Investing In The Data Systems We Need To Create The Health System We Deserve

The national COVID-19 public health crisis revealed stark truths about our nation’s health care system: who it is and is not designed to support; how its foundational technologies are often designed to advance private health system and vendor goals over the public health good; and how, in a market-based system with fragmented checks on costs and little visibility into spend, cost growth is inevitable. Provided the fleeting gift of perspective as we emerge from a year of retrenchment into a year of response and of investing in health, it is important to reflect on what we have learned and how we—individually and collectively—can confront the deficiencies in our health information infrastructure and support the creation of a more equitable, integrated, and sustainable system of health.

Racial Health Inequities Are Structurally Embedded In Our Society—And We Do Not Have The Data We Need To Address Them In Our Health Care System

According to data from the Centers for Disease Control and Prevention (CDC), Black and Hispanic Americans of similar age were nearly three times more likely than White Americans to be hospitalized with COVID-19 and twice as likely to die from the virus. The CDC acknowledges that “race and ethnicity are risk markers for other underlying conditions that affect health,” with many racial and ethnic groups more likely than White Americans to: be uninsured; have more limited access to health care services; work in lower-paying, less-stable jobs that are less likely to offer paid sick leave and more likely to be in occupations in which social distancing and working from home are not options; live in more crowded, multigenerational homes that make it more difficult to follow prevention guidelines; and be more likely to face eviction and homelessness in times of economic instability.

But the pandemic has only provided us a hastened view of what otherwise quietly unfolds every other year in the US. Black and Hispanic Americans and American Indians/Alaskan Natives are consistently more likely than White Americans to report poor or fair health, face high rates of infant mortality, contract and die from preventable viruses such as HIV/AIDS, and have and die from chronic health conditions such as diabetes. And, too often, the data we have to inform our understanding of these health inequities are limited to outcomes of our system’s failures.

Health payers, plans, and providers lack the complete, accurate, and standardized race and ethnicity (R/E) data needed to identify and address the multidimensional contributors to disparities. Efforts to collect data are frequently stymied by member reluctance to volunteer information, driven by system distrust and a lack of understanding of how such data will and will not be used; limited and uneven regulations that govern and incentivize R/E data collection across lines of business; and inconsistent use of R/E data standards among service partners, impacting collective data integrity and use. Even as late as 2015, most commercial and Medicaid plans, and more than a quarter of Medicare plans, had race data missing for a majority of their members; ethnicity data was even more incomplete.

We cannot fix what we cannot see, and our health care data has been blind to our racial and ethnic health inequities for too long. Addressing barriers to the acquisition of standardized race and ethnicity health care data is critical for ensuring all health care stakeholders have the information they need to contribute to resolving the deeply entrenched structural and programmatic barriers to better care and health for all Americans.

Our Data And Health Information Technology Systems Are Ill-Equipped To Share, Consume, And Link Patient Information

Our national COVID-19 response required our complex public health and health information technology (HIT) systems to exchange real-time information about how the pandemic was impacting our population and our health care providers. It required HIT systems that were more integrated, interoperable, connected, and scalable, and data that were more standardized, accurate, and complete than they were.

In a November report, the Congressional Research Service shared that “while Congress has long recognized the need…to monitor health care utilization and supplies [during a public health emergency], no federal data collection system for relevant information existed for the pandemic.” Furthermore, while the CDC had been working for decades to transition public data surveillance infrastructure to “more robust integrated electronic systems,” that process was not yet complete when the COVID-19 pandemic started. “Efforts to modernize public health data systems, while underway, are hindered by…a lack of standards that enable data sharing between health care entities and public health departments.” One study noted that more than 40 percent of hospitals reported that public health agencies lacked the capacity to electronically receive data. But while underinvestment in our nation’s public health infrastructure is not a new story, the same cannot be said for underinvestment in the information technology supporting our nation’s health care industry.

Health care’s HIT footprint is massive, complex, and fragmented. The electronic health record (EHR) market alone generates revenues exceeding $30 billion annually and comprises more than 700 vendors, each competing to provide clients with essential, baseline technology for collecting, storing, and managing patient data. And while the nation’s $30 billion Health Information Technology for Economic and Clinical Health (HITECH) Act-push to digitize our health records has largely been successful in terms of accelerating EHR adoption, it created modern castles of patient information without the roads—the infrastructure—to bind them, or the strong data-sharing and interoperability requirements needed to counter the development of a culture in which patient health information became a valuable, strategic, and private good for health systems and EHR vendors. New data from the Office of the National Coordinator for Health Information Technology (ONC) confirm that while more than 90 percent of nonfederal acute care hospitals nationally use certified EHR technology, only 55 percent used these systems to exchange patient data and 73 percent had challenges in electronically exchanging patient health information across different EHR systems.

While the recent interoperability rules from the Centers for Medicare and Medicaid Services (CMS) and the ONC promise to reduce data-sharing barriers, and America’s health information exchanges (HIE) remain well-positioned to bridge many of these chasms (as the Kansas Health Information Network [KONZA], Maryland’s Chesapeake Regional Information System for Patients [CRISP], and the Statewide Health Information Network for New York [SHIN-NY] all demonstrated through the pandemic), much work remains, and the costs of our past decisions continue to accumulate. The protectionism embedded in the design of our nation’s HIT infrastructure slowed the sharing of clinical information among health care and public health organizations—from patient treatment responses to laboratory and testing result reporting—and required the development of often-manual reporting workarounds.

We need HIT that supports the collective work of advancing a healthier nation and promotes data liquidity for HIEs, allowing protected patient information to safely, securely, and seamlessly travel between care providers; HIT that is scalable and versatile enough to support broader public health use cases and support our ability to address drivers of our health (which contribute to as much as 80 percent of our health outcomes); HIT that redirects market competition from data ownership to effective data use.  

Health Care Costs Continue To Rise—And We Do Not Have The Comprehensive Market Data We Need To Stem The Tide

During a year containing an unprecedented national shutdown, when more than half of adults reported they or their family members skipped medical or dental care between March and May 2020, and when the US gross domestic product fell by 3.5 percent, health services revenue still only fell by 1 percent. According to the Peterson-KFF Health System Tracker, despite an 8.9 percent drop in year-over-year health services spending during the second quarter of 2020, health services revenue rebounded by the fourth quarter (+3.4 percent year over year), leaving the sector a bigger share of our national economy than where it started—with a backlog of delayed care and vaccination spending still to be rendered.

Before the pandemic, CMS, in its National Health Expenditure Projections, estimated that national health spending would grow at 5.5 percent annually through 2027, when it would reach nearly $6.0 trillion and comprise nearly 20 percent of our economy. CMS estimated growth would be driven by demographics, as individuals age into Medicare and require higher-intensity services with higher service prices. This growth is unparalleled internationally, and unsustainable locally.

To put our global position in perspective, the United States spends about twice as much on health care per person as other wealthy countries do, with the difference only increasing over time. In 2018, total health care spending for a US family of four with employer-sponsored insurance exceeded $22,000, with nearly a third of that coming directly from family premium contributions and out-of-pocket costs, which continue to grow faster than wages. In other words, from our nation’s budget to our families’ bank accounts, health care spending is increasingly crowding out other priorities.

In lieu of federal action to stem broader market health care cost growth, state leaders—at insurance departments, Medicaid agencies, Marketplaces, and state employee health plans—often find themselves lacking the information they need to develop the cross-market policies required to contain spending in a dynamic market. Recognizing the public need, in March 2021, the Peterson Center on Healthcare and the Milbank Memorial Fund supported five states (joining three others) to establish health care cost growth benchmarking programs, which will provide their regulators with new data to understand cross-market health care spending trends. This new information—which may be paired with data from sources such as All Payer Claims Databases (APCDs)—will provide policy makers, regulators, and advocates with a critical baseline understanding of their markets, allowing for the development of more informed, cohesive, and effective cost-containment strategies while enhancing market transparency.

We need to understand the sources of health care cost growth to contain that growth. We need to have information to design the policy and program mechanisms needed to match—and check—the forces driving cost growth. Simply put, we need to equip state leaders with the data and resources they need to develop policies we need to reasonably contain health care costs.

To Strengthen Our Health Systems, Improve Data And Information Systems

As we emerge from this challenging period in our history and assess whether the systems of our society are supporting our collective cause, we should redouble our efforts to work together to strengthen our system of health by addressing the deficiencies in the data and information systems core to its efficacy. Several immediate actions we can take, under the new presidential administration, include:

Invest In R/E Data Standards, Acquisition, And Use

The Department of Health and Human Services (HHS) should work with states, health plans, providers, HIEs, and HIT vendors to advance a national data standard for R/E data collection that all stakeholders will be expected to use. Standards may build off of those established by the Office of Management and Budget (OMB), while allowing for disaggregation to identify and act on important and often hidden disparities within OMB’s broad classifications. HHS’s ONC has taken an important step in this direction, recently announcing that it would require use of the Centers for Disease Control and Prevention’s (CDC) Race and Ethnicity code set as part of its EHR certification requirements; CDC codes may be aggregated to match OMB standards. Payers, plans, and providers will play critical roles in implementing R/E data acquisition requirements—through self-reporting and indirect methods—and using such data to identify and address disparities, while HIEs may play a brokering role in filling R/E data gaps and managing a hierarchy of truth for R/E data accuracy. Health purchasers may build and incentivize the acquisition of R/E data and data use in their contracting, supporting the development of a proactive, responsive health care culture that recognizes our health inequities.

Establish A National Health Data Strategy That Meets Our National Health Data Needs

As we invest in our nation’s health data infrastructure through the Coronavirus Aid, Relief, and Economic Security (CARES) Act and the American Rescue Plan Act, we must have clear vision about what our nation’s health data and data-infrastructure needs are. We should expect nothing less than a seamless, connected, effective, and invisible data infrastructure where clinicians may readily access patients’ complete health history with permission; where health system information may be solicited and shared among public and private partners to support public health responses; where care managers and clinicians receive actionable event notifications about their patients’ needs, and have access to systems that traverse health care and health-supporting resources to connect patients to resources that address those needs; where payers and providers spend less time and fewer resources adjudicating health care claims in lieu of using the information gleaned from that data to support health interventions; and where patients do not need to be “empowered” with owning any part of a solution to an immensely profitable and frequently inefficient system.

HHS should establish a multi-agency, public-private-sector workgroup to advance a national health data strategy that identifies our health data infrastructure needs. The workgroup should identify where there are private or public-private solutions to address our country’s health needs—such as rigorously implementing our new interoperability requirements and data sharing (Fast Healthcare Interoperability Resources) standards within the new Trusted Exchange Framework and Common Agreement environment; advancing new public health-oriented EHR interoperability requirements; supporting the development and adoption of social determinants of health (SDOH) data standards (for example, HL7’s Gravity Project); addressing the legal and regulatory barriers to data exchange for physical and behavioral health care and SDOH data; and shaping a rapidly developing social service screening and referral industry to compete on how data may be meaningfully used, not on first-mover acquisition, while building on the experience of successful public-private partnerships. And where private or public-private partnerships are determined to be incapable of advancing the public good, public leaders should consider whether HIEs or other national data clearinghouses may be equipped to rationalize our data exchange system as stewards of our country’s clinical health information and, potentially, as common adjudicator(s) of provider claims against payer-derived standards.

Build State Data And Analytic Capacity To Monitor Market Performance And Cost Growth

If our country expects states to effectively govern their health care markets, we must provide our state health care leaders with the data capacity and resources (and regulatory authority) to effectively do so. According to a recent report by the Henry J. Kaiser Family Foundation, even our nation’s business leaders have strong consensus that our government must play a bigger role in “improv[ing] the transparency of prices and the total cost of care” in our opaque health care system. HHS should establish a cross-state workgroup to identify local data and regulatory needs to identify where local market oversight is beneficial and where cross-state or national action is required and to ensure activities are appropriately resourced. HHS and the cross-state workgroup should openly engage payers, providers, and health purchasers—including our nation’s major employers—in these discussions.

To promote state data capacity, HHS—potentially in partnership with the National Association of Insurance Commissioners, an existing and effective national regulatory and standards-setting organization—should work with states to establish national health care cost growth benchmarking data collection standards for states to uniformly employ to collect data about their markets’ health care cost centers and cost drivers, and ensure they are properly funded to sustain such programs. The Department of Labor, through the convening of its State All Payer Claims Databases Advisory Committee, should similarly establish uniform standards for the collection of all APCD data—not limited to self-insured group health plans—and recommend an efficient and rational mechanism by which payers may centrally submit this information nationally, while empowering states with the resulting data. States should assess how they may build on this data infrastructure to govern their health care markets more effectively, actively challenging those health care providers, plans, and other stakeholders responsible for driving up the cost of health care and driving out those who cannot afford it; confronting requests for consolidation with data on what such mergers might mean for health care spending; and focusing states’ collective authorities to swiftly and permanently address chronic market and system failures.

As we reflect on the past year, we each have an opportunity to contribute to creating a stronger, more connected, more sustainable, and more equitable system of health. Having the right data and information systems to support our journey is a necessary first step.