To Advance A National Health And Equity Infrastructure, Measure Drivers of Health

Twenty years ago, the Institute of Medicine published Crossing the Quality Chasm, a call to close the quality and safety gaps in health care, which spurred a proliferation of quality-focused measures, initiatives, departments, and organizations. While the health care sector has arguably made gains in patient safety, we are no healthier as a nation, with pervasive racial and ethnic disparities in health outcomes.

It is now time for our country to move from accountability for health care to accountability for health and health equity. To do so requires addressing the drivers of health (DOH): critical comorbidities such as food insecurity, housing instability, and access to transportation.

Calls to address these issues have emanated from across the sector. In pledging to tackle health inequities, Health and Human Services Secretary Xavier Becerra called for “examination of ways to address the social determinants of health,” citing their disproportionate impact on communities of color. As physicians and policy makers, we have witnessed the impact of these factors on patients, communities, and our colleagues.

Recognizing that what the health care system measures and pays for is the ultimate expression of what and who it values, we must account for drivers of health not just via pilots, projects, or special initiatives, but by integrating these factors into the regulatory framework of the health care system that moves close to 20 percent of gross domestic product.

We describe here an actionable DOH measurement vision and pathway—starting with the Centers for Medicare and Medicaid Services (CMS) enacting the first-ever federal DOH measures—to make visible and address the impact of these factors on health disparities, outcomes, and cost.

What Is A DOH Measure?

Consistent with CMS’s Accountable Health Communities model, state-based DOH initiatives, and leading DOH screening tools, we define a DOH measure as a person-level measure of food insecurity, housing instability, transportation problems, utility assistance needs, interpersonal safety, and other social needs that impact health. Absent such data, dual-eligible or socioeconomic status have been used as proxies for patient-level social needs, but they are often inadequate proxies and do not enable targeted individual- or community-level interventions to address DOH. Of note, we use the phrase “drivers of health,” to distinguish DOH from structural factors that require a different set of interventions and consistent with empirical data that show patients reject the language of “social determinants of health” as disrespectful and confusing.

To improve health outcomes and address inequities require both routinely collecting DOH data and stratifying it by race and ethnicity. This article focuses on the former, complementing other crucial efforts to address race and ethnicity data needs.

Growing Momentum: A Brief Timeline

Frontline physicians and other clinicians have long recognized the role of drivers of health. In a 2022 Physicians Foundation survey, 95 percent of physicians indicated their patients’ health is affected by at least one DOH, and 75 percent reported that integrating DOH into payment policy is important for improving health outcomes and ensuring high-quality, cost-efficient care for all. A 2019 JAMA study found that 24 percent of hospitals and 16 percent of physician practices were screening for all five DOH domains listed above, and 92 percent of hospitals and 66 percent of practices were screening for at least one domain. Commercial payers have similarly begun screening members for DOH and investing in access to healthy food, safe housing, and other DOH—but primarily for specific populations or as standalone pilots.

In 2016, CMS launched Accountable Health Communities, the first federal pilot to test the impact of social needs screening and resource navigation on health outcomes and costs. It has screened more than one million beneficiaries at 644 clinical sites in 21 states, with 34 percent screening positive for at least one social need and racial and ethnic minorities over-represented in the navigation-eligible population. In CMS’s Comprehensive Primary Care Plus model, 86 percent of approximately 1,500 Track 1 and 99 percent of approximately 1,500 Track 2 practices (together serving 2.4 million beneficiaries) implemented DOH screening, even though only Track 2 was required to do so. Other CMS models—including the Maryland Total Cost of Care and Comprehensive End-stage Renal Disease Care models—have also implemented DOH screening.

By 2020, CMS identified the development and implementation of “measures that reflect social and economic determinants” as a key measurement gap. More than half of states now require Medicaid managed care organizations or their contracted providers to conduct DOH screening, with North Carolina, California, and others leveraging DOH data to invest in addressing these factors at an individual and community level.

COVID-19 has further exacerbated the impact of DOH—with a disproportionate impact on communities of color. For example, the Census Bureau’s Household Pulse Survey revealed that, during the pandemic, food insecurity doubled and twice as many renters fell behind on their payments.

The impact of drivers of health-on-health outcomes, total cost of care, and health care expenditures is increasingly well-documented. For example, individuals with both diabetes and food insecurity cost public and private payers $4,500 more per member per year than diabetics with access to healthy food. Likewise, DOH contribute significantly to geographic variation in Medicare spending. Commercial payers have also begun to quantify the financial impact of DOH, with one recent analysis finding that pregnant Medicaid members reporting food insecurity had 81 percent higher pregnancy costs than those with no reported DOH, according to an internal analysis by the first author of this article.

Opportunity: Build A National Drivers Of Health Measurement And Data Pathway

Despite this, the health care system has not made drivers of health integral to its quality or payment programs, masking a primary contributor to disparities, poor outcomes, and costs. Currently, there is not one patient-level DOH performance measure in any major federal or quality payment program.

To improve health and tackle inequity, we must develop, implement, and improve DOH measures that are integrated into our core regulatory frameworks, consistent with the following goals:

• Evolve from process to health outcomes measures, using learning in the field to guide measure development.
• Commit to alignment of DOH measures across public and private purchasers, payers, and providers.
• Maximize use of existing measurement tools, platforms, and systems to avoid duplication, confusion, and administrative burden.
• Leverage DOH measures and data to deploy and quantify investments among providers and communities to address DOH.

National Drivers Of Health Measurement And Data Pathway: Four Key Actions

To accelerate the shift from accountability for health care to accountability for health and health equity, we have defined a national DOH measurement and data pathway including four key actions that build upon each other. CMS will be essential to enact these actions, while state policy makers and private-sector entities (including commercial payers and standard-setting bodies) have a crucial role in ensuring alignment and implementation.

First, secure adoption of foundational DOH measures. In 2021, the Physicians Foundation—whose directors are appointed by 21 state and county medical societies—submitted to CMS the first-ever DOH measures. Drawing from the Accountable Health Communities model, these two measures focus on screening for food insecurity, housing instability, transportation, utility help needs, and interpersonal safety and the screen positive rate for each domain.

In April 2022, CMS took a crucial first step by proposing these measures for the Hospital Inpatient Quality Reporting Program (IQR), which would incentivize hospitals accepting Medicare payments to screen all admitted adult patients for these drivers of health prior to discharge and report the screen positive rates. CMS will likely address these measures through the Merit-based Incentive Payment System (MIPS) in a subsequent proposed rule.

As the first DOH measures to be considered for a federal quality or payment program in the history of US health care—and the only patient-level equity measures now under review by CMS—these would send a powerful market signal of the move from health care to health and a systems-level commitment to operationalizing equity.

By enacting these measures, CMS would also create an early opportunity to align DOH measures across federal programs and bring a health lens to addressing risk adjustment, tackling disparities, and investing in communities. The screen positive rate measure is essential to make inequities visible and to link DOH data to other cost and quality indicators associated with health disparities.

These DOH screening measures and associated learning from implementation should be leveraged as the foundation for additional process and outcomes measures (see exhibit 1). As a next step, CMS should enact the first federal DOH “action” measures to ensure patients who screen positive are navigated to and secure the resources they need to be healthy.

Exhibit 1: Potential DOH and clinical measures

Source: Authors’ analysis of potential DOH and clinical measure categories.

Second, align drivers of health measures across CMS, states, commercial payers, and other industry actors. With providers and payers now scaling up DOH screening, there is a pressing need for federal DOH measurement standards and for stakeholders to coalesce around these measures to avoid proliferation and fragmentation, a major problem with existing quality and safety measures. To enable this, CMS should:

• Apply DOH measures to other CMS programs and Center for Medicare and Medicaid Innovation models and evaluations. CMS has signaled its appetite to integrate collection of patient-level DOH data into other models (such as ACO REACH) and programs (such as the Medicare Advantage requirement that all special needs plan health risk assessments “include one or more questions…on housing stability, food insecurity, and access to transportation”). At the same time, the Medicaid Child and Adult Core Sets include no driver of health measures despite this being identified as a critical gap for years.
• Leverage standard-setting bodies to enable DOH measure alignment and avoid fragmentation. The National Committee for Quality Assurance, for example, recently proposed a new “social need screening and intervention” measure that does not include reporting a screen positive rate, deviating from the proposed MIPS and IQR measures described above.
• Build DOH-level spending into the Health Care Payment Learning and Action Network measurement framework to quantify medical versus “health” spending.

Third, use driver of health measures and data to better understand clinical, social, and financial risk, and address inequities. Absent standard, patient-level DOH data, our understanding of risk will not improve. A Society of Actuaries study found promise in incorporating DOH into risk and payment models but noted “the lack of universal individual screening for DOH and lack of standardization in data collection limited the sample sizes for several of the analyses.” We can redefine risk if we:

• Systematically implement DOH measures and collect DOH data for all patients (versus defaulting to proxy-level data such as socioeconomic status or dual-eligible status).
• Integrate DOH as co-variates and complicating conditions in risk scoring/models through public processes. In the CY 2023 Medicare Advantage annual capitation rate announcement, CMS noted general agreement among stakeholders that factors related to social determinants of health should be incorporated into risk adjustment models to improve prediction of the relative costs of MA enrollees, while also noting that the necessity of “complete, reliable, and standardized data on SDoH” to do so. Commercial payers and state Medicaid agencies could likewise begin integrating DOH data into provider risk models.
• Integrate DOH measures into alternative payment models (such as accountable care organizations [ACOs]) and Medicaid managed care organization contracts. Including these measures in Medicaid contracts, provider contracts, and cost benchmarks will ensure DOH data are collected and factored into performance measures. CMS recently stated that “addressing social needs must be a central goal of ACOs going forward,” noting that “new ACO quality measures related to identifying and addressing social needs could support these initiatives.” Furthermore, its ACO REACH model begins to lay the foundation for integrating drivers of health and other health equity elements into alternative payment models more broadly.
• Align DOH measures with electronic DOH data and billing/coding sources and ensure data interoperability (for instance, the third version of Office of the National Coordinator interoperability standards).
• Ensure that DOH data and race and ethnicity data are both routinely collected to enable stratification and illuminate racial and ethnic health disparities.

Collecting patient-level DOH data should improve the accuracy of risk adjustment over time. As CMS recently noted, imputed or area-level geographic data may be a cost-efficient way to estimate social risk at a community level but should not be considered a replacement for patient-level data.

Fourth, leverage data from driver of health measures to invest in provider and community capacity to improve health. The Accountable Health Communities model demonstrated insufficient community infrastructure and resources to address DOH needs once identified. The health care system cannot address this problem alone, but across multiple payers and providers, it has the ability to generate and leverage population-level DOH data to guide upfront investments in the workforce, technology, and capacity needed to navigate patients to resources and assist communities in identifying where resources need to be targeted to address the underlying context that drives poor health and exacerbates inequity. As North Carolina and other states have demonstrated, DOH data can be leveraged by public and private institutions to direct dollars to providers and communities through the following mechanisms:

• Enable effective navigation and closed-loop resource referrals for patients (such as enhanced reimbursement of community health workers and paying for closed-loop resource referral platforms).
• Invest in community-based organizations by paying for services (for instance, by expanding Medicare Advantage supplemental benefits and Medicaid “in lieu of services”) and making “advanced payments” or pooling/braiding funding for information technology, data/accounting, or other infrastructure investments.
• Incentivize community investments by including these elements in the medical loss ratio numerator.

Enacting A National DOH Measurement Pathway

Each of the four key actions in the DOH measurement pathway—adopting, aligning, using, and leveraging standardized DOH measures—represent key steps in the pursuit of health and health equity. Exhibit 2 summarizes our recommendations for specific, timely, and pragmatic opportunities for CMS, states, commercial payers, and standard-setting bodies to operationalize these actions within their existing authority. To realize this pathway will require all these stakeholders to act, learn, and iterate—together.

Exhibit 2: Key actions to drive accountability to health

Source: Authors’ own analysis.

Authors’ Note

The authors thank Mark McClellan, MD, PhD; Robert Saunders, PhD; and Rebecca Onie, JD, for their support in development of the manuscript and for reviewing previous drafts. Shantanu Agrawal is an employee of Anthem, Inc. Gary Price is president of the Physicians Foundation and receives a stipend for service on its board of directors. He is also a committee member of the Connecticut State Medical Society. In addition, the Physicians Foundation has funded certain research referenced in this article, and both the Physicians Foundation and the Connecticut State Medical Society have taken public positions on policies referenced in this article. Alice Hm Chen is chief medical officer at Covered California; she is board chair of the Health Initiative. Rocco Perla is co-founder of the Health Initiative, which served as technical adviser on the Medicare drivers of health measures.