Introduction
In a previous module, we have explored, in detail, the concerns and considerations associated with implementing electronic health and medical records in clinics and hospitals. While these systems are largely commonplace across the US nowadays, there are many problems which plague the digital platforms, not the least of which is health information exchange (HIE). The exchange of highly personal and, to some extent, confidential records of patient health is a tedious yet critical component of any healthcare system. At a first thought, it seems like this is something that would be simple (perhaps even simpler than on paper) with the use of electronic medical records. In the age of information, the breadth of mankind's knowledge is merely a click and some WiFi away and at this point, the facility of these electronic interactions has seeped into nearly every corner of our daily lives. We communicate with one another daily on a plethora of websites and applications which are all housed on one of a handful of yet more diverse technology platforms; we upload and share video and music files from one device that can still be read and interpreted by nearly any other device; we rely on digital technology to facilitate practically every single bill-pay, banking, or other financial transaction; we pay our insurance, renew our licenses, and sometimes even purchase our groceries from these electronic portals that are all designed to communicate with one another to make our lives easier.
Indeed, in 2020, nearly every aspect of our personal and public lives is facilitated by digital software, and healthcare is no exception. As we have discussed previously, the passage of the 2008 HITECH app marked the stark and rapid transition from paper to computer book-keeping in the healthcare industry. Yet, more than a decade later, the exchange of healthcare information is exceedingly cumbersome in the US. Primary care clinics have difficulty communicating with hospitals who have a hard time demonstrating need to insurance companies which have a difficult time cooperating with pharmacies - and the headaches don't stop there. In an industry where relevant and up-to-date information is vital to the health and wellness of its patrons, why is it that the US healthcare system still struggles with fast and accurate information exchange? In the following sections, we are going to dive into the hurdles that are unique to the exchange of information in the healthcare industry as well as the current and future state of health information exchange.
Indeed, in 2020, nearly every aspect of our personal and public lives is facilitated by digital software, and healthcare is no exception. As we have discussed previously, the passage of the 2008 HITECH app marked the stark and rapid transition from paper to computer book-keeping in the healthcare industry. Yet, more than a decade later, the exchange of healthcare information is exceedingly cumbersome in the US. Primary care clinics have difficulty communicating with hospitals who have a hard time demonstrating need to insurance companies which have a difficult time cooperating with pharmacies - and the headaches don't stop there. In an industry where relevant and up-to-date information is vital to the health and wellness of its patrons, why is it that the US healthcare system still struggles with fast and accurate information exchange? In the following sections, we are going to dive into the hurdles that are unique to the exchange of information in the healthcare industry as well as the current and future state of health information exchange.
Barriers to effective HIE
There are several barriers to expanding and implementing effective health information exchange. One central issue in this discussion is the concept of data ownership. In today's world, data is one of the most important resources and it is difficult for hospitals and health care providers to simply share private health information with others without a significant incentive. In regards to health care data, much research and analysis must be done in order to figure out what improvements can be made on healthcare information pipelines; decision support systems can be built to facilitate the work of providers and user data can be used to construct better human-computer interaction tools to improve patient and provider satisfaction. All of which are profitable advantages but certainly do require a steep overhead in the development and implementation of new softwares. In general, however, it is important to think of these information exchange features as long-term investments. For example, a more well-connected decision support system that has access to patient data that is housed on multiple platforms could reduce the cost of medical errors and maybe even provide more inexpensive treatment options.[1] The outcome of such an investment would be the increase of hospital profits and reputation, allowing the hospital to attract more patients and high-quality providers.
In these considerations, it is also important to recognize that one of the parties which health information exchange platforms must consider is the patients themselves.[1] Raw medical records contain a plethora of biological, medical, and chemical jargon that is difficult to decipher without professional training. Learning to understand medical records is a burden and thus, HIE platforms must be designed with the intention in-mind to provide medical information to patients in a way that is accessible in lay terms.
Last but not least, data transfer standards and interoperability between platforms remains to be a challenge. Medical data is highly sensitive and any leaks can quickly turn into costly and embarrassing situations for healthcare providers. To this end, one complication that exists is the distrust between healthcare organizations for one organization may feel that the data security standards at another organization are sub-par and may, therefore refuse to exchange patient information with them.[1] For example, let's say healthcare agency A adopts standard X and healthcare agency B adopts standard Y. If agency A is asked to transfer medical records to agency B, there are three scenarios that are likely to take place: (1) the easiest case might just be that standard X is approximately the same as standard Y. In this case, agency A can transfer data to agency B easily. (2) In another case, standard X has a higher safety standard than standard Y which may cause agency A to refuse the data transfer. On the contrary, if standard X has overall lower security than standard Y, even though agency A is willing to transfer the data, agency B might refuse to accept, again, due to security concerns.
Overall, medical record transfer is a complicated issue because of the complexity and sensitivity of health data. These issues are further complicated by the fact that healthcare agencies do not have universal security standards and interoperability platforms. If these are the major barriers associated with HIE, then what does the current state look like? And how might we be able to help things migrate towards a better future?
In these considerations, it is also important to recognize that one of the parties which health information exchange platforms must consider is the patients themselves.[1] Raw medical records contain a plethora of biological, medical, and chemical jargon that is difficult to decipher without professional training. Learning to understand medical records is a burden and thus, HIE platforms must be designed with the intention in-mind to provide medical information to patients in a way that is accessible in lay terms.
Last but not least, data transfer standards and interoperability between platforms remains to be a challenge. Medical data is highly sensitive and any leaks can quickly turn into costly and embarrassing situations for healthcare providers. To this end, one complication that exists is the distrust between healthcare organizations for one organization may feel that the data security standards at another organization are sub-par and may, therefore refuse to exchange patient information with them.[1] For example, let's say healthcare agency A adopts standard X and healthcare agency B adopts standard Y. If agency A is asked to transfer medical records to agency B, there are three scenarios that are likely to take place: (1) the easiest case might just be that standard X is approximately the same as standard Y. In this case, agency A can transfer data to agency B easily. (2) In another case, standard X has a higher safety standard than standard Y which may cause agency A to refuse the data transfer. On the contrary, if standard X has overall lower security than standard Y, even though agency A is willing to transfer the data, agency B might refuse to accept, again, due to security concerns.
Overall, medical record transfer is a complicated issue because of the complexity and sensitivity of health data. These issues are further complicated by the fact that healthcare agencies do not have universal security standards and interoperability platforms. If these are the major barriers associated with HIE, then what does the current state look like? And how might we be able to help things migrate towards a better future?
HIE: Now what?
The success of the 2008 HITECH act is at its high adoption rate of EHRs among hospitals, clinics, and other healthcare organizations. As is discussed in a variety of other modules, we know that EHRs can solve some of the issues revolving around data storage, management, and (theoretically) accessibility but has not yet realized its full potential in improving healthcare safety and quality, and providing more cost-effective and patient-centered care. In this module, we focus on data interoperability in health information exchange which is one of the essential elements in fulfilling the goals of implementing EHRs and other health information technologies. The Office of the National Coordinator for Health Information Technology (ONC) has defined the following forms of HIE:
1. Direct exchange: Sending and receiving secure information electronically between care providers
2. Query-based exchange: Provider-initiated requests for information on a patient from other providers
3. Consumer-mediated exchange: Patients aggregating and controlling the use of their health information among care providers
Widespread usage of EHRs in most hospitals that lack interoperability hinders the data-intensive activities like public health surveillance which can leverage the electronic health information lied in EHRs. Except for public health surveillance, the lack of interoperability also impedes the principles of providing continuous, coordinated, and patient-centered care in patient-centered medical homes. On the other hand, the engagement of consumers in the healthcare delivery process mainly relies on patient portals but has limitations in accessibility, functionality, and once again interoperability. These limitations make its adoption rate relatively low. For the below paragraphs, we summarize the success factors and pitfalls to avoid facilitating health information exchange.
Success factors/pitfalls to avoid:
As mentioned in the article “The death by a thousand clicks”, one main reason that each healthcare organization serves as an isolated island without exchange of data between them is the lack of policy and motivation that incentivize them to do so. Without the funding from the federal or state government, organizations lack the incentive to share data with their patients or other hospitals because they might risk losing their patients that bring the data and information to other healthcare providers. To reach the clinical and economic benefits of health information technologies and EHRs, policy makers should allocate more energy and time in revising HITECH act to focus more on interoperability and health information sharing. The policies should also take into account the patient privacy, and equitable issues to maximize the benefits brought by health information exchange. In addition to policy, data standards and infrastructure are another two success factors that will facilitate health information exchange. Data standard that has the consensus of terminology, format and sharing approach will enable the data exchange not only limited to clinical equipment but also between different EHRs. With the help of data standards, EHRs should open access up to web-based applications and mobile devices that support a dynamic and nonvendor-constrained information infrastructure.
Implications for us:
We might not be the policy makers to revise the HITECH act; however, as Hersh et.al concluded in their literature review on outcomes of health information exchange, the lack of framework to guide the implementation, design, and assessment of the outcome of health information exchange result in the weak evidence regarding the positive effect of health information exchange in improving patient’s modality and mortality rate or in supporting physicians in decision making. One of the reasons that makes the measurement the effectiveness of health information exchange is so hard is that HIE is not specific to any diagnosis or health issue but an intermediate in improving healthcare delivery and increasing accessibility of data for physicians and patients. As an informatician, what we can do is to develop new frameworks or modify the existing ones to support the implementation and evaluation of technologies concerning data interoperability. Furthermore, we can also dedicate to creating the consensus data standards that support information exchange in a consistent format and terminology.
1. Direct exchange: Sending and receiving secure information electronically between care providers
2. Query-based exchange: Provider-initiated requests for information on a patient from other providers
3. Consumer-mediated exchange: Patients aggregating and controlling the use of their health information among care providers
Widespread usage of EHRs in most hospitals that lack interoperability hinders the data-intensive activities like public health surveillance which can leverage the electronic health information lied in EHRs. Except for public health surveillance, the lack of interoperability also impedes the principles of providing continuous, coordinated, and patient-centered care in patient-centered medical homes. On the other hand, the engagement of consumers in the healthcare delivery process mainly relies on patient portals but has limitations in accessibility, functionality, and once again interoperability. These limitations make its adoption rate relatively low. For the below paragraphs, we summarize the success factors and pitfalls to avoid facilitating health information exchange.
Success factors/pitfalls to avoid:
As mentioned in the article “The death by a thousand clicks”, one main reason that each healthcare organization serves as an isolated island without exchange of data between them is the lack of policy and motivation that incentivize them to do so. Without the funding from the federal or state government, organizations lack the incentive to share data with their patients or other hospitals because they might risk losing their patients that bring the data and information to other healthcare providers. To reach the clinical and economic benefits of health information technologies and EHRs, policy makers should allocate more energy and time in revising HITECH act to focus more on interoperability and health information sharing. The policies should also take into account the patient privacy, and equitable issues to maximize the benefits brought by health information exchange. In addition to policy, data standards and infrastructure are another two success factors that will facilitate health information exchange. Data standard that has the consensus of terminology, format and sharing approach will enable the data exchange not only limited to clinical equipment but also between different EHRs. With the help of data standards, EHRs should open access up to web-based applications and mobile devices that support a dynamic and nonvendor-constrained information infrastructure.
Implications for us:
We might not be the policy makers to revise the HITECH act; however, as Hersh et.al concluded in their literature review on outcomes of health information exchange, the lack of framework to guide the implementation, design, and assessment of the outcome of health information exchange result in the weak evidence regarding the positive effect of health information exchange in improving patient’s modality and mortality rate or in supporting physicians in decision making. One of the reasons that makes the measurement the effectiveness of health information exchange is so hard is that HIE is not specific to any diagnosis or health issue but an intermediate in improving healthcare delivery and increasing accessibility of data for physicians and patients. As an informatician, what we can do is to develop new frameworks or modify the existing ones to support the implementation and evaluation of technologies concerning data interoperability. Furthermore, we can also dedicate to creating the consensus data standards that support information exchange in a consistent format and terminology.
Want to learn more?
Here is a resource we used to write this article!
[1] EHRIntelligence. “What Are the Top Barriers to Health Information Exchange?” EHRIntelligence. 2013. ehrintelligence.com/news/what-are-the-top-barriers-to-health-information-exchange.
Personal Reflections
Oliver |
When I was reading this week’s articles, I thought about all the modules we have learned so far. If improving the quality of care, increase patient safety, provide patient-centered care, and increase healthcare accessibility are the ultimate goals, then I imagine EHRs,CPOE, and patient centered medical homes as the vehicles that bring us to those goals, whereas data standards and interoperability are like the wheels that make the vehicle move, and policy, framework, and funding are the fuel and materials that facilitate the creation of wheels and vehicles, and activate them. Another interesting thing comes to my mind is the emphasis of data sharing and interoperability within the healthcare and biomedicine domain that three of our required courses touch on this topic from different perspectives. BIME 530 focuses on data standards which I learned about DICOM, the standard for medical images. BIME 550 talks about ontology that is concerning how to model the knowledge in a way that supports automated reasoning and provides a consensus terminology to facilitate data exchange between researchers. I think this class looks at the issue from the healthcare perspective and helps me to understand how HIE can support HIT and the healthcare system to fulfill its potential.
|
Weipeng |
After writing the premier, I found HIE really has a lot of challenges. Data ownership, patients' actual benefits and security concerns are three major but also hard to solve problems. Nevertheless, I still think there are ways to address them. The data ownership problem is especially hard to solve. Entering the 21st century, we see the rise of many huge IT companies like Google, Facebook, Amazon, etc. Most of them have strong control over their data ownership, especially Google whose search engine not only provides a huge amount of data but is also built on a huge amount of data. Google's powerful search engine allows it to gain many users but also allows it to collect millions, if not trillions of data; the collected data can thus be utilized to enhance its search engine, which makes it even more powerful so more people are interested using it. This becomes a positive cycle. Thus we can see the importance of data ownership and it is hard for health care agencies to give it such a huge benefit.
|
Dakota |
I find myself, again, frustrated with this topic. In today's day and age, why is it so difficult to create digital health portals that are able to safely and effectively communicate with one another? Any phone in the world, can contact any other phone in the world, regardless of make. Any computer in the world can communicate with any other computer in the world - I can type up an email on a hotmail account on a macintosh computer operating on a MacOS and send it to be read by someone using gmail on a Dell computer that is operating Windows10. So many different technological systems are able to communicate with one another with ease so what is the difficulty with healthcare?
There is certainly an argument to be made that healthcare information is its own brand of information and has unique challenges associated with it. I would argue, however, that the banking industry is an excellent model for the healthcare industry. Credit card companies host online platforms now in which your card account can communicate with nearly any vendor in the world as well as being able to send and receive money to and from private banking institutions. These online platforms also have access to highly-sensitive (in fact, I would argue, more sensitive) information, just as the healthcare industry: credit card account numbers and banking numbers; SSN, income and salary, all expenses, and often even credit score. If the financial industry has been able to migrate all of the world's highly-sensitive transactions to online services, then why is it that the healthcare industry is struggling so much more significantly? |