Addressing Interoperability In Healthcare

By Michael Marchant and Mike Garzone

Healthcare organizations are notoriously complicated. Data pours in 24/7 from myriad sources, and each patient record incorporates information from numerous clinical applications, financial applications, and beyond. It’s critical these systems work together seamlessly to ensure record accuracy, and ultimately high-level care. The need for simple information sharing is clear, and yet, true interoperability in healthcare remains a challenge for several reasons.

Historically, there wasn’t a lot of incentive for vendors to make interoperability easy. Information (data) was captured, stored, and used based on the proprietary nature of each individual vendor’s clinical and financial application. Data models grew organically as these applications evolved, which were generally monolithic in nature; that is, one system and set of data serving multiple departmental functions.

As hospitals shifted to a more departmental best-of-breed approach, the need for data exchange became evident. Due to the variance between application architectures and the methods these applications used to store information, the need for translation and mapping of data as it moved between applications in complex interfaces emerged. For data to be interoperable, it would need to be used, understood, and stored in a common manner in both applications. This meant application vendors would need to cooperate and collaborate to make internal application modifications to enable a common data view.

Vendors are inherently incentivized to grow market share. When considering R&D allocation, many vendors continued to focus on developing unique differentiating features to improve the efficiency, safety, quality, and effectiveness of their applications as opposed to investments to improve the operability of third-party applications. Considerations for interfacing were included, but one could argue vendors may have lacked interest in allowing seamless integration with their competitors.

The data blocking currently being discussed is a design of the proposed loose standards and, in some cases, the intentionally closed design of the vendor applications used in most healthcare organizations. Opening the systems may eliminate the competitive advantage some vendors may have as a result of the breadth and depth of their offerings and their efforts to limit the use of best-of-breed type deployments in favor of one large vendor solution for the organization. This push and pull between an integrated versus interfaced approach has seen stems and tides in both directions over the past 20 years, from mainframe homegrown custom applications to best-of-breed and increasingly now the use of one large, often modular, software solution from a single company.

Secondly, governing bodies have tried to establish industry standards, but they’ve often lacked rigidity and have not been properly enforced. While standards bodies have existed for many years, most have had to include extensions to accommodate the significant variances between vendor applications. Early standards focused on packaging data, and these standards evolved to include interpretation of content and workflow use cases. Even so, organizations still needed to conduct significant mapping to make integration work.

In 2015, the Federal government published a shared interoperability roadmap, which has been a step in the right direction toward better connected care. The government’s involvement has meant certain mandates have been put into place, and the industry is seeing a shared interest in achieving interoperability. That said, challenges persist. In today’s healthcare environment, the key hurdles lie in the industry’s inability to identify use cases, recognize patients across entities, and make the exchange of information real-time and seamless across points of care.

There’s no argument achieving true interoperability presents complex challenges. But that’s no excuse for inaction. Healthcare organizations across the nation have a vested interest in improving system-to-system interoperability, and while challenges persist, there are certain steps that independent organizations — and the industry as a whole — can take to help move the process forward. The following explores some of the initial considerations for healthcare organizations striving to achieve true interoperability.

• Data Governance — Interoperability tends to be discussed when focusing on integration at the data level; however, data integration is not the issue. Interoperability must occur at the clinical and business level. Data must become information, and information must be used consistently across the entire continuum of care. Keep in mind that the majority of clinical workflow now spans multiple applications and multiple organizations. Our attention, therefore, must focus on how information is used at this level. This means both providers and payers must implement strict data governance policies and make smarter decisions about how information is managed at the enterprise level, and at a level that supports workflow that extends beyond their enterprise.

For example, what’s important and how do they manage those components? Implementing strict policies and educating all employees on how to follow them is critical, and employee education around data governance should be a top priority for healthcare organizations. Without strict data governance in place, integration across organizations is nearly impossible, as a lack of standards means discrepancies in individual policies and procedures.

The standards themselves must be dictated by a governing body, but each organization is responsible for making data governance a priority and educating its employees on how to adhere to those standards. Standards organizations also need to do a better job of harmonization across organizations and would be commended for providing industry governance on how organizations should implement those standards. Admittedly, healthcare has a gluttony of standards (ICD-10, LOINC, HCPCS, DRG, HL7, IHE, and EDI), but with often hazy implementation guidelines and no real harmonization or implementation guides for cross-standard implementations, many healthcare organizations are left to fend for themselves and make decisions that ultimately affect that organization’s ability to interoperate.

• Vendor Accountability — Many healthcare vendors claim to be interoperable, but there’s still a great amount of work to do. Vendors still don’t make it easy for practices across the U.S. to be interoperable, especially small private practices that lack some of the technical resources of larger organizations. Small practices normally have small (or nonexistent) IT staff and must rely heavily on their vendor partner to support interoperability and integration. Simultaneously, many vendors still struggle to achieve true plug-and-play integration, which is costly. The average interface from one of the leading vendors has an out-of-pocket implementation cost of approximately $10,000, with an additional annual operating cost of 10 percent of the implementation cost.

In addition, larger organizations have teams of people dedicated to handle integration. Small practices, on the other hand, lack the resources and budget necessary. Vendors must not only strive to simplify integration, but must also work to scale their cost structures so that all organizations, regardless of size, can integrate.

• Machine Learning — Achieving true automation and integration is dependent on computers. Machines must be able to facilitate record and information sharing, as well as ensure accuracy. Depending on humans to determine record accuracy slows processes down significantly; in order to speed efficiency, computers must be able to do this independently.

A machine should be able to understand and process an allergy notation on a medical record, for example, and ensure that medication is not dispensed for that patient without the clinician or pharmacist’s intervention. Decision support tools are efficient and effective programs within EHR software that enable such automation related to clinical content.

Similar automation within the interoperability framework and standards set capable of enabling the understanding of terminology, connectivity, and record provenance has the potential to enhance our ability to exchange patient records across organizations in an automated way. Technology enhancement at all levels and legislation requiring standard adoption and implementation requirements similar to HiTech and MACRA will most likely be necessary to facilitate this automated exchange of information and its implementation in clinical environments.

• Stay Active — Healthcare organizations must encourage their employees to be active about identifying interoperability standards and vigilant about enforcing them. Improvements will not happen passively — employees across healthcare organizations must take ownership, which means learning the standards, adopting them, enforcing them and voting on them, when applicable. Collaboration with professional organizations nationally as well as partnering with local healthcare provider organizations and vendors to facilitate conversations and programs that support this exchange of information is also needed. It will be increasingly important for those in the industry to join and participate in one of the many organizations that are pushing to drive standardization, such as: HL7 International (we are particularly impressed with the groundswell of interest in HL7 FHIR, one of the multiple HIMSS subcommittees that focus on interoperability, as well as Integrating the Healthcare Enterprise (IHE) and the Center for Medical Interoperability, among others). Whether point-to-point, regional HIE or NHIN, there are opportunities to influence the direction of the industry’s standards and to enhance organizational adoption and participation by being involved and informed.
   
• Patient Empowerment — There’s an assumption among patients that their healthcare records will be shared automatically, behind the scenes and without their input. While organizations do shoulder certain responsibility, they should encourage patient ownership, accountability, and awareness around medical records. Patients must be aware of the records kept about their health and understand how that information is shared. For those that can’t do this for themselves, a patient advocate is necessary. Healthcare organizations should implement educational sessions on healthcare records and work with patients to put more power and understanding in their hands — something which also improves accuracy.

There is no denying; however, this is a complex undertaking, particularly when the continuum of care spans multiple organizations and requires patients to interact with multiple portals in which their records are maintained — a reality that for the majority of consumers is not tenable.

One concept being discussed is the deployment of an iTunes/mobile app-like approach to help facilitate patients and record interaction using a familiar presentation method that will interact with ubiquitous patient data. Patients are aware that if they have music in their iTunes account and content on an application like Google/Yahoo etc., they are responsible for maintaining, reviewing, and responding to the information found in their account. The same principle can be applied to medical records. In this way, the mobile revolution is a good handmaiden to facilitate patient engagement and ownership of their health records.

This transition for healthcare, in which we give the patient the opportunity to be the lynchpin between organizations in relation to their health records — as opposed to having organizations attempt to make that happen in a black box behind the scenes — can expedite record exchange, especially for complicated patients with comorbidities or an extreme diagnosis where the accuracy of their record between organizations is paramount.

Addressing interoperability challenges in healthcare is no easy task. The healthcare industry is a complicated arena, with numerous players and many different data points to keep track of. Ultimately, achieving interoperability across vendors, systems, and organizations will benefit the sector as a whole, and there are certain steps that can be taken to help move the industry toward this end. Interoperability is certainly a challenge, but it is not impossible. Once achieved, it promises improved processes for healthcare organizations, and improved care for patients nation-wide.

About The Authors

Michael Marchant is the HIE and integration manager at the University of California, Davis Medical Center. He manages the integration team whose efforts assisted in earning UCDMC the Healthcare Information and Management Systems Society (HIMSS) Stage 7 award for the highest level on the HIMSS Electronic Medical Records Adoption Model and the HIMSS Nicholas E. Davies Award of Excellence, an award that honors organizations that use health information technology to improve patient outcomes and achieve a return on investment. Marchant is currently a member of HIMSS Interoperability and Health Information Exchange Committee, helping shape policy and advocacy efforts in these areas for HIMSS. Marchant has more than two decades of healthcare technology experience working with and for payers, healthcare organizations, solution providers and government agencies. Prior to his current role he led the team that built and deployed the enrollment portal for the Affordable Care Act’s EHR incentive program for California. During that time Marchant supported the portal, the state and its providers in submitting applications for more than 10,000 providers and paid more than $500 million in incentives during his tenure.

Mike Garzone is managing director at Computer Task Group (CTG), where he has responsibility for the firms healthcare technology solutions including interoperability, health information exchange, business intelligence, data warehousing, data center infrastructure, business continuity, security, cloud computing and mobility. He also manages the company’s corporate alliance program. This program creates solutions that empower healthcare organizations to successfully address a wide range IT demands that are imperative for success. A veteran executive with more than 30 years of experience in the technology and healthcare industries, Garzone previously served as chief technology officer for the U.S. healthcare division of Computer Sciences Cooperation, and vice president of healthcare technology services at First Consulting Group.