This blog was originally published by The Leadership Conference on Civil and Human Rights and co-authored with Julia Liou and Thu Quach.
People of different racial and ethnic backgrounds experience life differently. These differences can come in the form of joyful shared experiences and cultural practices and in the form of negative effects of historical oppression such as discrimination or racism. Acknowledging these differences and their impacts are critical to creating a better world for all of us.
In the health context, the experiences of patients of different racial and ethnic backgrounds are similarly not uniform. Measuring health outcomes in the aggregate masks the experiences of discrete communities and obscures the prevalence of health conditions that disproportionately impact certain populations. On the contrary, taking the experiences of people of different racial and ethnic backgrounds into account when planning for and delivering care has been shown to improve health outcomes and understanding of health care access.
Disaggregated Data as a Tool to Achieve Health Equity
Good data is paramount to advancing health equity. Collecting demographic information broadens our understanding of both positive and negative health occurrences in particular communities. Disaggregating demographic data involves breaking down larger, more commonly used racial signifiers, such as Asian or Latino, into more specific, granular subgroups, such as Vietnamese, Hmong, or Venezuelan. We have long collected data (though imperfectly) on larger racial and ethnic groups. We must move to disaggregating that demographic information to achieve the promise of equitable, individualized health care for all.
On the flip side, we have seen how aggregated data can be harmful. Lack of data on particular communities of Asian Americans, Native Hawaiians, and Pacific Islanders (AANHPIs) can mask disparities and further perpetuate a model minority myth — a damaging narrative that overgeneralizes and discounts the challenges faced by this population as well as within subgroups. In fact, data from the California Health Interview Survey show significant differences in health for select Asian subgroups, such as greater prevalence of high blood pressure, asthma, heart disease, and delayed medication use for Filipinos compared to Asians overall.
The COVID-19 pandemic made abundantly clear that access to disaggregated patient data can have an immediate impact by providing timely interventions and care to disproportionately impacted communities. For example, in late 2020, Santa Clara County in California collected disaggregated data on COVID-19 rates and found that Vietnamese and Filipino residents were being hit harder than any other Asian American groups. Similarly, around that time, Asian Health Services, a federally qualified health center in Alameda County, started collecting disaggregated data for those coming to receive COVID-19 testing and found that Vietnamese residents had nearly twice the case rates of the aggregated Asian Americans. This information informed the COVID-19 response team to conduct targeted in-language outreach and education in areas like Little Saigon of Oakland. Soon after, the positivity rates leveled out for this group, demonstrating the significance of granular data in informing real-time interventions and leading to major public health impacts.
State and Federal Agencies Must Commit Resources to Disaggregated Data Efforts to Realize the Promise of a Data-Driven Approach
Disaggregating demographic data in health programs requires thoughtful planning and implementation to be effective. State and federal agencies will need to invest resources in updating technology and infrastructure to reflect disaggregated demographic identifiers. Workers on the provider-, program-, or plan-side who interact with electronic health records, databases, or patient intake systems will need to be trained on how to effectively communicate with individuals about reporting disaggregated race and ethnicity in a way that protects their rights to privacy and how disaggregated race and ethnicity are shared publicly. Where there are small numbers of individuals in a certain subpopulation, sharing that information even as part of a de-identified public analysis can risk exposing personal information. Federal and state governments should issue updated privacy guidance and expectations for sharing public analysis of disaggregated data that reflect the risks of a more technologically advanced, racially and ethnically diverse world.
Plans, providers, and programs will also need to involve stakeholders from a diverse array of racial and ethnic communities on how to implement, develop, and update disaggregated data collection practices. Community feedback and engagement ensures that data collected is meaningful both to health systems and to the communities served by the data collection. Notably, there is high acceptance of demographic data collection and use to advance health equity among individuals of all races and ethnicities. We can further improve the agency and autonomy of individuals by sharing data with impacted communities and asking for input on data collection and reporting — while also improving buy-in to demographic data collection efforts and enhancing quality of data.
States and federal agencies have made recent strides to advance disaggregation of data. In California, AB1726 was signed into law in 2016. This bill required the California Department of Health to collect and release disaggregated data on AANHPI populations in everything from rates of major diseases to leading causes of death to pregnancy rates to housing status. New York followed suit in 2021, when Governor Hochul signed legislation requiring state agencies’ race/ethnicity questionnaires to include disaggregated response options for AANHPIs. Most recently, the Health Resources and Services Administration (HRSA), which oversees federally qualified health centers (FQHCs), has started asking FQHCs to collect “expanded sub-categories for race and ethnicity to better reflect the diversity of health center patients…”
Despite these developments, overall implementation of disaggregated data policies remains lacking. This could be a result of the dearth of investment in critical infrastructure, such as technology systems, training, resources, and community engagement, none of which is required by the aforementioned laws and policies. Costs, methodology, and capacity are often cited as barriers and undue burdens to effective implementation. Yet health care spending remains high, totaling 18.3 percent of our nation’s gross domestic product in 2021 — with estimates of health inequities accounting for $320 billion in health care spending with projections of $1 trillion in annual spending by 2040 if left unaddressed. Our health care system, and indeed our society, can no longer afford health inequities. It is imperative that we hold policymakers and agencies accountable to effectively implement these laws.
Shifting policies and programs from a “one size fits all,” aggregated approach to a more individualized, more equitable disaggregated approach will require surfacing these diverse differences. Appropriate and targeted allocations to meet community needs will not be realized until we can uncover these inequities. Providers, plans, and agencies must lean into the significant technological advances of our time to develop the capacity to uncover the multiple dimensions and nuances that are often hidden. In order to advance health equity, we must continue to lean into the power of disaggregated data to pave the pathway ahead to ensure long overdue resources to our communities who need them the most.