Evolution in the healthcare world requires never-ending implementation and leveraging

by John Glaser and Steve Flammini

As healthcare executives, we often consider our task to be serial implementation of applications - of laboratory, patient accounting, provider order entry systems and the like - with a goal of improved organizational performance. This view is not wrong, but given the goal, it may not be fully correct. This view often has three major limitations:

• It can focus management attention on the next implementation, and then the next and then the next. Consequently, we devote insufficient attention to broadly improving the organization's ability to implement systems and then manage their ongoing evolution to realize their full value.
• It can lead us to believe that once an implementation goes live, we are done with it and can move on to the next one. If our intent is improved organizational performance, gains are rarely fully present the day after go-live. In fact, the real work of performance improvement may have just begun.
• It can make us lose sight of the changing business environment around us. Locking in to a long-term series of application implementations limits an organization's agility. Often, the original premise of the application becomes dated long before installation is complete.

We might be better off if we focus on implementing and then leveraging an application foundation. A foundation provides the broad ability to perform a never-ending series of application-leveraged small, medium, and occasionally large advances and improvements.

Examples of application foundations

Provider order entry systems can be used as a foundation to improve physician decision-making. Once the system is implemented, the organization can introduce an unending series of decision-support rules and guides. These rules can address medication safety, the appropriateness of test and procedure orders, and the display to physicians of data relevant to a given order. No single rule will turn the tide in an organization's efforts to enhance care, but in aggregate, rules and guidelines can effect significant improvements.

Computerized medical records can be used as a foundation to enhance outpatient care processes. Once the record is implemented, the organization can continue to introduce modules that improve health maintenance, referrals, medication order effectiveness and capture of billing documentation. Together, these changes can significantly advance outpatient care.

Software supporting patient-provider communication can be used as a foundation to more fully engage patients in their care. It can support the development of communities of patients with a chronic disease and can, for example, assist in management of diabetes by helping patients take their medications correctly and provide information to address their healthcare questions.

Application software, by its very nature, lends itself to a foundation approach-more so than many traditional business assets in which previous versions must be discarded and replaced as the world evolves. If well architected, application software should be capable of being continually leveraged forward, with a need to "shed its outer skin" only as technology standards and business needs evolve.

Ramifications of foundations view

Clearly, there will be a flurry of intense effort as the foundation is laid. Introduction of provider order entry and computerized medical records is difficult work that requires great skill and significant resources. But once the foundation is in place, implementation is ongoing. In fact, implementation never stops, so management and clinical mechanisms must exist to manage it.

These mechanisms must identify the next area to be leveraged, ensure that required analyses are performed, install needed software modifications and enhancements, and reengineer relevant processes. In effect, these processes and mechanisms must continue the tasks that began before go-live.

Foundation replacement is rare if the application suite is effectively evolving as the organization changes and as discernment of effective system application matures. Typically, replacement would occur under one of four circumstances- all unusual:

• The vendor goes out of business and support disappears.
• A new product becomes available that has dramatically superior capabilities or lower costs.
• The foundation has seriously failed to incorporate new technologies.
• Radical changes in the care or business model upon which the foundation is based renders it useless.

The foundation must be able to evolve gracefully and support ongoing implementation. Tools that enable rule development, the safe addition of local modifications, incorporation of new data types and coding conventions, and efficient interoperability with other systems are essential. The foundation must be able to capitalize on new technologies efficiently, with minimal disruption.

Increasingly, application foundations should be thought of as a collection of services or components rather than a collection of applications or products. The best way to organize and leverage the application is as a suite of loosely coupled, standards-based services. A service-oriented architecture can greatly extend the useful lifetime of the foundation.

Technology inflection points

It is critical for organizations to recognize technology inflection points-points at which computing metaphors, or technology standards, change to such a degree as to warrant re-examination of the computing model on which the foundation is based. Failure to recognize inflection points can lead to a foundation that effectively becomes unleverageable. For many foundations, the service-oriented approach is such an inflection point. Robust, inexpensive grid computing may be another.

In many ways, technologies and tools that enable ongoing implementation are more important than is the present functionality of the application. The request for proposal (RFP) process implies that all functionality that will ever be needed must be understood and defined up front. While it is best to understand as early as possible as much as possible about the functionality needed, it is unlikely that anyone can anticipate all needs. Experience will be the teacher.

The RFP process and implementation are based on the belief that functionality must be implemented in large blocks-that implementation is a one-time event and when it is complete, you move on to the next application. The foundation view encourages implementing functionality in a series of modest-sized blocks, with learning taking place during and after each implementation. Organizations thus become smarter about the nature of the following blocks during this ongoing incremental-learning process. Over time and with more implementations, organizations will arrive at implementation of a large block of functionality. But that large block is likely to be rooted in a deep understanding of what is needed.

Final analysis

Assessing the return on investment (ROI) of a foundation during the process of deciding capital budgets is more difficult than determining the ROI of an application. Although it is essential to continue to evaluate the ROI, it is difficult to do because the path of evolution is indeterminate and implementation never-ending. Despite the challenges, organizations are much more likely to see a return if they view the task as the implementation of a foundation than if they view the task as the serial implementation of applications.

The goal of applying IT to improve organizational performance will be advanced if we consider our task to be implementing, and then perpetually leveraging, application foundations. Despite the challenges this view poses in today's industry, it should be at the center of information systems strategic thinking.

John Glaser, a member of the Healthcare Informatics editorial board, is vice president and CIO, and Steve Flammini is chief technology officer, Partners HealthCare System, Boston.

Copyright © The McGraw-Hill Companies, Inc.
Reprinted with permission from Healthcare Informatics, April 2003
Printed in the U.S.A.