When a patient chooses where to receive her health care, she may consider the online ratings of a doctor or hospital. Few people, however, are likely to take into account the technological advantages from one health-care setting to another.
But what if it’s not just the expertise of the clinicians that could affect quality of care? What if the technological tools that are selected by each hospital and doctor’s office could also have an impact on care?
Technology is playing an increasing role in the quality of health care provided. Case in point: Electronic Medical Records.
Electronic Medical Records (EMR) are digital versions of the paper charts in doctors’ offices and hospitals that contain notes and information collected by and used for health-care providers. Additionally, they have modules that provide decision support, alerts and the ability to digitally order and transmit prescriptions. An appropriately implemented system also can share digitally stored information across health settings.
Hospitals can choose to purchase all EMR modules in a suite from a single supplier (single-sourcing) or they can use multiple suppliers (multi-sourcing).
New research from the University of Notre Dame shows for the first time that the sourcing strategy chosen by hospitals has an impact on the quality of patient care. When hospitals move closer to a single-sourcing strategy, patients receive better evidence-based care.
With multi-sourcing strategies, mismatches between supplier modules can create situations where key medical concepts don’t share the same meaning, which can disrupt the exchange of data, potentially lowering the quality of care provided. For example, allergies could be defined and coded differently in EMR modules from different suppliers.
“Strategic Sourcing of Multi-Component Software Systems: The Case of Electronic Medical Records” is forthcoming in the Decision Sciences journal from lead author Kaitlin Wowak; Corey Angst and Ken Kelley, professors of IT, Analytics and Operations (ITAO) at Notre Dame’s Mendoza College of Business; and Sean Handley from the University of South Carolina.
Using data from U.S. hospitals that operated continuously from 2006 to 2013 and analyzing change over time, the team examined how a key dimension of the sourcing strategy for EMR systems — change in closeness to single-sourcing — affects conformance quality, a critical measure of hospital performance that assesses how frequently hospitals comply with evidence-based practices.
They found that moving closer to single-sourcing increases conformance quality, but the positive effect is reduced over time, meaning the benefits realized by hospitals using a single-sourcing approach become less substantial later in the observation period.
“Our study contributes much-needed insight into the performance implications of an organization’s sourcing strategy for multi-module software systems, which are increasingly prevalent in modern organizations,” Wowak says. “We show that how, not just if, EMRs are implemented is consequential to the quality of patient care provided.
“Administrators must consider how the number of different suppliers affects routines, interoperability (the capability of two different modules to receive and send data) and, ultimately, conformance quality.”
Wowak says aligning the sourcing strategy with technical and operational considerations is likely to result in better patient care. The study shows that the size of the effect has considerable practical significance, as a move toward single-sourcing corresponds to more patients receiving appropriate care as a result.
“To put this into perspective, during the timeframe of our study, U.S. hospitals admitted approximately 35 million patients per year,” Kelley says. “All else being equal, if hospitals moved slightly closer to single-sourcing in a given year, say 0.5 units, approximately 175,000 more patients per year would have received evidence-based care.”
Because hospital performance can often mean life or death for patients, the team hopes this study will help lay the foundation for future research on the relationship between health IT, sourcing decisions and hospital performance.
“Even though there are industry efforts to help translate key concepts between modules that are sourced from different suppliers to improve interoperability, EMR firms have little incentive to get their systems to work and communicate with competitor systems,” Angst said, “since integrating one system with another is time consuming, expensive and potentially error prone.”
The study showed that a single-sourcing strategy can improve patient care, but this effect is reduced over time. From a health-care provider perspective, Wowak says there are several pros and cons to maintaining either sourcing strategy as well as barriers to switching.
“The advantage of a single-sourcing strategy is that each module in an EMR is provided by the same company so all of the modules are designed to work together seamlessly. The downside of this strategy is that each individual module may not be the best-in-class, and it may not fit with the hospital’s processes, so health-care providers may have to modify what they do or perform workarounds,” says Wowak.
“On the other extreme end of the spectrum, you have a multi-sourcing strategy, which is when some modules in an EMR are provided by different suppliers,” Angst said. “The advantage of this strategy is departments can select a best-in-class module that most closely aligns with their processes. While modern EMRs are less modular and tend not to be hampered by as many interoperability issues, a downside of this approach can be an EMR system that has more limited functionality or less than optimal performance.”
Interoperability issues occur when modules from different suppliers don’t exchange information seamlessly with each other. This can result in health-care providers only seeing part of a patient’s medical history, which can lower the quality of care provided. In simple terms, data from one EMR may not end up in the correct place in another EMR, or it could be lost altogether. With something as important as medication and health histories, poor data exchange can be catastrophic.
The researchers spoke to IT companies, managers of hospital IT systems and medical providers to better understand the pitfalls and impact of a multi-sourcing strategy. One example that arose was if hospitals are using a multi-sourced EMR, translating the data from one module to another can be resource intensive.
“Translating EMR data places a heavy burden on people, which is both expensive and error prone. You can try to automate some things, but we aren’t to the stage yet where this is foolproof,” said Wowak.
For a practitioner, a multi-sourcing strategy can quickly become problematic from an operational standpoint. A doctor may have to access one module to see a patient’s medicines, another for their medical history, and possibly a third for additional information. Besides being burdensome to manage at once, it may also detract from the quality of care provided when a health-care provider has to switch back and forth between screens.
While this research shows there are benefits to a single-sourcing strategy, health-care organizations have to assess the costs of switching from a multi-sourcing strategy to a single-sourcing approach, which can be substantial. Like any organization, changing a software program from a legacy system comes with its own set of challenges: cost of a new system, training staff in the new system and potential disruptions while the new system is being implemented.
The study by Wowak, Angst, Kelley and Handley is just one way that analytics is being used in the health-care industry. In the health-care analytics space, there has been an explosion of growth in research activity, which is meant to help people understand how business analytics in health care has the potential to enhance patient care and provider performance. However, there is little consensus about what business analytics in health care is.
In another new paper published in 2023, Wowak and Angst collaborated with John Lalor, assistant professor of ITAO, and Sriram Somanchi, assistant professor of ITAO, to better define business analytics in health care. They reviewed more than 6,800 relevant research articles from top-tier business and medical journals. Doing so enabled them to provide insights into research trends overtime and develop a road map for future research for this rapidly growing field.
The paper, “Business Analytics in Healthcare: Past, Present, and Future Trends,” published in Manufacturing & Service Operations Management, defines business analytics in health care as well as the application of analytical techniques on health-care data to inform decision making and enhance outcomes for stakeholders involved in the business of health care.
In order to evaluate the papers via text analysis, the researchers developed an interactive article analysis web application (IWA). To create the IWA, the researchers identified the most common words each research article used to decipher the main topic. Through the app, users can have a dynamic look into the literature and examine how different topics coincide with each other and when topics increase in popularity.
One topic the IWA brought to light was how legislation, such as the Health Information Technology for Economic and Climate Health (HITECH) Act passed in 2009 to promote the adoption and use of health information technology, impacted what researchers examined. The act required health-care organizations to implement EMR systems. The IWA showed that before the act was passed, research was focused on the benefits of EMR systems, but after the act passed, research completely shifted to examining the challenges of implementing EMR systems.
This dynamic review of the literature can help those looking at business analytics and health care to understand what topics are typically analyzed together, such as health care and information technology, and also where gaps in the literature are.
“We hope our IWA will help inform future research on health-care analytics,” says Wowak.
The IWA is publicly available for scholars to identify opportunities in business analytics and health care for future research topics. The IWA can also be customized so scholars in any discipline can use it to dynamically review the literature to discover previously unidentified trends and important gaps in the literature.
With a combined 37 years researching EMRs at Notre Dame, Wowak is the Robert and Sara Lumpkins Associate Professor of Business Analytics; Kelley is the Edward F. Sorin Society Professor of IT, Analytics, and Operations; and Angst is the Jack and Joan McGraw Family Collegiate Professor of IT, Analytics, and Operations.
Illustration by Errata Carmona. Photo by Barbara Johnston.
KAITLIN WOWAK is the Robert & Sara Lumpkins Associate Professor of Business Analytics.
COREY ANGST is the Jack and Joan McGraw Family Collegiate Professor of IT, Analytics, and Operations with most of his research focused on the health-care industry.
KEN KELLEY is the Edward F. Sorin Society Professor of IT, Analytics, and Operations and senior associate dean for faculty and research.
“Strategic Sourcing of Multi-component Software Systems: The Case of Electronic Medical Records"
Data Sciences (2022)
Kaitlin Wowak, Corey Angst, Ken Kelley (University of Notre Dame), and Sean Handley (University of South Carolina)
“Business Analytics in Healthcare: Past, Present, and Future Trends"
Manufacturing & Service Operations Management (2023)
Kaitlin Wowak, Corey Angst, John Lalor and Sriram Somanchi (University of Notre Dame)