The Geography of Health: Rethinking Value-Based Payments

Article by Gurpreet Singh – Sr. Director, Product Management & Dr. Demitri Plessas – Principal Data Scientist

A growing body of evidence shows that where someone lives may be a more powerful predictor of their health outcomes than their genetic makeup or even their healthcare provider. Up to 60% of a person’s health status is influenced by their ZIP code and the associated social and environmental conditions it represents. As the industry continues its shift toward value-based care models, this reality can no longer be ignored.

The Impact of ZIP Codes on Health Outcomes

The correlation between an individual’s geographical location and their health has long been understood to some extent—but as recent analyses have shown, the impact of ZIP codes on patient health outcomes is far more significant than previously thought.

An employer health study compared emergency department visit frequency and per-member-per-month (PMPM) costs in Greenville, SC and Des Moines, IA. Greenville’s emergency department visit frequency was 309 per 1,000 members with PMPM costs of $543, while Des Moines averaged 114 ED visits per 1,000 members and PMPM costs of $354, or 53% less than Greenville. These variations aren’t random; rather, they reflect systematic differences in community resources, social support, and healthcare access.

The data tells a compelling story about how ZIP code shapes healthcare utilization and costs. When examining emergency department use across ZIP codes, research shows that residents of socially vulnerable areas have up to 39% higher rates of ED visits for conditions that could be treated in primary care settings. For Medicare and Medicaid dual-eligible beneficiaries, the impact is even more pronounced, with D-SNP members in underserved communities averaging significantly higher rates of preventable hospitalizations and emergency department visits compared to those in well-resourced areas.

These geographic patterns are both predictable and actionable. By analyzing hundreds of community-level factors—from food security to transportation access—healthcare organizations can now anticipate and address the socioeconomic factors driving costly utilization. Addressing social needs through targeted geographic interventions can yield significant returns: UI Health in Chicago found that providing permanent housing for homeless patients with medical and behavioral health needs reduced healthcare costs by 27%.

Why Geographic Insights Matter for Value-Based Care

Healthcare organizations implementing value-based care programs have begun to recognize that fair provider evaluation must account for these geographic disparities. The Centers for Medicare & Medicaid Innovation (CMMI) has acknowledged this reality by introducing social risk adjustments in new payment models, including adjustments for providers serving socially disadvantaged communities. This shift reflects a growing understanding that providers shouldn’t be penalized for caring for populations in high-need areas.

The impact is particularly pronounced in specialized programs like D-SNPs and MLTSS. For MLTSS populations, the ability to live independently is closely linked to the availability of community resources, from home caregivers to accessible transportation. In fact, recognizing the crucial role of place-based factors in health outcomes, over 91% of Medicaid managed care plans now report activities to address social conditions impacting their member populations.

ZIP Code Matters

A member’s ZIP code serves as a powerful indicator of social drivers (previously called “determinants”) of health, or SDOH, which offers crucial insights into the complex web of social, economic, and environmental factors that shape health outcomes. Through ZIP code analysis, healthcare organizations can assess critical factors including socioeconomic status, healthcare facility proximity, and environmental conditions that directly impact physical and mental health. Additionally, ZIP codes can reveal patterns in food security, with food deserts contributing to higher rates of nutrition-related illnesses, and transportation access, which affects healthcare utilization and medication adherence.

Beyond these fundamental indicators, ZIP code data illuminates housing stability patterns, where high rental rates and overcrowding can signal increased health risks and mental health challenges. Educational attainment and health literacy levels within specific ZIP codes correlate strongly with health behaviors and disease management capabilities, while public transit availability and car ownership rates provide insights into potential barriers to healthcare access.

Identifying these factors and their connection to health outcomes has historically been the remit of public health agencies, and their efforts have yielded a wealth of data. But accessing and leveraging that data has historically proven challenging for healthcare organizations, as we’ll discuss shortly.

Integrating SDOH Into Value-Based Contract Design

Payer actuaries can leverage SDOH data to enhance risk adjustment, cost benchmarking, provider incentives, and performance measurement to ensure contracts fairly account for non-clinical factors that impact patient health.

Adjusted Risk Models

Traditional risk adjustment models rely predominantly on claims and clinical data, often failing to capture the full scope of risk for populations with high SDOH needs. As a result, these models systematically underestimate the true cost of care for underserved communities. To address this gap, payers should begin incorporating non-medical data sources—such as housing instability, food insecurity, and transportation barriers—into their risk adjustment frameworks.

Recognizing this necessity, CMS has initiated testing of health equity risk adjustment factors within Medicare Advantage and ACO models to more accurately reflect the increased costs associated with caring for socially disadvantaged populations. Given the evolving landscape, commercial payers should proactively adopt similar methodologies to ensure fair compensation for providers serving high-need communities while improving health equity outcomes.

Informed Cost Benchmarking

Rather than relying solely on standard cost benchmarks, payer actuaries should incorporate SDOH segmentation to refine cost targets based on social risk factors. Provider cost benchmarks must be adjusted for ZIP codes with high poverty rates to ensure that those serving vulnerable populations are not unfairly penalized. Additionally, healthcare organizations should consider leveraging AI-driven geospatial analysis to uncover regional disparities in healthcare costs that can be attributed to SDOH to enable more precise and equitable payment models.

Provider Incentives & Payment Adjustments

Value-based payment (VBP) models should incorporate SDOH-driven incentive structures to ensure providers can fairly adjust practice patterns based on the social risk factors of their patient populations. To encourage proactive interventions, providers should receive higher payments for addressing key social drivers of health—such as screening for food insecurity and implementing targeted interventions.

To support providers serving high-need communities, VBP contracts should offer higher shared savings rates for safety-net providers, reflecting the additional resources required to care for disadvantaged populations. Additionally, provider performance scores should be risk-adjusted to account for the impact of social and environmental factors on health outcomes, preventing unfair penalization.

Medicaid Managed Care Organizations (MCOs) have already begun implementing equity-weighted incentives to drive better care delivery in socially vulnerable areas. This approach combines three key incentives:

• Risk-adjusted base payments to reflect the higher costs of serving high-SDOH-need populations
• Enhanced quality bonuses that reward providers for improving care outcomes in underserved communities
• Targeted support programs that fund interventions addressing specific barriers to care

For example, some MCOs offer a 15% higher quality bonus for improving diabetes management in high-risk ZIP codes, while others provide enhanced reimbursement for reducing emergency department utilization in socially disadvantaged areas. These financial incentives create a strong motivation for providers to invest in social risk factor mitigation, ultimately leading to better health outcomes and lower costs.

Commercial payers should adopt similar models, leveraging equity-based incentives to align provider payments with the realities of socioeconomic risk factors and ensure value-based care delivers equitable outcomes.

Alternative Payment Models (APMs)

Payers should embed SDOH factors into quality incentive programs, shared savings/risk arrangements, capitation, and bundled payment models to more accurately reflect the true cost of care for underserved populations. For example,

• Capitation models should incorporate SDOH-adjusted per-member-per-month (PMPM) payments to ensure adequate funding for social needs interventions, such as housing support, nutrition programs, and transportation services
• Episode-based bundled payments should expand reimbursement structures to include non-clinical services, recognizing that factors like stable housing and reliable transportation are critical to improving health outcomes and reducing avoidable costs
• Performance measurement frameworks should incorporate SDOH-adjusted readmission rates to reflect the increased likelihood of multiple comorbid conditions among historically underserved and high-need communities and the unique treatment challenges they can present

While these adjustments help create a more equitable payment landscape, careful calibration is essential to avoid unintended bias. Over-adjusting for SDOH could inadvertently lower expectations for care quality in disadvantaged populations, reinforcing disparities rather than addressing them. Therefore, models should strike a balance between fairness and accountability, ensuring high-quality care remains the standard across all communities.

Attribution & Performance Measurement

Payers should refine patient attribution models to better account for the impact of SDOH on healthcare engagement. Traditional attribution models primarily assign patients to providers based on medical complexity and claims data, often overlooking critical non-clinical factors that influence care access and utilization.

SDOH-enhanced attribution models incorporate additional factors such as housing stability, food security, access, transportation access, and social support networks to create a more accurate and equitable assignment of patients to providers. By integrating SDOH into attribution logic, healthcare organizations can improve continuity of care, reduce unnecessary utilization, and better align provider incentives with patient needs, ultimately driving better outcomes in value-based care models.

The Role of Data

As in most models, data plays a critical role in incorporating SDOH into VBC program design. Forward-thinking organizations are leveraging ZIP code-level insights in several ways. First, they’re using standardized ICD-10 Z codes (Z55-Z65) to document social needs in claims data, allowing for better tracking of geographic patterns in social risk. While utilization of these codes is still emerging—only about 1.6% of Medicare fee-for-service beneficiaries had a Z code on a claim in 2019—their use is steadily increasing as the industry recognizes their value.

Healthcare organizations are also enriching their analytics with public data sources. The Agency for Healthcare Research and Quality (AHRQ) maintains a comprehensive database that aggregates community-level indicators which can be linked by ZIP or county to patient data. This allows for more sophisticated risk stratification, identifying not just who has high clinical risk, but those at high socioeconomic risk based on their geographic location. The results of these data-driven approaches are compelling: housing interventions targeting high-utilization ZIP codes have shown up to 67% reductions in healthcare costs for enrolled individuals.

While valuable and wide-reaching data sources exist to inform geographic analysis, accessing and integrating them requires dedicated effort. The CDC’s Social Vulnerability Index provides detailed insights at the census tract level across 15 social factors. The Census Bureau’s American Community Survey, updated annually, offers extensive demographic and socioeconomic indicators.

Additional rich sources include the Bureau of Labor Statistics for employment data, FBI uniform crime reporting statistics, and the Robert Graham Center’s Social Deprivation Index. Centers for Medicare & Medicaid Services (CMS) also maintain valuable geographic data on healthcare access and utilization patterns. As noted above, however, the challenge lies not in the availability of data but in effectively combining these diverse sources into actionable insights.

Data availability and standardization remains a challenge, and SDOH data sources (e.g., community surveys, EHRs, census data) vary in quality and completeness. Organizations must invest in data integration capabilities and analytical tools to transform this wealth of community-level information into meaningful interventions that improve outcomes and reduce costs. We find that many of our clients are stuck at the crossroads of having identified the need but without the technology and analytics capabilities needed to support data integration and analysis.

Geography as a Core Component of Value-Based Payments

The healthcare industry is at an inflection point in how it views and uses geographic data. CMS’s 2021 guidance explicitly encouraged states to use available flexibilities to address social drivers of health in their programs. This policy direction, combined with advances in data analytics, is creating new opportunities for value-based care innovation.

The path forward is clear but requires a systematic approach. For healthcare executives and payers, the implications are significant. Success in value-based care increasingly depends on understanding and acting on geographic insights. Organizations that can identify high-risk ZIP codes, deploy targeted interventions, and measure the resulting impact on utilization and costs will be better positioned to succeed in risk-based contracts. More importantly, they’ll be better equipped to fulfill the promise of value-based care: delivering better outcomes while controlling costs.

However, organizations often find that building and maintaining the necessary data integration infrastructure can cost millions of dollars annually, requiring specialized expertise in healthcare data models, geographic information systems, and advanced analytics. Many have found that partnering with established healthcare technology vendors can provide a more cost-effective path to transforming this wealth of community-level information into meaningful interventions that improve outcomes and reduce costs.

The view that “your ZIP code matters more than your genetic code” is becoming a guiding principle for innovative health systems. By embracing geographic data as a core component of value-based care strategy, organizations can move from reactive to proactive care delivery, addressing social needs before they manifest as costly medical conditions. In doing so, organizations can help build a more equitable and effective healthcare system that truly delivers value for all communities.