The original sponsored article appeared on Healthcare IT Today.
Healthcare organizations are spending significant resources – time, money and human capital – to address social determinants of health (SDoH) and health equity. Effective execution requires going upstream of the programming itself to understand the social determinants of a population.
According to Data Bridge Market Research, the U.S. SDoH market is expected to grow with a compound annual growth rate (CAGR) of 22.9% between 2023 to 2030. Also, it’s forecasted to reach over $43 million by 2030 from more than $8 million in 2022.
And it’s been found that private insurance companies spend $1.87B on SDoH data, with the top six insurers accounting for 72% of that SDoH spend. And while the data is a few years old, a study found that 57 health systems invested at least $2.5B in SDoH over a two-year period. There’s a strong emphasis, and substantial funds are being spent in this area of healthcare – and it’s not slowing down.
As healthcare organizations continue to invest in SDoH, moving from analysis to impact requires operationalizing this data for coordinated efforts and long-term returns on investment. When considering SDoH, think about the following:
In this article we explore the top lessons learned to make SDoH data effective, and three actionable examples of organizations who have turned data to analysis and action.
Success Story #1: Moving Upstream to Identify Most Vulnerable Individuals to Prevent Homelessness
There’s a common metaphor in public health about the need to move “upstream” to invest in preventive health, rather than wait for people to present at the “bottom of the river” with severe chronic illness. Recognizing the challenge of identifying people experiencing homelessness, a western state Medicaid program partnered with LexisNexis® Risk Solutions to access extensive SDoH data and apply predictive analytics to identify and provide services to families before they experience severe homelessness. The goal was to reduce associated healthcare costs by connecting individuals experiencing homelessness to permanent supportive housing.
The state recognized that relying on homelessness data alone identified homeless individuals at the point of care. However, by combining state and homeless program data with the external SDoH data available from LexisNexis Risk Solutions, they were able to develop a Housing Stability Score that scored the entire population and prioritized those who may experience housing instability.
The state data combined with LexisNexis Risk Solutions data led to a 6.56% increase in the ability to predict housing instability compared to only state data. Stronger analytics also allowed the Medicaid program to use its limited resources to connect people to three levels of care coordination, based on the severity of the Housing Stability Score.
Success Story #2: Gaining Perspective on the “Whole Person” Patient Journey for Cancer Research
Understanding the impact of cancer extends far beyond the clinical pathway. Cancer researchers are investigating the impacts of cancer on long-term health outcomes, financial strain and household impact. To connect the full patient journey, epidemiologists collaborated with LexisNexis Risk Solutions on patient matching solutions to precisely link 6M individuals across dozens of cancer registries going back over 20 years.
Once the records were linked, researchers were able to enrich the “whole person view” with over 40 years of longitudinal data on address and migration, social determinants and financial history. This rich data provided insight into how pediatric cancers have a disproportionate disease burden, as they also impact household financial toxicity across the lifespan.
Success Story #3: Meeting Patients Where They Are to Improve Medication Adherence and Management.
A global healthcare organization developed predictive analytics to identify medication non-adherence patients to help inform care management programs. The lack of clinical data resulted in lower predictive power, especially for new members and newly diagnosed members who could benefit the most from care management programs.
Adding individual-level SDoH data helped fill in the gaps where clinical data was limited or unavailable. Data scientists compared two models: baseline models built on clinical, claims data, and basic patient demographics compared to the baseline model enhanced with LexisNexis® Socioeconomic Health Attributes layered to provide deeper insight to a patients’ social and economic situation.
Adding SDoH data attributes yielded a 20% increase in recall (correctly identifying non-adherent patients) at the beginning of the year when clinical data was sparse. By leveraging this SDoH data, the organization was able to identify eligible patients earlier in the process to optimize interventions with a greater chance of decreasing costs and improving health outcomes.
Turning SDoH Data into Insights and Action
The above examples show the added power and lift of adding individual-level SDoH data to improve predictive power, identify vulnerable patients and connect them to critical SDoH programs. Success requires a cross-functional approach to align cross-functional stakeholders and prioritize investment.
Where Else Does SDoH Data Have a Role to Play?
As SDoH programs become more sophisticated, maintaining a focus on data to drive and evaluate efforts is critical to success. The examples above show the importance of thinking upstream on how the right data and partnerships can make SDoH programs “smarter” to address the needs of the most vulnerable. Health equity and whole person care require a “whole person” view of the individual through as many different data points as possible to ultimately provide personalized care and improve health outcomes.