How Different Types of Patient Data Are Used to Improve Health Equity

Pharmacists at the Pharmacy Quality Alliance (PQA) 2023 Annual Meeting discussed the role of data collection and analysis in the movement toward health equity.

At the Pharmacy Quality Alliance (PQA) 2023 Annual Meeting, pharmacists talked about the different types of data that are collected and analyzed, and how data can be used to improve medication use quality.

The presenters were:

  • Mason Johnson, PharmD, executive fellow at Academy of Managed Care Pharmacy
  • Matthew Dinh, PharmD, senior director of pharmacy quality care and experience at SCAN Health Plan, and
  • Hannah Lee-Brown, PharmD, RPh, CPHQ, director of pharmacy benefits at Healthfirst.

Johnson kicked off the session by clarifying key terms often confused in discussions: health equity, social determinants, social risks, and health disparities. Social risks encompass adverse social conditions that contribute to poor health outcomes, such as food insecurity and housing instability. It's important to note that individuals may face multiple social risk factors, further increasing the likelihood of poor health outcomes.

Health disparities, on the other hand, focus differences in health outcomes affected by demographic factors such as race, gender, income, or geographic region. Johnson emphasized the significance of examining social determinants of health (SDOH), which encompass various conditions individuals are born into, including their work and living situations, age, economic, political, environmental, cultural, and social policies. As Johnson said, addressing these factors positively or negatively impacts health and ensures that everyone has a fair opportunity to achieve optimal health.

The Health Plan’s Role in Health Equity

Looking at the health plan’s role in health equity, Johnson stressed the importance of diversity in clinical data trials, as it enhances the applicability of findings to broader audiences.

“The more diversity that we have in clinical data, the more applicable it can be to all audiences that it reaches,” Johnson said. “We can also augment available data with diversity in that data, and with the use algorithms and artificial intelligence, we can also address it. There are sometimes unconscious or implicit bias that goes on with those, so acknowledging that there are some biases that may be an algorithms and artificial intelligence…is another key aspect.”

Aside from providing the right data, health plans are also responsible for several aspects of the formulary process, benefit design, and expanding patient access.

Under formulas processes, health plans should focus on improving data diversity in drug monographs, as well as specific education and considerations to pharmacy and therapeutics committees. For benefit design, Johnson said it’s important to consider health equity in the process and adjust cost sharing models based on income or disease states. Additionally, patient access is improved by use of automated tools, enhanced care coordination, and patient outreach programs.

So, where does the data come from? There are 2 ways.

Direct data is crucial and gathered directly from patients, serving as the “gold standard” for achieving health equity, according to Johnson. By actively listening to patients and understanding their unique needs and circumstances, pharmacists can provide focused and personalized interventions that can lead to even better impacts on the patient.

Indirect data refers to information inferred from external sources, such as zip codes and Census data. These sources provide details like income, education, language, and other specific factors associated with specific geographic areas. In one example Johnson gave, in California, their health plan utilizes community health advocates who serve each zip code. By analyzing the health disparities and SDOH within each zip code, they can effectively address the needs of the patients in that area. This approach enables population-level analysis and allows for a more targeted and personalized use of indirect data, such as zip code or census information.

Through these 2 types of data, a variety of information is collected, including race, ethnicity, language, SDOH, geography, disabilities, sexual orientation, and gender identification information. All of these factors play a role in narrowing the gaps in health care, and Johnson stressed that it is crucial to emphasize that these factors are equally important in achieving health equity in the United States.

Race, Ethnicity, and Language Data

After Dinh asked the crowded room of pharmacists to raise their hands if their organization is actively working with data surrounding health equity and SDOH, almost the entire room raised their hands.

Diving into SCAN Health Group’s specific journey with health equity data, Dinh mentioned SCAN’s 4.5 star rating under the CMS Star Ratings, which measures the quality of Medicare Advantage health plans and Prescription Drug Plans such as Part D. As he mentioned, SCAN has used race, ethnicity, and language data to guide much of their work in the past years in terms of identifying and addressing disparities in SCAN’s member population.

However, when SCAN used this data and stratified the population by race, that was not the case.

“We were not 4.5 stars when you looked at the different populations,” Dinh said. “Across all the measures, we had lowest performance among our Black members. Overall, in medication adherence, the lowest scores were seen with our Black and Hispanic population, and then our Asian and Black members rated us the lowest in terms of our survey measures. So, we saw these disparities and we thought, ‘this is a problem, we need to do something about it.’"

SCAN first focused on medication adherence. Among Black and Hispanic members, overall medication adherence in 2021 was about 85%, while all other members had a medication adherence of around 88%. SCAN then made it a goal to reduce that 3% gap by 25%. However, it wasn’t just a goal for the pharmacists—SCAN made it a corporate goal, tying executive bonuses to the goals and working together across the entire organization.

Within 18 months, SCAN ended up reducing the gap by 35%. How did they do this?

One notable approach was the use of race, ethnicity, and language data to match patients with appropriate health care staff, based on insights gained from interviews with Black and Hispanic members. The SCAN team learned that trust was a significant barrier to medication adherence for Black members, who expressed less trust in their physicians and more trust in other members of the health care team. For Hispanic members, language barriers were found to affect communication and the sharing of critical health information. To address these issues, SCAN Health Group stratified its member population and paired them with health care providers who could serve them in a manner that aligned with their preferences and needs. In short: patients need providers who are like them.

Dinh also mentioned the integration of race, ethnicity, and language data into medication adherence platforms, allowing for targeted interventions based on refill due dates and adherence risk. He said these platforms and algorithms are the same ones they’ve been using, just adding the extra data. SCAN also implemented cultural competency and humility trainings to ensure staff members were equipped with the skills to engage in effective conversations with members. Collaboration with the medical provider network also played a pivotal role in leveraging the collected data to guide their work.

Sexual Orientation and Gender Identification Data

While sexual orientation and gender identification (SOGI) data is relatively new to and has limitations, SCAN recognizes its importance for understanding patient representation.

Dinh explained that SCAN Health Group actively collects SOGI data through internal methods, incorporating it into their data warehouse to inform decision-making. Although disparities have not yet been identified, the integration of SOGI data demonstrates a commitment to inclusivity and addressing the unique health care needs of diverse populations.

Dinh also shared insights from a population of approximately 285,000 SCAN members, revealing that less than 1% identified as a gender other than male or female and around 20% disclosed their sexual orientation, with close to 2% of this subgroup of about 54,000 individuals identifying as members of the LGBTQIA+ community. According to Dinh, these figures emphasize the significance of capturing data to ensure representation and prompt action.

To address the specific needs of LGBTQIA+ individuals, particularly older adults facing discrimination, the organization introduced the Affirm Plan, which offers benefits such as virtual behavioral health services, companion care, legal service reimbursement, and lower co-pays for specialty medications like HIV drugs. The Affirm Plan, which currently serves around 500 members, shows how health groups can use data-driven insights to develop targeted interventions.

Social Determinants of Health

“Social determinants of health is a national priority,” Lee-Brown opened with. “In fact, it is 1 of 3 priorities for Healthy People 2030 along with health equity and health literacy.”

SDOH include non-medical factors that influence health outcomes, such as education access and quality, health care access and quality, economic stability, neighborhood and built environment, and social and community context.

While they are separate concepts, SDOH and geography are deeply interconnected, Lee-Brown explained. A patient’s socioeconomic status influences their residential choices, which, in turn, impact the educational opportunities available to them. These factors then affect their socioeconomic mobility, creating a full circle of influence. Lee-Brown stressed that, to understand how geography affects health outcomes, pharmacists must consider elements such as access to safe housing, transportation, education, job opportunities, nutritious foods, pollution levels, as well as language and literacy skills, as these factors are vital to comprehending the impact of geography on health care.

Focusing on medication quality, a central theme of PQA23, promoting healthy choices alone cannot eliminate disparities. Adherence to medication is crucial for positive health outcomes, but individuals who struggle with basic needs like finding their next meal or a place to sleep are unlikely to prioritize medication adherence. Lee-Brown emphasized that recognizing the impact of social determinants and geography is essential in addressing disparities and improving overall health care outcomes.

Lee-Brown went on to share insights from her experience working with Health First, a health care plan serving New York City residents.

Data from 2019 showed a poverty rate of 41% in New York City, highlighting the urgency to address social barriers and transform the environment for patients. According to Lee-Brown, clinical interventions alone are insufficient in a city like New York, and that health care professionals must also address the social determinants that impact access to care.

Breaking down poverty rates by factors such as race, ethnicity, borough, and educational attainment revealed uneven distribution. For instance, the Bronx had an average poverty rate of 60%, while Manhattan's rate was below 30%. Additionally, analyzing poverty rates solely based on boroughs and averages can also lead to oversimplified data, as a deeper analysis at the neighborhood and zip code levels additional disparities and localized challenges.

By recognizing the unique socioeconomic landscape of New York City and understanding the intricacies of poverty rates across neighborhoods, health care organizations like Health First can develop tailored interventions to address the specific needs of each community. The aim is to bridge the gap between health care services and SDOH, ensuring that all patients, regardless of their geographic location or socioeconomic status, have equitable access to quality care.

“The further you can dig into your data, the more granular you can get with this data, the more impactful and the more targeted you can be with your interventions,” Lee-Brown said.

There are 4 main ways that health plans can help their populations, according to Lee-Brown. These include universal assessment, predictive analysis, hot spotting, and group-based risk stratification. Lee-Brown went on to highlight the latter 2.

Health care organizations can use hot spotting to identify areas with greater needs, such as food deserts, even without member-level data. Partnering with community organizations allows for targeted interventions like providing fresh produce. While measuring direct impact can be challenging, building trust with the community is crucial.

Risk group-based risk identification relies on member-level data to stratify populations, enabling health care organizations to address specific needs related to social determinants and geography. However, limited resources and equal benefit application regulations set by CMS pose challenges in implementation. To overcome these challenges, CMS introduced innovative models like value-based insurance design (VBID), allowing for unequal benefits for beneficiaries. For instance, co-pays can be waived for low-income members or grocery benefits offered. The VBID program for 2023 involves 52 Medicare Advantage organizations and is expected to impact around 9.3 million members.

While outcomes reports are pending, early stages of these initiatives show promise. Professionals in the Medicare space are encouraged to familiarize themselves with VBID, explore research opportunities, and engage in discussions with colleagues. It is important to recognize the availability of untapped data, including large purchasable datasets, which can empower organizations to develop comprehensive programs that better serve their membership.


Disabilities have gained attention in the health care landscape, especially with the recent focus on health equity. According to the CDC, approximately a quarter of adults have a disability, and individuals with disabilities are more likely to experience higher rates of obesity, smoking, heart disease, and diabetes. Understanding this population is crucial for developing effective programs, Lee-Brown said.

Collecting disability data offers several benefits, including improving landscape assessments, comparing against state and national data, and identifying evolving patient needs for budget forecasting. The HHS data council has recommended standards for collecting disability status, emphasizing a functional perspective and tracking disparities between populations. However, organizations should aim beyond these minimum standards and explore additional sources such as plan enrollment data and medical records.

In closing, Lee-Brown said, “You need to incentivize the folks within your organization to really have these conversations with members and collect this data to build data governance programs, so that you can effectively design and implement programs that will help you improve the life of your members."

Related Videos
Sheila Garland, PhD, MSc, Memorial University
Andrew McHill, PhD, Oregon Health and Science University
Dayna Johnson, PhD, MPH, MSW, MS, Rollins School of Public Health at Emory University
Travis Brewer, vice president of payer relations, Texas Oncology
Ariel Williamson, PhD, DBSM, Children's Hospital of Philadelphia/Perelman School of Medicine
Judite Blanc, PhD, Miller School of Medicine/University of Miami
Mark Fendrick, MD
Dr Jason Westin, MD, FACP, MS
Related Content
© 2023 MJH Life Sciences
All rights reserved.