Framework to advancing Health Equity through Data and AI
4 min. read

Framework to advancing Health Equity through Data and AI

Health equity means that everyone has a fair and just opportunity to be as healthy as possible.

In the recent past, there has been renewed talk about health inequities and disparities. While health inequities and disparities existed for long, few reasons have brought the discussion to the fore:

1. The devastating effects of COVID-19 on Black, Indigenous, and People of Color (BIPOC) groups and

2. Social unrest in the past two years and the awareness it brought about the needs of these groups and vulnerable population.

Here are some unfortunate facts that show health inequalities:

Framework to advancing Health Equity through Data and AI

Healthcare organizations can no longer ignore the pervasive health disparities and outcomes within the members they serve. Organizations should remove the unconscious bias and consciously see every decision through the health equity lens.

The government and Center for Medicare and Medicaid Services (CMS) have been increasingly advocating to minimize health inequities and disparities. Multiple voices are pressing CMS to incentivize Medicare Advantage plans that meet certain health equity metrics. CMS has recently stated that its #1 priority is health equity.

Some healthcare organizations are tackling the issue of health equity head-on. They are looking to create products and services that will help to bridge the gap. There are teams in certain organizations that focus on health equity work, and some organizations have mandates from the C-level. While there is a lot of momentum, most teams are looking to start and how go about it. We see that these teams lack critical data, intelligence, and tools to take the right actions.

Our approach

Health equity is a complex issue. Advancing health equity needs a multifaceted approach involving data, analytics, AI, and design.

1. Understand the status quo

The first step is to understand the status quo. Build tools that will help visualize what inequities and disparities exist in the organization’s membership and uncover the facts. This involves aggregating, mapping, and transforming multiple internal (claims, conditions & comorbidities, cost, HEDIS, CAHPS etc.) and external (census, state and local governments, CDC, CAHQ, AHRQ etc.) datasets and building data assets, and developing cognitive BI solutions. In this step, organizations will identify insights such as:

  • Medical PMPM of African Americans is 20% higher compared to whites in Region 5 of South Carolina.
  • Diabetes medication adherence of females has been 7% lower when compared to males.
  • Arthritis screening for Spanish-speaking members was 15% lower when compared to English-speaking members in Arizona.

This will also aid in understanding the unconscious biases that exist in decision-making across organizations.

2. Understanding the “why”:

Health inequities and disparities result from multiple experiences a member has in their socio-economic and healthcare journey. Social determinants of health (SDOH), the community the member lives in, and structural factors play a significant role in one’s health outcomes.

 

Framework to advancing Health Equity through Data and AI

Figure 1: Causes of Health Inequities

Health equity is a complex issue.
Advancing health equity needs a multifaceted approach involving data, analytics and AI.

In this step, identify the drivers of inequities and disparities that are specific to the organization’s membership. This requires extensive data analysis, building advanced AI/ML models to pinpoint features and predict who would be more susceptible and which zip codes are more vulnerable. Analysis of multiple internal metrics such as claims, PMPMs, conditions, and community indicators such as social vulnerability index, social deprivation index, Gini index, etc., is critical here.

Framework to advancing Health Equity through Data and AI

Figure 2: Framework to understand the why and steps to drive actions


3. Strategize, execute and scale

With an understanding of drivers and insights acquired from previous steps, create health equity strategy and interventions considering the organization’s goals, internal and external resources available, and define how success will look like. Implement these interventions at the member level or community level, measure outcomes and scale.

It can take months and sometime years before organizations see the favorable outcomes. But not investing to advance health equity can fail not only health outcomes but also business outcomes of organizations. So, take action!

 

Authors

Framework to advancing Health Equity through Data and AI

Venkata Sivakumar Chippagiri

Principal Consultant

Framework to advancing Health Equity through Data and AI

Kishore Bharatula

Client Partner

Framework to advancing Health Equity through Data and AI

Abarna Priyaa

Engagement Manager

Framework to advancing Health Equity through Data and AI

Mithu Goswami

Senior Consultant

Framework to advancing Health Equity through Data and AI

Kaushal Sharma

Senior engineer

Framework to advancing Health Equity through Data and AI

Manish Shukla

Senior Engineer

 

Sources

https://www.nejm.org/doi/full/10.1056/NEJMsa2011686

https://jamanetwork.com/journals/jama/fullarticle/2766098

https://www.acpjournals.org/doi/10.7326/M20-2247