Where Challenges Lie in Health Disparities, Opportunities Exist for Health Equity

“What challenges have you encountered in collecting data on health equity?” the first speaker Christopher Diehl, PharmD, MBA, manager of clinical pharmacy programs, Excellus BlueCross BlueShield, asked attendees.

Hanna Lee-Brown, PharmD, CPHQ, director of pharmacy benefits, Healthfirst, also spoke during the morning session, “Challenges and Opportunities in Collecting Data to Address Health Disparities,” on day 1 of AMCP 2023, and together the pair addressed the difficulties inherent in collecting data on health equity and care disparities. They also proposed strategies for managed care organizations (MCOs) to consider that may aid them in gathering the data they need to start to overcome care gaps.

“What is health equity?” Diehl also asked. Everyone has their own definition of what health equity is, and because of this, health equity is going to look different to every person, he added. Therefore, it’s important that MCOs and the people they task with gathering data on health equity be open to change throughout the process.

How can data help to strengthen health equity and what types of data can have a meaningful impact? It’s first important to know where data come from and then where is it going.

Direct data, for example, are collected from the patient or whomever is observing them; this is the gold standard for health equity, Diehl explained. This type of data is best used for focused interventions and on the individual-patient level. Advantages include the accuracy and reliability of these and that they can be collected and adjusted as needed, with minimal bias. Disadvantages include the lack of a large sample size and generalizability to a wider population.

Indirect data (or imputed data) are data inferred from data already available, such as US Census data (ie, zip code, median income, education level, language) and a person’s name; this type of data is best used for population-level analysis and for identifying potential hotspots for intervention. Advantages include large amounts of data being readily available and the ability to carry out statistical analysis for a larger population health impact. Disadvantages are the potential for bias and that data may not be current or may require adjusting for usability.

Collecting data on race, ethnicity, and language is an area influenced by this conflict between direct and indirect data, Diehl explained, because there is a lack of data standardization and of infrastructure to use and analyze patient information. Enrollment files, provider electronic medical records, and health risk assessments are just a few examples of both direct and indirect data sources.

But therein lies opportunity for change, he emphasized. For example, consider following Office of Management and Budget (OMB) categories for data collection: American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or Other Pacific Islander, White, Hispanic or Latino, and Not Hispanic or Latino. Stakeholders may also want to consider expanding upon these categories, as long as they can ensure all responses can be mapped back to them.

Another opportunity for progress is to consider including multiple questions when screening for language, such as languages spoken in the home and level of English proficiency. Again, allow for multiple responses, such as if a member speaks several language or has a multiracial/ethnic background, Diehl stated. The OMB is considering combining race and ethnicity into 1 question and adding the categories Middle Eastern and North African, and the agency has already created an interagency group to improve federal race and ethnicity standards.

One final challenge Diehl noted is the infrastructure used to collect these data, because addressing and collecting health disparities data is so resource intensive, involving analytics support, data governance, and education and training. Opportunities for analytics support in the health equity space include educating analysts so they can become proficient in health equity data analysis and providing them with the infrastructure to receive and house the data they do collect. For data governance, it’s important to review data on a continuous basis and to consider having a committee that ensures appropriate access and use of the data collected. Lastly, for education, collaboration is key; for internal employees, this entails training them on the impact of health inequity and encouraging them to share ideas on how to use data, and externally, this encompasses leveraging relationships with community stakeholders.

Lee-Brown picked up the discussion from there, and she focused on the interrelationship between social determinants of health (SDOH) and geography, and how the 2 factors influence health equity outcomes. She started by noting the factors that are included under the umbrella of SDOH, which are education access and quality, health care access and quality, neighborhood and built environment, social and community context, and economic stability.

“This is a really big national priority,” she reiterated. “The conversations around SDOH, there is really a lot of exciting stuff going on here.”

Less is known about geography’s influence on health disparities, so to start, Lee-Brown provided some examples of factors influenced by geography that can go on to have a negative impact on health equity: food availability, or food deserts; demography (study of statistics); air and water quality; physical access to health care; access to public transportation; cultural practices; socioeconomic resources; attitudes; and lifestyle factors.

For example, according to 2019 data on poverty rates in New York City, the approximate 60% poverty rate in the Bronx, the highest of the 5 boroughs, far outran that of the 30% in Staten Island or the 40% in Queens. In addition, Hispanic individuals of any race had the highest poverty rate in NYC vs non-Hispanic White persons who had the lowest, and there was close to a 65% poverty rate among individuals with less than a high school degree vs almost 17% among those with a bachelor’s degree or higher.

Challenges exist, however, in pairing SDOH data with geographic data, Lee-Brown stated. “SDOH really influences a lot of geographic data. It’s really important that we focus on these 2 things together as a whole.”

For instance, SDOH and geographic data come from a wide variety of sources—such as health plan enrollment, accountable care organization member screenings, providers-reported data, and publicly available and purchasable data—and the collection of both is challenging. This is due to a lack of standardization, lack of provider participation, member/patient mistrust, internal readiness, and collection requirements.

Opportunity for change here exists in the utilization of robust data.

The examples of robust data that she specifically mentioned were universal assessment, hot spotting, predictive analytics, and group-based risk identification, all of which are very much tied together, she added. She further highlighted that within hot spotting, tackling food insecurity can be furthered along by partnerships with community organizations on zip code–focused and localized support initiatives. And for group-based risk identification, this might encompass utilizing the Medicare Advantage Value-Based Insurance Design Model to work toward making prescriptions affordable for a wider population and having targeted SDOH support.

Teamwork represents another opportunity for change, and it may be the most important one. To accomplish this, leverage all points of member contact and establish robust data governance; educate and incentivize provider partners; train on how to appropriately collect sensitive data; and always reinforce the importance of understanding and addressing inequity.

“It really boils down to teamwork and education; making sure that the people you work with understand they are your allies and partners,” Lee-Brown underscored. “Overcoming challenges takes teamwork.”