A Practical Approach to Interoperability: Base the Solution on the Problem
In the U.S., most patients are free to seek healthcare wherever they choose, and they often do so without notifying their primary care provider. This behavior presents a number of challenges for providers engaged in managing the care of high-risk patients, who can be extremely difficult to keep up with.
To make matters worse, patient data is not centrally stored. With more than 700 certified electronic health record (EHR) vendors, a patient’s information might be scattered across more than a dozen disparate health record systems. The software vendors that boast sky-high market share often have a significant share of only one specific segment of healthcare (inpatient, post-acute, ambulatory, home care, etc.), so they’re not holding data across the full continuum of care. In addition, these systems have their own proprietary features, unique nomenclatures, and disparate databases that create major obstacles to easy sharing of information between systems. And depending on the organization, an EHR might also be set up and used in several different ways, which dramatically slows and complicates data sharing between providers.
Patient consent for data sharing, governed by various federal and state laws, as well as organization-specific policies, can be another obstacle to information exchange. Moreover, while information blocking is now illegal under federal law, financial motives and territorial instincts can make some organizations reluctant to transfer patient data to competitors. When providers don’t share information, patient safety suffers. Patients wind up being put through duplicate or unnecessary procedures, receive incorrect prescriptions, are placed in a potentially inappropriate care setting, or worse.
How will these problems be solved?
Start with A Use Case
One of the most ambitious efforts to share data has been the development of public health information exchanges (HIEs), which exist in nearly every state. While some HIEs have succeeded by addressing the needs of the market, many of them began with an “if you build it, they will come” approach. That is, they pull together lots of data into one place, then consider which use cases the data can support.
This is in many ways a backwards approach. Great healthcare software starts with a clear and very specific problem that has been articulated by providers. Solutions are then built around that problem, iterated on, and then expanded to other use cases to solve more problems.
This is often referred to as the “bowling ball strategy” in software product development. You start by addressing one very specific problem for one very specific provider segment, then you expand to solve the same problem for more segments and more problems for the same segments, and so on. Amazon, for instance, started with a clear need (selling books online) and then expanded to music and everything beyond. eBay started with individuals obsessed with buying and selling collectibles and then expanded to everything else. Uber started with higher income riders looking for black cars, and then expanded to serving nearly all riders.
Health information sharing is no different. It needs to start with a clear, articulable problem. Many HIEs have begun to expand the services they offer by adding capabilities and partnering with specialty organizations focused on solving narrow but important problems.
Find out what providers need
To identify what kind of data is required to achieve a patient care goal, it’s important to understand what the clinician and the care team need to know right away. In the case of post-discharge care, for example, ambulatory-care physicians want to know “has my patient been discharged?” and “where is my patient now?” This lightweight—but critically important—information generally does not require patient consent to share (assuming the providers are attributed to the patient), so it can be quickly passed from one provider to the next within the right infrastructure.
To monitor and follow up with high-risk patients following hospital discharges, the most useful and valuable information is found in ADT (admit, discharge, transfer) feeds from acute-care hospitals. Access to real-time ADT data can help prevent patients from being readmitted by letting primary care physicians know when their patients have been discharged. Equipped with that knowledge, PCPs can make sure that high-risk patients visit them as soon as possible and monitor their care following discharge. By cooperating with outpatient providers to reduce readmission rates, hospitals can reduce or avert Medicare penalties. A win for all.
Improved care coordination after patient discharge also creates opportunities to reconcile medications, follow up on hospital test results, and ensure patients are assigned to the appropriate care settings. In addition, better care coordination can improve the quality of care and patient satisfaction and reduce leakage to providers outside the healthcare system or accountable care organization. All of this can be very helpful to provider organizations that are increasingly taking on financial risk for patient care.
Make the data easy and flexible to consume
Once the use case has been identified and the data has been gathered, it’s important to ensure that the data can be consumed easily by the provider. One way to do this is to build an application around use cases. Such an application can flag high utilizers, patients at risk for readmission and patients associated with different types of billing events (3-day waiver, transitions of care management, etc.). Most providers prefer to view the data in their EHR or their care management software, so it’s important that healthcare technology vendors work together to make access to important patient data seamless for the user.
Instead of waiting for the Holy Grail of interoperability to emerge, healthcare providers should find ways to adapt current technology to meet the key challenges they face as they try to improve the quality and reduce the cost of care. In evaluating new approaches, their first question should always be “what problem will this technology solve?”
About the author:
Brian Manning is SVP, Head of Growth at PatientPing, a Boston-based care coordination platform that reduces the cost of healthcare by seamlessly connecting providers to coordinate patient care. The platform allows providers to collaborate on shared patients through ‘Pings’– real-time notifications when patients receive care – and ‘Stories’ – important patient context at the point of care.
For more information, visit our Four Questions Series blog to learn how healthcare experts from across the continuum use technology to help solve care coordination challenges.