Demographic Data Collection in Healthcare: Best Practices for Race and Ethnicity
Accurate demographic data for health disparities populations is a fundamental component of healthcare equity and disparities resolution.1 Using demographic data to disaggregate key healthcare metrics unearths disparities in care quality, safety, and access among and between patient populations. Visualizing disparities provides information, which can be used to develop and test interventions and scale those with promising results.
Standardizing data for the purpose of reporting and exchange with other healthcare organizations is a desirable objective both nationally and regionally and is a top priority for the Centers for Medicare & Medicaid Services (CMS) and the Office of the National Coordinator of Health Information Technology (ONC).2,3 Of the many demographic data points we can collect, the healthcare industry is most advanced in its infrastructure for the collection of race, ethnicity, and language data. (REAL, REL, and REaL are all used in the field.)
Standardized minimum categories and best practices for REAL data collection are now widely accepted as industry norms.4 Though organizations throughout the industry have much work to do to standardize their categories and accurately collect the data.5 The collection of REAL data is the preferred and ideal starting point for most healthcare delivery organizations beginning the disparities-resolution journey.
For healthcare delivery organizations intending to advance REAL data efforts, there are three broad areas of consideration: 1) REAL categories, 2) data collection and validation, and (3) enterprise and patient communications.
REAL Categories: Minimum Categories, Subcategories, and Extraneous Categories
REAL data category options can be demystified by looking to the ONC as a source of guidance. The ONC’s guidelines start with minimum categories for ethnicity and race as defined by the Office of Management and Budget (OMB). The categories are often referred to as the OMB 2+5. Every racial and ethnic group in the world may be “rolled up” to the OMB 2+5. The rollup configuration is found within the Centers for Disease Control (CDC) Public Health Information Network Vocabulary Access and Distribution System (PHIN VADS) or PHIN Library.7
Office of Management & Budget Standard Categories (OMB 2+5)
Ethnicity:
- Hispanic/Latino
- Not Hispanic/Latino
Race:
- American Indian or Alaska Native
- Asian
- Black or African American
- Native Hawaiian or Other Pacific Islander
- White
In addition to the OMB 2+5, some organizations may want to offer patients subcategories. While having data on granular categories of race and ethnicity can provide health systems with additional insights when utilizing the data, there is not a one-size-fits-all solution. Many systems opt for a stepwise approach, first mastering collection of the minimum categories and then adopting carefully selected subcategories in later years.
There are two major considerations for determining when to adopt more granular race and ethnicity categories. First is evidence of diverse populations beyond the minimum categories within the healthcare catchment area or among the current patient population. Second are state, regional, or programmatic requirements or strong recommendations for collecting more granular data. When adding subcategories, the electronic medical record should be configured so the subcategories can be rolled-up to the OMB 2+5 for reporting, data exchange, and metric stratification purposes.
Additional acceptable categories include “Declined” to indicate when a patient did not want to answer and “Unknown” or “Unavailable” to indicate the patient was not able to communicate their preference.
Data Collection and Validation: Self-Reporting, Staff Training, and Validation
Patient self-reporting during the registration process is the gold standard for REAL data collection. Teaching staff about the value of REAL data and why it is collected provides the rationale for data collection and conveys that they are an important part of a broader enterprise effort to improve patient care. Additionally, staff should receive training on how to obtain data from patients. Many healthcare organizations provide staff with a script that helps them become acclimated to patient self-reported data collection. Staff training in REAL data collection should occur at orientation and then once per year thereafter.
The best method of validating REAL data is confirming race and ethnicity assignments during each patient visit. Be intentional about your validation method and note that collecting the data only once provides no opportunity to correct input errors and denies patients the opportunity to change their self-reported selection. Patients may change their REAL assignments for a wide variety of reasons.
Enterprise and Patient Communications: Convey Commitment from the Top and Transmit Intentions to Patients
Nothing speaks louder to health system employees than a communication from the CEO and the governing board. Executives who convey the importance of REAL data collection and health equity to providers and staff can expect more reliable and consistently collected data.
Finally, initiate an information campaign for patients conveying that your organization collects REAL data because the people in your organization care about achieving high outcomes for all patient populations. Many healthcare organizations have successfully deployed the “We Ask Because We Care” campaign developed in 2010 for The Robert Wood Johnson Foundation (RWJF). Campaign materials are still available free of charge on the RWJF Aligning Forces for Quality webpage.8 In addition, among organizations that have implemented patient campaigns, staff report feeling supported and not alone in their efforts to obtain REAL data from patients.
Notes
1. National Institute on Minority Health and Health Disparities. (2021, May 5). Overview. Retrieved from https://www.nimhd.nih.gov/about/overview/
2. Centers for Medicare and Medicaid Services. (2022, June 28). Promoting interoperability programs. Retrieved from https://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms
3. The Office of the National Coordinator for Health Information Technology. (n.d.). ONC’s Cures Act final rule supports seamless and secure access, exchange, and use of electronic health information. Retrieved from https://www.healthit.gov/curesrule/
4. James, C.V., Lyons, B., Saynisch, P.A., Hudson Scholle, S. (2021, October 26). Modernizing race and ethnicity data in our federal health programs. To the Point (blog), Commonwealth Fund. Retrieved from https://www.commonwealthfund.org/blog/2021/modernizing-race-and-ethnicity-data-our-federal-health-programs
5. Institute for Diversity and Health Equity. (n.d.). Why collect race, ethnicity, and primary language. American Hospital Association. Retrieved from https://ifdhe.aha.org/hretdisparities/why-collect-race-ehtnicity-language#:~:text=Most%20hospitals%20(82%20percent)%20currently,departments%20within%20the%20same%20hospital
6. Centers for Disease Control and Prevention. (n.d.). Public health information network vocabulary access and distribution system (PHIN VADS). Retrieved from https://phinvads.cdc.gov/vads/ViewValueSet.action?id=67D34BBC-617F-DD11-B38D-00188B398520
7. Aligning Forces for Quality. (n.d.). We ask because we care posters/tent cards. Robert Wood Johnson Foundation. Retrieved from http://forces4quality.org/node/4185.html
Lisa R. Sloane (lisa@moreinclusivehealthcare.com) is the founder and CEO of More Inclusive Healthcare (MIH). The company is a recognized leader in training and education solutions designed to accelerate the resolution of health disparities.