Patient choice of provider type in the emergency department: perceptions and factors relating to accommodation of requests for care providers
- 1Department of Emergency Medicine, University of Michigan, Ann Arbor, Michigan, USA
- 2Department of Emergency Medicine, University of Rochester Medical Center, Rochester, New York, USA
- 3Department of Biostatistics, University of Rochester Medical Center, Rochester, New York, USA
- 4Department of Neurosurgery, University of Pennsylvania, Philadelphia, Pennsylvania, USA
- Correspondence to Aasim I Padela, Robert Wood Johnson Foundation Clinical Scholars Program, 6312 Med Sci Bldg I, 1150 W. Medical Center Drive, Ann Arbor, MI 48109-5604, USA;
Contributors AIP was involved in conception of the study, survey design, database construction, data analysis and interpretation and manuscript preparation. SMS was involved in conception of the study, survey design, data acquisition, manuscript revision and provided funding. HH provided technical assistance, performed data analyses and critically reviewed the manuscript. ZA was involved with the conception of the study, survey design and database construction. TMR was involved with the conception of the study, survey design, data analysis and interpretation and critical review of the manuscript.
- Accepted 28 July 2009
Background Patient satisfaction is related to the perception of care. Some patients prefer, and are more satisfied with, providers of the same gender, race or religious faith. This study examined emergency medical provider attitudes towards, as well as patient and provider characteristics that are associated with, accommodating such requests.
Methods A survey administered to a convenience sample of participants at the 2007 American College of Emergency Physicians Scientific Assembly. The nine-question survey ascertained Likert-type responses to the likelihood of accommodating patient requests for specific provider types. Statistical analyses used Wilcoxon rank-sum, Wilcoxon signed-rank and Cochran's Q tests.
Results The 176 respondents were predominately white (83%) and male (74%), with a mean age of 42 y. Nearly a third of providers felt that patients perceive better care from providers of shared demographics with racial matching perceived as more important than gender or religion (p=0.02). Female providers supported patient requests for same gender providers more so than males (p<0.01). Provider race, practice location, type and duration did not significantly affect the level of accommodation. When requesting like providers, female patients had higher accommodation scores than male patients (p<0.001), non-whites than whites (p<0.05), with Muslim patients (male or female) most likely to be accommodated (p<0.01).
Conclusion Accommodating patient requests for providers of specific demographics within the emergency department may be related to provider characteristics. When patients ask for same gender providers, female providers are more likely to accommodate such a request than male providers. Female, non-white and Muslim patients may be more likely to have their requests honoured for matched providers.
- Patient choice
- cultural accommodation
- healthcare disparities
- gender concordance
- patient satisfaction
- clinical assessment
- environmental medicine
- emergency department management
- quality assurance
Delivering culturally sensitive high-quality health care to a diverse population presents many challenges. When the physician and patient are of differing backgrounds miscommunication due to language barriers, misunderstandings stemming from cultural variances and stereotyping by both the patient and physician may present obstacles to equitable care. Thus healthcare disparities seen within minority populations may, in part, result from the effects of racial, ethnic and cultural mismatching upon the clinical encounter.1 Increasing workforce diversity has been cited as a means to reduce disparities since physicians and patients that share common background and culture may engender greater trust and develop more stable therapeutic alliances.2–4
Within this era of diversity, minorities seek out and may prefer familiar providers due to perceived discrimination, cultural familiarity and religious ethics.5–7 It follows then that some patients view clinical encounters with like providers more positively.8 9 Illustratively, a nationwide survey found black patients with black physicians more likely to rate their physicians as excellent, and Hispanics patients with Hispanic physicians more likely to be satisfied with care.10 Similarly in an emergency department (ED) study, female patients were more satisfied with same-gender patient-physician pairings and felt female providers showed more concern and were more trustworthy.11
Given that the ED is an entry point into the healthcare system for immigrant populations and is disproportionately used by racial and ethnic minorities for many aspects of health care, patient requests for like providers likely occur.12 However, the affect of gender, race, religious and ethnic concordance upon patient satisfaction, quality of care and as a mediator for disparities in the emergency setting has not been studied. Our study assessed ED provider attitudes toward patient requests for same-gender, race or religion providers. We also investigated how patient and provider demographics influence the accommodation of such requests. We hypothesised that providers of minority backgrounds would be more likely to accommodate patient requests, and patients from minority backgrounds would have higher accommodation scores, with same-gender provider requests having the greatest acceptance. As patient satisfaction may be related to the perception of providers and departments accommodating patient needs, focussing on such requests provides insight into the larger ‘culture’ of patient-centered adaptations within the ED. The factors that play a role in accommodating such an ‘extreme’ request may play a role in smaller, yet still significant, practice modifications.
A nine-question survey comprised of seven questions investigating perceptions of patient requests for same gender, race or religion providers and two case scenarios were administered to all participants. The vignettes described a patient presenting to the ED with a non-emergent complaint (upper respiratory infection or abdominal pain with stable vital signs) requesting to see a provider of the same gender, race or religion. The patient race and religion and/or gender descriptor was varied in the vignette such that each respondent received the entire range of possible demographics. Respondents were asked whether they would try to accommodate the request on a 5-point Likert type scale from which was derived the mean accommodation score based on each patient demographic (1=never accommodate to 5=always accommodate). Pilot testing was performed via an iterative process of feedback and survey administration to a group of physician assistants and nurse practitioners at an academic university ED. The survey was approved by the Research Subjects Review Board at the University of Rochester.
Study setting and population
The study was conducted at the 2007 American College of Emergency Physicians Scientific Assembly with a convenience sample of emergency providers visiting a designated research booth.
Completed surveys were entered into a database by two of the authors and verified for accuracy independently. Data analysis was conducted using SAS version 9.1. Descriptive statistics were reported for continuous variables in terms of means and standard deviations, and were reported for categorical variables using frequencies and proportions. For comparisons of mean accommodation scores, the non-parametric Wilcoxon rank-sum test for independent measurements and Wilcoxon signed-rank test for correlated measurements were used due to lack of normality. The Cochran's Q-test was used to compare differences for three-matched sets of frequencies.
Characteristics of survey respondents
There were 176 respondents and their demographics are summarised in table 1. Respondents were primarily white, male and EM certified.
A significant difference was noted between the percentage of respondents who felt that patients perceived care delivered by a concordant provider to be better (24% for same gender vs 32% for same race and 30% for same religion; p=0.02). Nearly twice as many providers would always accommodate same-gender provider requests compared to race or religion requests (32% vs 16% and 17%, respectively; p<0.0001). Paradoxically, more providers stated they would never accommodate requests for gender concordant providers than for race or religion (38% vs 31% and 28%, respectively; p<0.0001) (table 2).
Provider factors related to accommodation of requests for a same-gender provider
Gender concordant provider requests by stable patients with abdominal pain were met with different levels of acceptance by providers. For both patient scenarios (one male and one female), only respondent gender, but not race, practice location, type and years in practice, affected accommodation scores significantly. Female providers would accommodate same-sex provider requests more so than their male counterparts (p<0.01 and p<0.0001, repectively) (table 3).
Patient factors influencing accommodation level when requesting a same gender, race or religion provider
A case scenario where patient demographics of race, gender and religion were varied with each patient requesting a provider of the same background as themselves yielded notable findings. Overall, patients of minority backgrounds had significantly greater accommodation scores than those from majority backgrounds. Female patients had greater accommodation scores than male patients (3.12 vs 3.02; p<0.0001) and non-whites than whites (3.08 vs 3.03; p<0.05). Furthermore, paired comparison across different patient demographics revealed that Muslim male patients had statistically higher scores than Black, Hispanic or Jewish male patients and Muslim females had the highest score across all demographics (table 4).
This exploratory study represents the first attempt to understand patient and provider characteristics that affect clinical accommodation of patient requests in the ED. By focussing on patient requests for accommodation through same-gender, race or religion providers, we start to define the ‘ED culture’ of patient-centered accommodation of requests. How do providers respond when they are asked to change their normative practice style and make compromises to meet patient needs? If a provider type is willing to accommodate a seemingly ‘extreme’ request by a patient—that is, to be taken care of by a provider of the patient's demographic—it speaks highly of a willingness to accommodate patients in smaller but significant ways. Since patient satisfaction is associated with concordant providers for certain patients, a willingness to accommodate such requests is important. Further, by studying which patient type is likely be accommodated, and by whom, may inform health disparities in minority care. Our provider-based survey yielded several interesting findings.
With regard to the care provided within patient-physician pairings of shared gender, race or religion, a minority of the surveyed healthcare providers felt that patients perceive it to be better. Patient level data yield a different view, as patients from minority backgrounds consistently seem to be more satisfied when receiving care from providers with similar characteristics. Black patients tend to rate Black providers higher in various measures of quality and Hispanics are more satisfied with care received from Hispanic providers.5 9 13 Similarly, a large percentage of Asians seek care from providers of the same background revealing a tendency to prefer concordant relationships that may stem from perceptions of better care.14 While a complex interplay of geographics, prior experiences of discrimination, feelings of lack of cultural sensitivity and language difficulties play a role in some patients preferring providers of similar backgrounds, and a myriad of interpersonal characteristics lead to higher satisfaction in such relationships, our survey reveals that emergency practitioners may not be aware of these relationships.15 This knowledge gap may affect provider behaviour towards patient requests and strategies aimed at improving workforce diversity.
When asked whether providers would accommodate patient requests for providers of the same sex, each category (always, sometimes and never accommodate) received similar percentages (between 32 and 38%). The responses to gender concordant patient requests were significantly different than those for racial or religiously matched providers (p<0.01), with more respondents either always accommodating or never accommodating with regards to same gender requests. Exploring this further through the case scenarios of patients requesting same gender providers, we found that only provider gender but not provider race, practice type or years of training significantly affected accommodation scores; female respondents were more likely to accommodate these requests. This finding adds to the literature about differential patient care dynamics based on provider gender. Female providers have greater gender role sensitivity and may be more empathic during clinical encounters;16 hence, they may be more attuned to patient needs for, and responsive to, patient requests for same-gender providers. Qualitative investigation may help us to understand gender differences in response to patient requests.
On the patient side, patients from minority backgrounds may be more accommodated when requesting providers of their racial, religious or gender demographic. Given that our respondent population contained only 26% female providers and 16% non-whites, this result is not attributable to providers of minority backgrounds rating higher levels of accommodation for patients from their own background. Rather, the results show a willingness to accommodate females and minorities by the dominant white male provider population. It may be that it is more common for female and minority patients to request same gender or racial/ethnic providers and, thus, providers are more sensitive to their needs. Alternatively, providers may be more willing to respond to requests from traditionally disadvantaged backgrounds. It is important to note that while the difference between accommodation scores was statistically significant, the clinical practice significance of these behavioural attitudes needs further exploration.
When further comparing accommodation scores across patient demographic lines, Muslim patients had higher scores than other groups. This may reflect that providers are more likely to alter practice styles or make accommodations in response to patients' religious values. Alternatively, providers may be more attuned to the needs of Muslim patients given the increased media attention paid to this population in the current geopolitical context.
Our study is the first to explore provider decision-making and perceptions about accommodating patient choice of provider in the ED and has some limitations. As with any non-validated survey, different question order or wording may yield different responses. Furthermore, convenience sampling at a national emergency medicine conference was used in order to generate hypotheses for further study. As such, selection bias may limit generalisation. Yet, the demographics characteristics of our sample do match those of American College of Emergency Physicians (ACEP) attendees and US emergency medical providers at large. For example, the mean age of survey respondents was 42 years, while nearly 50% of ACEP attendees were 40 y old or younger. Similarly, over one-third of respondents had practiced emergency medicine for 5 y or less, while 41% of ACEP attendees had less than 6 y of experience.17 Lastly, while our sample had a small percentage of non-whites (16%), national workforce data demonstrate that over 80% of practicing US emergency medical physicians are white.18
Minority patient populations receive an inferior quality of care across the spectrum of medical treatment, from preventive measures to the management of chronic conditions and emergency care.19 The causes of these disparities include factors related to the provider, the patient and the healthcare delivery system. Multiple organisations champion workforce diversity as part of the solution to disparate care.2 3 5 They hold that physicians and patients who share common values and language are more likely to develop stable healthcare relationships. Further, such relationships improve patient satisfaction, which is receiving increased importance as a marker of quality.20 This trend is witnessed to by the reliance on Press-Ganey patient satisfaction scores to compare EDs.21 Thus, given the benefits in patient-doctor relationships and in patient satisfaction, should efforts at workforce diversity take on a more systematic approach based on improving provider-patient concordance? While concordance may have something to offer, the affect on clinical outcome is not fully clear. Further research must better define this relationship.
Lastly, within health care at large, and particularly within the ED, provider and patient matching is not entirely possible nor in line with our value system. A better approach is to enhance cultural sensitivity and compassionate care. However, cultural sensitivity in certain situations may dictate trying to clinically accommodate a concordant provider request, as patient distress and lack of partnership may ensue from a discordant provider. Our study starts to address the culture of accommodation within the ED.
Patient-centered care through cultural accommodation improves patient satisfaction and may increase adherence.22 Since some patient populations prefer providers of their own background or gender, accommodating these requests, when possible, has the potential to enhance patient satisfaction and healthcare partnership.
Our study demonstrates that only a minority of emergency medicine providers believe that patients perceive better care from providers of their own background. When facing a patient request for a provider of the same gender, provider gender but not race, practice type or years in practice, influences likelihood of accommodation. Female providers are more likely to accommodate such requests than male providers. From the patient side, patients from minority backgrounds (female and non-white) may be more likely to have their requests for like-providers accommodated. Further study into how provider attitudes translate into adaptation of clinical practices in response to patient requests is needed to understand delivery of patient-centered care within the ED.
This study was presented in part and in poster form at the American College of Emergency Physicians Scientific Assembly in Chicago, IL 2008. The authors thank Jay Keyes, Alan Teigman, Bridget Flickinger and Lars Kristofer-Peterson for manning our research booth and to Dr Manish Shah for encouragement and manuscript review.
Funding Department of Emergency Medicine, University of Rochester Medical Center.
Competing interests None.
Ethics approval This study was conducted with the approval of the University of Rochester Research Subjects Review Board.
Provenance and peer review Not commissioned; externally peer reviewed.