Eric P. Delozier (epd103@psu.edu) is Senior Assistant Librarian and Virginia A. Lingle (lingle@hslc.org) is Associate Librarian at the George T. Harrell Library, Hershey Medical Center, Pennsylvania State University, Hershey, PA.
In the face of budget cuts, rising publisher costs, and the introduction of new technologies, librarians continue to provide quality services to their respective constituents through innovative alliances. This paper describes a unique cooperative venture among health sciences libraries of the Committee on Institutional Cooperation (CIC) to identify, organize, annotate, evaluate, and provide access to health-related resources on the Internet. Established in 1958, the CIC is an academic consortium of the Big Ten athletic conference plus the University of Chicago. It develops and implements a variety of programs and activities in support of higher education in participating institutions. Recognizing the value of information on the Internet and a need to offer a convenient interface to health-related information for health care practitioners, educators, students, and consumers, a group of health sciences librarians from CIC institutions founded HealthWeb (http://hsinfo.ghsl.nwu.edu/healthweb/) in 1994.
The political climate for libraries these days seems to be very positive. A recent issue of American Libraries(1) discusses the proposed U.S. budget for fiscal year 1998 which calls for increased monies amounting to $500 million for Educational Technology to help meet President Clinton's goal of linking every school to the information superhighway by the year 2000. Budget requests for federal library programs see an overall increase as well. The President twice referred to libraries in his State of the Union address and repeated his pledge to connect every school and library to the Internet. In the same news bulletin(2), the American Library Association's Executive Director praises the Secretary of Education for his planning for full library involvement in his campaign to make every child a reader.
Indeed, the overall attitude regarding libraries and librarians has taken an interesting positive turn with the glut of information that has exploded on the Internet and the Web. A well-known article that recently appeared in the New York Times(3) highlights the expanded role that librarians can now have in "a raft of new career opportunities." The article goes on to say "while it is still a discipline dedicated to finding, filtering, organizing, evaluating, and presenting information, it now requires a great deal more technical expertise with electronic information and computer networks." With the Internet and the Web, the need for those skills to "find, filter, organize, evaluate, and present information" is greater than ever - including for resources related to health information.
The HealthWeb project that I have come to discuss today, was conceived to help to meet that need. Health information on the Web has evolved and is continuing to develop and multiply at a phenomenal rate. Web pages are being created by almost everyone it seems-from well-known organizations to individual health care providers. In a recent issue of the ACP Observer(4), for example, a number of Web-based physician directories, such as DoctorNet (http://www.doctornet.com), are described that allow a physician to list practice information on a Web page. Numerous resources that traditionally have appeared in print are being made available on the Web.
In another article in the same ACP Observer(5), a well-know medical informatician, who is interviewed, sees health information evolving from printed materials digitized over the Internet, to become "specialized products" that will be very different from traditional archival publications. So much so, he believes that "the hospital library is in peril unless librarians find a way to add value to their new 'middleman' role." The HealthWeb catalog, created by librarians, is, certainly, one tool that does add value to the health information on the Web and enhances the role of librarians.
What is HealthWeb?
HealthWeb is an annotated catalog of current health-related Internet resources that uses the World-Wide Web as an interface to the end user. The catalog is maintained by a cadre of professional librarians affiliated with twelve institutions that are members of the Committee on Institutional Cooperation (CIC), the academic arm of the Big Ten Athletic Conference, plus the University of Chicago. Eleven of the twelve are also participants in the Greater Midwest Region of the National Network of Libraries of Medicine (NN/LM). Each participating library is responsible for the identification, organization, and annotation of one or more health-related subject areas, or, areas of excellence, from a list of seventy-two possible topics. The subjects of HealthWeb are discussed in greater detail later. HealthWeb utilizes a distributed interface model in that there is no centralized Web server; although the project does have a homepage (http://hsinfo.ghsl.nwu.edu/healthweb). Consequently, individual subject pages are housed on a diverse collection of systems.
HealthWeb Purpose
HealthWeb's purpose, or, mission, is based on four core areas: collaboration, identification, organization, and evaluation. Given this, HealthWeb will:
The origins of HealthWeb began in 1993 when two health sciences librarians from the Universities of Minnesota and Michigan discussed their Internet teaching experiences. Realizing that the number of resources and services on the Internet was growing at an enormous pace and that there was a lack of network literacy among users, they foresaw an opportunity for collaboration with other librarians. From these discussions, a meeting was organized in conjunction with the Virtual Electronic Library Project of the Committee on Institutional Cooperation (CIC) libraries in 1994. Out of this meeting grew the Health Sciences Libraries Internet Working Group (HSLIWG), a special interest group of the CIC. The group agreed that the Internet lacked a mechanism for the organization and access to meaningful health-related resources and began brainstorming on a variety of strategies. The group also agreed that the most appropriate way to present this information was through the World-Wide Web (WWW). During the later part of 1994 and into 1995, the group identified seventy-two subjects and divided them among the twelve institutions. The project achieved a milestone in 1995 when the first collection of prototype Nursing pages was made available to the Internet community. This was soon followed by pages on minority and consumer health. As of January 1997, HealthWeb boasted over twenty subject pages completed and twelve in progress.
As with most cooperative undertakings, responsibilities are distributed among participants. With HealthWeb, those responsibilities are assumed by a steering committee comprised of at least one person from each institution, along with a collection of subgroups, or working groups, who are charged with carrying out the various administrative responsibilities for the group as a whole.
Subgroups For Planning And Implementation
The five subgroups include: communications, content, design, planning, and technical. The communications group develops guidelines for dealing with publicity-related issues that the project might encounter. For example, it creates and designs publicity statements and materials for use on all HealthWeb pages and documentation. It also prepares appropriate sections of grant applications; investigates methods for marketing HealthWeb; and establishes requirements for organizing and archiving documentation pertaining to HealthWeb.
The content group proposes guidelines, or criteria for evaluating resources being considered for inclusion into the HealthWeb catalog. The group also considers the evolution of HealthWeb as an opportunity for developing and enhancing existing library resources. The group examines the following questions when measuring the worthiness of an Internet resource:
The planning group monitors the progress of HealthWeb by coordinating the work of the other groups and developing strategic initiatives for ensuring that the project remains a viable service. It's activities have included: allocating responsibilities for various tasks; monitoring the activities of related projects; and conducting strategic and long-range planning. The group is heavily involved in pursing funding opportunities through member fees and with the preparation of grant applications.
The technical group addresses the technological concerns that impact HealthWeb. Initially, it confronted the issue of whether HealthWeb should pursue a central or distributed model for content administration. Based on the results of an extensive investigation that considered funding and capabilities within each library, the group opted for the distributed model. The technical group is currently exploring a strategy for incorporating a search engine that will interact with the various servers that house the HealthWeb content.
Members (12 + the CIC)
The libraries that contribute to HealthWeb inherit a relationship based on their respective parent institutional affiliation with the CIC. Since 1958, the CIC has developed and implemented a variety of programs and activities in support of higher education among its participating institutions. By all indications, HealthWeb is the first collaborative undertaking to specifically bring together the health sciences libraries of CIC academic health centers. Of the numerous innovative programs of the CIC, the Virtual Electronic Library (http://cedar.cic.net/cic/velnew.html) and the Electronic Journals Collection (http://ejournals.cic.net/) epitomize how technology can enhance the resource sharing among institutions. The twelve institutions of the CIC are presented in Table 1.
Table 1
Committee on Institutional Cooperation Institutions
| Indiana University | Michigan State University |
| Northwestern University | Ohio State University |
| Pennsylvania State University | Purdue University |
| University of Chicago | University of Illinois - Chicago |
| University of Iowa | University of Michigan |
| University of Minnesota | University of Wisconsin Madison |
The underlying framework of the HealthWeb catalog is the division of the overall discipline of the health sciences into seventy-two distinct subjects, or areas of excellence (See Table 2). Individual libraries assume the responsibility for developing and maintaining resources for one or more areas of excellence. Currently, all but five areas of excellence (Cardiology, Nephrology, Occupational Medicine, Optometry, and Urology) have been allocated. Areas of excellence were assigned according to institutional strengths, faculty expertise, and library collections. By assigning areas of excellence in this manner, librarians could work closely with and have convenient access to faculty within their respective institutions. As of January 1997, over twenty-three areas of excellence were fully developed, thirteen mini, or abridged subjects were produced; approximately five were close to being completed, and twelve were in progress. Other subject areas are being considered for inclusion. They include education, general reference, and forensic medicine.
Table 2
HealthWeb Areas of Excellence
| AIDS | Allergy | Allied Health | Alternative Medicine |
| Anatomy | Anesthesiology | Biochemistry | Biomedical Ethics |
| Cardiology | Cell Biology | Chiropractic Medicine | Consumer Health |
| Critical Care | Dentistry | Dermatology | Diabetes |
| Emergency Medicine | Endocrinology | Family Practice/Primary Care | Gastroenterology |
| Genetics | Geriatrics & Gerontology | Health Administration | Hematology |
| History of Medicine | Immunology & Microbiology | Infectious Diseases | Internal Medicine |
| Laboratory Medicine | Medical Informatics | Microbiology & Immunology | Minority Health |
| Molecular Biology | Mortuary Science | Nephrology | Neuroanatomy |
| Neurology | Nuclear Medicine | Nursing | Nutrition |
| Obstetrics and Gynecology | Oncology | Ophthalmology | Optometry |
| Orthopedic Surgery | Osteopathic Medicine | Otolaryngology | Pathology |
| Pediatrics | Pharmacology | Pharmacy | Physical Medicine and Rehabilitation |
| Physical Therapy | Physiology | Preventive Medicine | Primary Care/Family Practice |
| Psychiatry | Public Health | Pulmonary Medicine | Radiology |
| Rheumatology | Rural Medicine | Speech Pathology/Audiology | Sports Medicine |
| Substance Abuse | Surgery | Telemedicine | Toxicology |
| Transplantation | Urology | Veterinary Medicine | Women's Health |
HealthWeb is an evolving project and overtime several issues emerged that called for key members from the various working groups to work together. As a result, two task forces were created; one to investigate the feasibility of integrating a search engine into the HealthWeb interface and another to compare other Web catalogs that were competing with HealthWeb.
It soon became clear that the interface needed a search engine. As the number and size of the subject pages grew, the group realized that the current interface, which only permitted users to browse, was too restrictive. In order to analyze, test, and implement a search engine that could span distributed servers, the group had to come up with more explicit guidelines for writing and storing Web pages. Working in tandem, several key members of the design and technical groups found that indexing software from Open Text Corporation (http://www.opentext.com) met pre-defined criteria. They then defined and distributed to the group a set of criteria for the use of directory names, HTML tags, headers, body, and footers for subject pages.
Comparing HealthWeb to similar projects on the Web was part of the project's strategic initiatives since its inception, but an organized comparison did not get underway until mid-1996. In response to a call for participants, several members of the design working group volunteered to form an ad-hoc task force. The task force solicited recommendations from the entire HealthWeb group for other similar Web sites to evaluate. This resulted in a list of thirteen other similar projects, which are present in Table 3.
Table 3
Selected HealthWeb Competitors
| CliniWeb | CyberStacks |
| Hardin Meta-Directory | HealthAtoZ |
| Medical Matrix | MedWeb |
| Netmedicine | OMNI |
| McKinley Group | MedMark |
| MIC-KIBIC | NOAH |
| WebDoctor |
Like any project, the progress of HealthWeb has been hindered due to anticipated circumstances. Just as all libraries struggle within fiscal constraints, so too does HealthWeb. Financial support for content maintenance and travel for group meetings have been absorbed by participating institutions. Although several attempts to acquire seed funding from private and government sources have been unsuccessful, the group is developing new strategies for seeking financial support. For example, the project may have a greater opportunity for financial support if it incorporated some kind of innovative research or discovery component, e. g., integrating the Unified Medical Language System (UMLS) Metathesaurus in the HealthWeb interface, into a grant application. While the appropriate working groups continue to seek grant funding, it looks as though an annual dues structure is likely. This will help fund the costs of trademarking the HealthWeb name and contribute to the design of a logo.
With well over seventy librarians from twelve libraries spanning eight states, face-to-face communications is a luxury. As a result, it is sometimes difficult to reach a consensus on critical issues, such as preparing a grant proposal, that require communication in real-time and in person. The group does conduct formal meetings twice a year in conjunction with the annual meetings of the Medical Library Association and its Midwest Chapter. The group continues to exploit e-mail and the Web as the backbone for conducting business activities. It also relies extensively on the use of L-Soft's LISTSERV® mailing list software and the Web for distributing correspondence and project documentation.
The distributed model of HealthWeb is, for the most part, dictated by a lack of funding necessary to implement a centrally-managed system. The search engine could have been implemented much more rapidly if the project was able to financially sustain a centralized system and technical expertise devoted to the project.
The Web is a time-sensitive and intensive publishing and distribution medium. Furthermore, its rate of growth was practically impossible to predict. As a result, one can imagine the difficulty in attempting to identify, let alone organize what is available in not only the health sciences, but any subject area for that matter. Add to this a volunteer staff that devotes their time as schedules permit and one soon appreciates the complexity of the project and the progress that has been made.
HealthWeb continues to grow, in large part, because of the dedicated perseverance of a collective group of librarians who continuously create and update its subject areas. It's continuity is also attributed to the group's ability to strategically plot important directions, e.g. the search engine and meta-directory task forces discussed earlier, the project should follow in order to remain competitive. The HealthWeb steering committee and planning group are considering additional strategies, such as: expanding the scope of participating libraries; increasing the awareness of the project; trademarking the HealthWeb name; and investigating a database format for subject content.
The level of participation among the current group of HealthWeb libraries has been constrained by local resources and commitments. As a result the pace at which content gets added to the catalog is slower than what the group had hoped for. With this in mind, an invitation to other libraries, such as those from the Greater Midwest Region of the NN/LM, to participate is currently under consideration. A Memorandum of Understanding is under development and is pending formal approval by the steering committee.
Despite the depth and quality of the HealthWeb catalog, there is still a need to increase the project's awareness. One way to accomplish this is by gradually continuing to create new content. The group is also investigating the prospect of increasing awareness by exhibiting at professional association meetings and conferences. The project can also be promoted by creating pamphlet, flyers, and other educational materials.
Investigations into creating a fully qualified domain name (FQDN), e. g., healthweb.org, uncovered another Healthweb organization. Recognizing the novelty of the project and importance of protection against infringement by or upon the rights of others, the group has decided to examine the feasibility of trademarking the HealthWeb name.
Almost without exception, the HealthWeb pages are static and somewhat labor intensive to maintain. A more practical arrangement would be for the group to investigate a database model that would allow subject Web pages to be created and modified dynamically, or "on the fly." This issue is currently on the technical group long-range planning activities.
What initially started as an electronic survey to identify who was doing Internet instruction and whether they were interested in sharing their expertise has matured into a unique resource on the Web. The attributes that make it a unique resource include its emphasis on selectivity, cooperation, and design. Also significant are the technical operations and the development of guidelines shared among twelve institutions. The selectivity is seen by the critical evaluation that is done cooperatively with librarians and subject experts at their respective institutions. Collaboration is the major element that has made the creation of HealthWeb possible. Following a distributed model, the twelve institutions share the workload and cost to support a common set of principles. The continually evolving design of HealthWeb has been and will continue to be crucial to its success. A key goal in displaying information on the Web is to get users to be only two clicks away from the home page to the information they want. This is an ambitious goal that needs continual work.
All in the project believe that HealthWeb is distinctive by its emphasis on health-related disciplines, by the selectivity among the many available Internet and Web resources, and by its efforts to link resources via a consistent interface. We also recognize that the effort to build HealthWeb will eventually aid and complement projects like OCLC's Internet Database and NLM's Finder Project. Each of these projects replicates the traditional library concepts of identifying, selecting, collecting, organizing and diseminating information. Certainly, the idea of collaboration among libraries is not new, but HealthWeb is a very visible demonstration of the new technologically sophisticated environment to which these concepts must be applied.