Donald M. Voltz, MD
Health care depends on the coordination of a great number of activities, communications and data. Anesthesiology sits at a unique intersection within health care, responsible for the coordination of resources from surgeons, primary care physicians and other health care providers. Anesthesiologists are also responsible for the collection of large volumes of physiologic, pharmaceutical and other patient information during the perioperative period, using a combination of paper forms and data from electronic health records (EHRs) and anesthesia information management systems (AIMS).
One of the greatest points of frustration when using health information technology (HIT) is encountering problems when entering and accessing our patients’ data; we would hope the process would be efficient, safe and timely, particularly given the acuity of our patients. It is easy to get lost in the argument for more technology regarding the need to collect and store great volumes of data.
The design and development of complex health care data systems, such as EHRs and AIMS, depends on the creation of concepts and assumptions about how we interact with the information stored in these applications. Even within a single health system, data can be stored in any number of locations within a single or many health care data applications, requiring users to navigate multiple screens or even switch between multiple programs to gain a complete understanding of a patient.
This searching for information creates inefficiencies and frustration, but more importantly, it adds a cognitive burden to care, increasing the chance for error. “Interoperability,” the concept of connecting technology and sharing data between two separate systems, is a critical problem in HIT. According to the Healthcare Information and Management Systems Society, interoperability is the extent to which devices and systems exchange information. A more important definition comes from the Institute of Electrical and Electronics Engineers, which defines the term as the ability of a system or product to work with other systems without special effort on the part of the user. From this standpoint, the interconnection of the technologies many of us use during the care of our patients falls short.
Extending this idea further, in health care to “interoperate” is to function as a whole, not simply exchanging information but also using the information collected in different systems in an actionable and coordinated way without the user being required to stitch the different pieces together. Caring for patients depends on this integration of data in order to gain a complete understanding of past history, clinical investigations and other providers’ insights, along with the patient’s presenting signs, symptoms and the various collected data to deliver and/or modify care for both acute and chronic conditions.
Care Delivery Models Drive the Need for Greater Data Integration
To function as a whole goes beyond a single patient encounter or interaction. Changes in the way care is being delivered and the emergence of care teams require a more integrated and collaborative model for sharing information and insights between care team members and using automation where appropriate to reduce the expense generated by redundancies or missed information. Changing health care delivery models also requires real-time and efficient flow of data that can be accessed by patients, providers, administrators, payors, regulators, other systems and medical applications. Given the functional requirements of such a complex system, it is not surprising that such a problem will not be solved effectively with a single software application.
As we have learned from other data-intensive industries, such as telecommunications and finance, standardization and transmission of data between systems does not completely solve these problems. To interoperate in medicine is to build connections between people, processes and data. Recently, a study of health information exchanges (HIEs) showed little evidence of impact on either cost or quality of care.1 Although most physicians and health care providers do not have an understanding of this technology, we have an obligation to become involved in redesigning the systems we use in order to safely and effectively deliver care to our patients.
Middleware as Alternative to HIEs
Interoperability, from a purely technical aspect, focuses on the sharing of data. Building HIEs, which are large data stores that aggregate data from various health care data applications so that they can be accessed by authorized people, is the result of this focus. On the surface, this might appear to be a better solution for all of us in health care. If all patient data were collected in a single warehouse, and we had access to it, then we would have addressed the problem. But we don’t need to look too much further to realize that it would be not unlike our current process of searching through yet another system to piece together the digital story of our patients. This is the simple explanation of why HIEs are a challenge. Other issues such as data duplication and corruption, and not fully understanding the provenance of the data, add to the challenges of using HIEs for solving health care’s informational ills. It is not only about centralizing the data—much more potential comes from understanding the flow of data, such as who is generating the information and who is accountable for following up, and then building connections between patients, providers, and processes and systems.
Black Book, a company that is focused on collecting the perceptions of health care leaders, recently released a survey addressing HIEs. The survey found that 90% of hospitals and 94% of independent physicians don’t trust the business model of public HIEs and have concerns about cost, causing a contraction in the HIE market.2
Are there other ways to look at this problem and perhaps propose a different solution than moving data from one system to another? I believe there is, and it not only addresses the issues of accessing patient information but brings the potential to solve related problems as well. “Middleware”—software that acts as a bridge between different products, applications or databases—has been used successfully for many years in other industries. Technology will continue to change, and the systems we are using today will be upgraded and replaced. When this happens, data collected will still need to be accessed, requiring a connection into old, legacy systems as well as those yet to be developed. Middleware offers the ability to do exactly this.
Health care 2.0 innovators are beginning to offer middleware solutions that have proven themselves in other industries. Vertical markets such as retail, banking and transportation have solved interoperability with middleware solutions, demonstrating that they are both reliable as well as cost-effective in data-intensive and data-dependent businesses.
Looking outside of health care, we have seen the development of sophisticated platforms upon which applications have been built to support access to and use of data across disparate systems. The mortgage and loan industry is a classic example where various pieces of information need to be collected, analyzed and acted on during the loan approval process. Without the ability to aggregate data from different sources, this would be a laborious and manual process, fraught with error and risk. Bringing these ideas to health care opens up our ability to move beyond being dependent on technology to instead harnessing its power, designing solutions that are safer and of higher quality, and are more cost-effective. Middleware brings the ability to design custom and standardized interfaces irrespective of which system is actually storing the data. With such a platform in place, physicians, other health care professionals and patients can become active participants in the design of tools, technologies and processes that work with and support EHRs and AIMS.
A simple solution to a challenge many of us deal with in the operating room is the administration of medications. Even with the use of computerized AIMS, we often need to enter data on antibiotics into both the AIMS as well as the hospital EHR, not to mention entering patient information into the medication dispensing system. However, we should be able to design solutions in which these systems are all connected. Medication dispensing, tracking and billing were linked to the interaction we had in the AIMS, and certain facts could be recorded in the hospital EHR, where other professionals could access the information for the postoperative care of the patient.
It is encouraging to look forward to the opportunities available for us to create solutions that work interdependently in the HIT ecosystem. I feel we are on the verge of having access to platforms upon which we can build solutions to affect and change the course of health care.
Becoming involved in the solution and not mired in the problem is our responsibility, albeit still a challenge for all of us. Medicine has always been a challenging field; this is exactly what has attracted many of us to it. The new era brings great challenges and opportunities. It just depends on how we choose to look at things.
The author is an anesthesiologist and Medical Director of the Main Operating Room at Aultman Hospital, Canton, Ohio, and is Assistant Professor of Anesthesiology at Case Western Reserve University and Northeast Ohio Medical University. He is board-certified in anesthesiology and clinical informatics.
- Rahurkar S, Vest JR, Menachemi N. Despite the spread of Health Information Exchange, there is little evidence of its impact on cost, use, and quality of care.Health Aff. 2015;34:477-483.
- Black Book™ Market Research. Health Information Exchanges.http://blackbookmarketresearch.com/health-information-exchange/. Accessed May 15, 2015.
Leave a Reply
You must be logged in to post a comment.