HEALTH POLICY CENTER Support for this research was provided by the Robert Wood Johnson Foundation. The views expressed here do not necessarily reflect the views of the Foundation. Racial, Ethnic, and Language Concordance between Patients and Their Usual Health Care Providers Dulce Gonzalez, Genevieve M. Kenney, Marla McDaniel, and Claire O'Brien March 2022 Historical medical mistreatment of Black Americans and other people of color, including the provision or withholding of treatment on the basis of erroneous beliefs about biological differences by race (Hoffman et al. 2016) and ignoring such patients' health concerns (Institute of Medicine 2003), has contributed to a mistrust of health care providers within these groups. 1 Perceptions of a shared identity between patients and their health care providers could be one way to improve the patient-provider relationship and foster trust and better communication. Although the available evidence is not conclusive, having a health care provider of the same race or who speaks the same language has been associated with a greater likelihood of patients agreeing to and receiving preventive care (Alsan, Garrick, and Graziani 2019), better patient experience ratings (Takeshita et al. 2020), and higher ratings on patient-reported measures of care quality (González, Vega, and Tarraf 2010). Drawing on the April 2021 round of the Urban Institute's Health Reform Monitoring Survey (HRMS), we provide a current snapshot of racial, ethnic, and language concordance between nonelderly patients ages 18 to 64 and their usual providers. 2 We find the following:  Among non-Hispanic/Latinx adults with a usual provider, Black adults were less likely to report racial concordance with their health care providers (22.2 percent) than were white adults (73.8 percent) and adults of additional races (34.4 percent). * * The racial and ethnic categories examined in this brief are Black, Hispanic/Latinx, white, and additional races. Respondents who are Black, white, or additional races did not identify as Hispanic/Latinx. We use "Hispanic/Latinx" to reflect the different ways people with Latin American ancestry self-identify. Many see "Latinx" as more inclusive; unlike "Latino/a," it is not gender specific. The term used in the HRMS is "Spanish, Hispanic, or Latino." Adults of additional races are Asian; Native Hawaiian; Pacific Islander; American Indian or Alaska Native; or more than one race, including white or Black and another race. Survey sample sizes do not allow for reporting disaggregated results for this diverse group of people. Though interpreting results for these adults is difficult, we include them for completeness.  Among Hispanic/Latinx adults with a usual provider, nearly 1 in 4 (23.1 percent) reported racial, ethnic, and language concordance with their provider. About 3 in 10 Hispanic/Latinx adults (29.2 percent) reported ethnic concordance with their providers, and 23.7 percent reported both racial and ethnic concordance with their providers.  As we discuss further in the Data and Methods section, estimates of racial and ethnic concordance between patients and their usual providers drawn from the HRMS are generally similar to those based on the 2018 Medical Expenditure Panel Survey (MEPS). One limitation of our analysis is that it excludes the experiences of people who do not have a usual source of care or a usual provider at their usual source of care. Hispanic/Latinx adults were much less likely than other adults to report having a usual provider and a usual source of care, so they are particularly underrepresented in our analysis. We describe differences in the share of adults reporting having a usual health care provider and a usual source of care by select characteristics in the Data and Methods section. In the same section, we also discuss other limitations of our study. To the extent that patient-provider concordance matters to patients and could help improve health outcomes by improving the patient-provider relationship, changes in policy and practice will be needed to facilitate access to the types of providers patients prefer to see. However, providers of color are underrepresented in the health care workforce, so improving the diversity of the health care workforce will likely require dismantling historical and ongoing barriers to medical education among underrepresented groups. Achieving equitable patient outcomes may also depend on health insurance payers (e.g., Medicaid or private insurers) holding health care providers, irrespective of their race or ethnicity, accountable for delivering culturally and linguistically effective evidence-based care to all of their patients. Finally, racial, ethnic, and language concordance are not the only factors that shape the patient-provider relationship. Other dimensions of identity, including socioeconomic status, religion, and sexual orientation, could also be important but are not covered in this study. Further research is needed to understand how patients' unique identities and preferences intersect to shape their provider choices, health care decisions (e.g., seeking less care because of negative experiences with health care providers), and health outcomes. Results In this section, we first focus on racial concordance for non-Hispanic/Latinx adults. We then assess racial and ethnic concordance jointly for Hispanic/Latinx adults, because such adults can be any race. Racial Concordance among Non-Hispanic/Latinx Adults Among non-Hispanic/Latinx adults, Black adults reported the lowest rates of racial concordance with their providers (22.2 percent). In comparison, 73.8 percent of white adults reported racial concordance with their usual providers. More than one in three adults of additional races (34.4 percent) reported racial concordance with their usual provider. Overall, 61.8 percent of non-Hispanic/Latinx adults who had a usual provider reported that their provider was their same race (figure 1). 2 RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS FIGURE 1 Racial Concordance between Non-Hispanic/Latinx Patients and Their Usual Providers, Overall and by Race, April 2021 73.8%*** 61.8% 34.4%*** 22.2% All non-Hispanic/Latinx Black adults Adults of additional races White adults adults URBAN INSTITUTE Source: Health Reform Monitoring Survey, April 2021. Notes: Adults who are Black, white, or additional races are not Hispanic/Latinx. Adults of additional races are Asian; Native Hawaiian; Pacific Islander; American Indian or Alaska Native; or more than one race, including white or Black and another race. Survey sample sizes do not allow for reporting disaggregated results for this diverse group of people. Though interpreting results for these adults is difficult, we include them for completeness. */**/*** Estimate differs significantly from that for Black adults at the 0.10/0.05/0.01 level, using two-tailed tests. Racial, Ethnic, and Language Concordance among Hispanic/Latinx Adults Among Hispanic/Latinx adults, close to 3 in 10 (29.2 percent) reported they consider their usual providers to share their same ethnicity (figure 2). Nearly 1 in 4 (23.7 percent) reported they were the same race and ethnicity as their usual provider, and a similar share (23.1 percent) reported they were the same race and ethnicity as their provider and that their provider speaks to them in the language they prefer. RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS 3 FIGURE 2 Racial, Ethnic, and Language Concordance between Hispanic/Latinx Patients and Their Usual Providers, April 2021 29.2% 23.7% 23.1% Provider is same ethnicity Provider is same race and ethnicity Provider is same race and ethnicity as as respondent as respondent respondent and speaks to respondent in their preferred language URBAN INSTITUTE Source: Health Reform Monitoring Survey, April 2021. Notes: Only Spanish-speaking or bilingual Hispanic/Latinx respondents were asked questions about language concordance. We assume Hispanic/Latinx adults who only speak English speak to their providers in the language they prefer. Conclusions We find that among non-Hispanic/Latinx adults, Black adults were less likely than other adults to report racial concordance with their usual health care providers. Additionally, only about one in four Hispanic/Latinx adults reported that their provider is their same race and ethnicity and speaks to them in their preferred language. Our analysis focuses on concordance between patients and their usual health care providers, meaning we do not know if respondents without a usual provider see providers of their same racial, ethnic, and linguistic background when they seek health care. As shown in table 1, about 30 percent of all respondents and 45 percent of Hispanic/Latinx respondents indicated they did not have a usual health care provider. This likely reflects systemic factors including lower access to affordable employer coverage due to the nature of a person's job, eligibility restrictions and immigration concerns for the subset who might be immigrants or in immigrant families, and a lack of services in the preferred languages of patients with limited English proficiency (ASPE 2021; Bernstein et al. 2021; Broder, Lessard, and Moussavian 2021; Dubay et al. 2020; Fortuny and Pedroza 2014). For patients who have experienced discrimination or unfair treatment based on their race or other aspects of their identity, concordance with their provider on one or more dimensions of identity could shape their health care experiences, decisions about provider selection, and health outcomes (Malat and van Ryn 2005). Though race, ethnicity, and language are important dimensions of identity, other dimensions, like gender identity, sexual orientation, class, and immigration status, could also intersect in unique ways to form patients' perceptions of closeness and rapport with their providers. In addition, 4 RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS some patients may not have a preference regarding the race of their providers or cannot access providers who look like them or share similar cultural experiences or languages (Malat and van Ryn, 2005; McDaniel et al. 2021). But to the extent patients prefer providers of their same race or who speak to them in their preferred language, barriers to medical education for people of color and barriers to services in patients' preferred languages will need to be addressed to improve access to concordant providers. Access to Medical Education and a Diverse Health Care Workforce Limited access to medical education for providers of color has driven an underrepresentation of Black Americans in the physician workforce. The exclusion of Black students and other students of color from predominantly white medical schools is one barrier; as of 1948, about a third of all medical schools did not accept Black students, and even in 1964, 10 years after the Brown v. Board of Education Supreme Court decision, five medical schools still had not admitted Black students (Institute of Medicine 2003; Sullivan and Mittman 2010). Closures of Black medical schools following the release of the Flexner report in 1910 also greatly reduced the number of Black medical school graduates in the 20th century (Harley 2006; Institute of Medicine 2003; Sullivan and Mittman 2010). Campbell and colleagues (2020) estimated that 35,000 additional Black students would have graduated medical school by 2019 had those Black medical schools not closed, which would have raised the number of Black medical school graduates by 29 percent that year. Because Black medical institutions play a key role in training Black providers and other providers of color, creating and supporting medical schools at historically Black colleges and universities and other minority-serving institutions could help increase the diversity of the health care workforce. Other barriers for Black people and other people of color who apply to medical school include the high cost of applying to and attending medical education programs, limited guidance on how to navigate admissions, and limited support systems to ensure students have the resources they need to succeed in their programs (AMSNY 2020). Cost barriers are particularly salient, given that the median cost of attending an in-state four-year medical school is about $260,000 for public institutions and more than $350,000 for private ones. 3 Other costs, including application and exam fees, travel costs for admissions interviews, and housing and other living expenses, also add to the financial barriers. 4 Tuition-free programs, debt-free medical education programs for students qualifying for financial aid, and scholarships and grants for underrepresented students are promising strategies for lowering the price of medical education and increasing the representation of Black students and other students of color (Kang and Ibrahim 2020; Toretsky, Mutha, and Coffman 2018). 5 Holistic screening practices and diverse admissions committees are other strategies shown to help increase admissions rates for Black people and other people of color pursuing medical education (Robinett et al. 2021). Medical professionals such as registered nurses, nurse practitioners, and physician assistants are a significant part of the health care workforce that also lacks diversity. Thus, extending the strategies above to these professions will also help increase diversity in the health care workforce (Salsberg et al. 2021; Willis et al. 2020). Diversifying the teams who facilitate or provide care to people of color and RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS 5 patients with limited English proficiency could further improve culturally and linguistically effective care. 6 For example, Kaiser Permanente's Qualified Bilingual Staff Model seeks to improve linguistic capabilities not just among doctors and nurses but also among front office staff and medical assistants (Kaiser Foundation Health Plan 2014). 7 Given the greater diversity among health care technicians and support staff, such as medical assistants, providing more pathways for these professionals to obtain medical degrees or other advanced training could be an efficient way to diversify the physician workforce (HRSA 2017). Access to Medical Interpretation Services Patients with limited English proficiency often experience worse health outcomes than people who speak English proficiently (Dubard and Gizlice 2008; Shamsi et al. 2020). However, despite federal regulations requiring health care providers to ensure access to language services for people with limited English proficiency, some health care providers are not meeting these regulations (Chen, Youdelman, and Brooks 2007; Diamond, Wilson-Stronks, and Jacobs 2010; Schiaffino, Nara, and Mao 2016; Shah, Velasquez, and Song 2020). One strategy that has been shown to improve health outcomes by bridging language gaps is providing professional interpretation services (Karliner et al. 2007), but providers may not offer such services because of the associated costs. Most Medicaid programs reimburse for professional medical interpretation, but reimbursement ranges from $30 to $50 per visit, and interpretation for an encounter could cost up to $200; private insurers seldom reimburse for interpretation services (Shah, Velasquez, and Song 2020). Stronger enforcement of language access regulations and higher reimbursement for medical interpretation services by health insurance payers could improve language access in health care (Chen, Youdelman, and Brooks 2007). Improving multilingual capabilities among health care workers is another way to bridge language gaps. Although some medical schools recommend students be fluent in a language other than English, making this a requirement or a heavily positively weighted factor for medical school admissions could also help diversify the languages spoken in the health care workforce. 8 Offering bonuses and higher pay for health care workers who speak multiple languages could also incentivize providers to add staff who speak languages other than English and to improve their communication with patients whose primary languages are not English. 9 Addressing the underrepresentation of Black people and other people of color in the health care workforce will not be a rapid process. And even if such gaps are closed, some patients still may not come across health care providers who look like them because of the demographics of providers available where they live. Thus, for physicians to meet the needs of different patients, they will need training that equips them to provide culturally and linguistically effective care. However, medical school curricula appear to fall short in this area; for example, only about 40 percent of US medical schools' curricula incorporate racial health inequities (White and Ojugbele 2019). Finally, it will be important for institutions, such as health insurance payers and hospital systems, to identify mechanisms that can hold all health care providers accountable for providing culturally and linguistically effective care. Steps toward this goal could include identifying and addressing instances of unconscious bias; incorporating greater use of patient satisfaction surveys; and making patient-centered techniques, such as 6 RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS encouraging patients to ask questions, a part of the workplace culture (White and Stubblefield-Tave 2017). Data and Methods This brief draws on data from the April 2021 round of the Urban Institute's Health Reform Monitoring Survey, a nationally representative, internet-based survey of adults ages 18 to 64. Launched in 2013, the HRMS was originally designed to provide timely information on the Affordable Care Act before federal survey data became available. In 2020, the Urban Institute updated the design and content of the HRMS to focus on the impact of the COVID-19 pandemic and support analyses of vulnerable populations so that the survey continues to provide timely data on critical policy issues. A total of 9,067 adults participated in the most recent round of the HRMS between April 2 and 20, 2021, and 82 percent of respondents completed the survey by April 8. For each round of the HRMS, we draw a stratified, random sample of nonelderly adults from Ipsos's KnowledgePanel, the nation's largest probability-based online panel. Members of the panel are recruited from an address-based sampling frame covering approximately 97 percent of US households, including those without internet access. If needed, panel members are given internet access and web- enabled devices to facilitate their participation. The April 2021 round of the HRMS includes oversamples of adults in low- and moderate-income households, nonwhite and Hispanic/Latinx adults, and young adults. Survey weights adjust for unequal selection probabilities and are poststratified to the characteristics of the national nonelderly adult population, based on benchmarks from the Current Population Survey and the American Community Survey. Participants can take the survey in English or Spanish, and the survey takes a median of 15 minutes to complete. The margin of sampling error, including the design effect, for the full sample of adults is plus or minus 1.2 percentage points for a 50 percent statistic at the 95 percent confidence level. Additional information about the HRMS can be found at hrms.urban.org. Analytic Approach Respondents to the April 2021 HRMS who reported having a usual source of care that is not the emergency room and a usual provider were asked the questions below to determine racial, ethnic, and language concordance with their usual providers.  Racial concordance. Do you consider your personal doctor/your personal health care provider to be the same race as you are?  Ethnic concordance. Is your personal doctor/your personal health care provider Spanish, Hispanic, or Latino?  Language concordance. Does your personal doctor/your personal health care provider speak to you in the language you prefer? RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS 7 The HRMS defines usual providers as the person at the respondent's usual source of care who they consider to be their personal doctor or health care provider. In a follow-up question, respondents could select one of the following as their usual provider:  a general doctor  a nurse practitioner, physician assistant, or midwife  a doctor who specializes in women's health (an obstetrician/gynecologist; only asked of women)  a mental health professional, such as a psychiatrist, psychologist, psychiatric nurse, or clinical social worker  a specialist (other than a psychiatrist or obstetrician/gynecologist)  a physical therapist, speech therapist, respiratory therapist, audiologist, or occupational therapist  other type of provider (respondents can specify in a text box)  not sure Most respondents (74.3 percent) reported that their usual provider was a general doctor. Another 14.2 percent reported that their provider was a nurse practitioner, physician assistant, or midwife. Each of the other response categories were selected by less than 3 percent of respondents (data not shown). The HRMS asks only Hispanic/Latinx adults who speak Spanish or are bilingual in English and Spanish about language concordance with their providers. We assume Hispanic/Latinx adults who only speak English have providers who speak to them in their preferred languages. We began by focusing on the racial, ethnic, and language concordance between patients and their usual providers. Given that Hispanic/Latinx adults can be any race, we first estimated the share of non- Hispanic/Latinx adults who reported racial concordance with their health care providers overall and separately for adults who are Black, white, and additional races. Next, we estimated the shares of Hispanic/Latinx adults who are ethnically concordant with their providers; both racially and ethnically concordant with their providers; and racially, ethnically, and linguistically concordant with their providers. Finally, we estimated the share of all adults ages 18 to 64 who had a usual source of care and a usual provider overall and by race or ethnicity and other characteristics (table 1). Limitations This study has several limitations. First, we cannot determine whether respondents who did not have a usual source of care typically see providers of their same race or ethnicity when they interact with health care providers. Even among adults with a usual source of care, some may have multiple providers with varying racial, ethnic, and language concordance. However, we only know concordance for the personal health care provider a respondent sees most often. Additionally, a respondent might not know or might mischaracterize their provider's race or ethnicity. Further, our survey is only fielded in English 8 RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS and Spanish, so we do not capture language concordance for adults who speak other languages but may face significant language access barriers when accessing health care. Adults with a Usual Source of Care and a Usual Provider Nearly 8 in 10 adults reported having a usual source of care (79.0 percent; table 1). Access to health insurance coverage is a key underlying factor; uninsured adults were much less likely than adults with public coverage and adults with private coverage to report having a usual source of care (46.5 percent versus 81.9 percent and 83.6 percent). Adults who do not have a chronic condition or a disability were also less likely than their respective counterparts to report having a usual source of care. Younger adults, adults with family incomes below 250 percent of the federal poverty level, men, and Hispanic/Latinx adults were also less likely to report having a usual source of care. Although the levels differ, similar patterns hold for the share of adults who reported having a usual health care provider. Because our analysis focuses on adults who have a usual health care provider and a usual source of care, we exclude the experiences of a large portion of people who are uninsured and people who may have a lower perceived need to interact with the health care system (e.g., adults without disabilities, adults without chronic conditions, and younger adults). Additionally, Hispanic/Latinx adults were less likely than adults of other races to report having a usual source of care and a usual provider. However, these differences likely reflect long-standing, systemic racial and ethnic inequities in access to health insurance coverage, which in turn affect access to timely, high-quality health care (ASPE 2021; Bernstein et al. 2021; Broder, Lessard, and Moussavian 2021; Dubay et al. 2020; Fortuny and Pedroza 2014). TABLE 1 Share of Adults Ages 18 to 64 Who Had a Usual Source of Care and Share Who Had a Usual Health Care Provider, Overall and by Select Characteristics, April 2021 Percent Had a usual source Had a usual of care provider All adults 79.0 66.9 By health insurance coverage type Uninsured^ 46.5 28.5 Public insurance 81.9*** 68.3*** Private insurance 83.6*** 72.8*** By chronic conditions Does not have a chronic condition^ 72.0 57.2 Has one or more chronic conditions 87.3*** 78.4*** By disability status Does not have a disability^ 78.0 65.9 Has a disability 86.9*** 75.5*** By age 18 to 34^ 71.6 56.5 35 to 49 79.1*** 66.6*** 50 to 64 87.5*** 79.2*** By family income RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS 9 Had a usual source Had a usual of care provider < 250% of FPL^ 74.4 60.4 250% of FPL or higher 82.3*** 71.6*** By gender Male^ 76.2 64.4 Female 81.7*** 69.4*** By race or ethnicity Hispanic/Latinx^ 73.1 55.8 Black 79.8** 65.6*** Additional races 78.9** 68.9*** White 80.7*** 70.4*** Source: Health Reform Monitoring Survey, April 2021. Notes: FPL = federal poverty level. Public coverage includes Medicaid, Medicare, the Children's Health Insurance Program, and any state- or government-sponsored assistance plan for which eligibility is based on income or a disability. Private coverage includes employer-sponsored insurance, military coverage, and private nongroup coverage purchased through or outside the Marketplaces. Adults who are Black, white, or additional races are not Hispanic/Latinx. Adults of additional races are Asian; Native Hawaiian; Pacific Islander; American Indian or Alaska Native; or more than one race, including white or Black and another race. Survey sample sizes do not allow for reporting disaggregated results for this diverse group of people. Though interpreting results for these adults is difficult, we include them for completeness. */**/*** Estimate differs significantly from that for the reference group (^) at the 0.10/0.05/0.01 level, using two-tailed tests. Comparison of Patient-Provider Concordance in the HRMS and the MEPS We also compared estimates of patient-provider concordance among nonelderly adults from the April 2021 HRMS with such estimates from the 2018 MEPS. The MEPS asks about patient-provider concordance as follows: Respondents can report whether their usual provider is a person, a person in a facility, or a facility. The MEPS only asks respondents who report their usual provider is a person or a person in a facility and who have a usual source of care about the races and ethnicities of their providers. To determine concordant pairings, we matched patients' and usual providers' races and ethnicities. See the Analytic Approach section above for wording of the patient-provider concordance questions in the HRMS. Despite different wording for patient-provider concordance questions in the HRMS and the MEPS, we find estimates of patient-provider racial and ethnic concordance are generally similar between the two surveys (figures 3 and 4). 10 RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS FIGURE 3 Racial Concordance between Non-Hispanic/Latinx Patients Ages 18 to 64 and Their Usual Providers, Overall and by Race, in the April 2021 HRMS and the 2018 MEPS HRMS MEPS 79.9% 73.8% 67.8% 61.8% 34.4% 26.8% 22.2% 21.3% All non-Hispanic/Latinx Black adults Adults of additional races White adults adults URBAN INSTITUTE Source: Health Reform Monitoring Survey, April 2021, and Medical Expenditure Panel Survey, 2018. Notes: HRMS = Health Reform Monitoring Survey. MEPS = Medical Expenditure Panel Survey. For MEPS estimates, respondents who only report Asian or Pacific Islander as their race are considered racially concordant with their provider if they report that their provider is Asian or Pacific Islander and no other race. Respondents who identified as multiple races were considered racially concordant with their providers if the races they reported for their providers were the same races they reported for themselves. Adults who are Black, white, or additional races are not Hispanic/Latinx. In the HRMS and the MEPS, adults of additional races are Asian; Native Hawaiian; Pacific Islander; American Indian or Alaska Native; or more than one race, including white or Black and another race. Survey sample sizes do not allow for reporting disaggregated results for this diverse group of people. Though interpreting results for these adults is difficult, we include them for completeness. RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS 11 FIGURE 4 Racial and Ethnic Concordance between Hispanic/Latinx Patients Ages 18 to 64 and Their Usual Providers in the April 2021 HRMS and the 2018 MEPS HRMS MEPS 32.0% 29.2% 25.3% 23.7% Provider is same ethnicity as respondent Provider is same race and ethnicity as respondent URBAN INSTITUTE Source: Health Reform Monitoring Survey, April 2021, and Medical Expenditure Panel Survey, 2018. Notes: HRMS = Health Reform Monitoring Survey. MEPS = Medical Expenditure Panel Survey. We do not compare estimates of language concordance in the HRMS and the MEPS given that (1) the MEPS asks whether providers offer translation services for patients and whether providers speak to patients in the languages they prefer within the same question, and (2) the respondents who receive the language concordance questions in the MEPS and the HRMS differ significantly. Notes 1 Martha Hostetter and Sarah Klein, "Understanding and Ameliorating Medical Mistrust among Black Americans," Commonwealth Fund, January 14, 2021, https://www.commonwealthfund.org/publications/newsletter- article/2021/jan/medical-mistrust-among-black-americans. 2 The HRMS only asked Hispanic/Latinx adults who speak Spanish or are bilingual in English and Spanish about language concordance with their providers. It also did not ask respondents whether they have access to medical interpreters when seeking care. We assume Hispanic/Latinx adults who only speak English have providers who speak to them in the languages they prefer. See the Data and Methods section for question wording. 3 "Medical Student Education: Debt, Costs, and Loan Repayment Fact Card for the Class of 2021," Association of American Medical Colleges, October 2021, https://store.aamc.org/downloadable/download/sample/sample_id/468/. 4 Emma Goldberg, "'I Have a Ph.D. in Not Having Money,'" New York Times, November 25, 2019, https://www.nytimes.com/2019/11/25/health/medical-school-cost-diversity.html. 5 James Doubek, "NYU Medical School Plans Free Tuition for Those Studying to Be Doctors," NPR, August 17, 2018, https://www.npr.org/2018/08/17/639467023/nyu-medical-school-says-it-will-offer-free-tuition-to-all- students; and Alyssa Sunkin-Strube, "Weill Cornell Medicine Announces Debt-Free Medical Education," Cornell Chronicle, September 16, 2019, https://news.cornell.edu/stories/2019/09/weill-cornell-medicine-announces- debt-free-medical-education. 6 "How to Improve Cultural Competence in Health Care," Tulane University, March 1, 2021, https://publichealth.tulane.edu/blog/cultural-competence-in-health-care/. 12 RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS 7 "Kaiser Permanente: Qualified Bilingual Staff Model," Robert Wood Johnson Foundation, accessed January 31, 2022, https://www.rwjf.org/en/library/research/2008/09/the-national-health-plan-collaborative- toolkit/section-5-case-studies/kaiser-permanente-qualified-bilingual-staff-model.html. 8 "Eligibility and Recommendations," Stanford Medicine, accessed January 31, 2022, https://med.stanford.edu/md-admissions/academic-requirements.html; "MD Admission FAQ," Emory University School of Medicine, accessed January 31, 2022, https://med.emory.edu/education/programs/md/admissions/faq.html#General_Questions3; "Prerequisites," University of Minnesota Medical School, accessed January 31, 2022, https://med.umn.edu/admissions/how- apply/prerequisites; and "ACE-PC Admissions," University of California, Davis, School of Medicine, accessed January 31, 2022, https://health.ucdavis.edu/mdprogram/ACE-PC/admissions.html. 9 "Salary and Benefits for House Staff," Harbor-UCLA Medical Center, accessed January 31, 2022, https://www.harbor-ucla.org/gme-resources/salary-benefits/; "Bilingual Incentive Pay Program Policy," Dallas County, Texas, January 17, 2017, https://www.dallascounty.org/Assets/uploads/docs/human- resources/Bilingual-Pay-Incentive-Policy-Court-Order-2017-0091.pdf; and Adria Iraheta, "Hillsborough County Agency Rewarding Its Bilingual Employees," Bay News 9, November 14, 2021, https://www.baynews9.com/fl/tampa/news/2021/11/04/hillsborough-county-agency-rewarding-its-bilingual- employees-. References Alsan, Marcella, Owen Garrick, and Grant Graziani. 2019. "Does Diversity Matter for Health? Experimental Evidence from Oakland." American Economic Review 109 (12): 4071–111. https://doi.org/10.1257/aer.20181446. AMSNY (Associated Medical Schools of New York). 2020. Addressing the Challenges to a Diverse Physician Workforce. New York: AMSNY. ASPE (US Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation). 2021. Health Insurance Coverage and Access to Care among Latinos: Recent Trends and Key Challenges. Washington, DC: ASPE. Bernstein, Hamutal, Michael Karpman, Dulce Gonzalez, and Stephen Zuckerman. 2021. "Immigrant Families Continued Avoiding the Safety Net during the COVID-19 Crisis." Washington, DC: Urban Institute. Broder, Tanya, Gabrielle Lessard, and Avideh Moussavian. 2021. "Overview of Immigrant Eligibility for Federal Programs." Los Angeles: National Immigration Law Center. Campbell, Kendall M., Irma Corral, Jhojana L. Infante Linares, and Dmitry Tumin. 2020. "Projected Estimates of African American Medical Graduates of Closed Historically Black Medical Schools." JAMA Network Open 3 (8): e2015220. https://doi.org/10.1001/jamanetworkopen.2020.15220. Chen, Alice Hm, Mara K. Youdelman, and Jamie Brooks. 2007. "The Legal Framework for Language Access in Healthcare Settings: Title VI and Beyond." Journal of General Internal Medicine 22 (Suppl. 2): 362–67. https://dx.doi.org/10.1007%2Fs11606-007-0366-2. Diamond, Lisa, Amy Wilson-Stronks, and Elizabeth A. Jacobs. 2010. "Do Hospitals Measure Up to the National Culturally and Linguistically Appropriate Services Standards?" Medical Care 48 (12): 1080–87. https://doi.org/10.1097/MLR.0b013e3181f380bc. DuBard , C. Annette, and Ziya Gizlice. 2008. "Language Spoken and Differences in Health Status, Access to Care, and Receipt of Preventive Services among US Hispanics." American Journal of Public Health 98 (11): 2021–28. https://doi.org/10.2105/ajph.2007.119008. Dubay, Lisa, Joshua Aarons, Steven Brown, and Genevieve M. Kenney. 2020. How Risk of Exposure to the Coronavirus at Work Varies by Race and Ethnicity and How to Protect the Health and Well-Being of Workers and Their Families. Washington, DC: Urban Institute. Fortuny, Karina, and Juan Pedroza. 2014. Barriers to Immigrants' Access to Health and Human Services. Washington DC, Urban Institute. RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS 13 González, Hector M., William A. Vega, and Wassim Tarraf. 2010. "Health Care Quality Perceptions among Foreign- Born Latinos and the Importance of Speaking the Same Language." Journal of the American Board of Family Medicine 23 (6): 745–52. https://doi.org/10.3122/jabfm.2010.06.090264. Harley, Earl H. 2006. "The Forgotten History of Defunct Black Medical Schools in the 19th and 20th Centuries and the Impact of the Flexner Report." Journal of the National Medical Association 98 (9): 1425–29. Hoffman, Kelly M., Sophie Trawalter, Jordan R. Axt, and M. Norman Oliver. 2016. "Racial Bias in Pain Assessment and Treatment Recommendations, and False Beliefs about Biological Differences between Blacks and Whites." Proceedings of the National Academy of Sciences of the United States of America 113 (16): 4296–301. https://dx.doi.org/10.1073%2Fpnas.1516047113. HRSA (Health Resources and Services Administration, Bureau of Health Workforce). 2017. "Sex, Race, and Ethnic Diversity of U.S. Health Occupations (2011-2015)." Washington, DC: HRSA. Institute of Medicine. 2003. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: National Academies Press. https://doi.org/10.17226/12875. Kaiser Foundation Health Plan. 2014. "Kaiser Permanente's Qualified Bilingual Staff Model." Kaiser Permanente Policy Story 3 (2). Kang, Yoon, and Said A. Ibrahim. 2020. "Debt-Free Medical Education-A Tool for Health Care Workforce Diversity." JAMA Health Forum 1 (12): e201435. https://doi.org/10.1001/jamahealthforum.2020.1435. Karliner, Leah S., Elizabeth A. Jacobs, Alice Hm Chen, and Sunita Mutha. 2007. "Do Professional Interpreters Improve Clinical Care for Patients with Limited English Proficiency? A Systematic Review of the Literature." Health Services Research 42 (2): 727–54. https://dx.doi.org/10.1111%2Fj.1475-6773.2006.00629.x. Malat, Jennifer, and Michelle van Ryn. 2005. "African-American Preference for Same-Race Healthcare Providers: The Role of Healthcare Discrimination." Ethnicity & Disease 15 (4): 740–47. McDaniel, Marla, Audrey Richardson, Dulce Gonzalez, Clara Alvarez Caraveo, Laura Wagner, and Laura Skopec. 2021. "Black and African American Adults' Perspectives on Discrimination and Unfair Judgment in Health Care." Washington, DC: Urban Institute. Robinett, Kathryn, Raushanah Kareem, Kristin Reavis, and Sandra Quezada. 2021. "A Multi-pronged, Antiracist Approach to Optimize Equity in Medical School Admissions." Medical Education 55 (12): 1376–82. https://doi.org/10.1111/medu.14589. Salsberg, Edward, Chelsea Richwine, Sara Westergaard, Maria Portela Martinez, Toyese Oyeyemi, Anushree Vichare, and Candice P. Chen. 2021. "Estimation and Comparison of Current and Future Racial/Ethnic Representation in the US Health Care Workforce." JAMA Network Open 4 (3): e213789. https://doi.org/10.1001/jamanetworkopen.2021.3789. Schiaffino, Melody K., Atsushi Nara, and Liang Mao. 2016. "Language Services in Hospitals Vary by Ownership and Location." Health Affairs 35 (8): 1399–1403. https://doi.org/10.1377/hlthaff.2015.0955. Shah, Shivani A., David E. Velasquez, and Zirui Song. 2020. "Reconsidering Reimbursement for Medical Interpreters in the Era of COVID-19." JAMA Health Forum 1 (10): e201240. https://doi.org/10.1001/jamahealthforum.2020.1240. Shamsi, Hilal Al, Abdullah G. Almutairi, Sulaiman Al Mashrafi, and Talib Al Kalbani. 2020. "Implications of Language Barriers for Healthcare: A Systematic Review." Oman Medical Journal 35 (2): e122. https://dx.doi.org/10.5001%2Fomj.2020.40. Sullivan, Louis W., and Ilana Suez Mittman. 2010. "The State of Diversity in the Health Professions a Century after Flexner." Academic Medicine 85 (2): 246–53. https://doi.org/10.1097/ACM.0b013e3181c88145. Takeshita, Junko, Shiyu Wang, Alison W. Loren, Nandita Mitra, Justine Shults, Daniel B. Shin, and Deirdre L. Sawinski. 2020. "Association of Racial/Ethnic and Gender Concordance between Patients and Physicians with Patient Experience Ratings." JAMA Network Open 3 (11): e2024583. https://doi.org/10.1001/jamanetworkopen.2020.24583. Toretsky, Christopher, Sunita Mutha, and Janet Coffman. 2018. "Breaking Barriers for Underrepresented Minorities in the Health Professions." San Francisco: University of California, San Francisco. 14 RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS White, Augustus A., and Beauregard Stubblefield-Tave. 2017. "Some Advice for Physicians and Other Clinicians Treating Minorities, Women, and Other Patients at Risk of Receiving Health Care Disparities." Journal of Racial and Ethnic Health Disparities, no. 4: 472–79. White, Stephanie, and Olutosin Ojugbele. 2019. "Addressing Racial Disparities in Medical Education." Curriculum in Context 6 (2): 1–6. Willis, Joel, Brian Antono, Andrew Bazemore, Anuradha Jetty, Stephen Petterson, Judy George, Bedda L. Rosario, et al. 2021. The State of Primary Care in the United States: A Chartbook of Facts and Statistics. Washington, DC: Robert Graham Center, IBM-Watson Health, and American Board of Family Medicine. About the Authors Dulce Gonzalez is a research associate in the Health Policy Center at the Urban Institute. She forms part of a team working on the Urban Institute's Well-Being and Basic Needs Survey. Gonzalez conducts quantitative and qualitative research focused primarily on the social safety net, immigration, and barriers to health care access. Her work has also focused on the impacts of the COVID-19 pandemic on nonelderly adults and their families. Before joining Urban, Gonzalez worked at the Georgetown University Center for Children and Families and at the nonprofit organization Maternal and Child Health Access. Gonzalez holds a BA in economics from California State University, Long Beach, and a master's degree in public policy from Georgetown University. Genevieve M. Kenney is a vice president and senior fellow in the Health Policy Center. She is a nationally renowned expert on Medicaid, the Children's Health Insurance Program (CHIP), and health insurance coverage; health care access and quality; and health outcomes for low-income adults, children, and families. She has played a lead role in several Medicaid and CHIP evaluations, including multiple congressionally mandated CHIP evaluations, and has conducted state-level evaluations of the implementation of managed care and other service delivery reform initiatives and policy changes in Medicaid and CHIP. Currently, she is leading a project for the Robert Wood Johnson Foundation focused on health equity that involves working with a community advisory board. In other work, she is assessing Medicaid policies aimed at improving outcomes in the postpartum period and increasing receipt of evidence-based treatment for substance use disorder. Her prior work has used mixed methods to examine Medicaid expansions for pregnant women, parents, and children; Medicaid family planning waivers; and a range of policy choices related to Medicaid and CHIP. She received a master's degree in statistics and a doctoral degree in economics from the University of Michigan. Marla McDaniel is a senior fellow in the Urban Institute's Center on Labor, Human Services, and Population at the Urban Institute whose research examines racial and ethnic disparities; low-income children, youth, and families; and the programs and policy environments that touch families' lives. She is interested in the relationships between inequities across multiple domains-including health, education, and employment-and their compounding effects on overall health and well-being. McDaniel holds a BA in psychology from Swarthmore College and a PhD in human development and social policy from Northwestern University. Claire O'Brien is a research assistant in the Health Policy Center, where she uses quantitative analysis to study the impact of health reforms. This includes using Medicaid claims to evaluate integrated care RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS 15 plans for dually eligible Medicare and Medicaid beneficiaries and to study racial disparities within the Medicaid program. Additionally, O'Brien takes part in the implementation and analysis of the Urban Institute's Health Reform Monitoring Survey. Finally, O'Brien monitors changes in the Affordable Care Act's Marketplaces and models what the effects of a public option would be. O'Brien has a BA in economics and applied math with a minor in poverty studies from the University of Notre Dame. Acknowledgments This brief was funded by the Robert Wood Johnson Foundation. The views expressed do not necessarily reflect the views of the Foundation. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Funders do not determine research findings or the insights and recommendations of Urban experts. Further information on the Urban Institute's funding principles is available at urban.org/fundingprinciples. The authors would like to thank Eva H. Allen, Michael Karpman, and Kimá Joy Taylor for their helpful comments and suggestions on this brief. ABOUT THE ROBERT WOOD JOHNSON FOUNDATION The Robert Wood Johnson Foundation (RWJF) is committed to improving health and health equity in the United States. In partnership with others, we are working to develop a Culture of Health rooted in equity, that provides every individual with a fair and just opportunity to thrive, no matter who they are, where they live, or how much money they have. ABOUT THE URBAN INSTITUTE The nonprofit Urban Institute is a leading research organization dedicated to developing evidence-based insights that improve people's lives and strengthen communities. For 50 years, Urban has been the trusted source for rigorous analysis of complex social and economic issues; strategic advice to policymakers, philanthropists, and practitioners; and new, promising ideas that expand opportunities for all. Our work inspires effective decisions that advance fairness 500 L'Enfant Plaza SW and enhance the well-being of people and places. Washington, DC 20024 Copyright © March 2022. Urban Institute. Permission is granted for www.urban.org reproduction of this file, with attribution to the Urban Institute. 16 RACIAL, ETHNIC, AND LANGUAGE CONCORDANCE BETWEEN PATIENTS AND PROVIDERS