Plans for the Development of a Medical Library Network, 1965
Page  1

PLANS FOR THE DEVELOPMENT OF A MEDICAL LIBRARY NETWORK Martin M. Cummings, M.D.* [L.B. - get slides to go with this speech for tomorrow - N. G. trip- M.C.] [*Presented at the Second Institute on Information Retrieval, University of Minnesota, Minneapolis, Minnesota, November 13, 1965.] During recent years, many plans have been submitted for the development of national medical library systems and networks. Paradoxically, medical librarians who should know most about their needs and requirements have not developed their own national plan. Unfortunately, many of the existing proposals completely disregard existing libraries and the role they can play in the development of a national information network. The National Library of Medicine believes that any national system should build upon existing resources, utilizing new techniques and equipment wherever possible to improve the flow of information throughout the network. We are in the process of developing the principles and specifications for a national medical library system for the health sciences. The planned network was given major impetus by the recent passage of the Health Sciences Library Assistance Act, which was signed into law by the President on October 22nd. The NLM staff are now in the process of collecting the necessary data and performing systems analysis work for the development of a medical library network. I will report in general terms the overall framework of a medical library system which we believe should be implemented if it is economically feasible. Undoubtedly changes will occur as workon the study progresses. The objectives of a national medical library plan should be: 1. To increase the resources of published information necessary for the advancement and application of sciences related to health.

Page  2 2. To relate strengthened resources to the introduction of new technplogies for development of a coordinated network of services. 3. To offset inequalities of geographic distribution of information resources in an attempt to guarantee to all engaged in the national health effort equal and prompt access to the published information which they require in their work. 4. To serve mission or subject oriented groups as well as medical libraries. The following principles are basic to the development of this plan: 1. The major capital investments in existing resources should be utilized optimally. Institutions which have developed information service functions are capable of modification but need not be replaced by new institutional forms which involve duplication of capital costs. 2. Modification of traditional service patterns necessitates close cooperation between government and private agencies. The Federal function should be one of leadership, persuasion and support for modernizing library functions. 3. Universities and industry as major users of information should play an important role in the development and implementation of the system. 4. To realize their full potential new technologies of information storage and retrieval must be developed in close association with library-based document retrieval systems.

Page  3 5. The development of a network of medical information resources involving libraries, while fundamental to improved medical communication, must be supplemented by programs for the dissemination of these resources to specialized interests involved in the national health effort. 6. The system should be compatible with other major national efforts in other scientific disciplines. OVERVIEW OF THE TOTAL SYSTEM (Slide 1.) Basic input to the system would be medical literature published throughout the world. The National Library of Medicine would continue its traditional function as a central collecting and switching mechanism by comprehensively acquiring this literature and processing it for the widest possible dissemination. Distribution of this information to users at the regional or local level would be accomplished in two ways: 1. Information would be distributed geographically through a network of regional libraries serving local institutions and individual users. 2. Information would also be repackaged for specialty groups where the medical literature is segmented to meet the particular needs of a subject field or institutional mission. Let us now turn our attention to the three major parts of the system. Centralized processing at NLM, geographic dissemination, and missionoriented dissemination.

Page  4 CENTRALIZED FUNCTIONS OF THE NETWORK (Slide 2) The National Library of Medicine has traditionally functioned as the center of an informal network of medical libraries in this country. NLM will continue to serve in this role, and with the advice of its Board of Regents will be the responsible agent for planning the specifications for the network, including selection of existing centers of excellence to serve as regional medical libraries and computer search centers. Once the network is operational, NLM will coordinate the network activities and evaluate the services being provided by the member institutions. The regional medical libraries will be supported financially by the NLM. We do not agree with Kilgour's assumption that the Federal government should support the development of independent regional plans and then leave them to languish because universities do not have funds to operate them. NLM engages in a comprehensive acquisitions program to procure published medical literature throughout the world. This necessitates a constant search for all literature in the medical field and frequently requires changes of policy on scope and coverage to encompass new interdisciplinary fields, e.g., biophysics and environmental health. For economy and standardization of format, we believe that bibliographic processing should be done centrally by NLM which has engaged in indexing and cataloging the world's medical literature for almost a century. All new books will be cataloged and substantive journal articles will be indexed in depth. Bibliographic data so generated will be entered into the NLM computer (MEDLARS System) and stored on magnetic tape. These tapes will then be made

Page  5 available to other libraries and user groups for regional and local services. Since the Library has a statuatory requirement for archival preservation of all the material in its collection, it will be necessary to record most of the new acquisitions permanently in microform. Information in microform will then be provided for strengthening the resources of local medical libraries. In those cases where materials cannot be purchased from commercial sources, a film file of current material could also be useful in an inter-library loan system covering material not available in the regional or local medical libraries. NLM will attempt to maintain a continuous liaison with publishers in order to discuss and develop improved methods of publishing which can aid the bibliographic processes. For example, as more and more of the publishers begin to use automated processes in their printing plants. magnetic tapes or paper tapes of the full text of documents might be available for possible experimental use in automatic indexing. Although the Library will serve as a "wholesaler" of information to components of the network, it will also continue to serve to some extent as a "retail" outlet providing local services to other Federal agencies, to professional societies and to medical libraries in the Middle Atlantic States. The NLM would also provide supporting services to the network through its newly approved grant programs for the development of facilities, resources, secondary publications, and the support of local research and training activities

Page  6 GEOGRAPHIC DISSEMINATION (Slide 3) Geographic dissemination of published information would be accomplished through the mechanism of regional medical libraries. NLM would periodically update the information stores of the regional medical libraries with bibliographic data on magnetic tape, printed bibliographies, and microforms of selected publications which were not commercially available. The regional medical libraries would be responsible for providing services to local institutions including: 1. Literature searches by computer. 2. Computer generated current-awareness listings. 3. Copies of documents or in some cases abstracts of the original documents in either hard copy or microform. 4. Reference services. 5. Training and orientation of medical librarians. 6. Support of specialized information centers. NLM is currently involved inviting industry to perform serveral duties to determine adequate standards for regional medical libraries and to pinpoint the services which can be most effectively provided through this mechanism. The development of a formal network without such information can only lead to an artificial paper network. MISSION-ORIENTED DISSEMINATION (Slide 4) It is recognized that the regional library complex cannot meet the total needs of all special interest groups in the health sciences. In the NLM plan, specialized bibliographic data will be repackaged in machine-readable form or printed lists for evaluation and use by these mission-oriented groups.

Page  7 For example, bibliographic data will be provided to other Federal agencies to provide services to their own medical and scientific personnel. The NLM has already worked with the Veterans Administration for the establishment of a specialized MEDLARS Literature Search Center to provide demand bibliographies to VA hospital staff members. Similar agreements are being negotiated with several other of the Federal health agencies. The Library has also been involved for the last two years in working with professional societies and other national specialty organizations in providing recurring bibliographies to their members. Through the MEDLARS computer system, master film copies of several such specialized bibliographies have already been prepared. Examples of these include the semimonthly Index of Rheumatology, printed and distributed by the American Rheumatism Association; the Index to the Dental Literature, printed and distributed by the American Dental Association; and the Cerebrovascular Bibliography, jointly sponsored by the National Institute of Neurological Diseases and Blindness and the National Heart Institute. Specialized private industrial organizations such as drug companies may also acquire published data in machine readable form from the medical library network. Preliminary negotiations have been held with such organizations as the Pharmaceutical Manufacturers Association for the distribution of MEDLARS magnetic tapes to member companies. As mentioned previously, regional medical libraries may also support mission-oriented information dissemination by providing intensive bibliographic services for specialized information centers in their geographic

Page  8 locality. For example, the MEDLARS Center at the University of California, Los Angeles, will be providing magnetic tape files of citations of particular interest to the Brain Research Information Center located at UCLA. An important cycle of feedback to NLM can be effected as seen in this list of examples is by no means comprehensive but is indicative of the type of specialized repackaging of information required to meet the interests of the many mission-oriented groups. (Slide 5) EXPERIMENTAL LINKAGES Information will flow throughout the national network in a variety of forms. Materials acquired from publishers may take the form of printed documents, microform, and machine-readable tapes used in automatic typesetting systems. Bibliographic data can be transmitted from the central NLM facility to regional libraries and mission-oriented users via printed bibliographies, the mailing of magnetic tapes, or through electronic linkages utilizing teletype, voice-grade, or microwave circuits. Hard copy may be transmitted from NLM to regional libraries and within the regional complexes themselves via mailing of original documents, mailing of microforms or photocopies of the original, and possibly by facsimile transmission. NLM with the assistance of industry intends to experiment with the selection of the appropriate linkages for various parts of the network through contractor studies as well as through its own intra-mural research program.

Page  9 Factors which must be considered in determining the appropriate means of transmitting information are: 1. Cost 2. Volume of. data to be transmitted 3. Throughput time requirements 4. Transmission quality requirements. (That is, minimum allowable noise levels). 5. The state of the art of the technologies for electronic linkage. Some of these questions may be answered by using computer simulation techniques based on mathmatical models. Others can only be solved by the test and demonstration technique such as has been employed successfully in the decentralization of MEDLARS. CONCLUSION In summary, this report has attempted to outline a framework of a national library network. The plan is by no means complete and considerable work must be done by the staff of the National Library of Medicine with the cooperation of many institutions throughout the U.S. Other considerations not covered in this report must be explored also. For example, how should a medical library network relate to proposed networks in other disciplines and to an overall national information system? Should the network be eventually expanded to include international institutions to foster cooperation with organizations such as the World Health Organization? How will NLM's experiences with mechanization be utilized by other research libraries such as the Library of Congress and the National Agricultural Library? How will independent developments such as the Columbia-Harvard-Yale effort fit into

Page  10 a National network? NLM intends to pursue its study vigorously. Its aim is to develop a comprehensive plan designed to facilitate the dissemination of scientific and medical information among health scientists, medical educators, and practitioners. The scientific revolution of the Twentieth Century must be accompanied by library technological developments which makes the knowledge generated by research available to all who have need to use it. However, we will not attempt to do this at any cost. We intend to utilize the test and demonstration technique as we have with MEDLARS before implementing a national library network. The network may be built from an apex downward or from a base upward--we are convinced however that it cannot be built from uncoordinated efforts outside of the triangle of existing local, regional and national resources.