Participants described race-related issues that may affect patient and physician communicating patterns, and be a barrier to active patient involvement in shared decision-making. Although a range of viewpoints were expressed in both focus groups and in-depth interviews, different themes predominated in these two settings. Within the in-depth interviews, patients were more likely to report that race does not influence patient/provider communication, and among those stating that race may affect communication, patient factors (e.g. limited health knowledge, “bad attitudes” and internalized racism) were thought to be barriers to SDM. No one reported experiencing negative race-related communication encounters or discrimination (although many reported hearing about such occurrences), and most believed that their self-efficacy and communication style accounted for their success at shared decision-making.
In contrast, the majority of patients in the focus groups indicated that race does influence patient/physician communication and shared decision-making, and primarily described physician factors (e.g. discrimination and cultural discordance) as the origin of communication disparities. Within every focus group, participants discussed negative communication encounters between white physicians and themselves, family members and/or close friends that they attributed to race.
The reasons for the difference in the predominance of themes between the focus groups and the in-depth interviews are unclear, but likely reflect an array of complex sociopolitical and interpersonal dynamics. First, the differential dynamics of one-on-one versus group encounters could have played a significant role (Lewin & Gullickson, 1997
). Despite the use of a race-concordant interviewer and assurances of confidentiality, participants in the in-depth interviews may have felt uncomfortable discussing topics potentially portraying their physician unfavorably, particularly with a researcher affiliated with the health system in which they received their care. In such settings, participants may have felt more compelled to conform to societal norms about patient perceptions of care (Ajzen & Fishbein, 1980
). In contrast, focus group participants may have felt more anonymous and more empowered, within a group of persons with similar backgrounds, to speak more openly about healthcare experiences, particularly after someone else had “normalized” and validated perceptions of discrimination within the group (Kitzinger, 1994
; Morgan, 1995
Second, our findings may be interpreted in light of Attribution Theory, wherein people explain, or attribute, their own and others' behaviors in ways that help maintain a positive self-image (Bem, 1972
). Persons who have negative experiences are likely to assign external causation (i.e. environmental factors), and persons with positive experiences are likely to assign an internal causation (i.e. personal attributes) (Bem, 1972
; Jones & Davis, 1965
; Kelley, 1967
). We found that persons who denied having negative communication experiences usually attributed their successful interactions to their own self-efficacy, and often attributed negative communication encounters of others to patient-related factors. Similarly, persons who reported negative race-related communication experiences (all of which were reported within focus groups) were usually persons who attributed communication disparities externally to physician-related factors such as discrimination. This external attribution may have been reinforced by group dynamics wherein group norms prohibit the attribution of responsibility to a given member of the group.
Finally, differences in ages between interview and focus group participants may partially explain our findings. Participants in the focus groups were somewhat younger than those in the in-depth interviews. Previous studies have shown that African- Americans that grew up during the civil rights era have more racial centrality/group identification than African-Americans born prior to World War II (Jackson, 1987
). Researchers have shown that racial centrality/group identification is positively associated with reports of perceived discrimination, and thus, generational differences may account for some of the higher reporting of negative communication experiences and discrimination within the focus groups (Sellers & Shelton, 2003
Regardless of the reasons for the differential predominance of themes between the focus groups and the in-depth interviews, our participants identified a range of patient and physician-related factors that may lower the quality of patient/physician communication and be a barrier to shared decision-making between African-American patients and their physicians. Many participants perceived these race-related communication barriers as being rooted in issues of discrimination, cultural discordance and internalized racism.
In order to fully understand the relationship between race and shared decision-making, it is important to place it within the larger context of race and health within the U.S. Jones developed a framework for understanding how race and racism contribute to health disparities (Jones, 2000
). This framework consists of three levels of racism—institutionalized, personally-mediated, and internalized racism (Jones, 2000
). Institutionalized racism, defined as “differential access to the goods, service and opportunities of society by race” (Jones, 2000
), has had manifestations within the field of medicine that include unethical experimentation (Gamble, 1997
; Washington, 2006
), disparities in healthcare (Hasnain-Wynia, Baker, Nerenz, Feinglass, Beal, Landrum et al., 2007
; Schneider, Zaslavsky, & Epstein, 2002
; Smedley, Stith, & Nelson, 2002
), and unequal access to resources such as health insurance (Doty & Holmgren, 2004
; Shi, 2001
). Institutionalized racism affects African-Americans' expectations about the quality of care they will receive from clinicians and health systems (Boulware, Cooper, Ratner, LaVeist, & Powe, 2003
; Jacobs, Rolle, Ferrans et al., 2006
). Participants in this study noted that differential access to resources such as health insurance and income directly influenced the ability of African-Americans to participate fully in shared decision-making.
It is the other two levels of racism— personally-mediated and internalized— that are particularly relevant to the findings in this study. Internalized racism, defined as “acceptance by members of the stigmatized race of negative messages about their own abilities and intrinsic worth” (Jones, 2000
) was reflected in the in-depth individual interviews, where participants perceived that deficient characteristics of other African-Americans (e.g. not “speaking well” or having limited health knowledge) were the reason for communication disparities and less shared decision-making between African-American patients and their physicians.
In contrast to the in-depth interviews, within the focus groups, personally-mediated racism, defined as “prejudice [differential assumptions about the abilities, motives and intentions of others according to their race] and discrimination [differential actions towards others according to their race]” (Jones, 2000
), was a prominent theme, and participants disproportionately attributed communication disparities to physician-related factors such as discrimination. A growing body of literature documents the use of negative stereotypes about African-American patients by physicians and healthcare disparities that may result from such implicit bias (Bogart, Kelly, Catz, & Sosman, 2000
; Finucane & Carrese, 1990
; Green, Carney, Pallin, Ngo, Raymond, Iezzoni et al., 2007
; Rathore, Lenert, Weinfurt, Tinoco, Taleghani, Harless et al., 2000
; van Ryn & Burke, 2000
; van Ryn, Hanan, Burke, & Besculides, 1999
), and the clinical encounter may create conditions that heighten the use of stereotypes and unconscious bias. Situations with time pressure, high cognitive demand, limited resources and uncertainty (all found in clinical settings) increase the likelihood of using cognitive shortcuts (e.g. stereotypes) to make decisions (Hamilton, 1981
). The most potent forms of discrimination currently experienced by African-Americans are the subtle and unconscious forms of discrimination experienced regularly (Banks, Kohn-Wood, & Spencer, 2006
; Harrell, 2000
We found that participants primarily discussed race-related communication problems within the context of white physicians, although racially unmatched relationships with foreign-born Asian and Indian physicians were reported as particularly problematic. No one in our study reported race-related communication barriers (e.g. discrimination) from African-American physicians. This is an important finding because there is literature suggesting that African-Americans may anticipate and/or report discrimination from same-race physicians in addition to different-race physicians (LaVeist, Rolley, & Diala, 2003; Malat & Hamilton, 2006).
Thus, the lingering vestiges of racism—institutional, internalized and personally-mediated— on both African-Americans and whites, both patients and physicians, may have the potential to influence patient/physician communication and shared decision-making. Within each SDM domain (information-sharing, deliberation/physician recommendations, and decision-making), participants identified areas in which race may have a negative influence.
Within the information-sharing domain (where patients and physicians discuss symptoms, diagnoses and lifestyle issues), participants felt that African-American patients may be less likely to share information with their providers, particularly about health behaviors and medication usage, and physicians may be less likely to share information with their African-American patients, including important information about patient's illnesses and the results of diagnostic tests. Participants also noted that physicians may be less likely to actively and patiently listen to African-American patients in comparison to whites. Our findings are particularly important in light of a previous paper where we reported that information-sharing may be the most important SDM domain to African-Americans, and that the need to “tell their story and be heard” is a crucial experience for this population (Peek, Quinn, Gorawara-Bhat et al., 2008
Within the ‘deliberation/physician recommendation’ domain (which focuses on exploring treatment options), participants believed that physicians may be less likely to review treatment options with African-American patients (versus whites), and described how African-Americans may be less willing to “speak up” to their doctors and question physician treatment recommendations because of an exaggerated deference to physicians rooted in internalized racism.
Finally, within the decision-making domain, participants believed that physicians were more likely to be domineering about treatment decisions and less likely to share in the decision-making process with African-American patients. Study participants also described patient-related factors. We previously reported that African-American diabetes patients conceptualize the decision-making so that treatment “noncompliance” was as a viable means of exerting control over treatment decisions and actively participating in their own care (Peek, Quinn, Gorawara-Bhat et al., 2008
). In this study, we found that non-adherence may be driven by racial dynamics such as physician mistrust and low self-efficacy to “stand up” to authority figures such as physicians. The phenomenon of verbally agreeing to treatment (e.g. taking insulin) but being non-adherent may be partially explained by the African-American adaptation of presenting one identity to whites (oftentimes a deferential one) and a different identity at home (that may disregard the opinions of whites). This adaptation has its origins in slavery and legalized segregation, which allowed African-Americans to maintain a positive sense of self, exert control over their lives and communities, and avoid physical harm (Pittinsky, Shih, & Ambady, 1999; Banks, Kohn-Wood, & Spencer, 2006
; Bogle, 1992; Blassingame, 1979).
In summary, our research suggests that all aspects of shared decision-making— information-sharing, deliberation/physician recommendations, and decision-making— have the potential to be negatively influenced by race, through mechanisms of cultural discordance, patient beliefs arising from internalized racism, and unconscious stereotyping/bias (personally-mediated racism). Such influences serve to exacerbate the inherent power imbalance that exists between patients and their physicians.
This study has several limitations. First, it took place in an urban academic medical center within the midwest region of the United States; the majority of our patients were women and nearly half of patients were retired. As such, our findings may not be generalizable to all African-Americans with diabetes. Second, this research utilized a purposeful sample of patients. Consequently, patients who had particularly strong and/or negative communication experiences with their physicians may have decided to participate in the study at higher rates (in order to express these strong emotions) or at lower rates (to avoid re-experiencing unpleasant encounters) than other patients. And finally, this study did not specifically explore gender, education, age, income, diabetes severity/illness status, or other sociodemographic variables that may influence patient/physician communication patterns. Because race interacts with many of these variables in a variety of settings and social contexts, it is likely that such interplay also occurs within shared decision-making. However, our goal of this study was to focus on the complex issue of race and the various ways that it may influence shared decision-making patterns between patients and physicians. Future studies should build upon this work and explore how race, and its influences, may interact with and be affected by other social variables within the setting of shared decision-making.
Our study has several strengths. First, we were able to obtain in-depth, rich information about a phenomenon that is challenging to explore—the perceived influence of race and racism on communication patterns between patients and physicians. Second, our study utilized a multi-method approach that enhanced our ability to arrive at valid conclusions.
This study suggests that race-related barriers may exist to shared decision-making between African-American patients and their physicians. Finding innovative ways to address such communication barriers and enhance SDM among African-Americans is an important area of research. Because barriers to shared decision-making may exist for patients and providers alike (Peek, Wilson, Gorawara-Bhat, Odoms-Young, Quinn, & Chin, 2009), communication strategies should be developed for both groups. Extending cultural competency training and general communication training for physicians to address potential race-related barriers to SDM has the potential to enhance shared decision-making among African-Americans. While patient communication interventions can be effective, only one study has examined the efficacy of such an intervention among African-Americans and it had no demonstrable effect (although it increased patient communication among whites), a finding the authors attribute to the intervention's lack of cultural appropriateness (Greenfield, Kaplan, Ware, Yano, & Frank, 1988
; Post, Cegala, & Marinelli, 2001). Systematic reviews and meta-analyses of diabetes interventions indicate that culturally-tailoring interventions and incorporating behavioral and pyschosocial strategies (versus information transfer alone) have the potential for greater effectiveness (Peek, Cargill, & Huang, 2007; Peyrot, 1999; Anderson, Funnell, Butler, Arnold, Fitzgerald, & Feste, 1995; Brown, 1999). A recent study reported an increase in diabetes self-efficacy among Hispanic Americans when family members were used to support communication efforts with physicians (Coffman, 2008). Thus, culturally-tailored communication training for patients may be an effective strategy for increasing SDM among African-Americans with diabetes, particularly if it incorporates patient beliefs and cultural norms, and addresses potential issues of internalized racism.