1997 Joseph Leiter Lecture

Positioning the library at the epicenter of the networked biomedical enterprise*

By William W. Stead, M.D., Associate Vice Chancellor for Health Affairs, Professor of Medicine and Biomedical Informatics, Director of the Informatics Center, Director, Eskind Biomedical Library, Vanderbilt University Medical Center, 2209 Garland Avenue, Nashville, Tennessee 37232-8340

Introduction
It is a pleasure and a privilege to present the 1997 Leiter Lecture. From his perch at the National Library of Medicine, Joseph Leiter took a broad view of librarianship. I understand he was quite controversial in the way that he presented his views. He focused on the objectives of the profession, not the individual tasks that librarians do. Therefore, change came naturally in his view.

Today I want to ask you to look beyond the uncertainty caused by the changing economic environment in which our biomedical enterprises operate. I want to encourage you to focus on the unparalleled opportunities that are presented to medical librarianship by those changes. The members of the profession can assume new and essential job roles if certain challenges can be overcome.

I want to be clear from the beginning that I do not consider myself to be an expert on medical librarianship. I have roots in informatics and in clinical medicine, and I have the real pleasure of being able to coach the remarkable team at Vanderbilt University Medical Center's (VUMC) Eskind Biomedical Library. But the thoughts that I am going to share with you today come from sitting at the table with the VUMC leadership team as we decide how to navigate the turbulent waters in front of us while trying to move toward goals such as achieving a top ten ranking for the School of Medicine.

The Opportunity
Before we can think about how to place the library at the epicenter of the networked biomedical enterprise, we need to identify the strategies that the enterprise can use to become more effective. The first strategy involves evidence-based medicine, an effort to bring the principles of continuous quality improvement to bear on the patient care process instead of viewing practice as an art form, where each patient is treated as an experiment with an n of one. We try to get agreement on best practice, document that practice as a pathway, and then to manage patients on those pathways. We monitor variance to provide a feedback loop that indicates when the process needs modification. Each modification makes the process better for the next patient. Using these processes at Vanderbilt, we have reduced the cost of some high-volume procedures by almost 40% while documenting an increase in patient satisfaction [1].

Patient empowerment is a strategy at the other end of the spectrum. The idea is to move away from prescriptive medicine, which we do "to" a patient, toward participatory medicine where we seek to understand a patient's value system and involve the patient in the decision making process. We end up with a plan that the patient feels is tailored to him or her. The patient is more likely to carry out such a plan, instead of being non-compliant with something that somebody is trying to do "to" them.

Other strategies go beyond the clinical environment that is driving much of the change today to our educational and research missions. Take true lifelong learning as an example. The idea is to move away from a dedicated episode of learning that is preparatory to a career, toward just-in-time instruction. Instruction takes place at "teachable moments," when the learners are really ripe to learn a concept, and includes access to up-to-date facts.

These strategies are representative of the ones that are going to enable the networked biomedical enterprise. Every one of these strategies involves either a new type of shared information, which must be managed, or access to information in new ways or by different people-tasks with which librarianship is involved directly. Librarians are therefore placed perfectly to provide new products and services that support these strategies.

Records management is one such product. It involves archiving records about what we do for patients, for students, and for each other in our work processes. Outcomes measures can then be added. These records can be used as a time lapse camera to show us which interventions work and which don't, whether they are educational, clinical, or research [2]. Information filtering and interpretation is a second area in which new products are needed. At this time, information technology is making the information explosion worse. It is causing people to have to sort through more and more. We need to develop services that let us filter, focus, and interpret to get people exactly what they need to know. Publication at the source is a third area for product development. We need to move from the practice of giving away knowledge and then buying it back from publishers. Instead, our libraries could publish the artifacts that are produced in their institutions and make them available for universal access. Information producers would get access to each other's artifacts through a barter arrangement. People who were not information producers would have to pay for access with dollars.

The Challenge of Aligning with the Enterprise Business Strategy
The first challenge that the library must meet if it is to be at the epicenter of the enterprise is to align the library's strategy with the enterprise business strategy. We must begin by recognizing that the enterprise is a business. Whether the enterprise is a school, a university, a not-for-profit hospital, a for-profit hospital, or a health system, it is a business. Now that we can no longer bill and be reimbursed for whatever our costs are, we have to learn how to act like a business. We have to develop a business strategy. The process is quite simple. First, we look at the threats that face the enterprise-what is going to make something we are doing unprofitable or make people stop coming to us for it. We then look at opportunities-services that would allow us to create new revenues. Next, we look at our core competencies. What do we as an enterprise do better than anybody else? Based upon these insights, we can identify strategies for differentiation of our enterprise from the competition and build plans for development of new products. In this way, the enterprise business strategy highlights what is needed for success. The library business plan can then flow from that strategy by showing how the library will provide critical elements. To the degree that we can focus the library's resources on products and services that are necessary for the enterprise to meet its business strategy, nobody will question investing in the library. The investment is a critical step toward production of revenue, not as a "nice to have" that can be done without when times are tight.

Let me use VUMC as an example. It is clear that we face a threat that payers will stop paying what it costs us to provide what we consider to be quality service. We have an opportunity because we provide more than thirty services that are not provided any place else in the region. We have the core competency of developing knowledge and rapidly moving it from the bench to the bedside. One strategy for using the opportunity and competency to overcome the threat involves development of a payor-neutral network that is second to none—one that every panel will be obligated to include. This idea is simple, but hard to achieve. What do we need to do to help providers who integrate with Vanderbilt to be more effective than ones who do not integrate with Vanderbilt? The library has core competencies that can help: the ability to access, filter, and interpret information. If we can get information to people more efficiently than they can otherwise get it, we can make them more effective.

The Challenge of Scale
That brings us to a second challenge, the challenge of scale. Quality is necessary, but quality is not sufficient. We have to be able to produce that quality in a quantity that is sufficient to enable a business strategy. To continue with the Vanderbilt example, we cannot afford to hire librarians to answer any question posed by every physician across the middle Tennessee region. A person-to-person service model would not scale-up to have a significant effect. On the other hand, we can have a core of people who do such filtering and interpreting work, and who then record that work in a knowledge base of frequently asked questions that is then reused across the region [3].

Let me give you some examples of strategies that can lead to services that can scale-up to support the enterprise. First is the area of patron education. We should raise the bar every year as to what we expect our clients to be able to do on their own, freeing up library staff to help them to the next stage.

Second, we should "mainstream" client training and support. Library staff should prepare instructional content, but the in-service mechanisms of our hospitals or clinics and the curricula of our schools should be used to deliver that content.

Third, we need to develop information resources and services as reusable components. The data resources that are being created through the Human Genome Project are an example [4]. One set of databases can be accessed through two different user interfaces. One supports the researcher and the other supports K-through-12 students and other consumers. This approach required much less effort than would have been required for two separate systems. Staff resources can be leveraged in a similar way. Each member brings a set of talents to the table. We should apply those particular talents to several project and service teams. Products also provide an opportunity for re-use. For example, information filtering and interpretation is a service that adds value and that can be billed as fee for service. In addition, it is a building block for a training program, which can constitute another revenue-producing product.

A final strategy for achieving scale involves integration of information access directly into work processes. Where that can be done, people don't need to know how the information access works, and they don't need to interrupt their routines to make it happen. For example, care providers at Vanderbilt use a direct decision-support device as they browse information about the patient, make decisions about what they want to do, and capture their decisions as care orders [5]. Since diagnoses and orders are mapped to the Unified Medical Language System Metathesaurus, a push of a button retrieves the fifty most relevant articles based upon the patient's record.

Combining Alignment and Scale
Before discussing the third challenge, let me give you examples of the ways in which the team at VUMC's Eskind Biomedical Library is responding to the first two challenges. We have implemented a customer-centered service model [6]. This model has replaced an approach based upon divisions of the library in which we handed our patrons around and made them deal with us in the way we were organized. In the new model, we try to package what we do the way the customer wants to see it. We try to have exactly the right level of skill dealing with a problem-no more and no less. We have tried to build staff growth and development into the delivery of each unit of service. Those are the objectives. The strategies that we have used include cross-training of everyone involved in coverage of the service points so that people can handle a request, whether it is one that would traditionally have been part of reference, or whether it is one that would have traditionally been part of circulation or document delivery. The first person that a patron sees owns solving their problem. That person may or may not solve the problem, but if they cannot solve it, they take the patron to the person who can solve it, and they learn how to solve it. In that process, the patron does not become lost, and the next time that problem comes to that staff member, they will know the answer. We have established a three-tiered career ladder for our health information analysts that allows them to grow and be promoted as they learn these various skills. We intend for the model to make the library the single source to information from Vanderbilt. We're not yet there, but we would like to reach the point where, if you come to us with a question, and the answer lies in a faculty member's head in biochemistry, we will be the bridge to that faculty member's head.

We have created differentiated product lines to expand the depth and breadth of our services (Figure 1). These product lines are built upon a core of resources including our facility, the expertise of the informatics faculty and staff, our print collection, our electronic databases, our artifacts, and our archives. These core resources are funded through allocation by the schools and hospital, and there are a limited number of dollars that can be obtained from these sources. Our strategy is to layer onto that core value-added products on a fee-for-service basis. We can deliver a service that utilizes core resources for less than it would cost if those resources were not available. At the same time, those fees allow us to expand the resources beyond what would be possible without them.

The Records Center is an example [7]. We decided to streamline movement of material into the archives by implementing good record management processes throughout VUMC. The program included a record retention plan and provision for warehousing, indexing, and retrieval services. This program was set up to keep important material from being discarded and to have it "archives ready." Once the program was in place, we noticed the amount of space throughout our institution that was occupied by boxes of files that people don't look at very often. We decided that we could "build" space very inexpensively at home base by moving those boxes to a warehouse facility and retrieving them on demand. The strategy that began as an effort to build the archives saved the institution money, and generated a revenue stream that could, in part, support the archives. As people discover that we know how to run a cost-effective archive, VUMC is being selected as the archival site for resources from various professional societies, etc. In this way, Vanderbilt is getting original information that does not exist elsewhere. This information is not something that can be bought in electronic database format. It is the kind of thing we should be collecting at Vanderbilt.

Our professional development strategy provides a second example. It begins with a taxonomy of needed skills [8, 9] and goes on to create a nurturing, learning work environment for each member of the staff. The model is similar to the clinical clerkship and house-staff experience where you work under the guidance of a mentor and learn as you do. The customer-centered service model, the Journal Club, and Information and Education Service's "challenging search of the week" are examples of learning opportunities. We have taken a next step by placing library interns and staff into clinical rounds on certain of our key patient care units. They are involved in discussions and decisions at the time care is being provided, not in a morning report setting. They work as part of the larger expert informatics team, filtering, interpreting, and bringing information back into the care process. They build a knowledge base of frequently needed information as a byproduct. In this way, one service provides professional development, value at the point of care, and a new information resource. In the process, people throughout the enterprise become aware of the resources and talents that make up the library.

The Challenge of Measuring Effectiveness
Let me step away from Vanderbilt to the third challenge, the challenge of measuring effectiveness. We have gotten into trouble in health care because we have said we have quality, but have not proven it. If we want the library to be at the epicenter of the enterprise business strategy, we are going to have to prove its effectiveness and contribution to that strategy. I do not think that we know how to begin to tackle this problem. Instead, we measure process. How often is a resource available? How responsive are we to a question? What is the rate of use by resource? What is the cost by unit of service? What is the satisfaction by unit of service? I think we have to go beyond looking at process if we are going to stand up to the scrutiny that will be focused upon a critical element of a business strategy.

The definition of the library's products and services as a part of the business strategy presents the opportunity of assessing the degree to which we actually achieve the effect that we say we can achieve. If we are going to raise the level of clients' self-sufficiency a rung each year, we should measure their level of understanding. We should report the percentage of our clients who know what they need to know, not just the number of people who have attended class. If we commit to increasing the efficiency with which people access information, we should report the number of times a resource is used out of the set of instances in which benefit would have occurred if it had been used. At the same time, we should report the number of times that an information resource is accessed unnecessarily. If we pursue publication at the source, we should report the percentage of original material produced in our enterprise that is archived and accessible.

We should also try to measure the effectiveness of those activities of the enterprise that might be affected by the library. This type of assessment would be analogous to clinical outcomes, where you seek to determine the impact of an intervention on quality of life. It would also be analogous to evaluating an education strategy by assessing the lifetime productivity of people who were exposed to a program as contrasted to those who did not participate. The number of times the work of a faculty member is cited would be one such measure. Alternatively, if we were to move to publication at the source, we could measure the balance of trade—how many people come into an enterprise's archives versus how often its staff goes into other archives. Finally, to the degree that the library is essential to a new business product, the success or failure of that product would be proof of effectiveness.

Conclusion
With this framework of three challenges in front of you, let me go back to the first one. How do we align the library strategy with the enterprise business strategy? I will show a short clip from a video that depicts how focus groups at VUMC think an integrated health care delivery system ought to work [10, 11]. That video segment is a statement of VUMC's business strategy. It does not mention the library. But very few of the functions that it depicts can be done without involvement of the library. How are we going to get the information to the clinical teams so that they can work together to agree on best practice? How are we going to manage, update, and access the library of pathways? Consider the records management that will be required to support an outcomes respository. Think about the implications of patient empowerment. How are we going to get information into a form that patients can use? Will we need a library of personal value models? Finally, think about process redesign. How are we going to document the enterprise's processes so that they do not have to be recreated from the beginning each time? These are all areas where the library can play a critical role, and yet the library was not mentioned.

So I challenge you. When your business puts together its business strategy, I doubt that the leaders are going to recognize the importance of the library. It is your job to put that issue on the table. It is your job to say, "This is something we can do. We can do it better than anybody else can and it will provide an essential step towards your business strategy. Therefore, it needs to be included in the price of that business strategy, and you need to invest in our capability to do it."

The Eskind Biomedical Library stands in the geographic center of Vanderbilt University Medical Center. At night, it shines like a beacon drawing people to its warmth and resources [12]. Librarianship, like it, and you as individuals, like it, are at the right place at the right time to make a difference. In the words of my mentor, Dr. Ike Robinson, onward and upward.

Acknowledgments
The author thanks Nunzia Giuse, M.D., Jeffrey Huber, Ph.D., Marcia Epelbaum, Patti Lee, T. Mark Hodges, Frances Lynch, Randy Miller, M.D., and Ed Shultz, M.D., for their suggestions during the preparation of the lecture. Thanks to the staff of the Eskind Biomedical Library for bringing the ideas to life as they go about their work.

Notes

* Stead WW. Positioning the library at the epicenter of the networked biomedical enterprise. Bull Med Libr Assoc 1998 Jan;86(1):26–30.

References

1. Koch MO, Seckin B, Smith JA Jr. Impact of a collaborative care approach to radical cystectomy and urinary reconstruction. Urology 1995 Sep;154(3):996–1001.
2. McNeer JF, Starmer CF, Bartel AG, Behar VS, Kong Y, Peter RH, Rosati RA. The nature of treatment selection in coronary artery disease. Experience with medical and surgical treatment of a chronic disease. Circulation 1974 Apr;49(4):606–14.
3. Giuse NB. Advancing the practice of clinical medical librarianship [editorial]. Bull Med Libr Assoc 1997 Oct;85(4):437–8.
4. Schuler GD, Boguski MS, Stewart EA, Stein LD, et al. A gene map of the human genome. Science 1996 Oct 25;274(5287):540–6.
5. Geissbuhler A., Miller RA. A new approach to the implementation of direct care-provider order entry. Proc AMIA Annu Fall Symp 1996:689–93.
6. Epelbaum M, Lynch F, Carlton B, Martin S, Pfeiffer J. A team approach to customer-centered service. Presented at Medical Library Association Annual Meeting. Seattle, Washington, May 25, 1997.
7. Jones R, Banta B. Records management in health care—challenges, innovations, and opportunities. Presented at Association of Records Managers and Administrators Annual Meeting. Denver, Colorado, October 1996.
8. Giuse NB, Huber JT, Kafantaris SR, Giuse DA, Miller MD, Giles DE, Miller RA, Stead WW. Preparing librarians to meet the challenges of today's health care environment. J Am Med Inform Assoc 1997 Jan–Feb;4(1):57–67.
9. Giuse NB, Huber JT, Stead WW. Project-based management training as a tool for proactive learning. Presented at Medical Library Association Annual Meeting. Seattle, Washington, May 27, 1997.
10. 10. Olsen AJ, Stead WW. Strategic planning for information management: what are the deliverables? Proc Annu Symp Comput Appl Med Care:785–9, 1995.
11. Olsen AJ., producer. Creating our future in the information age: seamless healthcare delivery [video, 7 min. 26 sec.]. 1995.
12. Hodges TM. Technical scale and high-level detail: Vanderbilt's award-winning Eskind Biomedical Library. Bull Med Libr Assoc 1995 Jul;83(3):333–45.

Received August 1997; accepted September 1997 ,/p>