Karen De Salvo in todays Health Affairs on ONC's and CMS's proposed regulations and the Nationwise interoperability Roadmap
The Nationwide Interoperability Roadmap And The ONC’s And CMS’s Proposed New Regulations: Are We Halfway There Yet?
Karen DeSalvo, Mark Savage
In 2015, the ONC’s Connecting Health and Care for the Nation: A Shared Nationwide Interoperability Roadmap outlined the steps necessary to implement a nationwide “learning health system” by 2024. This system would be capable of improving individual and population health, empowering consumers, driving innovation, addressing health disparities, advancing precision medicine, and supporting value-based care and reimbursement. The 21st Century Cures Act—bipartisan congressional legislation passed in 2016—reinforces the aims of the roadmap by requiring, for example, that application programming interfaces (APIs) allow complete access, exchange, and use of all electronically accessible health information on a national scale. The 21st Century Cures Act made interoperability and patient access national imperatives and targets improving the quality and coordination of health care, population health, health disparities, and health information technology to enable care and monitoring in homes and communities.
While the digital health ecosystem has progressed rapidly over the past five years, recent regulations proposed by the ONC and CMS push the public and private sectors to take an overdue leap forward on the roadmap. Some may suggest that the draft regulations move too fast in some areas, while others may urge that we catch up in key areas to reach the roadmap’s goal of nationwide interoperability by 2024. We suggest using a set of four core use cases to assess the sufficiency of the draft regulations and to mark our progress on the road to an interoperable nationwide learning health system.
The Nationwide Interoperability Roadmap, 2015–24
The ONC’s 2015 roadmap synthesized the input of stakeholders and experts to outline the path to an interoperable health ecosystem of individuals, care providers, communities, payers, public health agencies, researchers, and innovators. This new ecosystem would prioritize making reliable and meaningful data available to the right people at the right time, with the appropriate consent.
The roadmap set goals and milestones; carefully mapped out and coordinated strategies for 2015–17, 2018–20, and 2021–24; and called for action in the private and public sectors. By the end of 2017, practitioners and experts could send, receive, find, and use priority data domains to improve health care quality and outcomes. By the end of 2020, the nation should have an expanded set of data sources and users, and the beginnings of an interoperable health care ecosystem. By 2024, we would have a nationwide learning health system.
One year later, Congress reinforced the roadmap’s vision and passed the 21st Century Cures Act by an overwhelming margin. The Cures Act made secure interoperability a national imperative and affirmed the importance of “complete access, exchange, and use of all electronically accessible health information for authorized use under applicable State or Federal law...without special effort on the part of the user.” Congress required a trusted exchange framework to facilitate coordination and communication between health information networks. The ONC subsequently proposed a single on-ramp to electronic health information (EHI)—a learning health system—for authorized providers, individuals, payers, public health agencies, federal agencies, health information networks, and technology developers.
Since the roadmap’s publication in 2015, there has also been increasing attention to social determinants of health, person-reported outcomes and person-generated health data, shared care planning, and health care beyond the clinical setting. The health care sector has also been influenced by broader discussions about the power of big-data analytics and data from the Internet of Things. New digital diagnostics and therapeutic (dD&T) software applications are steadily expanding diagnosis and treatment outside of the clinical setting and into homes and communities while enabling real-time measurement of health status and outcome variables. Every day we recognize the need for interoperability and the multidirectional exchange of health information. But, there can be significant lag between the needs we face and the interoperable systems we must build to meet them. Fortunately, the roadmap anticipated these developments and provided a comprehensive structure to create and coordinate the necessary capabilities.
Now the ONC and CMS have released intricate draft regulations to implement key portions of the 21st Century Cures Act. With so much detail, one might forget to step back and check these new initiatives against the interoperability roadmap. What is our polestar, and are we halfway there?
Four Use Cases To Track Progress Toward Nationwide Interoperability In 2024
As stakeholders and experts wrestle with whether we are moving too fast or too slow, we suggest using four core use cases to assess interoperability proposals and to monitor our progress on the road to an interoperable nationwide learning health system. These four use cases are national priorities for interoperability and should help the ONC, CMS, and stakeholders align abstract discussions with real and immediate needs. They help explain and assess how a proposal would work in real-world situations and allow us to identify the opportunity costs of a road not taken.
1. Individuals’ And Patients’ Electronic Access
Congress declared individuals’ access to their EHI a national imperative. People must be able to not only view their EHI, but also use it—to download it, transmit it to any provider or person, and use it with their apps of choice. Individuals cannot effectively manage their health and health care without ready and convenient access to information about their medications, health status, diagnoses, and treatment instructions whenever and wherever needed. This use case is not limited to certified electronic health records (EHRs) and the view/download/transmit/API criterion—it applies across diverse modes and settings of access and use, from personal health records to home monitoring devices; from community health centers to urgent care clinics to nutritionists.
The ONC’s proposals to move from merely open APIs, to open, standardized APIs, and to export all of a patient’s EHI, are significant leaps forward on the roadmap. CMS’s proposal to require payers to provide access to claims, encounter data, utilization history, and clinical health information with standardized APIs is an equally significant leap forward in the transition to value-based care. As the interoperability roadmap stated, “a learning health system must converge on a limited set of standard APIs to support a core set of services that enable EHI to flow when and where it is needed.” However, the proposal to wait until January 2022 before standardized APIs must become operational lags two years behind the interoperability roadmap, at significant cost. The ONC and CMS could design incentives to implement standardized APIs earlier.
2. Shared Care Planning And Coordination
Better care, health, and value depend upon better communication and coordination among providers, patients, family caregivers, and others who coordinate a person’s care outside the clinical setting. Shared care planning and information coordination are essential. This must go beyond the static, episodic “plan of care” buried in one provider’s EHR. Instead, patients need a dynamic, longitudinal shared care plan and planning platform that allows everyone involved to update the plan electronically and in real time. Providers cannot succeed under new models of care without empowering patients to engage actively in shared care planning and decision making.
The ONC affirms that a dynamic longitudinal care plan and shared care planning are “critical” for managing a person’s health across the continuum of clinical and non-clinical settings. However, the draft does not go beyond requesting comment on the maturity of existing and forthcoming technical specifications, and only applies to opioid use disorder prevention and treatment. The interoperability roadmap, in contrast, planned for individuals to be able to “seamlessly integrate and compile longitudinal electronic health information across online tools, mobile platforms and devices to participate in shared decision-making with their care, support and service teams” by 2024. We need to do far more, now, than collect comments on the maturity of standards, if we are to meet this “critical” interoperability goal by 2021–
3. Person-Generated Health Data
Providers, patients, researchers, and payers all recognize that one-way access to health information is not enough. Access, interoperability and data portability must be bi-directional, so that patients have access to their electronic health data and providers have access to real-time patient-reported outcomes and environmental health data. Accountable care organizations (ACOs), precision medicine initiatives, and efforts to reduce health disparities all depend on the ability to know and integrate patient-reported outcomes and patient-contributed health data.
The ONC’s proposed move to open, standardized APIs is again a significant step forward on the roadmap. However, the ONC proposes to continue requiring only “read” APIs, until Fast Healthcare Interoperability Resources (FHIR)-based APIs are widely adopted in 2022. To align this goal with the roadmap, the ONC need only specify that new APIs provide both “read” and “write” access. This would make a far more significant leap forward on the roadmap. As the roadmap notes, “Changing the paradigm to a person-centered ecosystem is vital to improving health, given that an individual’s actions inside and outside the clinical care delivery system greatly impact health outcomes. Moving forward, the health IT ecosystem needs to put greater focus on (1) incorporating patient-generated health data across health IT products and services....” The ONC might also prohibit API technology suppliers from charging patients a fee for “write” access as well as “read” access, and require real-world testing of this criterion, too.
4. Social And Environmental Determinants Of Health
Medical care delivery determines an estimated 10–20 percent of health status. Factors outside the clinical setting, including socioeconomic factors and the patient’s physical environment, determine the remaining 80–90 percent. By 2024, a nationwide learning health system must integrate non-clinical data and facilitate interoperability with non-clinical settings to account for sociodemographic factors such as the physical environment, nutrition, education, housing, and access to social services.
The interoperability roadmap targeted the broader learning health system of clinical and non-clinical data and services precisely because individual and population health depend upon much more than clinical care. Integrating social and environmental determinants is critical. The ONC’s proposed move to open, standardized APIs and to enable the export of EHI for population health services are steps in the right direction. But, equally critical next steps are missing. For example, the ONC proposes to add new elements to the US Core Data for Interoperability. Given the extraordinary significance of social determinants of health, the ONC could add them now without further delay. The 2015 edition already includes a voluntary health IT module with standards for eight categories of social, psychological, and behavioral data. We must begin including social determinant data now to learn how best to integrate and use them on the road to a nationwide learning health system by 2024.
These four use cases help evaluate the important new proposals from the ONC and CMS and whether they go far enough to keep us on track to reach nationwide interoperability by 2024. Over the next five years, stakeholders must consider how new initiatives align with established interoperability goals and whether the benefits of those initiatives outweigh the costs of implementing or not implementing them.