A Study of National Physician Organizations’ Efforts to Reduce Racial and Ethnic Health Disparities in the United States

Abstract

Purpose

To characterize national physician organizations’ efforts to reduce health disparities and identify organizational characteristics associated with such efforts.

Method

This cross-sectional study was conducted between September 2009 and June 2010. The authors used two-sample t tests and chi-square tests to compare the proportion of organizations with disparity-reducing activities between different organizational types (e.g., primary care versus subspecialty organizations, small [<1,000 members] versus large [>5,000 members]). Inclusion criteria required physician organizations to be (1) focused on physicians, (2) national in scope, and (3) membership based.

Results

The number of activities per organization ranged from 0 to 22. Approximately half (53%) of organizations had 0 or 1 disparity-reducing activities. Organiza-tional characteristics associated with having at least 1 disparity-reducing effort included membership size (88% of large groups versus 58% of small groups had at least 1 activity; P = .004) and the presence of a health disparities committee (95% versus 59%; P < .001). Primary care (versus subspecialty) organizations and racial/ethnic minority physician organizations were more likely to have disparity-reducing efforts, although findings were not statistically significant. Common themes addressed by activities were health care access, health care disparities, workforce diversity, and language barriers. Common strategies included education of physicians/trainees and patients/general public, position statements, and advocacy.

Conclusions

Despite the national priority to eliminate health disparities, more than half of national physician organizations are doing little to address this problem. Primary care and minority physician organizations, and those with disparities committees, may provide leadership to extend the scope of disparity-reduction efforts.

In the United States, racial and ethnic health disparities exist across a broad range of clinical conditions and health care settings.^1–6 Socioeconomic, political, environmental, and health care system factors all contribute to disparities.^1,7–13 Health care disparities have been attributed to racial and ethnic minorities having unequal access to comprehensive, high-quality health care, inadequate emphasis on health promotion and prevention, and aspects of the patient–provider relationship (e.g., patient distrust, communication barriers, lack of provider cultural competence, physician bias or discrimination).^1,14 Many physician organizations have contributed to these health disparities by promoting or countenancing racial and ethnic bias in health care delivery through actions such as training health care providers in racial eugenics, excluding racial and ethnic minority physicians from physician organizations, and segregating minority patients and providers into health systems with inadequate resources.^12,15–17 The Institute of Medicine (IOM) has proposed several strategies for reducing, and ultimately eliminating, health disparities.¹ These recommendations include monitoring quality and equity measures (e.g., tracking racial differences in diabetes care) and heightening awareness of health disparities. The IOM also recommends increasing workforce diversity within health care and enhancing professional training directed at reducing health disparities. The IOM recommendations also address patient issues, suggesting that health care providers support patient decision making through culturally and linguistically appropriate care and patient education.

Although there is a robust literature about interventions that target patients, individual physicians, and health systems to improve minority health outcomes and reduce disparities, little is known about the efforts of physician organizations to reduce racial and ethnic health disparities.⁵ Yet, physician organizations are in a unique position to implement policies and programs consistent with the IOM recommendations to address health disparities. These societies are able to use their infrastructure, resources, human capital, and prestige to influence health care providers, patients, and other key stakeholders through a range of educational, research, clinical, and advocacy roles. For example, physician organizations can influence undergraduate, graduate, and continuing medical education; promote health care workforce priorities; create and disseminate patient-education materials; identify and promote volunteer opportunities for their members; set research priorities; fund research and education; and advocate for health policy and legislation. Physician organizations increasingly play these roles in efforts to reduce health disparities. For example, the Commission to End Health Care Disparities, a coordinated effort of over 35 physician organizations, seeks to address health disparities through 10 priority areas, including increased awareness, patient–physician communication, enhanced quality, and focused research.¹⁴ To date, however, there has been no national assessment of physician organizations’ efforts, as a whole, to reduce racial and ethnic disparities in health.

To better characterize the role of physician organizations, a subcommittee of the Society of General Internal Medicine’s Disparities Task Force conducted a large, cross-sectional study with the primary goal of characterizing the magnitude, frequency, and scope of national physician organizations’ efforts to reduce health disparities, and to identify organizational characteristics associated with such efforts. We hypothesized that primary care organizations, organizations consisting primarily of racial and ethnic minority physicians, organizations with health disparities committees, and large organizations would be more likely to have activities in place to address racial and ethnic health disparities.

Method

Identification of organizations

We used three sources to identify eligible organizations: the American Medical Association (AMA), the Council of Academic Societies (CAS), and MedlinePlus. The AMA, through their “Federation of Medicine,”¹⁸ provides a listing of national medical specialty organizations represented in the AMA House of Delegates. CAS¹⁹ includes 94 academic societies devoted to biomedical and behavioral research, medical education, and patient care. MedlinePlus²⁰ brings together information from the National Library of Medicine, the National Institutes of Health, and other health-related organizations to provide a listing of organizations providing health information to the public. We supplemented this list through snowball sampling²¹ by inviting organizations we identified through the initial process to provide information about other potentially eligible physician groups.

We then systematically reviewed the organizations in this list to create a final sample. To be included in the study, organizations had to be (1) focused on physicians (the majority of members were physicians, and the organizational focus was on physician-related issues, such as clinical care), (2) national in scope (local and regional organizations, such as state medical societies, were excluded), and (3) membership based (organizations allowing physician participation without membership, such as foundations and institutes, were excluded). Two members of the research team independently reviewed each organization for eligibility and then met to resolve differences of opinion. The mean interrater agreement for the four pairs of investigators was 93%, with a range of 90% to 95% (kappa statistic). We also iteratively assessed eligibility of organizations through a series of conference calls to ensure group consistency. Of the 640 organizations originally identified, 167 (26.1%) were eligible for inclusion (see Supplemental Digital Appendix 1, http://links.lww.com/ACADMED/A90). The research protocol was reviewed by the institutional review board of the University of Chicago and deemed exempt from review.

Organizational taxonomy

We created a taxonomy of organizational activities (e.g., education, research) and themes (e.g., access to care, language barriers) based on literature reviews, experience of the research team, and a detailed review of the Web sites of two organizations—the AMA and the National Medical Association. We selected these organizations because they have membership across medical specialties and have undertaken significant work to address health disparities.^14,22 The final taxonomy of organizational activities (e.g., education of physicians and trainees) and the thematic domains within each activity (e.g., cultural competency) are found in List 1. During the data collection process, we applied the taxonomy to assess eligible organizations’ activities to address health care disparities. In addition, we collected the following information about eligible organizations: year established, membership size, racial and ethnic minority membership focus, primary care versus subspecialty focus, and whether the organization had a health disparities committee.

List 1.

Taxonomy of Organizational Activities and Thematic Domains of U.S. National Physician Organizations’ Efforts to Address Racial and Ethnic Health Disparities

Organizational activity types

Education

• Physicians and trainees (e.g., medical students)

• Patients and the general public

• Diversity pipeline programs*

Clinical care†

Position statements

Advocacy

Data monitoring‡

Research

Mentorship (of health disparities researchers)

Thematic domains

Racial and ethnic health disparities

Cultural competency

Access to health care

Communication skills

Language barriers

Health literacy

Patient discrimination

Physician discrimination

Workforce diversity

^*Diversity pipeline programs seek to enhance the representation of underrepresented minorities along the educational continuum to becoming a physician.

^†Clinical care activities include the direct provision of medical care (e.g., special health screening events), temporary placement of members in medically underserved areas, etc.

^‡Data monitoring includes collection/analysis of internal data about the organization’s racial and ethnic composition.

Data collection

Our primary data sources were organization Web sites. To ensure consistency in data abstraction and classification, all members of the research team independently reviewed three organization Web sites and then met to discuss coding discrepancies and revise the taxonomy categories. This process was repeated until no new taxonomic categories arose and the team was consistent in data abstraction and classification.

One member of the research team (S.W.) then reviewed each of the 167 organization Web sites to abstract information about organizational activities and thematic domains. This information was captured in two ways: (1) dichotomous data (i.e., binary codes for the presence or absence of an activity within a given theme) for quantitative analysis, and (2) descriptive data for qualitative analysis. To confirm data accuracy and completeness after we collected data from organization Web sites, two rounds of letters were sent to the president, executive director, and public relations or communication director of each organization explaining the study, providing abstracted data, and asking for additions and revisions.

The team used the Delphi method (a validated process for obtaining consensus opinion from multiple experts) to identify 23 key organizations to contact by telephone if we had not received responses to our written queries.²³ These organizations were selected from the original 167 organizations; each member of the research team was asked to create a list of 10 organizations that they felt were key to include in the analysis (no specific selection criteria were given). These individual lists were combined and discussed as a group. The resulting list contained 23 “key” organizations that the team planned to contact by mail and telephone if necessary to ensure accurate capture of data.

One hundred nine organizations (65%) responded to our queries, and 15% of the total data we analyzed were obtained from the organizations themselves in response to written and/or telephone queries. We abstracted original data from Web sites between September 2009 and February 2010. We obtained additional data by phone from the organizations between March 2010 and June 2010.

Data analysis

We tabulated the descriptive characteristics of the organizations and calculated their rates of use of activities to address racial and ethnic health disparities (both mean number and dichotomized at none versus any) and the frequencies of different thematic domains and activity types. We used two-sample t tests and chi-square tests of proportions to compare the proportion of organizations with disparities-reducing activities between the following organizational types: primary care versus subspecialty organizations; organizations consisting primarily of racial and ethnic minority physicians versus other organizations; organizations with a health disparities committee versus other organizations; and small (<1,000 members) versus medium (1,000–5,000) versus large (>5,000) organizations. We conducted our analyses using STATA 10.0 (StataCorp, College Station, Texas), and we defined statistical significance as a two-tailed P value of <.05.

Results

Organizational characteristics

The majority of the 167 physician organizations included in this study were medium to large (>1,000 members) subspecialty organizations (Table 1). Few consisted primarily of minority physicians, and less than 25% (n = 40) had health disparities committees.

Table 1.

Characteristics of 167 U.S. National Physician Organizations

Specialty
Primary care	9 (5)
Subspecialty	158 (95)
Membership size*
<1,000	12 (11)
1,000–5,000	47 (43)
>5,000	50 (46)
Minority physician organization
Yes	12 (7)
No	155 (93)
Disparities committee
Yes	40 (24)
No	127 (76)

^*Data for this variable were not available for all 167 organizations.

The number of activities to address health disparities within a single organization ranged from 0 to 22. Approximately one-third (n = 53; 32%) of physician organizations had no such activities, 35 organizations (21%) had 1 activity, 42 (25%) had 2 to 5 activities, and 37 (22%) had more than 5 organizational activities to address health disparities (17 [10%] had 6– 9 activities; 20 [12%] had 10–22 activities).

Organizational characteristics that were statistically associated with efforts to address health disparities included membership size and the presence of a health disparities committee (Table 2). Large organizations (>5,000 members) were statistically more likely than small organizations (<1,000) to engage in efforts to address disparities, as were organizations with health disparities committees (compared with those without such a committee). Primary care (versus subspecialty) organizations and societies primarily consisting of racial and ethnic minorities were also more likely to engage in efforts to address health disparities, although these findings did not reach statistical significance.

Table 2.

Characteristics of 167 U.S. Physician Organizations Associated With Efforts to Address Racial and Ethnic Health Disparities

Variable	No activities to address health disparities, no. (%)*	$1 activity to address health disparities, no. (%)*	P value
Specialty			.16
Primary care (n = 9)	1 (11)	8 (89)
Subspecialty (n = 158)	53 (34)	105 (67)
Membership size†			.004
≤1,000 (n = 12)	5 (42)	7 (58)
1,001–5,000 (n = 47)	19 (40)	28 (60)
>5,000 (n = 50)	6 (12)	44 (88)
Minority physician organization			.07
Yes (n = 12)	1 (8)	11 (92)
No (n = 155)	53 (34)	102 (66)
Health disparities committee			<.001
Yes (n = 40)	2 (5)	38 (95)
No (n = 127)	52 (41)	75 (59)

^*Percentage values reflect the use of row totals as the denominator.

^†Data for this variable were not available for all 167 organizations.

Organizational activities by theme

One hundred thirteen physician organizations (68%) had at least one activity to address health disparities. Among these 113 organizations, activities addressed themes including access to health care (n = 42; 37%), the general topic of racial and ethnic health disparities (n = 37; 33%), workforce diversity (n = 33; 29%), language barriers (n = 31; 27%), health literacy (n = 16; 14%), cultural competence (n = 11; 10%), patient-directed bias or discrimination (n = 11; 10%), and physician-directed bias or discrimination (n = 9; 8%).

Organizational activities by strategic method

We also categorized organizational activities by strategic method (Table 3). Of the 113 organizations with efforts to address disparities, 59 (52%) had educational activities for physicians and trainees (e.g., medical students), 45 (40%) had advocacy efforts, 33 (29%) had educational activities that targeted patients or the general public, and 27 (24%) physician organizations had position statements addressing racial and ethnic health disparities. Research (n = 15; 13%), diversity pipeline programs (n = 11; 10%), tracking of membership data by race or ethnicity (n = 9; 8%), clinical care (n = 7; 5%), and mentorship of young disparities researchers (n = 4; 4%) were less commonly used strategies for addressing disparities.

Table 3.

Comparisons by Strategic Methods of National Physician Organizations’ Activities to Address Topics Related to Health Disparities

Strategic method†	No. (%) of institutions offering activities*	No. (%) of activities offered by topic
		Racial and ethnic health disparities	Cultural competence	Access to health care	Communicatio n skills	Language barriers	Health literacy	Patient discrimination	Physician discrimination	Workforce diversity
Education
Physicians and trainees	86 (52)	38 (44)	28 (33)	15 (17)	36 (42)	13 (15)	15 (17)	4 (5)	4 (5)	21 (25)
Patients/publi c	49 (29)	8 (16)	8 (16)	7 (14)	6 (12)	33 (67)	7 (14)	2 (4)	0 (0)	3 (6)
Diversity pipeline programs	16 (10)	1 (6)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	15 (94)
Clinical care	9 (5)	92 (22)	1 (11)	3 (33)	2 (22)	1 (11)	2 (22)	0 (0)	0 (0)	1 (11)
Position statement	40 (24)	10 (24)	6 (15)	21 (53)	2 (5)	2 (5)	3 (8)	9 (23)	8 (20)	12 (30)
Advocacy	66 (40)	26 (40)	12 (18)	45 (68)	9 (14)	7 (11)	5 (8)	5 (8)	1 (2)	25 (38)
Data management	13 (8)	1 (8)	1 (8)	2 (15)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	1 (8)
Research	22 (13)	10 (46)	0 (0)	3 (14)	2 (9)	1 (5)	2 (9)	0 (0)	0 (0)	2 (9)
Mentorship of junior disparities researchers	6 (4)	1 (17)	0 (0)	0 (0)	1 (17)	0 (0)	0 (0)	0 (0)	1 (17)	1 (17)

^*Reflects the total number (and percentage) of organizations that had any activities (e.g., CME lectures) within a given method (e.g., education of physicians and trainees) out of the 167 organizations in the study.

^†Reflects the distribution of activities, based on their thematic domain, among those organizations with activities in a given strategic method (i.e., the denominator is the number of institutions offering activities for each of the topic cells in a given row).

Education of patients/general public

The most common strategy behind patient-education efforts was the provision of educational materials in a language other than English (usually Spanish); we categorized these efforts as “addressing language barriers” (Table 3). Other examples include providing educational materials about health disparities, educational materials about how to access medical care, and culturally tailored educational materials.

Education of physicians and/or trainees

Eighty-six organizations (52%) addressed health disparities by educating physicians and/or trainees. Racial and ethnic disparities, cultural competence, and communication skills were the most commonly addressed educational domains. Patient-level discrimination was rarely covered. Educational efforts were presented in a variety of venues and formats, including online materials and resources, one-time meetings or symposia, recurring conferences, special sessions within an organizational conference, or awards at national meetings that highlight the importance of disparities-related work.

The presence of a subcommittee devoted to disparities education efforts was often associated with a higher number and a broader assortment of disparities education interventions.

Position statements

One-fourth (n = 40; 24%) of the organizations that had efforts to reduce health disparities had position statements, which we defined as beliefs or principles that an organization upholds (e.g., equal health care access for all), but which do not necessarily translate into action through programs or projects. Over half of the societies had statements that addressed health care access (n = 21; 53%), and approximately one-third had statements about workforce diversity (particularly as related to the organization’s membership inclusiveness) (n = 12; 30%). Approximately 20% of the statements addressed patient-targeted discrimination (n = 9; 23%), physician-targeted discrimination (n = 8; 20%), or specifically racial and ethnic health disparities (n = 10; 24%). Position statements were often the precursor to subsequent advocacy work.

Advocacy

We defined advocacy as policy, legislative action, regulations, and/or codes of an organization. Forty percent (n = 66) of organizations that had efforts to reduce health disparities included physician advocacy among their strategic initiatives. Health care access (n = 45; 68%) and racial and ethnic health disparities (n = 26; 40%) were the most common focus for these advocacy efforts; other common themes included communication skills (n = 9; 14%), cultural competency (n = 12; 18%), and workforce diversity (n = 25; 38%).

Advocacy regarding specific legislation (both state and federal) was common and was primarily accomplished through statements of support on organizations’ Web sites; congressional briefings and testimony were less common. Four of the organizations provided advocacy training for their members. Several partnered with other organizations or joined a coalition to engage in advocacy. Whereas some organizations advocated for general health care access (e.g., publicly supporting the Patient Protection and Affordable Care Act), others specifically addressed access to their professional members’ services.

Discussion

Effectively addressing racial and ethnic health disparities in the United States will take a comprehensive, multifaceted, evidence-based approach that involves policy makers, communities, health systems, providers, and patients. Although physician organizations, through their work with each of these stakeholders, are uniquely positioned to make significant contributions to reducing health disparities, this study suggests that their efforts fall short in many areas.

To our knowledge, this is the first study to assess national physician organizations’ efforts to reduce racial and ethnic health disparities in the United States. We found that over half of the organizations in our study (53%) had little or no efforts (0–1 activities) under way to address health disparities. Moreover, most of the efforts were led by relatively few organizations: 12% of organizations were involved in more than 10 strategic initiatives to reduce disparities. Large organizations (>5,000 members), which likely have more infrastructure and resources, were more likely to engage in disparities-reducing efforts, as were organizations with disparities committees. Primary care organizations and racial and ethnic minority physician organizations were also more likely to have disparity-reducing initiatives, although their small numbers in our sample may have left us with inadequate power to detect a difference that met statistical significance.

The organizations’ emphasis on position statements, advocacy, and education, with relatively less effort devoted to clinical care and research, may reflect the inherent infrastructure and capacity of physician organizations. That is, these organizations are better equipped to educate and advocate than to provide direct clinical care or research. We purposely distinguished between position statements and other forms of organizational activity. Although position statements reflect the beliefs, values, and principles of an organization, they do not necessarily translate into action. We found that, although position statements were commonly used, they were usually just one component of a diverse portfolio of organizational activities and often formed the basis for subsequent advocacy.

Overall, the most common themes of organizational activities included health care access and the general topic of health disparities. Communication skills, language barriers, cultural competence, and workforce diversity were also frequent themes. Despite the recent attention to racial and ethnic discrimination in health care as a contributor to health disparities,^1,24–27 few organizations identified addressing health care discrimination as a means of reducing health disparities. Those that did typically had position statements against patient-targeted racial and ethnic discrimination in health care. In general, themes varied based on the type of activity. For example, workforce diversity was most commonly addressed in educational efforts of health providers and in diversity pipeline programs; it was rarely the focus of patient-education efforts. Access to care was most commonly addressed through advocacy efforts and not through the research mentorship or pipeline programs we reviewed.

It is important to note that although this study focuses on physician organizations’ efforts to address disparities among racial and ethnic minorities, many of the underlying contributors to such disparities are based on socioeconomic factors, such as income and access to health insurance. Current national healthy policy debates about health care reform and its implementation, which focus on improving the access of low-income, uninsured individuals to health care, will certainly have a direct and disproportionate influence on the health of racial and ethnic minorities. Because of the interrelationship between race/ethnicity and socioeconomic factors, we included thematic domains such as “access to health care” in our analyses as ones which could potentially reduce racial and ethnic health disparities.

Our study has some potential limitations. First, despite our attempts to generate a comprehensive inventory of national physician organizations, our list may be incomplete. However, it is unlikely that the number of missing organizations is large enough to change the overall pattern of results. Second, our primary means of data collection was through a standardized review of organization Web sites. Small or resource-constrained organizations may have had incomplete or outdated Web pages. However, when available, ancillary data collected from the organizations’ leadership did not differ appreciably from the information we obtained originally from Web sites. We interpret our findings with caution. Although the presence and number of disparities activities provide some indication of an organization’s commitment to disparities reduction, we do not have information on the intensity or effectiveness of the efforts that are documented. Future research into the scope and effectiveness of physician organizations’ efforts to reduce health disparities is warranted. Finally, we did not examine the activities of other health professional organizations, though there is evidence of ongoing disparities work in dentistry, pharmacy, and nursing, among other fields.^28–30

Our study also has several strengths. First, we used multiple sources of data to enhance the accuracy and comprehensiveness of our dataset. Second, we examined a broad spectrum of organizations that included primary care and multiple subspecialty organizations. Third, the data obtained in this study may be useful for examining changes in the disparities-reduction activities of physician organizations and their relationship to health care inequities and health disparities.

In summary, despite the national priority and pressing need to reduce and eliminate racial and ethnic health disparities in the United States, more than half of national physician organizations have relatively few activities to address this problem. We did find evidence that physician organizations, as a whole, are implementing a wide range of IOM disparities-reduction recommendations, including education, research, advocacy, clinical care, and mentorship. Primary care and racial and ethnic minority physician organizations and societies with health disparities committees may be uniquely poised to provide leadership and serve as national resources to extend the scope and reach of disparities-reduction efforts. Understanding these ongoing efforts may help organizations, policy makers, and other key stakeholders to identify ways to enhance existing programs and reduce gaps in efforts to achieve health equity. Future research should investigate the effects of physician organizations’ efforts on health outcomes and health disparities.