By Drew Ivan
What does “interoperability” mean? How will we know we have achieved it? The answers to these questions depend on who you ask and what their incentives are.
It’s common for hospitals to have efficient interoperability among their internal systems, such as their registration, billing, scheduling, and EMR systems. It’s also common for hospitals to have efficient interoperability between their billing system and the payers who reimburse their services. From a hospital IT department’s perspective, interoperability may be a problem, but it’s one that has known solutions and mature tools. That’s largely because hospitals have business incentives to connect their systems internally and externally.
By contrast, moving data between care settings, for example: from a hospital to a long-term care facility, is less common because the organizations that need to exchange data might not have an existing business relationship or a compelling incentive to make the data flow easily. Interoperability across organizations is hit-or-miss, in spite of the hope that the Meaningful Use program would make health records portable at the same time it made them electronic.
HIPAA requires data holders to provide patients with a copy of their medical records upon request; however, the process is often difficult, frustrating, and fragmented. The best case is usually that a patient can log into specific providers’ patient portals to see their records; however it can be difficult to download and impossible to aggregate the data. From the patient’s point of view, good interoperability is the exception rather than the rule.
Interoperability is strong in some settings and weak in others. As a way to address this disparity, two agencies within the department of Health and Human Services (HHS) released proposed rules recently. One notice of proposed rulemaking (NPRM) came from ONC and another came from CMS. Both agencies in part deal with the definition of interoperability and improving the level of interoperability overall. When these NPRMs have worked their way through the process and become regulations, stakeholders throughout the health care system will have new incentives and penalties to move toward a more connected health care ecosystem.
Among the many provisions of ONC’s NPRM, two sections related to interoperability stand out.
1. ONC’s NPRM defines information blocking by describing what it is not. For example, a provider or vendor is not considered to be blocking information if they fail to share electronic health information (EHI) in order to prevent harm, promote privacy or security, or because their system is down. The proposed rule defines seven very broad exceptions, but it says in all other cases EHI must be shared in accordance with applicable law.
2. The proposed rule from ONC also defines how health IT vendors must expose data through APIs, including how they can license and monetize access to the APIs. In short, the goal is to expose a minimum standard data set (the US core data for interoperability, or
The NPRM from CMS contains a different set of interoperability initiatives, including:
- Sending notifications of admission, discharge and transfer (using HL7 2.5.1) to providers that have an established relationship with the patient
- Exposing APIs for data related to claims paid by CMS
- Sharing data among health plans when a member switches coverage
- Better sharing of directory data (e.g., provider names and Direct addresses)
Both NPRMs specify that organizations should be prepared to participate in TEFCA, when that network for national health data interchange is established.
The 60-day comment period for both NPRMs ends in early May, and then there are more steps beyond that before the proposed rules are enforceable. The impact of these rules will not be felt in the near term, but their purpose is clear: to raise the level of interoperability in places it is needed most by introducing regulations and penalties that rebalance incentives for stakeholders.
Drew Ivan is Executive Vice President of Product and Strategy for Rhapsody.