In the aftermath of Hurricane Katrina, the need for a true health information exchange in Mississippi cannot be denied.
After Hurricane Katrina destroyed the medical records of thousands of evacuees, the state of Mississippi decided to create networked electronic records that would not be lost during the next major emergency. Governor Haley Barbour created the Mississippi Health Information Technology Infrastructure Task Force in March 2007 to develop strategies for healthcare IT adoption and health information exchange (HIE) development. The task force’s action plan called for a “proof of concept” project for the six-county area where Katrina had the greatest affect. This led to the creation of the Mississippi Coastal Health Information Exchange (MISCHIE).
In the aftermath of Hurricane Katrina, the need for a true health information exchange in Mississippi cannot be denied.
Funded by a $4 million grant from the federal government, MISCHIE must hew to guidelines specifying that the grant’s purpose is to help restore health information systems damaged by Katrina. However, we at Information and Quality Healthcare (IQH), the federally designated quality improvement organization for Mississippi that is tasked with administering MISCHIE, believe that, in the long run, MISCHIE could also be the cornerstone for a statewide information exchange that would improve healthcare, lower costs and enable Mississippi healthcare organizations to cope with the next natural disaster.
When IQH, with the help of the state IT agency, began to seek an HIE vendor, the organization knew that time was short to set up MISCHIE since the grant was available only for one year. That’s one reason why IQH selected Medicity and Perot Systems, two organizations that had already built a successful, statewide HIE in Delaware. Also, they had been chosen as the architects of CalRHIO, the statewide exchange that is under construction in California. Since the firms were used to working together, we anticipated that they could implement our HIE rapidly and effectively.
A well-designed HIE should enable bi-directional integration between practice and hospital EMRs, as well as interoperability with other physicians’ EMRs or outside systems, such as reference labs. Patients should also be able to integrate their medical histories with a personal health record (PHR), such as Google Health or Microsoft’s HealthVault.
Sustainability is Key
Like other HIEs across the country, MISCHIE faced the challenge of developing a sustainable business model. The value propositions include better quality of care, overall cost reductions for the healthcare system and administrative savings for providers. In the quality arena, information exchange can lead to fewer medical errors, fewer adverse drug events, a better ability to manage a patient’s chronic conditions, better coordination of care across care locations and physicians, and fewer admissions. Clearly, all of this will improve efficiency in the healthcare system and reduce the friction that currently exists when a patient moves between care settings.
Unfortunately — or fortunately — Hurricane Katrina re-invigorated our focus and attention on the need to put medical records online and establish connectivity amongst providers across care locations. But every community should have an HIE, regardless of whether it is located in an area prone to natural disasters.
The MISCHIE solution will be centrally hosted at Perot Systems’ secure facility, and Medicity will provide the clinical interoperability platform and the end-user application infrastructure for physicians to deliver high-quality, cost-effective care throughout the state.
Where’s the Value for Doctors?
For more information on
Medicity’s information
exchange solutions
More HIEs and regional health information organizations (RHIO) become operational each year (42 in 2008 from 32 in 2007, according to the eHealth Initiative); however, the increase may not be as substantial as many had hoped. Several experts say the main obstacle to increasing the number of HIEs and RHIOs is the lack of a sustainable business model. While financial sustainability is a significant issue in many cases, the problem of financing these organizations may be related to a more fundamental issue. That is, most providers have not seen a compelling need to use their services, much less to pay for them.
At a high level, the reasons to exchange clinical data are obvious: Enhance patient safety with better access to comprehensive clinical information about the patient; optimize clinical workflow by delivering clinical results more quickly to clinicians; reduce costs by eliminating redundancy in clinical tests and procedures; reduce the friction that currently exists in healthcare delivery by enabling physicians to collaborate on patient care; and, improve handoffs across providers and care locations.
Clinical interoperability, whether at the local, regional or state level, can lower the costs of healthcare delivery. Moreover, some observers believe that physicians are more likely to adopt electronic medical records (EMR) if those systems automate the exchange of patient information with the clinical systems of hospitals, labs, pharmacies and other healthcare providers, as those systems and care locations contain the essential data they need to populate their EMRs and, hence, enable them to deliver higher quality, more cost effective patient care.
Workflow Solutions
For purposes of treating patients, physicians require access to key clinical information about patients, such as problems, medications, clinical results, procedures, immunizations and allergies. Ideally, this information would be accessible by physicians directly within their EMR. However, not all physicians have EMRs in place and, therefore, to serve the many physicians who are not currently using EMRs, an HIE should enable access to the patient-centric clinical information via the Internet or mobile device so that it is readily accessible from any care location. Such access to clinical information should ideally empower physicians with a longitudinal and chronological view of every patient across inpatient and ambulatory care settings. Clinical results should be delivered directly to physicians in the Web-based solution so that doctors can quickly be alerted to new results.
Several experts say the main obstacle to increasing the number of HIEs and RHIOs is the lack of a sustainable business model. While financial sustainability is a significant issue in many cases, the problem of financing these organizations may be related to a more fundamental issue. That is, most providers have not seen a compelling need to use their services, much less to pay for them.
Additionally, physicians should be able to immediately navigate from a given result to the overall patient chart view, so that a comprehensive assessment of the patient can be made. Moreover, physicians should be in control of how and where they want results delivered — at the point of care, in their office, at home, while rounding or to their mobile devices.
Other essential features in a Web-native, physician-centric solution include automated electronic prescribing, chart signature, order initiation, and referral and consult management, so that clinical workflow is optimized and physician efficiency is enhanced.
A well-designed HIE should enable bi-directional integration between practice and hospital EMRs, as well as interoperability with other physicians’ EMRs or outside systems, such as reference labs. Patients should also be able to integrate their medical histories with a personal health record (PHR), such as Google Health or Microsoft’s HealthVault.
Making the Business Case
HIEs have been shown to improve healthcare and lower costs. For example, Shared Health, a subsidiary of Blue Cross and Blue Shield of Tennessee, has developed a RHIO using claims and clinical data that has yielded a 17 percent efficiency gain in the treatment of chronic disease. Using Shared Health data, EDs have performed 40 percent fewer services, resulting in a 21 percent reduction in cost. In another study, the company’s researchers demonstrated that patients of doctors who viewed Shared Health data received 5 percent more screening mammograms, 7.7 percent more colon cancer screenings and 13.2 percent more flu shots than did patients in practices that did not use Shared Health’s information.
Regardless of the demonstrable benefits, Shared Health has struggled to spread its message to doctors and the same is true for other HIEs and RHIOs throughout the country. To achieve their goals, the organizers of information exchanges must make it easy for physicians to consume communitywide clinical data. Training physician offices to change their workflows to some extent may include combining electronic data with a paper-based workflow or connecting their EMRs to other providers, such as hospitals, physicians and reference labs. These modifications will deliver more value to the physician and empower the broader HIE, as well as improve their clinical effectiveness and operational efficiency.
As physicians find that an HIE helps them do their work more effectively and improves patient care, they will use it often and might even be willing to contribute to the financing of the solution.
James S. McIlwain, M.D., is president and CEO of Information & Quality Healthcare. Kipp Lassetter, M.D., is CEO of Medicity. Contact them at [email protected] and [email protected] .
Data Governance Models for HIE Initiatives
Three prominent models for data governance exist today: centralized, federated and hybrid. In all three cases, healthcare organizations’ internal policies for data sharing and enterprise security are taken into account during implementation to ensure the data gets stored, secured, exchanged and managed appropriately. This is critical to gaining support for the exchange initiative.
Centralized Data Repository
Data Storage and Security: A centralized model offers providers advantages over the federated model including economies of scale for the technical infrastructure and enhanced data security. All participant-contributed data gets stored in the same physical repository with logically (or “virtually”) segmented participant data to prevent co-mingling of information. All management of participant connections to the repository take place over secure channels with only authorized clinical and administrative users allowed access to a patient’s community health record.
Role of a Master Patient Index (MPI) and Record Locator Service (RLS): All of the data governance models connect with an MPI as a shared service that layers patient metadata over the underlying HIE data repositories to resolve patient identities across care locations and systems. Given the centralized deployment, an RLS can typically be bypassed; however, the role of an RLS becomes critically important in a federated or hybrid model.
Management: This option offers the lowest amount of administrative work for the participants. Typically, Service Level Agreements (SLA) are established to ensure reliable security, connectivity and availability of the centralized infrastructure.
Federated Architecture
Data Storage and Security: The appeal of the federated architecture (also known as peer-to-peer or distributed architecture) lies within each provider retaining control over its own data and not co-mingling their data with other data providers. This approach reassures providers who are concerned about the security and privacy of data in community exchanges and allows each provider to retain its competitive advantage with data ownership. In the federated model, data is shared across the HIE, but the data is not stored in one central repository.
Role of an MPI and RLS: In the federated model, an RLS is essential to identifying the location of a specific patient’s information across the network, while an MPI reconciles the patient’s identity.
Management: The HIE depends on a highly reliable network with an advanced security framework that requires connectivity with every clinical system inside all participant organizations, and includes a stringent set of SLAs for query response times. Autonomous software “agents” can also be deployed to the edges of the network, facilitating the exchange of clinical data among HIE stakeholders.
Hybrid
Data Storage and Security: To address providers’ fear that their competitors will see their data within a centralized model, an advanced clinical data repository can be segmented so that each provider’s data is separated from data provided by other healthcare systems and physicians. Providers who want to control access to their data can use edge servers that are part of the HIE infrastructure, but are hosted by the facility where the data resides. This hybrid model has gained significant traction in advancing several successful statewide HIE initiatives.
Role of an MPI and RLS: Every HIE needs a “community” MPI that exists independently of the disparate MPIs of participating providers and their information systems so that patient identity can be reconciled across multiple care locations and providers. A true community MPI should support a wide variety of patient, provider, health plan and other identifiers, and have configurable algorithms that can be customized to the contributing data source. If the central repository keeps individual databases separate in a hybrid model or links edge servers together, it will also be necessary to use an RLS. This utility creates a virtual patient record composed of metadata descriptors that can be queried to retrieve information across data stores.
Management: A hybrid model, as the name suggests, offers the benefits of a centralized model with the security of a federated deployment. This deployment has a slightly higher amount of maintenance activities than that of a centralized model, but nothing as complex or sophisticated as a true federated approach.
The debate over the best repository model continues. All three models enable HIE stakeholders to access information via Web-enabled devices, such as tablet PCs and PDAs. Information also can be pushed to data consumers in “offline” modes, such as auto-print or auto-fax.
During natural disasters like Hurricane Katrina, pandemics or bioterrorism, all of these deployment models enable providers to continually provide care when and where it is needed by exchanging information in a highly secure, scalable environment that ensures the preservation and accessibility of the patient identity and medical history.
