A core objective of the national health care information technology strategy is achieving interoperability — the free flow of patient information across organizational, vendor, technology and geographic boundaries.
In the national discussion, interoperability is seen as the flow of information between electronic health records and as the exchange of clinical data like problem lists, medications and laboratory results. But it also encompasses the flow of information between personal health records, public health agencies, health plans and organizations conducting clinical research.
There has been some progress in advancing interoperability. As of 2014, 80 percent of nonfederal acute care hospitals can electronically query patient health information from external sources — an increase of 30 percent from the previous year. In 2014, 48 percent of hospitals routinely queried patient health information from outside their organization or system.
Unfortunately, the gains have been uneven and difficult to achieve. One may assume that the interoperability efforts of the next several years will be shaped by pursuing the path that has guided the country the past several years: defining data standards, advancing the reimbursement incentives that encourage exchange, ensuring privacy, certifying the interoperability capabilities of EHRs and addressing issues such as data blocking.
While we should focus on this path toward interoperability, we should understand the forces that will shape the country’s interoperability agenda in the years ahead. These forces — some new, some not — are likely to lead us to alter our path.
For several years the federal government has been interested in advancing interoperability. The original Health Information Technology for Economic and Clinical Health Act legislation called for the “meaningful use of interoperable electronic health records.” The resulting Centers for Medicare & Medicaid Services meaningful use incentives have certainly accelerated the adoption of EHRs. But the progress on interoperability has been less spectacular.
Earlier this year, Congress proposed legislation, contained in bills to accelerate biomedical innovation, that addressed perceived barriers to interoperability. The House and Senate took on topics such as data blocking, standards development and ratings of vendor products.
None of this legislation passed (or is likely to pass in the remaining months of this Congress), but we can expect that legislation will be proposed when a new Congress convenes in 2017. We will have a new administration, new senators and new congressmen; hence, it is difficult to predict the form of new legislation. But we can expect that Congress will be interested in furthering interoperability to support both value-based reimbursement and progress in biomedical research. This interest may focus on outcomes of care and punishing bad behavior, or it may focus on interoperability per se; the former may be more effective than the latter.
In addition, we should anticipate that the Department of Health and Human Services will continue to focus on the topic even with the arrival of a new secretary and political appointees for HHS agencies. The meaningful use program will effectively sunset shortly, but proposed Medicare Access and CHIP Reauthorization Act regulations have shown us that CMS will remain focused on advancing interoperability.
The Food and Drug Administration has entered the interoperability arena by publishing guidelines on device interoperability. The Office of the National Coordinator for Health Information Technology will likely continue its standards development work, publishing national interoperability strategies and plans.
Some state governments have begun to address interoperability by incorporating it into Medicaid requirements, evaluating technology certification and filing anti-trust actions against vendors.
As has been true for several years, the federal and state governments will be a major force in the evolution of interoperability, but there is some uncertainty about the specifics.
Collective industry actions
There are several industry groups that are tackling various aspects of interoperability. The CommonWell Health Alliance, a group of more than 50 vendors, is collectively developing patient identification, record locator and consent management services. The Sequoia Project is advancing the development of interoperability frameworks and assessment programs. The Argonaut Project, with vendor and provider participation, is developing application programming interface profiles based on the emerging Fast Healthcare Interoperability Resources standards. The Electronic Health Record Association, the Healthcare Information and Management Systems Society, KLAS and the eHealth Initiative have all developed position papers and guidance to advance EHR interoperability. The Center for Medical Interoperability is working with its members to develop standards for device interoperability. The Global Standards Initiative on Internet of Things is in the early stages of proposing standards for all industries for the emerging internet of things.
Collective industry actions have been instrumental in developing standards adopted in other industries. The SWIFT network standards — SWIFT stands for the Society for Worldwide Interbank Financial Telecommunication — were established by a global interbank consortium to enable the flow of funds between banks in different countries. The Universal Product Codes (the bar codes that we see on a wide range of retail items) were created by IBM through a contract with the Uniform Grocery Product Code Council, whose members were a collection of grocery chains and large suppliers such as Johnson & Johnson and Kraft.
In these two examples, the business case was compelling to all participants, and they realized that it could be accomplished only through collective industry collaboration. While one could argue that their tasks were easier than our efforts to achieve interoperability in health care, they were complicated enough that those efforts can teach us much and show us the power of collaboration.
While the industry groups in health care interoperability have not had the impact of their predecessors in other industries, they are making progress and have begun to combine efforts. The success of these efforts remains to be determined.
Government and private purchasers continue to transform the business model of providing care by developing value-based reimbursement approaches. When any industry undergoes significant business model changes, the industry restructures itself.
The number of accountable care organizations, willing to assume multipayer risk, continues to grow, approaching 800. The boundaries between providers, health plans and public health are blurring. Retail pharmacies and stores offer clinics to serve patients with semiurgent needs. Provider networks have expanded to include social care organizations. The decline in the number of small, independent physician practices is accelerating.
This restructuring is accompanied by the introduction of new or advanced cross-organization care processes such as care coordination, referral management and health management. With both come needs for new IT applications that span the diverse array of care participants and support these new processes. These applications support both the interoperability of data but also move health care toward process interoperability.
Health IT and policy company Audacious Inquiry outlined several of the IT applications that will be needed to support the advanced payment models proposed by CMS. These applications and capabilities include interoperability between EHRs and community-based social service organizations, care plans that function for different settings and with the patient, care event notification (such as when a patient seen in the emergency room) in various care settings, interoperable and multi-institution provider directories, and integrated multipayer and clinical databases.
The sector restructuring in health care has been a major driver of efforts to advance interoperability, ensuring the flow of patient information data through an integrated system of care. To this core need, the restructuring is adding:
- Organizations that need to be made part of interoperability efforts (e.g., social services).
- Services and applications (e.g., provider directories and care plans) that traverse organizations and provide a new form of interoperability — shared, multi-institution applications.
- New types of data (e.g., care events, multipayer claims and social services data) to the mix of data that must be interoperable.
These additions are being recognized, but initiatives that factor them into the national interoperability discussion still are embryonic.
Expanding the range of data
The types of data that must be exchanged go well beyond any core set of clinical data such as problems, medications and laboratory results.
The Institute of Medicine discussed a wide range of social and behavioral data to support the provision of care and the management of health: employment, education, stress, negative mood and affect, dietary patterns, social connections, exposure to violence, and neighborhood and community compositional characteristics. These data are important, but many of them lack standards, and the best approaches to data capture have yet to be defined.
As advances in precision medicine continue, we will need to add data describing aspects of a person’s genome and proteome.
The meaning of the term data will also expand. Currently, we view data as being discrete fields that describe an important aspect of a patient, his or her health, the care the patient has received and the settings it occurred in.
In addition to this, population health applications will need to share a précis of a patient’s overall plan for health with the EHR of a specialist treating a specific problem of a patient: “plan for health” interoperability. This can record that something happened or didn’t happen (e.g., a patient failed to pick up a prescription); this uses a different kind of data than data that describe an attribute of a patient.
The electronic health record is a platform composed of application modules: documentation, engines that process decision-support algorithms, databases, interface technologies and security capabilities. Interoperability has been targeted between EHR platforms.
"Parallel platforms" have recently emerged. Population health applications consist of modules, logic engines, databases, interface technologies and security. Population health platforms often sit on top of EHRs, extracting data from those records (and many other sources) and pushing back into the records information on gaps in care and dashboards from disease registries.
The internet of things increasingly uses a platform. This platform communicates with multiple devices, analyzes data from those devices and integrates device data with nondevice sources of data such as the EHR. The internet of things platform will send summaries of device data to the EHR and receive data from the EHR that help interpret the signal. For example, knowing that a patient has heart disease helps interpret the meaning of a blood pressure reading.
Fast Healthcare Interoperability Resources application programming interface standards effectively position EHR, population health and internet of things platforms to enable an ecosystem. These standards open up these platforms to a rich source of innovative third-party applications that will strive for a wide range of uses such as novel ways to present data to patients, addition of sophisticated clinical decision-support algorithms, and workflow applications to support telehealth. FHIR, while in its early stages, may enable these platforms to develop ecosystems just as the iPhone has enabled an innovation ecosystem. Interoperability is essential for an ecosystem.
These different platforms exist because they serve different needs. For example, a population health platform must support care management, disease registries and the ingestion of a large amount of diverse data. These needs are different from those of the electronic health record.
The platforms must operate with each other, but their interoperability needs are different. Population health platforms must provide data on gaps in care to EHRs. Internet of things platforms must provide signal analysis to EHRs. EHRs must provide a population health platform with data on a patient’s visit schedule, data that is irrelevant to the internet of things platform.
The patient or consumer
Patients and consumers clearly have an interest in the free flow of their health information. They want a complete and accurate picture of their health and their plan for health, they need information to make care decisions, and their care team needs to know what is going on and to control the flow of data about them.
These interests are leading to the creation of mobile health platforms that may emerge as potent parallel platforms. They will shape the direction of standards development, the form of privacy provisions and most of the other forces in the field.
Patients or consumers are the sleeping giant among the forces that will shape interoperability. The degree to which that giant will exercise its power is unclear.
The work ahead
Advancing interoperability in health care has been difficult. Progress has been made, but gains have been hard-fought, and there is much work to be done.
The work that has been done to date must continue: developing and testing standards, advancing incentives that encourage interoperability, protecting privacy, and developing standard business agreements between organizations that exchange data.
Health care interoperability, however, which so far has centered on exchanging clinical data between EHRs, faces several major and interacting forces that add to the complexity and uncertainty of progress toward broad interoperability. These forces are:
- Government actions.
- Collective industry actions.
- Industry restructuring.
- Expansion of the range of data.
- The advent of parallel platforms and the emergence of standardized application programming interfaces that allow for interactive interoperability.
- The patient or consumer.
A short list of clear and decisive steps to harness these forces does not exist. Rather, in our collective efforts, we must account for these factors, mold them as we can, and be prepared for the forces to alter our strategies and tactics.
John Glaser, Ph.D., is the senior vice president of population health with Cerner in Kansas City, Mo. He is a regular contributor to H&HN Daily.