On paper, the strategies behind health information exchange (HIE) formation seem solid enough, but not without the requisite bumping and grinding that accompanies behavioral change management and process improvement.
Creating a system that integrates information flow, analytics and workflow for population health initiatives extends beyond typical plug-and-play tactics. Still, under the relatively short period of time HIEs have been in development and operational within the auspices of the Affordable Care Act and industry reformation efforts, providers and other healthcare organizations have been learning some valuable lessons.
Looking back at health information exchange development, what are some of the opportunities and challenges they revealed, and how might healthcare organizations move forward with that intelligence to redefine and redesign what the industry – and patients – really need?
Health Management Technology set out to examine the progress of HIEs through the viewpoints of a group of industry experts.
HMT: In what specific ways have HIEs improved clinical and financial operations for healthcare organizations?
| Kathleen Aller, HealthShare Director, InterSystems |
Aller: By connecting the dots between clinical, administrative and patient data, HIEs enable both providers and health plans to deliver more proactive, more effective and more coordinated care. A great example is automated clinical alerts based on information exchange.
Alerts allow instant notifications to both medical professionals and health plan care managers when a patient enters an emergency room or is admitted to or discharged from a hospital. Primary care providers can use these to better manage care transitions and reduce readmissions. Health plans can assist members to better navigate the referral system and to optimize follow-up care plans. This rapid information sharing and intervention is especially critical for patients with complex or chronic conditions.
| Greg White, Senior Vice President and General Manager, TouchWorks Business Unit, Allscript |
White: Clinically, the sharing of information will lead to safer and more effective care. By facilitating better and more transparent communication, there’s more opportunity to focus on quality of care. Clinical improvements fundamentally lead to reduction of medical errors and improvement with regard to appropriate treatment delivery, all of this in anticipation of creating a more cost-effective healthcare system with less waste.
In particular, one of our Allscripts clients has earned a national reputation for its visionary approach to health information exchange and population health management. Over the past three years, they have:
1. Initiated data sharing across a diverse community that relies upon more than a dozen electronic health record (EHR) systems
2. Identified “hot spots” within its service area where specific disease states are prevalent – and launched targeted programs to engage affected populations; and
3. Become one of the first healthcare organizations in the country to meet the electronic transitions of care requirement for Meaningful Use Stage 2 incentive payments.
This California-based client has partnered with Allscripts to effect real change in how they treat and manage chronic conditions in their service area. They are currently planning a multi-faceted population health program using the aggregated data and advanced analytics captured by our Allscripts dbMotion community connectivity platform. This information helps provide the clinical intelligence and insight to develop appropriate outreach programs. This will then be layered with patient engagement strategies with the goal of changing behaviors that contribute to chronic disease.
| John Kelly, Principal Business Advisor, Edifecs |
Kelly: If healthcare stakeholders were polled in March 2015, I believe most would reply that investments in HIEs have not returned any substantive value. Healthcare information exchange was intended to be an activity, not an organization. HIEs are essentially replacing phone, fax, mail and Post-It notes while making patient information ubiquitous at all points of care. In an Internet Age where music, movies and opinions can be transmitted “virally” using light, local apps and cheap bandwidth, the idea that HIEs have been built using hub-and-spoke frameworks with transaction fee-based business models is both troubling and perplexing. In hindsight, while rapidly consolidating integrated delivery systems struggled to get their various clinical systems to talk to each other, and population health proponents sought the holy grail of the big clinical database in the sky, it is perhaps not surprising that the HIE model that has emerged mimicked the proprietary private network models and clinical-data repositories that evolved prior to the Health Information Technology for Economic and Clinical Health (HITECH) Act. It’s also not surprising, though, that last year’s JASON report indicted the industry for blindly ignoring a decade of telecom and info-tech advancements while going on a health IT spending spree with perhaps $50B-$75B of public and private money.
What has gone right is that some of that money has been spent to learn about the use of electronic patient information to support better, more efficient care. Those lessons are slowly percolating through the industry consciousness. Perhaps the best result is that for the first time in 40 years of failed starts at creating electronic patient records, a critical mass of healthcare providers is using electronic medical records (EMRs) to generate structured clinical information. As the business cliché goes, “you can’t manage what you can’t measure.” With that structured information beginning to come online over the next couple of years, we can start to measure process and outcomes without resorting to proxy data from claims and secondary process measures. As every industry has found since Deming’s first efforts with the automotive industry, once you can measure, you can effect real change. The real challenge now will be for the provider community to accept the inevitable changes in the way healthcare is delivered. Physicians haven’t seen change of such scope since states began building medical schools in the wake of the Flexner Report of 1910.
| Michael Macaluso, Director, Product Management, Document Management and Intelligent Coding, McKesson |
Macaluso: It has caused healthcare organizations to examine their policies and behaviors when it comes to medical information originating outside of their healthcare organization. In the past, healthcare information was either not available, difficult to find or not from a trusted source. This caused many providers to immediately order potentially duplicative tests. With HIEs, a provider now has a reliable source for finding healthcare information, which is starting to reduce the incident of duplicative test.
It has also caused healthcare organizations to examine their policies and behaviors on how they share their own information. In the past, many healthcare organizations viewed the clinical information acquired through patient encounters as something not shared outside their organization. However, if clinicians are to take advantage of regional healthcare information, all providers in that region must share their information for any organization to receive value. This has caused many organizations to soften their policies and begin sharing the information externally.
Traish: Clinical and Financial operations can benefit from HIEs, though results vary depending on the HIE business model and architecture. A small private HIE for The Washington Hospital and Physician Organization in Pennsylvania demonstrated value in both. Paper reports were turned off, faxing was eliminated between organizations, pre-operative clearance was simplified. Secure messaging, care transitions and access to results were significantly faster and with greater integrity. With pre-operative testing moved to the hospital, it was better for patients and revenue operations. Exchange of information was paid for through grant funding, hospital-paid interfaces and physician organization funding.
It’s a good step in the right direction – though they identified the same challenges every exchange faces today: The exchange only meets the needs of providers, operations and patients in the network. Step out of network and you’re back in the data vacuum. Data standards are a challenge, too. They developed for the needs of the 12 EHRs they use. It’s unclear how this can work with the regional or national exchange hypothesis. Patient consent and secure transactions still needs to be solved on a larger scale.
CMIO,
Nuance Communications
Terheyden: The most important clinical benefit is arming clinicians with patient information across sites or state lines. That benefit was realized by Norman Regional Health System when a Level 5 tornado tore through several hospitals in Oklahoma, injuring hundreds and wiping out medical records when patients were actively being treated. Tapping into the state HIE helped clinicians keep patients safe, providing a portable patient record of patient histories, medications and allergies.
As a physician, if a patient has been seen by another physician earlier today – or even last week – I want and need that information to deliver quality patient care. The idea that physicians don’t always have this kind of information is troubling to say the least.
From a financial aspect, HIEs and information sharing can help reduce unnecessary healthcare procedures, specifically in regards to imaging and testing. If you do two chest X-rays or conduct two lab tests that show the same thing, that’s unnecessary waste and somebody is paying for it. Unfortunately, HIEs don’t handle images well today but they should and can.
Another benefit in all this is consumer education. Historically, physicians gave patients information, and most patients forget a lot of what they were told, particularly in stressful situations. HIEs give patients the ability to review information, confirm its accuracy, and keep them involved in their own health. This kind of engagement and education can help reduce the amount of care needed, while still improving quality and value.
Systems Made Simple
Underhill: HIEs improve clinical operations through a better understanding of a patient clinical complaint, condition and care pathway, leading to improvements in healthcare business practices, such as care planning, delivery and co-ordination between care givers.
Patient outcomes are improved through better safety, care giver decision making as a result of access to a broader health record, timely and efficient medical record exchange and care co-ordination between providers.
IT Infrastructure is improved through availability of a standards-based set of data services that are readily available and can be consumed by any applications within your enterprise.
Interoperability is enhanced though data standardization and best practice implementation guidelines.
Increased data liquidity enables data analytics leading to improvements in patient care and financial health via financial operations by automated exchange of information, standardized structuring of information, reducing waste and abuse of services.
HMT: What’s at stake for the credibility and effectiveness of HIEs if more don’t succeed or if HIEs don’t become the predominant model?
Aller: We should distinguish between private HIEs and public HIEs. Regional and statewide HIEs may succeed or fail based on factors unrelated to their underlying technology. Theirs is a mixed report card: some winners, some losers. Private HIEs, in which data is shared within and among health systems and affiliated organizations, are a different story. These HIEs are essential to coordinated, patient-centric care and to success in a value-based payment environment.
A case in point is a large health system in New York state that was relying on a regional HIE to monitor high-risk obstetrics patients. The HIE technology became so essential to ensuring comprehensive, up-to-the-minute information for better and more connected care that the organization took the technology in-house as an interoperable platform for its own strategic and clinical initiatives. It has since become a central component of its response to risk-based reimbursement models.
White: Because every health plan, hospital, clinic, lab, pharmacy and government agency has disparate IT practices and capabilities, creating an HIE that will work for every member of a healthcare ecosystem can be difficult. The model or moniker for data sharing across a diverse community is seemingly irrelevant – the notion of managing the health of populations, surrounding an organization and extending beyond relative borders (whether they be practice, health system, surrounding community, etc.) should be the top priority.
Connectivity platforms such as dbMotion can be used to share and aggregate data among a wide variety of disparate systems. This type of EHR-agnostic interoperability engine semantically harmonizes patient information from different clinical systems into a central data repository. The platform then makes this information available within a provider’s native workflow. This ability to harmonize the data gives providers the power to integrate a large number of practices and have all of a patient’s data readily accessible.
This collaboration and integration helps to improve quality and care coordination, and access to information has been shown to enhance provider satisfaction and loyalty.
Macaluso: The current trend in HIEs is to focus on the discrete clinical data flowing from the patient’s medical record. This information is incredibly important, but this discrete data does not tell the entire story of the patient’s health history. Many parts of the patient’s medical record are tied to unstructured data (i.e., scanned documentation, electronic content from ancillary systems, etc.) and are thus not available through an HIE. This provides clinicians with only a partial view of the medical information for a patient under their care.
The other major challenge to the effectiveness of an HIE concerns the ability for a clinician to easily find the relevant information for their care decisions. With HIEs already starting to acquire data at a rapid rate, patients with frequent healthcare encounters will provide a challenge for clinicians to focus on useful data. If physicians spend too much time filtering out information hunting for what they deem important, their frustration levels will rise and their adoption of the system will fall off.
Kelly: Since the general perception is that HIEs have not emerged as sustainable entities as they are currently configured, the good news is that they are in a state of credibility reformation. They are doing that within the context of intense public scrutiny that extends across Congress, the White House, provider and payor lobby organizations, as well as the press at large. Just as clearinghouses are facing the intense pressure of disintermediation, the HIE vendor community is realizing that the technology they sell must fundamentally change to focus on four key things:
- Scalable trust models that use a shared definition of the process for certifying identity and authorization for basic access;
- Lightweight edge clients that are integrated into EMRs and handle packaging and transport of payloads as attachments that integrate with the local EMR message workflow. This should be no more complicated than buying a PC and having Outlook loaded and ready to connect to an ISP;
- Clearly identified capabilities that acknowledge the difference between person-to-person communication (i.e., provider communicating visit results to a referring clinician) and machine-to-machine automated workflow applications that generate clinical alerts and do bulk sharing of lab results. With the advent of investments in FHIR, HIEs and EMRs may now bridge the gap between SMTP and RESTful Web services; and
- EMR and HIE vendors, with the aggressive public focus on easy interoperability, must implement business models that reward themselves for creating value from the information being exchanged, not for the mere transport of the payloads.
Traish: Healthcare organizations were and are eager to obtain MU 2 dollars – this required a certain level of data exchange. Participation in an HIE is appetizing financially even without the potential clinical wins. HIEs without a strong business model that also fails to accommodate the national framework for data exchange standards will likely struggle and challenge the credibility of others with a similar model.
With non-profit, public-utility, MD-Payer collaborative and for-profit exchanges, there are different priorities and values offered to member organizations. The debate will likely be less about if HIEs will become a predominant model, though instead, which business models will adapt to the policy changes, data standards and vendor specifications.
Even if MU Stage 3 brings a 50 percent requirement for data exchange by 2017, this will be accomplished through many means including HIEs. The market has not matured to reveal how transaction-fee based exchanges can offer ROI in congruence with federal and state dollars to enable broader connectivity.
Van TerHeyden: There’s a lot of credibility on the line for HIEs if they aren’t able to become more successful. Although there was a lot of excitement when HIEs were funded initially, when it came to the business model, few have found a way to sustain. HIEs still haven’t answered the clear question of where the value is and who’s going to pay for that value.
Hospitals need HIEs to work for population health, but time and adoption are waning. Standardization and governance are desperately needed for clinical information to be meaningful – especially with so many systems feeding data in and out – but lack of standardization has made it difficult for HIEs to accomplish what they set out to do. This can be especially challenging for health systems, such as Banner Health, that care for patients across multiple states and have to be part of numerous HIEs due to the lack of standardization.
In terms of security and privacy, healthcare records are a bigger target than any other data that’s on the Web – they’re worth about $50 a piece for a single medical record. The big challenge is credibility and the effectiveness of securing that information both from a patient trust perspective as well as earning clinicians trust. Both care deeply about protecting personal information and keeping it secured.
Underhill: Some of the HIE advances in care and efforts to reduce healthcare cost will be throttled if HIEs don’t succeed. Even if the potential is not fully realized, we will see fewer research dollars, less population health improvement, and a slowdown associated with the personalized care movement. Clinical and administrative errors and abuse will increase and efficiencies will be lost.
20 trends in HIE development
What are some of the most influential trends in health information exchange (HIE) development today, and how do they matter? HIE experts share their insights with Health Management Technology.
Kathleen Aller, HealthShare Director, InterSystems
- Coordinated, patient-centered care. Successful healthcare organizations recognize the need to create and sustain engaged communities of clinicians, patients, families and other healthcare participants. At the center of those communities is comprehensive, shared medical information that is structured around the patient.
- Data reigns supreme. Now more than ever, data matters. Patients, providers, researchers and care managers need consistent, current and comprehensive information about the whole care process, even for a geographically dispersed patient population. Assembling normalized and enhanced clinical and claims data of all types, from disparate sources, is the foundation for virtual care delivery, observational research and population health management.
- The beginnings of real patient engagement. Patient engagement is not about creating another patient portal. Patients and their and families want a single point of access to all relevant information – medical records, prescriptions, appointments and claims – to help them participate more fully in their own care. That requires the right information platform to bring all the pieces together.
- Staying nimble to support new kinds of collaboration. Whether building a clinically integrated network, affiliating, merging, acquiring or creating new business models, robust information exchange is a critical success factor. With interoperability at their core, HIEs give member organizations the flexibility to lead care transformation and seize market opportunities as they arise.
- A focus on value. Rapidly changing incentives, razor-thin margins and ever-increasing transparency leave healthcare organizations no time to wait for return on IT investments. A robust interoperability platform supports rapid connections across new accountable care organizations. It reduces the need to rip and replace existing technologies, and, by ensuring rapid information integration, it lets providers and payors better navigate culture and process changes.
Greg White, Senior Vice President and General Manager, TouchWorks Business Unit, Allscripts
- Expansion of HIE – analyzing data from multiple sources – contingent upon sustainability beyond the federal budget. A standards development process and education of stakeholders is necessary to properly convey the anticipated benefits of sharing data.
- Increased adoption of connected healthcare solutions – via the promise of reducing the number of unnecessary tests, thus, lowering healthcare costs and increasing clinical efficiencies.
- Increased access to data, tests and imaging – fundamentally reducing the likelihood to have repeat imaging tests. This timely access and sharing of patient records gives providers better, more complete insight into a patient’s health status.
- Cloud-based and mobile health applications – due to the ability to dynamically scale technical resources and pay for those being used to connect myriad healthcare stakeholders with divergent needs and interoperability requirements.
- Growth of hybrid, private HIE models – due to increased and enhanced control of patient data and compatibility of multiple interfaces within a healthcare ecosystem.
Michael Macaluso, Director, Product Management, Document Management and Intelligent Coding, McKesson, Enterprise Information Solutions
- Cross-enterprise patient identification.
- Security concerning what is released, what is viewed and how it is managed.
- Opt-in/opt-out policies that vary locally and regionally.
- How organizations will protect themselves from lawsuits related to information originating outside of their organization.
- How can patients view a comprehensive list of information disclosures of their healthcare information.
John Kelly, Principal Business Advisor, Edifecs
- Fast Healthcare Interoperability Resources (FHIR). One might say “FHIR and forget” everything else we did before. While the demise of DIRECT, CDA R.2 and XDR/SOAP are greatly exaggerated, the fundamental mindset that FHIR proponents set forth is to unlock the data. Companies like Epic will continue to look for models that maintain a Chinese wall between the EMR and a FHIR resource server, but the pressure from across the industry to find a good compromise between the “push” and “pull” models for clinical data exchange seems to be moving the ball in the right direction.
- Data federation. The privacy concerns associated with consolidating data on large populations just won’t go away. As the models and capabilities supporting interoperability quickly advance, the technology exists to support population-based queries without necessarily persisting large bodies of data that incur huge costs and create big risks. We live in a Google world. We pose a question to the ether, information is gathered in real time from across the globe, the answer appears, we can drill down, then magically, nothing substantive is persisted. There will continue to be specialized, purposed databases for population data management, but the pressure to maintain most data in a federated model will only grow.
- Edge-based integration. The pressure for the process community we call healthcare to share information across enterprise boundaries will grow exponentially. Patients are customers shared simultaneously among organizations using a wide spectrum of technologies and applications. Being able to quickly establish information exchange arrangements with trading partners that are flexible, configurable and don’t require huge technology implementation budgets will be critical to the success of establishing collaborative relationships between and among all the healthcare stakeholders.
- Cloud. An old trend that isn’t going away. The perceived privacy risks of cloud-based infrastructure far exceed the reality. Major cloud vendors like Amazon, Microsoft and EMC are probably more secure than most local hospital IT networks. The CDC has been operating in the Amazon cloud with great success. The cost advantages of doing business in a cloud infrastructure are undeniable and there will be a threshold effect that kicks in when the late-to-adopt stakeholders suddenly find their IT cost structure puts them out of business overnight.
- Collapsing of the clinical and administrative data streams. “Claims data bad, clinical data good.” Ever since ePatient Dave challenged John Halamka and Google Health to stop including claims data in his personal health record (PHR), the general perception has been that claims data is useless as a tool for aiding care or assessing quality and outcomes; at best ICD and CPT are proxy measures for a process from which some useful information can be extracted. In reality, early work by Medecision to create useful PHRs and the incredible work BCBSLA did after hurricane Katrina to pull together a basic patient record that aided providers and patients in crisis stand as evidence that there is important information in billing instruments. If we are truly moving to value-based purchasing where we effectively leave behind the question of whether we are getting good care or bad care and view healthcare reimbursement as a function of whether we are getting what we are paying for, then we need a complete picture of both the financial and clinical data. HIE infrastructure that can, in real time and through a common interface, ingest all patient transactions, both clinical and administrative, and cross correlate x12 and HL7, will be essential in building an accurate and defendable evaluation of the ratio between quality and cost. Additionally, the ability to look for subtle patterns and trends, as well as clinically meaningful triggers, is greatly enhanced when two sets of data representing the same event are correlated and analyzed as one.
