Exploring Health Plan Perspectives in Collecting and Using Data on Race, Ethnicity, and Language
July 12, 2012, 12:00:00 AM
Julie Gazmararian, PhD, MPH; Rita Carreón, BS; Nicole Olson, MPH; and Barbara Lardy, MPH
Over the last few decades, there has been increasing attention by leading public and private organizations1-3 and federal agencies4,5 on improving the collection of data on an individual’s race, ethnicity, and language preference (REL), assessing the underlying differences in quality of care among diverse populations,6 and monitoring the progress of culturally and linguistically appropriate programs and services.6-8
Many entities play key roles in obtaining REL data for improving healthcare equity, including health plans, hospitals, providers, community health centers, and employers/plan sponsors.9 Collaborative efforts will be important as the healthcare sector begins to adopt new models of care and increase access based on the Patient Protection and Affordable Care Act of 2010 (PPACA).
Since 2003, America’s Health Insurance Plans Foundation (AHIPF) has collaborated with the Robert Wood Johnson Foundation (RWJF) to assess the progress a number of health plans have made in the collection and use of REL data to reduce racial and ethnic disparities and improve healthcare quality.6,7,10,11 These assessments were conducted through national health plan surveys in 2003, 2006, 2008, and 2010. Based on the 2008 survey findings, health plan in-depth interviews were conducted in 2009 to explore challenges with data collection and use and develop successful strategies to improve care. The interviews provide key stakeholders with a better understanding of the health plans’ decisions to collect and utilize REL data for quality improvement and disparity reduction efforts.
Selection of Plans
Twenty health plans were selected to participate in in-depth interviews based on specific criteria referenced in Table 1. Fifteen health plans, representing the total enrollment of almost 53 million covered lives, agreed to participate and were divided into 2 groups: Health plans that indicated in the 2008 survey that they do collect and use data on the race and ethnicity of their members (n = 10 plans), and those plans that do not collect REL data (n = 5). Participating plans were classified by product (8 commercial, 4 Medicaid, and 3 Medicare plans), as well as size of membership enrollment (Table 2).
Development of Interview Guides
The interview guides included a series of themed questions about REL data collection efforts, leadership support, collaboration with external partners, and challenges to and opportunities for the collection and use of REL information (Table 3). For health plans that were not currently collecting REL data, questions were also asked regarding their reasons to forgo data collection and any existing health equity efforts. Prompts were developed for each question so that consistent probes would be used for each interview.
How Interviews Were Conducted
Interview questions were sent in advance to prepare the appropriate person(s) for the 1-hour interview. The interviews, conducted between June and August 2009, frequently were conducted by a comprehensive team including medical directors, analysts, and communications staff. All health plans were asked if the interview could be audiotaped; 14 of the 15 plans agreed to being recorded.
A summary report for each interview identified key themes based on leadership, organizational infrastructure, data collection and use, and partnerships. Any discrepancies in the interpretation of results were resolved between team members. Since the results are based on qualitative data, the findings are not presented numerically. However, to provide some estimate of how frequently particular themes appeared, terms like most (representing more than half of the participating plans), many (just under half of plans), and several, some, and few (mentioned by at least 2 plans) are used in the Results section.
Plans That Collect REL Data
Commercial, Medicare, and Medicaid plans that collect REL data varied dramatically in population and geographic region served; from 48,000 to more than 26 million members and from a region within a state to nationwide coverage. Collection of REL data began as early as 1985 in 1 health plan and as recently as 2007 in several others (Table 4).
Organizational Leadership and Infrastructure
We found that support from senior leadership is instrumental in initiating and sustaining the health plans’ REL data collection efforts. Advances in data collection often evolved at health plans whose efforts first focused on improving their workforce diversity from activities like conducting cultural competency training or establishing language access services. From these experiences, health plans recognized the importance of gaining a better understanding of their membership, in terms of both how their members seek healthcare, and what cultural norms and beliefs influence their health behaviors and attitudes. Health plans acknowledged that obtaining REL data was an essential step toward advancing quality improvement (QI).
Most of the interviews revealed that considerable time and effort was spent researching, planning, and laying the groundwork for a plan’s REL data collection activities. These efforts included identifying health information technology (IT) capabilities, appropriate software and analytical tools, databases, and storage requirements; determining data categories; establishing privacy and protection policies; training staff to obtain data from members and through proxy methods; and developing health equity reports.
Most health plans also discussed internal policies that support their commitment and accountability to workforce diversity and health equity. For example, 1 plan incorporates health disparities improvement measures into a corporate scorecard that forms the basis of employees’ bonuses and incentives, and into the company’s overall strategic plan. Another plan has cultural competency guidelines to inform culturally appropriate services, language access services, and accommodations
for members with special healthcare needs. Many plans indicated that they have internal policies on confidentiality and use of REL data.
REL Data Collection
Health plans indicated that they used both direct methods, data self reported by members, and indirect methods, such as data available from third parties (eg, Medicaid agency) or from geocoding and surname identification softwares. A few plans mentioned working with external organizations to strengthen indirect estimation methods. All health plans indicated that they consider selfreported REL data to be the “gold standard.” However, most plans reported that it is more difficult to collect REL data directly from their membership since members often do not understand the need for and use of this information.
Health plans described various strategies used to directly collect REL data. Methods include obtaining it at enrollment; during member participation in the plan’s disease management, wellness, case management, and dental programs; in health risk assessments; during calls to obtain missing laboratory information or encourage preventive screenings; during hospital admissions; from member contacts with customer service; and through the member portal of the health plan’s website. One plan mentioned that online newsletters and pop-up messages from its website encourage members to report REL data. Several plans indicated that they provide training to employees with direct member contact (eg, customer service, disease management) to teach them how to collect data and respond to members’ concerns about providing this information. A few plans discussed their efforts to incentivize providers to collect REL data from patients.
All health plans interviewed experience both internal and external challenges to collecting REL data. These challenges include regulatory and enterprise hurdles that prevent employer/plan sponsors from sharing enrollee data; the plan’s IT system capability to store and reconcile multiple data categories; the ability to share REL information across divisions within the plan and more broadly across the healthcare system; legal issues (real and perceived); member privacy concerns; the lack of provider understanding of the importance of collecting REL data; and requirements from compliance organizations. Many plans expressed a strong desire to educate the general public about the importance of collecting REL information. Furthermore, a few plans mentioned that competing priorities for limited resources and the impact of the recent economic decline impede data collection efforts.
The health plans expressed a strong interest in having standardized categories and consistent data metrics so data would be comparable, meaningful, and actionable across the healthcare sector. Most health plans also stated that they could not accurately determine the validity or completeness of REL data from available secondary sources. The lack of accurate and reliable REL data collected or obtained by federal and state agencies diminishes the plans’ ability to use these data to effectively
identify and act on disparities.
Use of Data to Improve Quality and Healthcare Equity
In 2009, most of the participating plans were in the early stages of using REL data to identify health disparities. The health plans studied noted that the limited availability of accurate REL data affects their ability to differentiate health outcomes by subgroups. Plans used Healthcare Effectiveness Data and Information Set (HEDIS, registered trademark of the National Committee for Quality Assurance) measures, stratified by race and ethnicity, to help prioritize and develop appropriate QI initiatives. For example, a commercial plan incorporated REL data into electronic medical records, linking quality measures to identify persistent disparities among racial and ethnic groups. As a result, numerous small-scale interventions to improve chronic care management and preventive screenings have demonstrated improvements in diabetes management and mammography screening rates among Latinos and reduced emergency visits among Latino members with asthma. Another commercial health plan found differences in the severity of diabetes illness among African American members after in-depth data analysis. As a result, the plan implemented a program to improve diabetes screening rates among African American members using provider-based interactions and culturally appropriate member information. Other examples of REL data use and performance measurement data analyses included targeted intervention programs for breast, cervical, and colorectal cancer screening, pediatric asthma, blood pressure, and cholesterol.
Language Access Services
Interviews found that most health plans support members’ language needs using several approaches, including access to telephonic interpreting, use of bilingual and multicultural staff, availability of member materials in several languages, and community outreach support. Health plans with advanced language access programs have evaluated the quality of these services and strengthened their workforce diversity programs. For example, a few plans offer basic Spanish classes to customer service representatives. One plan indicated that evaluation of its interpreter services led to improvement in quality and efficiency of services, which reduced overall costs. Another plan that primarily serves Medicaid beneficiaries assessed the quality of the plan’s interpretation services through provider and patient feedback about their experiences and satisfaction with interpreter services (contracted and in-house). As a result of these and other quality assessment activities, the plan shifted from contracting some of its language services to hiring and training interpreters for commonly spoken languages, which resulted in reduced costs and improved quality fromterminating poor-performing contractors.
Most plans indicated they are working with external partners to improve REL data collection and reduce gaps in care. The health plans collaborated primarily with non-profit community groups serving uninsured or minority populations, such as federally qualified community health centers and faith-based organizations, in addition to universities and professional associations. Through these external partnerships, health plans have increased awareness about the need to collect and report REL data for QI efforts. Furthermore, some plans stressed the importance of sharing best practices and lessons learned from program interventions with key stakeholders. Although few interviewed health plans have successfully collaborated with employers/plan sponsors on health disparities, plans expressed a strong interest in working with these groups to obtain employee REL data. Several plans suggested increasing purchaser engagement by providing detailed information about the significant costs of healthcare disparities. One plan suggested that employer engagement could lead to changes in enrollment forms to substantially increase the REL data collection rate.
Health Plans That Do Not Collect REL Data
Three of the 5 “non-collecting” plans represented small to medium-size membership at the time of the interview and all (n = 5) are distributed across several regions (Table 4).
Organizational Leadership and Infrastructure
Executive leadership at all 5 health plans realized the value of obtaining REL information to identify and address health disparities. The primary reason these plans do not collect data is that they currently serve a relatively homogeneous membership. However, the organizations are closely monitoring changes in their market demographics and external demands from key stakeholders, such as state agencies and accrediting bodies. Recognizing that plan membership composition rapidly changes, the majority of “non-collecting” health plans are beginning to explore REL data collection. These plans are currently examining if, when, and how to best collect these data, as well as identifying resources required for integrating such efforts within the organization’s infrastructure.
Initiating REL Data Collection
Plans that currently do not collect REL data are using US Census data, surveys, and other local data to help strengthen their language services and identify resources to better serve their memberships. Health plans that are in the initial stages of assessing their infrastructure capabilities have determined that obtaining stakeholder buy-in and collecting sufficient data will require multiple approaches. Since the 2008 survey and during the interview process, we found that some health plans have decided to begin to collect these data at enrollment, during health risk assessments, and/or with support of their provider networks. Other health plans were determining the best method for collecting these data and how to simplify the process through partnerships with providers and state agencies. We provide examples of their initial efforts.
Plans that do not collect REL data expressed concerns similar to health plan “collectors.” These challenges include costs of adapting IT systems; protecting members’ confidentiality; meeting new state and federal agency requirements; establishing policies and procedures; legal issues; and other competing priorities that limit allocation of resources and dedicated staff. Many health plans anticipate resistance from stakeholders and seek assistance in finding ways to ease member, employer, provider, and even staff concerns. Clear communication about the importance of collecting these data can boost support for REL data collection and use efforts.
Several health plans expressed a strong interest in learning about best practices regarding the most efficient and costeffective methods for collecting REL data, as well as specific challenges that other health plans have faced in data collection and strategies to overcome barriers. These plans believe that sharing this information is essential, especially among small or regional plans faced with limited resources.
Efforts to Improve Quality and Healthcare Equity
“Non-collecting” health plans demonstrated a longstanding commitment to ensuring that members receive highquality healthcare, despite not being able to identify health disparities using self-reported data. One health plan, which at the time of the survey did not collect data, is now using geocoding software to identify gaps in access to care within its market, and then link these data with quality measures to identify opportunities for improvement. The plan is confident that this activity will prepare the company to take appropriate action when it begins collecting REL data.
Health plans are interested in linking REL data with quality measures to identify gaps in both preventive services and chronic care, as well as to increase access to preventive screenings in specific communities through partnerships and outreach efforts. One health plan reaches out to providers to assess how the plan can best support and connect providers and members with community resources.
Language Access Services
All health plans interviewed provide language access services, such as translating print materials, providing telephonic interpreting, supporting cross-cultural training and language classes for employees, and recruiting multicultural staff. Some plans also support members’ language and cultural needs by increasing access to ethnically diverse network providers that have the linguistic capability to better serve members with limited English proficiency.
Several of the plans demonstrate their commitment to serving low-income and vulnerable populations, primarily through involvement with community-based organizations and other external partners. One plan works closely with providers to help individuals from immigrant communities learn how to navigate the US healthcare system. Another plan indicated that it interacts with employer groups regularly by performing and sharing its analyses on specific disease and immunization rates among employer populations. Others collaborate with community-based organizations, community health centers, and local elementary schools to support health equity and prevention/wellness initiatives.
Industry Demands and Health Reform
Lastly, all health plans that participated in the in-depth interviews were asked to discuss internal or external demands on their organization, including the debate on health reform and resource limitations. All plans mentioned closely monitoring federal and state regulatory and accreditation bodies for changes in requirements related to REL data collection and use. The health plans also expected that state legislative activities will increase awareness and potentially ease the public’s legal concerns. Several plans indicated that improving their accreditation survey score was a key driver to initiating their REL data collection efforts. Formalizing a plan to meet accreditation standards also helps the organization address the diversity of its provider network. Some plans also mentioned participation in the National Business Coalition on Health’s eValu8 tool—an assessment instrument for purchasers to examine how health plans are improving quality of care for their employees—as another incentive to engage in health disparities efforts, and potentially demonstrate a competitive advantage in the marketplace.
The health plans were asked if the debate on health reform affects the way their organizations are approaching disparities efforts. One plan mentioned using the experience of Massachusetts to assess how individuals from diverse backgrounds and socioeconomic status, who were previously uninsured, are now accessing the healthcare system. Because these individuals have diverse quality issues and variable costs, these models gave the plan a better understanding of how to adapt approaches to help members better navigate the healthcare system.
The interviews provide important and insightful information about the health plans’ data collection activities in 2009, prior to the passage of the PPACA (Table 5). Several of the themes identified have been substantiated in other research, particularly the need for standardized categories for REL2,5 and the lack of valid and reliable data.2 However, our data provide new insight surrounding leadership support, strategies to mitigate disparities, and challenges to REL data collection efforts, such as establishing work processes and organizational infrastructure to collect and store information to be used in quality improvement initiatives. Our study also revealed that all health plans were closely monitoring environmental changes—both geographic and regulatory—in preparation for allocating the necessary resources needed to collect data on race, ethnicity, and language. Many plans are concerned about consumer sensitivity around collecting this information and expressed a desire to have other organizations—such as providers, employers/plan sponsors, community-based organizations, and government agencies—assist with communicating the need for this information. A national educational public health campaign to inform consumers and their families about the importance of collecting demographic information to address disparities was a consistent recommendation both from plans that currently collect and from plans that are preparing to collect these data. Furthermore, all participating health plans demonstrated commitment to improving healthcare equity. The health plans analyze the utilization of services and quality measures stratified by REL characteristics to identify disparities and target culturally and linguistically appropriate interventions.12
Despite the valuable information obtained from this study, it should be noted that findings are based on a small number of health plans. Due to time, funding, and available resources, we reached out to 20 health plan respondents of the 2008 survey to volunteer their time to share their experiences, and 15 agreed. While the qualitative study represents the experiences of 15 health plans that represent a total of about 53 million covered lives, the findings cannot be generalized based on the 10 health plans collecting data and 5 plans that reported not collecting data on race, ethnicity, and language at the time of the interviews.
Moving forward, our shared learnings from the in-depth interviews highlight the need for new partnerships and coordinated efforts to improve healthcare equity through disseminating best practices and tools to help expand data collection and disparity-reduction activities.13,14 The interviews and health plan surveys also noted barriers similar to those identified by employers and providers, including the cost associated with adapting IT systems to accommodate new functions, such as new data fields, appropriate software and analytical tools, and the lack of standard codes for race and ethnicity.15-17 Efforts to standardize collection and reporting of data should
be aligned with recent national reports, federal and state activities, and new voluntary standards developed by HHS and accrediting bodies.5,18-20 The implementation of PPACA and other ongoing activities by federal and state agencies and national and local learning networks provide a tipping point for developing a consensus on how data should be collected and offer opportunities to collaboratively eliminate health disparities.21,22 With the recent release of guidelines for collecting data on race, ethnicity, language, sex, and disability status for population surveys and the Secretary’s Report to Congress on HHS’ evaluation of such data and approaches to address health disparities, as required by section 4302, health plans can work with federal and local state agencies to collect data, measure progress, and reduce disparities through innovative and effective programs and interventions.20,23
Lastly, the health plans’ willingness to collaborate with new partners and share strategies to collect REL data and reduce disparities was evident from the interviews. Opportunities exist to collaborate with employers and purchasers to improve the extent and quality of REL data and lead to implementing culturally appropriate programs in the workforce.2 Multiple efforts can foster employer involvement, such as educating employers and employees about racial and ethnic healthcare disparities; promoting the business case for disparities reduction; emphasizing the ethical underpinnings and legality of race and ethnicity data collection; and creating more visible and cohesive national leadership around these issues.2,241. Smedley BD, Stith AY, Nelson AR, eds. Unequal Treatment: Confronting Racial and Ethnic Disparities in Healthcare. Institute of Medicine, Washington, DC: The National Academies Press; 2002.
2. Higgins PC, Taylor EF. Measuring racial and ethnic disparities in health care: efforts to improve data collection (policy brief). Mathematica Policy Research, Inc. http://www.mathematica-mpr.com/publications/pdfs/health/disparitieshealthcare.pdf. Published May 2009. Accessed May 2011.
3. Regenstein M and Endres E. Aligning forces for quality: local efforts to transform American health care. Aligning Forces for Quality National Program Office, Center for Health Care Quality in the George Washington University’s Department of Health Policy. http://www.rwjf.org/files/research/55128.pdf. Published August 2010. Accessed January 2012.
4. Agency for Healthcare Research and Quality. 2010 National Healthcare Disparities Report. Rockville, MD: US Depart of Health and Human Services, Agency for Healthcare Research and Quality; March 2011. AHRQ Pub. No. 11-0005.
5. Ulmer C, McFadden B, Nerenz DR, eds. Race, Ethnicity, and Language Data: Standardization for Health Care Quality Improvement: Subcommittee on Standardized Collection of Race/Ethnicity Data for Healthcare Quality Improvement. Institute of Medicine, Washington, DC: The National Academies Press; August 2009.
6. America’s Health Insurance Plans. Health Insurance Plans Address Disparities in Care: Highlights of a 2004 AHIP/RWJF Quantitative Survey Collection and Use of Data on Race and Ethnicity. http://ahip.org/Race-Ethnicity-Language. Published 2004. Accessed January 2012.
7. America’s Health Insurance Plans. Collection and use of race and ethnicity data for quality improvement: 2006 AHIP-RWJF survey of health insurance plans. http://www.rwjf.org/files/publications/other/2006AHIP-RWJFSurvey.pdf. Published November 2006. Accessed January 2011.
8. Robert Wood Johnson Foundation. Finding answers: disparities research for change: grants portfolio. http://www.solvingdisparities.org/media/file/finding_answers_grants_portfolio.pdf. Published November 2011. Accessed January 2012.
9. High-Value Health Care Project. Lessons in the acquisition of race, ethnicity, and language data by health plans. Washington DC: Engelberg Center for Health Care Reform at Brookings Institution; February 2010.
10. Escarce JJ, Carreón R, Veselovskiy G, Lawson EH. Collection of race and ethnicity data by health plans has grown substantially, but opportunities remain to expand efforts. Health Aff (Millwood). 2011;30(10):1984-1991.
11. Lawson E, Carreon R, Veselovskiy G, Escarce JJ. Collection of language data and services provided by health plans. Am J Manag Care. 2011;17(12):e479-e487.
12. Angeles J, Somers SA. From policy to action: addressing racial and ethnic disparities at the ground level. Center for Health Care Strategies.
http://www.chcs.org/publications3960/publications_show.htm?doc_id=519202. Published August 2007. Accessed August 2010.
13. Martin C. Reducing racial and ethnic disparities: a quality improvement initiative in Medicaid managed care. Center for Health Care Strategies, 2007. http://www.chcs.org/publications3960/publications_show.htm?doc_id=440684. Accessed August 2011.
14. Hasnain-Wynia R, Pierce D, Haque A, et al. Health Research and Educational Trust Disparities Toolkit. http://www.hretdisparities.org/. Published 2007. Accessed August 2011.
15. Higgins PD, Au M, Taylor EF. Reducing racial and ethnic disparities in health care: partnerships between employers and health plans. Policy Brief. Mathematica Policy Research, Inc. http://www.mathematicampr.com/publications/pdfs/health/reducedisparities.pdf. Published July 2009. Accessed July 2011.
16. America’s Health Insurance Plans. Health Insurance Plans Address Disparities in Care: Challenges and Opportunities, http://ahip.org/Race-Ethnicity-Language/. Published 2004. Accessed January 2012.
17. National Health Plan Collaborative. The National Health Plan Collaborative Toolkit. http://www.rwjf.org/qualityequality/product.jsp?id=33960. Published September 2008. Accessed August 2010.
18. National Quality Forum. A Comprehensive Framework and Preferred Practices for Measuring and Reporting Cultural Competency: A Consensus Report. Washington, DC: National Quality Forum; 2009.
19. National Committee for Quality Assurance. Multicultural health care distinction. http://www.ncqa.org/tabid/1195/Default.aspx. Published July 2010. Accessed August 2011.
20. US Department of Health and Human Services. Office of the Assistant Secretary for Health, Office of Minority Health. Notice of Availability of Proposed Data Collection Standards for Race, Ethnicity, Primary Language, Sex, and Disability Status Required by Section 4302 of the Affordable Care Act. Federal Registry. June 30, 2011:76(126). http://www.gpo.gov/fdsys/pkg/FR-2011-06-30/pdf/2011-16435.pdf. Accessed August 2011.
21. National Health Plan Collaborative. Reducing racial and ethnic disparities and improving quality of health care: phase one summary report. http://www.rwjf.org/files/publications/other/NHPCSummaryReport2006.pdf. Published November 2006. Accessed June 2011.
22. Patient Protection and Affordable Care Act of 2010, Pub. L. No. 111-148 (March 23, 2010) and Healthcare and Education Reconciliation Act of 2010, Pub. L. No. 111-152 (March 30, 2010).
23. Sebelius, K. Report to Congress: Approaches for identifying, collecting and evaluating data on health care disparities in Medicaid and CHIP. Washington: U.S. Department of Health and Human Services. Published September 2011. Accessed January 2012.
24. National Business Group on Health and HHS Office of Minority Health Launch Initiative to Reduce Racial and Ethnic Health Disparities [press release]. http://www.businessgrouphealth.org/pressrelease.cfm?ID=101. Published February 11, 2008. Accessed August 2011.Acknowledgment
The authors would like to thank the health plan professionals who took the time to participate in the interviews, German Veselovskiy, senior policy research associate to the project, and Claire Gibbons, project officer, at the Robert Wood Johnson Foundation for her guidance and support in preparing this analysis.
Author Affiliations: From Department of Epidemiology (JG, NO), Rollins School of Public Health, Emory University, Atlanta, GA; America’s Health Insurance Plans (RC, BL), Washington, DC.
Funding Source: This paper is part of the AHIP Foundation’s project, Health Plan Collection and Use of Race, Ethnicity and Language Data: Challenges and Opportunities, funded by the Robert Wood Johnson Foundation.
Author Disclosures: Ms Carreón and Ms Lardy report employment and receipt of grants from America’s Health Insurance Plans. The other authors (JG, NO) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (JG, RC, BL); acquisition of data (JG, RC, NO); analysis and interpretation of data (JG, NO); drafting of the manuscript (JG, RC, NO); critical revision of the manuscript for important intellectual content (JG, NO, BL); statistical analysis (JG); obtaining funding (BL); administrative, technical, or logistic support (JG); and supervision (JG, BL).
Address correspondence to: Julie Gazmararian, PhD, MPH, Associate Professor, Department of Epidemiology, Rollins School of Public Health, Emory University, 1518 Clifton Rd, Atlanta, GA 30322. E-mail: email@example.com.