Why can we run our entire financial lives with a few smartphone apps, a couple of plastic cards, and an ATM network, while so many of our interactions with the health care system still rely on phone calls, copiers, fax machines, and even the occasional multi-part form? Why is the routine exchange of health care information still so difficult compared with the routine exchange of financial information? One reason is that banking has solved some foundational interoperability issues that health care still struggles with. The obstacles are numerous, but they include a lack of agreement on standard methods of information sharing, a habit of regarding patient information as a competitive advantage to be jealously guarded, and a fragmented system of reimbursement that doesn’t encourage interoperability. By gleaning three lessons from banking, the health care can overcome these obstacles.
Why can we run our entire financial lives with a few smartphone apps, a couple of plastic cards, and an ATM network, while so many of our interactions with the health care system still rely on phone calls, copiers, fax machines, and even the occasional multi-part form? Why is the routine exchange of health care information still so difficult compared with the routine exchange of financial information?
One reason is that banking has solved some foundational interoperability issues that health care still struggles with. As the longtime CIO at Partners HealthCare in Boston and, more recently, as part of my work at Cerner, a major vendor of electronic health records, I’ve watched this struggle up close for decades. The obstacles are numerous, but they include a lack of agreement on standard methods of information sharing, a habit of regarding patient information as a competitive advantage to be jealously guarded, and a fragmented system of reimbursement that doesn’t encourage interoperability.
The widespread adoption of electronic health records should theoretically mean that we don’t have to repeat our medical history for every new doctor, any emergency room (ER) ought to know we’re allergic to penicillin even if we arrive unconscious, and we’ll automatically get a heads-up if we have a diagnosis that qualifies us to participate in a clinical trial.
Patients who seek their care from a single health care system with advanced information technology, such as Kaiser Permanente or Intermountain Healthcare, may see at least some of these benefits now. But most of us don’t. The records of our checkups and blood test results are with our doctor, the information on our knee surgery or chemotherapy treatment is at the hospital, and only our pharmacy knows if we have actually filled our prescriptions.
What lessons can we learn from banking to help bring health care’s information-handling capabilities into the 21st century?
Lesson 1: Limited interoperability can address a broad range of needs.
Banks move money among themselves with the Society for Worldwide Interbank Financial Telecommunications (SWIFT) messaging system, which was established by 293 banks in the late 1970s and is now used by more than 11,000 banks, brokerage houses, securities dealers, and other institutions. They pay a membership fee and a per-transaction fee based on message volume; in exchange they get a method of quickly and securely transmitting instructions for payments and other transactions using a standardized system of institution codes and structured messages. The system delivers more than 34 million messages every day.
The health care industry has a more complicated information-sharing task than the financial industry. The parties involved may seek information for patient care, billing, clinical research, or public health. And in addition to numerical values and fixed data fields, the information may take other forms such as images, heart monitor tracings, and free-text notes. We become daunted by the magnitude of the task. But like the banking industry, we may be able to accomplish a major part of what we need by concentrating on focused and highly impactful forms of interoperability. Our initial to-do list might include notifying the care team whenever a patient receives any health care services, regardless of provider or location; automating the “prior authorization” exchange between providers and insurers; and ensuring that all providers know about a patient’s history of opioid abuse.
Lesson 2: Interoperability will only succeed when it makes business sense for the participants.
SWIFT members can use the system to quickly debit and credit the accounts of their customers, an ability that they all benefit from. However, they set their own policies for when to clear the transactions and credit the money to the relevant accounts. And they keep sole custody of their customers and their customers’ money. If you use an ATM to withdraw $ 20 from Bank A and you only have $ 10, Bank A can’t seamlessly draw that $ 10 from your account at Bank B, even if you’d really like it to. While it’s feasible technically, the knowledge that Bank A gains about your account might help it persuade you to abandon Bank B altogether, and Bank B understandably does not care for that idea.
Likewise, even though easy access to all available information on a given patient would be best for the patient, it hasn’t (up until now) been best for providers. The business case for interoperability in health care is taking shape with the adoption of value-based reimbursement models, where providers are rewarded for delivering higher quality care at lower cost and keeping patients healthier. Under a value-based system, it’s more important for providers to be able to track complete information about a patient than it is to keep that patient’s data imprisoned. A provider may risk losing money under value-based arrangements if its clinicians don’t have a clear picture of a patient’s overall condition, or if the provider can’t manage the patient’s care across multiple settings. Thus, the financial health of all providers is coming to depend on access to one another’s data.
More than a third of payments to providers are now made under some value-based arrangement, though those arrangements right now represent only a modest total increase or decrease in income for most providers. For example, while a third of a provider’s claims may be covered by value-based contracts, the amount the provider stands to gain or lose based on its performance may be only 2% of that revenue, so the total bottom-line impact of these arrangements is less than 1% either way.
The percentage of revenue at risk is likely to increase as we gain experience with value-based care. As it does, the business case for easier data sharing will solidify. Information exchange efforts will follow the specifics of value-based contracts and government regulation. They will not be driven by an altruistic regard for the patient’s well-being.
Speaking of patients… They may help drive the development of this business case for providers, as David Blumenthal, president of the Commonwealth Fund, observes. A provision in the 21st Century Cures Act requires providers and vendors of health information technology to include open application programming interfaces (APIs) with their products. Open APIs are the magic that enables a universe of apps that talk among themselves without human intervention. Once patients start to see this magic applied to their health data, they will seek out providers who use it well — just as the banks with the best mobile apps now enjoy a competitive edge.
Lesson 3. Interoperability must be driven by industry-wide efforts.
Health care needs its own SWIFT: a single organization supported by the whole industry that develops and refines interoperability business cases, establishes data and transaction standards, allocates costs, monitors industry progress, addresses any legal or regulatory barriers, and develops procedures for participation.
The industry may be about to get such an organization. In the United States, the Office of the National Coordinator for Health Information Technology expects to appoint a governance organization sometime this fall to be the “Recognized Coordinating Entity” (RCE). The RCE’s mandate will be to bring the health care industry together to establish the technical, business, and legal mechanisms and agreements necessary to significantly advance interoperability in health care. The RCE will focus on establishing a “network of networks” that links existing, fragmented health-information-exchange efforts.
This move addresses another provision in the 21st Century Cures Act. While participation is voluntary, so is participation in SWIFT. The proposed RCE will have a big job, and it may take years to know whether it’s effective, but something like it is a necessary ingredient for the health care industry to achieve the level of interoperability it needs.
Despite a decade of rapid progress on adopting electronic health records, the health care industry still has a lot to envy when it looks at the ease of data-sharing in industries like banking. But I believe the combination of an effective governing organization and compelling business cases will propel us toward a world where managing our health information is as easy as managing our finances.
Interoperability in health care is a perennial priority, but up until now the industry has taken action only sporadically and sometimes under duress. Technology and business imperatives may finally be converging to get it done.