 2); however, no significant correlations were seen.

Table 2.

Ranking of medical school, residency and fellowship prestige categorized by patient's education level

Patient Education	Medical School Prestige (scale 1-10)	Residency Prestige (scale 1-10)	Fellowship Prestige (scale 1-10)
< High School	4.0	4.1	4.2
High School	4.48	4.8	4.7
Technical School	4.91	4.7	5.2
4-year College	4.86	5.4	4.9
Graduate School	5.34	5.6	5.6

Academic Production

We used number of publications as an estimate of academic production. Participants were asked to rate on a scale of 1 to 10 (1 being not important at all and ten being very important) how important the amount of research articles published by their theoretical surgeon was. The income level distribution of our cohort was less than 50K (45.0%), 50-100K (17.6%), 100-250K (9.0%), greater than 250K (4.2%) and the rest had no answer (24.2%). The trend in mean values of importance showed no significant correlation (Table 3).

Table 3.

Trend in mean values for importance of number of medical publications by treating physician

Patient Income	Mean Rating Value of Importance	Number of Patients Participating in this Question	Standard Deviation
<50K	3.8	215	3.4
50-100K	4.4	85	3.0
100-250K	3.8	45	2.9
>250K	4.6	21	2.7

Patient Education Level

Participants were asked if they had a preference for their theoretical surgeon's amount of years of experience, and if so, how many they would prefer. The largest group of participants, 155 (31.0%), preferred 11-20 years in practice, however, the next largest group, 110 (22.0%), had no preference for years of experience. No significant difference or correlation was seen with patient education.

Discussion

Diversity within the orthopaedic surgeon population has the potential to improve patient care. Studies have shown that patients who identify with their physicians have quicker recovery rates and better adherence^3-5. This emphasizes a pertinent question: what can we do to make patients more comfortable with their orthopaedic surgeons?

Although several investigations analyzing patients' preferences for general practitioners have been done^6-8, few studies have considered patients' preferences for surgeons¹¹, and to the best of our knowledge this topic had not been explored in orthopaedic surgery.

We were able to draw several conclusions on patient preferences within our diverse patient population. With regards to patient and physician race, our data showed a majority of patients do not particularly have a preference for their surgeons' race. However, patients demonstrating a preference for race tend to choose surgeons of a similar ethnic background. According to Simon et al., this is attributed to a sense of comfort patients feel when communicating with someone of a similar racial background⁹. Patients' experiences with illness and treatment are unique and often influenced by their beliefs, culture, and family; thus, a physician with similar values or those that are sensitive to such values can better care for and treat such patients.

When analyzing patient and surgeon gender, our data showed that most patients do not place emphasis on the gender of their surgeon. Yet, amongst those patients with a preference, both male and female patients preferred male surgeons. Communication is certainly important in determining patient outcomes. Weisman et al. discuss several methods in which physician gender may affect the patient-physician relationship¹⁰. First, gender differences in nature and attitude may affect the physician-patient interaction: females tend to be more nurturing and sensitive while males are more reserved. Second, physician gender influences patient expectations: female physicians are expected to “sugar-coat” situations while males are expected to be direct. Finally, same-sex physician patient relationships are presumed to result in greater status equivalence¹⁰. However, no definitive conclusions proving these factors affect the patient-physician relationship have been reported. Nonetheless, studies suggesting female patients prefer female physicians have been published; the female patients in these studies were choosing physicians for physical exams or procedures of women-related health issues¹¹. This begs the following question: are patients more comfortable with physician genders stereotyped to certain fields? For example, orthopaedics has traditionally been a male dominated specialty, and the percentage of female orthopaedic surgeons is the smallest of any surgical subspecialty. This observation may lead to the impression that orthopaedic procedures are best suited to male surgeons¹². This could potentially explain why both female and male patients in our cohort who had a preference in surgeon gender choose male surgeons. However, more studies must be carried out to investigate this suggestion further.

Our next analysis focused on the relationship between patient and surgeon age. The data showed that most patients had no preference for the age of their surgeon. For those who displayed preference, a non-significant trend was seen demonstrating older patients prefer surgeons similar in age. Jung et al. also saw this trend in a study observing patient preferences for primary care physicians¹³. In a study looking at age preferences for emergency physicians, the majority of patients preferred physicians in the 30-40 age group; 41-50 was favored next; however the age of the patients were not recorded and therefore no correlation was tested. Overall, 79% of patients favored emergency medicine physicians between 30-50 years of age; 31.4% of our cohort favored orthopaedic surgeons in the 30-50 age group. Studies have regarded these findings to be due to patients' desire of having a physician with many years of experience who also remains up to date with the current knowledge on newer treatment methods^14,15.

When investigating the relationship between patient and surgeon religion, we found the majority of participants had no preference for the religious background of their surgeon. However, a strong positive correlation was seen for patients with a religious preference indicating patients preferred surgeons of their own religious background. No studies regarding this association have been published to our knowledge, however, studies that discuss end-of-life care do show patient preferences for physicians with a spiritual background of any kind¹⁶. We propose the positive correlation seen in our data further reflects a patient's desire for commonality; common religious background may render a physician more apt in understanding values and beliefs heavily influencing the type of care one requires.

A positive correlation was seen between increasing patient educational background and emphasis on prestige of surgeon medical school, residency, and fellowship, however this was not significant. We hypothesize that patients who have achieved higher levels of education place an increased value on the quality of training their surgeon has received, however this proposition is not supported by the literature. The relationship between patient income levels and number of surgeon publications showed no discernable trend of importance.

Finally surgeon experience as a predictor was explored. The majority of participants preferred surgeons with 11-20 years in practice; however, the next largest group had no preference for years of experience. No significant difference or correlation was seen with patient education. This demonstrates that regardless of education level, patients seek a physician with many years of experience to optimize their quality of care^14,15.

The limitations of this study include the survey design. The survey has not been utilized by any other study, and has not been validated. Although anonymity was protected, some patients may have answered questions thinking that their care could have been affected and thus biased the results.

In summary, our results neither support nor refute the need for diversity in the population of orthopaedic surgeons. However, evidence suggests that a strong physician-patient relationship yields superior patient outcomes. As such, further investigation identifying how patients identify their ideal orthopaedic surgeon is warranted. Although some studies have demonstrated patients select physicians based on race and gender, other studies have shown that physician capability, skill, and compassion are the most important deciding factors¹⁷. Despite some gains, orthopaedic training programs demonstrate one of the lowest female and minority resident percentages compared to other fields^9,18,19, resulting in potential barriers in care of diverse patient populations. This study makes no judgment regarding the quality of care received by the patients. While there are many reasons for diversification in all fields of medicine, patient perceptions about their orthopaedic physicians seems to be based upon the quality of education and experience within the field rather than age, sex and race of the provider.