By Jeff Desmarais & Ranjani Paradise
In honor of National Minority Health Month, ICH is exploring the complexities of race and ethnicity data collection. Any program, organization, or institution seeking to identify and address racial and ethnic health disparities must collect race/ethnicity data on the individuals they serve. However, in practice this proves challenging, as individuals often do not identify with the racial and ethnic categories provided.
ICH partners with the Cambridge Health Alliance (CHA) Zero Disparities Committee to ensure that race and ethnicity patient data are both accurate and meaningful for patients and the hospital. We assist CHA with patient demographic data collection and analysis, which helps CHA address disparities in service utilization and ensure that all patients receive linguistically and culturally appropriate care. We collect race data by asking patients to choose one or more of the categories defined by the federal Office of Management and Budget (OMB):
- American Indian or Alaska Native
- Black or African American
- Native Hawaiian or Other Pacific Islander
These OMB categories are required for state and national reporting and allow us to easily compare the race distribution of our patient population with other healthcare systems’ or communities’. However, CHA’s patient population is extraordinarily diverse, and the concept of race varies widely across cultures. Many of our patients simply do not relate to the OMB race definitions and categorizations. For example, Latino patients and Middle Eastern patients often express confusion when asked to select a race, as many do not see themselves fitting any of the given categories. And while it may seem clear that patients from India should fall under the Asian category, some Indians disagree – to them, Asian indicates East Asian and is distinct from South Asian. Overall, more than 25% of CHA’s patients do not identify with any of the OMB categories and choose “Other” as their race, which calls into question the usefulness of the OMB categories in our increasingly diverse community.
Because of these issues, we also collect detailed ethnicity data to better characterize our patients’ cultural backgrounds. Patients are asked to self-identify their ethnicity (or ethnicities) from a list of more than 150 options (e.g., Algerian, Bangladeshi, Egyptian, Greek, Nepalese, Syrian, etc.). We strive to make these ethnicity options specific enough that patients can select one or more with which they truly identify. Overall, we find that ethnicity data can be far more useful than race data for ensuring we provide culturally appropriate care to everyone we serve.
ICH also works as the evaluator of the Youth First Initiative in Holyoke and Springfield, MA. Youth First is a community-wide teen pregnancy prevention initiative, which was developed with grant funding from the Centers for Disease Control and Prevention. The Initiative is a collaboration between the Massachusetts Alliance on Teen Pregnancy (MATP), the YEAH! Network, and many stakeholders within the two communities, including schools, community organizations, and health clinics. Youth First began in 2010 and its goal is to reduce teen births by 10% by 2015 in Holyoke and Springfield, MA. While the overall teen birth rates in these communities are significantly higher than the state’s, there are also large racial/ethnic disparities within the communities.
Thus, collecting meaningful race/ethnicity data is critical to both improving and documenting improvement in teen pregnancy prevention efforts.
One of the key components of the Youth First initiative is implementing evidence-based sexual health education programs for adolescents. As a part of reporting requirements for funders, we collect race/ethnicity data from program participants via surveys. Collecting this data allows us to understand the populations we are reaching. In addition to the OMB race categories, the CDC requires data for the following ethnicity categories:
- Hispanic or Latino
- Not Hispanic or Latino
Cultural practices, beliefs, values and behaviors, all of which can reflect a person’s identity, can have an impact on a community’s health. Collecting data that taps these identities, if done meaningfully, can serve as an indicator to better understand, explain and ultimately address health inequities. With large Spanish-speaking communities in Springfield and Holyoke, it is critical to use measures that capture their unique and specific identities. Using the race/ethnicity categories often required by funding organizations can prohibit a comprehensive understanding of how communities might view themselves. For example, a Dominican or Puerto Rican teen might not identify with any of the OMB racial categories. Additionally, using the ethnicity category “Hispanic or Latino” can mask differences within Spanish-speaking communities. A teen that grew up in rural Mexico might have very different experiences, behaviors and values than a Puerto Rican teen that grew up in Holyoke. In this project, in order to ensure identity is accurately captured, we added an additional question that includes a much more comprehensive list of 29 different ethnicity categories (for example: Puerto Rican, Caribbean Islander, Dominican, Korean), thus expanding our understanding of identity among youth enrolled in evidence-based programs.
Capturing race and ethnicity data is essential not just for program improvement purposes, but also for ensuring that public health interventions are actively addressing health inequities. In developing alternative ways of measuring this data, it is critical to engage the community during the instrument development phase. It’s often necessary to go beyond the basic race/ethnicity reporting requirements, and delve deeper into communities’ understanding of race, ethnicity, and identity, in order for the data to be meaningful and actionable.
The views expressed on the Institute for Community Health blog page are solely those of the blog post author(s), and do not necessarily reflect the views of ICH, the author’s employer or other organizations with which the author is associated.