Communications of the ACM

Virtual extension

Advancing Information Technology in Health Care

By Steven M. Thompson, Matthew D. Dean
Communications of the ACM, June 2009, Vol. 52 No. 6, Pages 118-121
10.1145/1516046.1516077
Comments

The health care industry represents a major part of the U.S. economy. In 2007 total U.S. health care spending comprised 16% of the country's GDP, amounting to $2.3 trillion.⁷ According to the National Institute of Standards and Technology, approximately 20% of health care expenditure, some $460 billion in 2007, was dedicated to the tasks of processing, storing, and disseminating information,⁸ signifying the importance of these tasks in health care delivery. In fact, over the past ten years health care organizations have increasingly adopted information systems to improve numerous business processes such as patient scheduling, laboratory reporting, medical record keeping, billing, and accounting. Even though information systems have become more commonplace in health care, many opportunities to improve efficiency, productivity, and quality remain.

A number of factors, ranging from limited resources to heavy government regulation, contribute to the slow adoption of information systems by health care organizations. However, one of the greatest obstacles preventing large-scale health care IT initiatives is the nature of the industry itself. From an information systems perspective, health care lacks industry standards at the most basic levels. The lack of standards can be attributed in large part to a mostly fragmented industry comprised of a number of "local" organizations. In addition, for many health care systems, a clear return on investment is difficult to find. Improvements in productivity can be documented for certain systems, such as those dealing with billing and inventory. Although there is typically a time lag between when the technology investment is made and when productivity and financial gains are realized.⁴ However, clinical systems typically result in hard to measure improvements such as an increase in quality of care rather than a directly measurable decrease in the cost of providing care.

The end result is that most information systems currently used in health care organizations are designed to support ancillary services (such as laboratory reporting, logistics and inventory management, billing, and medical records). They do not serve the overarching process of providing patient care. As a result, while health care organizations accrue some benefits under the current state, many additional opportunities abound.

Back to Top

A Push into the 21^st Century

The use of IT to improve the productivity and efficiency of health care delivery continues to be a hotly debated topic. IT advocates such as Dr. David Brailer, the former national coordinator of Health Information Technology at the U.S. Department of Health and Human Services, point out that seemingly "simple" strategies can yield tremendous benefits. One example espoused by Dr. Brailer involves providing each individual in the U.S. an electronic health record and then creating a linked network across which these records can be accessed. Dr. Brailer asserts that such an initiative has the potential to eliminate as much as $600 billion peryear in costs.⁶ The savings reflect the elimination of redundant tests and procedures performed when physicians cannot access information in a timely fashion, shortened hospital stays, and improved care of the chronically ill.

The potential for IT to rein in the rapidly rising U.S. health care costs has garnered some high powered support for technological initiatives in health care. Congress, with the support of President Bush, is debating several bills designed to promote so-called "e-health care." Indeed, much of the current push for e-health care receives bipartisan support. For example, former Republican Senate Majority leader Bill Frist of Tennessee and Democratic Senator Hilary Clinton of New York put forward joint legislation in 2005 for comprehensive health IT and quality.⁴ Not everyone agrees with the infiltration of IT into the health care industry. Polling shows that a substantial portion of the public expresses concerns about the privacy and security of electronic health records, and any eventual health IT solutions will be created and implemented in this context.¹⁰

Back to Top

E-Health care 2.0: Systems Modeling and Decision Support

While IT holds great appeal and potential, cost remains a significant barrier for many health care providers. Due to the highly fragmented nature of the industry, most providers are relatively small and unable to achieve the economies of scale needed to make complex systems cost-effective. However, the potential for IT to transform the health care delivery process can change that value proposition by improving productivity, resource utilization, and quality of care. Decision support systems (DSSs) provide one of many instances where IT has accomplished that goal in the context of managing patient flow through a hospital.

The potential benefit of DSSs as tools that can improve patient care and the bottom line can be illustrated by the authors' experiences at Wind-ham Community Memorial Hospital (WCMH) in Willimantic, CT Periodically an increase in demand for inpatient beds forced WCMH to board admitted patients in the emergency department. Subsequently, WCMH would often divert ambulances to other facilities as a result of the overcrowding conditions. This fundamental problem faced by WCMH is a problem experienced by the majority of hospitals in the U.S.¹

We developed and implemented a DSS at WCMH to manage available beds, enabling the hospital to maintain the flow of patients from the emergency department to the patient floors, even when census levels were high.¹² The system worked by allocating patients within the hospital such that relatively more patients were assigned to "slow" floors and beds were proactively made available on "busy" floors (provided the assignments were clinically sound). The rationale was that while some patients could be assigned to one of many floors others, due to the nature of their medical problems, could only be assigned to a specific floor. The patients presenting with diagnoses that gave the bed manager little flexibility in terms of where to place the patient would often be held in the emergency department. While the patient was being held, already admitted patients were reallocated within the hospital in order to make room. The DSS proved effective because, by trying to keep any floor from reaching maximum capacity, the placement of new patients could proceed in a timely manner.

Figures 1 and 2 show the impact of using IT to modify the process of managing inpatient bed capacity. Figure 1 shows nearly a 50% decrease in the transfer time from the emergency department to the floor. Figure 2 shows that this improved transfer time was sustained even when as few as four beds remained vacant. With a total of 86 inpatient beds at WCMH, the DSS maintained patient flow as bed utilization approached 95%. Prior to the DSS, delays occurred when bed utilization was as low as 75%.

These results represent a significant improvement in a key quality of care metric that is also associated with financial benefits. By improving patient flow, WCMH is able to avoid ambulance diversion and reduce the number of patients that leave the emergency department without being seen by a physician (because of the long wait). By accommodating more ambulances and providing service in a timely manner that reduces walk-outs, WCMH is not only providing better care it is providing more health care services for which it is reimbursed.

Back to Top

Moving Beyond the Hospital

The same strategy of managing patient allocation can be viewed at a broader level, where bed utilization is managed across a geographic area. Hurricane Katrina provides the most recent backdrop for how delays in information gathering, processing, and dissemination can result in heavy costs in terms of lives, property, and money. In this case, as in most disaster response efforts, managers were simply unable to access key pieces of information regarding resource availability. Hundreds of patients languished in New Orleans' hospitals without power or water for days as responders tried to figure out where and how to move them.

Even after disaster response managers became fully aware of the extent of the problem, solving it was complicated. The sickest patients, those requiring ventilator support and cardiac monitoring, proved particularly problematic. Response managers needed to know which surrounding hospitals contained capacity for different types of patients and what resources (helicopters, ambulances, boats, health care professionals, etc.) were available to make the transfer. None of that information was readily available.

While information systems have proven successful at improving patient flow for individual hospitals and hold great potential for large scale efforts such as disaster response, the development and wide-spread use of such tools is problematic. The inability to support such strategic IT initiatives is largely due to difficulties with respect to internal and external system integration.⁹ The root cause of the system integration problem is the dizzying array of systems and practices that characterize the health care industry. In the absence of standards, especially in the technological sense, it becomes very costly to develop, implement, and link such specialized applications.

Back to Top

Making the transition: We can get There From Here

In order to move forward with large scale IT initiatives, the health care industry needs at least two things: standards and money. Standards for both data and metadata are desperately needed. The ideas outlined in the previous section show that standards are critical to maximizing the benefits of a health care information network. Fortunately, as illustrated in Table 1, there are several initiatives underway by various organizations attempting to tackle the tasks of creating industry-wide American National Standards Institute (ANSI) approved standards.

While these efforts to establish standards are laudable, they are taking place in a fragmented manner that is characteristic of most health care initiatives. Achieving the highest level of interoperability requires a more unified approach which will help avoid the conflicts that frequently arise when standards compete (for example VHS vs. Betamax, 8-track vs. cassette tapes, the current high-definition video disc format war of Blu-ray vs. HD-DVD). The U.S. government responded to the interoperability issue by authorizing the Commission on Systemic Interoperability. The Commission released its report³ to Congress and the American public on October 25, 2005 detailing its strategy to make health care information instantly accessible at all times by consumers and their health care providers.

The Commission affirmed some of the ideas presented here, acknowledging that interoperability "makes possible a powerful public-health resource against bioterrorism, the spread of disease, and other nationwide medical concerns."³ The Commission also acknowledged over 300 interoperability projects are currently underway nationwide. In fact, one of the projects of the Center for Health Information and Decision Systems at the Robert H. Smith School of Business at The University of Maryland tracks the various federal, state, and private health IT initiatives across the U.S.¹¹

The initiatives for creating ANSI-approved standards are a good start. The fact that there are over 300 current interoperability projects underway is encouraging. However, in order to actually bring the ideas presented here and those presented by other proponents of a health care information network to fruition, more work is needed. To start, a better understanding of what the health care network will be able to do is needed. To actualize some of our ideas, for instance the sharing of the number and type of beds available in a hospital, new standards are needed beyond those currently in development.

However, even a clear vision that guides the creation and compatibility of standards is not enough to bring such a grand mission into reality. The benefits attributed to a health care information network can only be realized if the relevant organizations participate. To move from an idea on paper to an implemented, functional health care information network will require money...lots of it.

Back to Top

Show Me The Money!

While the U.S. is not structured as a single-payer health care system, the Federal Government does represent the single largest payer in the health care system. As such, it stands to gain the most from future cost savings initiatives. This is an important observation because while creating a unified network that enables the access and transfer of individual health information has obvious benefits in terms of quality of care and cost savings, the development and transition costs are not trivial.

The big question then becomes: "Who should pay?" The U.S. Federal Government wants to see a decrease in health care costs in order to control Medicare payouts. However, since the U.S. health care system is not a single-payer system all stakeholders will, theoretically, have to financially invest in order to turn the dream of networked e-health care into reality. Despite the promise of future cost-savings, not all health care providers, even if they are financially capable of making the up-front investment, are inclined to do so. This reluctance may be related to having been "burned" too many times in the past by applications that did not deliver results. It is also possible that the smart money is sitting on the sidelines waiting for standards to be developed and an actual network implemented before committing resources.

This is where the Federal Government needs to intervene. A national health care IT infrastructure is a classic example of a public good. Economists define a public good as a good that is non-rival and non-excludable. Non-rival means the use of the good does not deplete the good. For example, breathing clean air does not leave less air for others. Non-excludable means that once the good is created its use is not restricted, again clean air comes to mind. Other examples of public goods that the Federal Government has played an active role in developing and financing are the national highway system and the Internet. In both cases, the Federal Government had an ulterior motive. The national highway system was developed to enable commercial shipping and military troop transport. The Internet was developed to share information and provide a robust infrastructure to maintain command and control capability in the event of a massive attack on U.S. communication infrastructure. So what does the Federal Government stand to gain by supporting (and influencing) the creation of a national health care infrastructure?

First, since the Federal Government receives the bill for all Medicare services, it has a direct interest in controlling the cost and improving the quality of health care. Second, the Federal Government is very concerned with a number of problems that are actually highly dependent on data from the health care industry. Important functions such as disaster response, disease surveillance, and matters important to homeland security and public health depend heavily on information derived from the health care industry. By taking a highly active role, the Federal Government has the opportunity to influence the final shape of the health care IT infrastructure and ensure that its own interests are supported. Left to the private sector, no such assurances are in place.

Back to Top

A Federal Mandate

Suggesting that the Federal Government needs to play a key role in advancing the development and implementation of a health care information network is not to say that it should take complete ownership of the project. The Federal Government is not in the business of large scale systems development and implementation. Likewise, a successful system should address as many requirements as possible across all stakeholder groups, not just the requirements of the Federal Government. The Federal Government should, in conjunction with funding, create a mandate that clearly lays out the requirements of what constitutes a feasible solution. In this manner, the Federal Government can ensure that the outcome will satisfy all of its requirements while at the same time utilizing the speed, innovation, and development and implementation expertise of the private sector.

The Federal Government has a strong motivation to see a health care information network born into existence. The fact that other institutions will also benefit, even if they are not able to contribute to the funding of the network, can be seen as a long-term investment; the viability of such health care providers serves to strengthen the health care safety net. While President-Elect Barack Obama's health IT plan allocates $10 billion per year over five years, most industry analysts expect the total cost of upgrading the U.S. Health IT infrastructure will cost much more. An under-funded mandate will only serve to extend the implementation timeline and encourage incomplete, inadequate systems. With so much on the line it is important that health care IT advance in the most rapid and coordinated manner possible.

Back to Top

Back to Top

Back to Top

Back to Top

Back to top

The Digital Library is published by the Association for Computing Machinery. Copyright © 2009 ACM, Inc.

---

No entries found