Health equity is an increasingly important topic within the healthcare sector. Analyzing and understanding information related to not only clinical history, but also social context can provide illuminating insights into disparities in health outcomes. Within Oliver Wyman Actuarial’s Health Practice (Oliver Wyman), our Predictive Insights (PI) group strives to address this question by bridging the gap between traditional actuarial methods and predictive analytic techniques to create robust recommendations and lasting solutions.

Oliver Wyman’s PI group has partnered with HealthWise Data to establish the critical connection between traditional clinical data and consumer information that may impact health outcomes and disparities. In an analysis of more than 40,000 Medicare-covered people in central Florida, Oliver Wyman was able to link consumer-level social determinant of health (SDoH) attributes provided by HealthWise Data to clinical level information at a rate of nearly 80%. All data and exhibits that follow are based on the combined view of SDoH information provided by HealthWise Data to Oliver Wyman as well as traditional clinical information available via administrative claims data.

HealthWise Data, a leading healthcare data and analytics firm, is at the forefront of bridging the gap between clinical outcomes and SDoH domains for these healthcare entities. With a comprehensive consumer database of almost 250 million US adults with hundreds of attributes spanning all domains of SDoH, plus a robust set of health equity scores, health behavioral and adherence propensities, and healthcare outcome predictions, HealthWise Data can empower the industry with the tools to analyze and address SDoH outcomes as they relate to population health outcomes.

The importance of SDoH as it relates to health outcomes

SDoH represent environmental and societal conditions that have an impact on overall health and quality of life. There are five major domains within SDoH: Education Access and Quality, Economic Stability, Social and Community Context, Health Care Access and Quality, and Neighborhood/Built Environment. Some common attributes within each of these domains contributing to the overall health of the population are highlighted in Exhibit 1.

Recognizing that providers have the largest influence in improving access, quality of care, and overall community health, the Centers for Medicare and Medicaid Services (CMS) recently released the calendar year 2024 Medicare Physician Fee Schedule Proposed Rule, which includes several enhancements and physician incentives to address SDoH. Following up on CMS’s proposed rule, in this article we illuminate the importance of SDoH as it relates to health outcomes and recommend that SDoH be used in provider/payer contracts to appropriately compensate providers for managing patients with social risk factors, thereby driving better health outcomes and lowering total cost of care.

SDoH studies show imbalances and impact on patient health

SDoH data has become increasingly popular within the healthcare industry. Studies continue to show the importance of SDoH imbalances and their impact on patient health. Further, we’ve seen SDoH features leveraged in a handful of areas in the current healthcare landscape. First, federal and state regulators are exploring the use of select SDoH attributes within risk adjustment programs which, in their current state, tend to skew results in favor of people with less social determinant imbalance. We’ve also seen the importance of SDOH as it relates to provider network sculpting. Overlaying SDOH attributes such as transportation access and geographic makeup can help entities sculpt their networks to maximize health equity across the country.

Imbalances between healthcare delivery and SDoH is the next hurdle

Understanding that there is a connection between SDoH and clinical outcomes is only half the battle. Addressing imbalances between healthcare delivery and SDoH is the next hurdle, and that starts with the provider. Typically, healthcare providers negotiate with insurance entities and are reimbursed based on services rendered. However, these reimbursements do not reflect a sliding scale for services rendered to a population with SDoH imbalances, even though there is support to the contrary.

For example, we found that healthcare expenditure predictive ratios — the ratio of predicted expenditure relative to actual expenditure — are significantly lower via traditional risk adjustment mechanisms for lower income levels (measured as a percentage of the Federal Poverty Line, or FPL) than for risk models that incorporate key SDoH attributes (see Exhibit 2) provided by HealthWise Data, as shown in Exhibit 3 below.

Further, we found that lower market home value coupled with high levels of food insecurity (a high-level proxy for neighborhood quality) drive higher healthcare costs on average. Exhibit 4 isolates average healthcare cost for Medicare by overall home market value and levels of food insecurity, two data points provided by our HealthWise Data partners. As shown, populations subject to high levels of food insecurity have healthcare costs that are roughly 10-15% higher than individuals with better access, likely due to the emergence of chronic conditions associated with poor nutrition, such as diabetes and heart disease. We also see that people in areas with both low real estate value and high levels of food insecurity drive the highest healthcare spend, with average costs more than 10% higher than total average cost for this population and over 40% higher than individuals not subject to these factors.

Equipping healthcare providers with the power of SDoH data is key

This underpredicted morbidity combined with higher-than-expected healthcare spend for patients with SDoH has a direct impact on provider payments. First, artificially low risk scores for patients with SDoH can dampen any pass-through savings to providers from health plans based on risk-sharing arrangements (for instance, high costs and low risk scores may cause the provider to look inefficient at managing care).

Second, providers that predominantly service member populations in lower income, impoverished areas will naturally see higher costs for their patients than their counterparts without reimbursement adjustments reflecting this, as illustrated in Exhibit 4 above. While there are likely numerous drivers of these higher costs, we often see a higher prevalence of behavioral health issues and lower medication adherence within these populations that can cause additional comorbidities and complications.

Equipping healthcare providers with the power of SDoH data has two critical outcomes: a) helping them identify and understand the level of health inequity in their population that need the most additional support, and b) providing leverage for value-based contract negotiations with payers. Both outcomes go hand-in-hand: if providers can demonstrate the need for higher reimbursement to effectively treat patients with SDoH, the additional revenue from payers can be used to improve resources and care, thereby supporting the reduction of overall healthcare cost. For example, the additional revenue may enable provider systems to hire more licensed clinical social workers and community health workers who have a “boots on the ground” approach of addressing social needs that affect the health of these patients.

As a secondary outcome, documentation and insights provided to physicians by social and health workers represent a bridge between provider medical records and the 360-degree view of clinical disease history of patients. Improved visibility into patient history in turn drives better coding and electronic health record documentation of conditions, and more appropriate risk adjustment payments for the payers servicing areas with a high proportion of members with SDoH imbalances.

Oliver Wyman’s three-pronged approach to leverage the power of SDoH

Through our partnership with HealthWise Data, Oliver Wyman has developed a three-pronged approach for assisting provider systems in leveraging the power of SDoH to transform care for their patients. Our platform ingests data on enrollment, claims, and member engagement with provider system and SDoH, and conducts a rapid assessment by running the data through our machine-learning-enabled data extraction and enrichment pipeline. The infographic below illustrates our approach:

Conclusion

SDoH has a critical impact on health equity. The data illustrate this and, as the gatekeepers of care, provider systems need to be armed with data-driven solutions that will help drive change in the industry. Through our partnership with HealthWise Data and predictive analytics capabilities, Oliver Wyman can give providers the tools and assistance they need to negotiate contracts with SDoH stipulations. This will provide more appropriate funding for the treatment of current patients as well as funding for enhanced care strategies and outreach that will lead to better health, lower costs, and improved quality of life for patients in the long term.

Additional contributors: HealthWise Data, LLC and Nilabh Sanat