Purpose of the Directory

The Medical Technology Assessment Directory is the first substantial published guide to the field of medical technology assessment. It is intended to better define the substance of medical technology assessment, and to crystallize the emerging network of organizations, institutions, and individuals that generate and use technology assessments. The Directory is a reference to medical technology organizations, assessments, and information and data resources. It covers activities in the U.S. and certain other countries, including those of government agencies, private companies and associations, and research and educational institutions. It represents the first indexed bibliographic compilation devoted to medical technology assessments produced, in progress, or planned by a broad set of assessment programs spanning the public and private sectors. The Directory was developed to increase access to valuable health care technology assessment information, much of which is found outside standard published sources.

This is a product of the Information Panel of the Council on Health Care Technology. One of four panels of the council, the role of the Information Panel is to oversee the development and operation of an information clearinghouse for health care technologies and health care technology assessments, as called for in the congressional mandate for the council. Although the various constituencies of the council have different sets of needs regarding technology assessment, nearly all have expressed a desire for the council to collect, organize, and disseminate technology assessment report information.

As implied by the term ''Pilot'' in its subtitle, the Directory takes a leading step toward a more comprehensive capacity for information services in medical technology assessment. Though it has immediate utility in its own right, the Directory is an experimental compilation of information in an emerging field. The assessment report citations and other categories of information included here represent a cross-section of the field, and cover assessment activity of the recent several years of a selection of programs in the U.S. and abroad. One of the major functions of the Directory is to act as a hardy-copy precursor of a computerized technology assessment information base. The development and testing of the Directory will help guide the Information Panel's clearinghouse plans.

With its extensive descriptions and contact information, the Directory provides ready access to assessment people, programs, and resources. It is intended for users who wish to identify available assessments and contact people currently conducting assessments of a particular technology before making a decision on medical practice, coverage, reimbursement, acquisition, or standard-setting.

Private and government third party payers, HMOs, and self-insured employers currently reviewing technologies for coverage or inclusion in their health plans may use the Directory to locate assessments of these technologies by regulatory bodies and physicians' groups. Physicians seeking medical consensus regarding appropriate use of technologies may find guidance from assessments indexed here. Pharmaceutical and health device and equipment makers may use the Directory as a reference to the processes and scope of activity of programs that are likely to assess their products. Procurement decisions for health care facilities may be more informed by reviewing assessments of devices and equipment, and by consulting organizations currently assessing new technologies. Assessors seeking information concerning such areas as postmarketing surveillance, drug utilization, or new product development may use the Directory to locate appropriate resources.

The Directory may help to advance the field in many other ways. Organizations that conduct or are considering initiating assessment activities may discern ways of strengthening their programs by examining the attributes of profiled programs. By identifying, describing, and providing contact information for assessment organizations, the Directory should raise the level of, and render more efficient, communication in the assessment community. Knowing the types of technologies generally assessed by programs, as well as methods used and other program characteristics, those seeking assessment information may prompt appropriate programs to undertake assessments to address specific needs. Those in the academic, research, and public policy sectors who study the field of technology assessment may contrast and compare assessment processes and consider how better to set assessment priorities and otherwise strengthen the assessment enterprise. Thus, the Council on Health Care Technology and others may use the Directory information base to review the capacity and products of technology assessment and to discern trends, gaps, and redundant efforts in the field.

Given that this edition could not be a comprehensive rendering of medical technology assessment and that this is an emerging field, wide dissemination of the Directory is crucial for building and updating its information base. The council hopes that, by reviewing the Directory , assessment programs and other resource organizations not already included will recognize that they conduct activities similar to certain of those described here, and will contact the council about being included in forthcoming editions.

Gray Literature for Assessment

Certainly an important step in carrying out most assessments of medical technologies is searching an information base covering the relevant peer reviewed literature in health and biomedicine. Perhaps the foremost example of such an information base is the online bibliographic citation base MEDLINE, a major component of the National Library of Medicine's MEDLARS system. MEDLINE is one of more than 70 information and data resources described in Part 3 of the Directory. It contains all of the citations published in Index Medicus. MEDLINE covers more than 5 million articles from some 3,200 biomedical journals, out of the more than 25,000 cataloged by the library.

Yet, given the rapid evolution of health care's technological portfolio and the breadth of assessment perspectives on medical technologies, valuable technology assessment information may be found outside the mainstream of published periodical literature available from traditional sources. This "gray" or "fugitive" literature is found in industry and government monographs, regulatory documents, professional association reports and guidelines, market research reports, policy and research institute studies, spot publications of special panels and commissions, conference proceedings, online services, and other sources.

Although users of the gray literature must consider that much of it has not been subject to the quality screen of peer review, it often has great utility for assessment given its scope beyond traditionally selected sets of journal literature and a currency not subject to the lag time of peer review. The border between gray and standard literature is not distinct; what is gray and what is standard may differ according to one's familiarity with and access to various sources. To be sure, some gray literature eventually passes into the standard literature. Many of the assessment reports generated by organizations profiled in this Directory, such as the American College of Physicians Clinical Efficacy Assessment Project, American Medical Association Diagnostic and Therapeutic Technology Assessment Program, and the National Institutes of Health Consensus Development Program, do appear in the peer reviewed literature, though months or perhaps one or two years after initial release in some gray form.

Conducting a good assessment may depend not only upon awareness of key gray literature sources, but one's ability to access them. Desired reports may be available through press releases, journal articles, or books; provided free of charge or at a fee; made available in open libraries or held as proprietary. For instance, the comparative medical device and equipment assessments published by the independent evaluator ECRI are not made available to open libraries such as the National Library of Medicine (NLM), but only to subscribing hospitals, HMOs, and other institutions because ECRI's revenues could be eroded by free or low-cost (e.g., by photocopying) access to its publications. Many of the Food and Drug Administration's (FDA) technical reports on drugs and devices are available only upon request from the FDA or from the National Technical Information Service. Certain reports on emerging and new medical technologies generated by market research firms are publicly available, though at costs ranging from a few hundred to tens of thousands of dollars.

Rather than devoting considerable effort to already organized and indexed assessment reports appearing in the peer reviewed journal literature, much of the Directory is addressed to gray sources, including many of the profiled assessment programs and the information and data resource descriptions. Thus, this Directory is the first attempt to organize gray literature sources in the field of medical technology assessment.

Major Parts of the Directory

Part 1 of the Directory includes profiles of 68 assessment programs and a total of approximately 3,200 citations of their respective assessment reports. These assessment programs are located in a variety of public and private sector organizations, including professional, scientific, and industry associations, biomedical research institutions, educational institutions, government advisory bodies, regulatory agencies, third party payers, policy institutes, and for-profit corporations. The profiled programs are primarily from the United States; among other countries represented are Canada, the Netherlands, Sweden, and the United Kingdom. These are ongoing programs that generate medical technology assessments. For the purposes of this Directory , a medical technology assessment is a study or inquiry, the object of which is to provide information regarding the effects of a technology that is intended to maintain or improve health, or be used as part of an intervention for such purpose, whether or not the study evaluates these effects directly.

Using a common multidimensional framework, these categorical profiles represent a systematic characterization of a diverse community of assessors. This framework was adapted from the one used to profile 20 assessment programs in the 1985 Institute of Medicine study Assessing Medical Technologies. The profiles address the purpose, types of technology assessed, topic selection, assessment methods, properties assessed, report dissemination, budget, contact information, and other characteristics of each program. Each profile is followed by a listing of bibliographic citations of the program's recent assessment reports (over approximately the last five years). Following the profiles is a subject index to the citations. Each of the assessment report citations is indexed using from one to five terms. For instance, a report titled "Dialysis treatment for end stage renal disease (outpatient)" is indexed under the terms Dialysis; Kidney failure, Chronic; and Outpatients.

Part 2 has two major sections: a technology thesaurus and a cross-listing of the assessment report citations by technology. The technology thesaurus was developed for this Directory to help users search for assessment reports by technology topic. It was generated based upon the technologies appearing in the titles of the assessment report citations. Where appropriate, it uses the NLM's MeSH (Medical Subject Headings) vocabulary terms. To enhance searching by all types of users, the 1,200-term thesaurus includes many synonyms, acronyms, and cross-references, and indicates broader and narrower terms to help clarify relationships among technology terms.

The cross-listing arranges the assessment report citations by technology, e.g., Apnea monitoring, Coronary artery bypass grafting, and Infusion pumps, so that all citations for each type of technology may be viewed together. Under any particular technology heading, the citations are listed in reverse chronological order to provide the newest information first. This listing is analogous to printouts of computer searches for citations involving particular technologies. In addition to bibliographic information, most citations note the general type of assessment method used, e.g., clinical trial, group judgment, and information synthesis. Because not all of the 3,200 assessment report citations have a technology term in their titles, Part 2 contains about 2,500 citations arranged by technology. Of course, the subject index to report citations at the end of Part 1 uses technology and other terms to index all 3,200 citations with respect to their source assessment programs.

It is important for users to take advantage of the technology thesaurus for searching the cross-listing of assessment report citations by technology. Unless a report citation has more than one technology in its title, it is shown only once in the cross-listing, usually appearing under its most specific technology term. For citations listed in this section, distinctions generally are made between procedures and the equipment used for them. For example, headings include Computed tomography as well as Computed tomography scanners, and Fetal monitoring as well as Fetal monitors.

Part 3 describes 73 public and private sector information and data resources that may be of value for assessing medical technologies. Arranged alphabetically by name of resource, these are of many types, including online citation bases and services; data files; compendiums, inventories, and directories; market research reports; and others. Each is described in terms of source, subject, content, compilation, and access information. The resource descriptions are preceded by a categorical chart indicating for each resource its file type (directory, bibliographic information, data set, etc.), format (hard copy, online, etc.), and degree of access (public or limited). At the end of this part is a subject index to the information and data resources.

Part 4 provides 300-word descriptions and contact information for 72 organizations that are active in affairs related to medical technology and are stakeholders in medical technology assessment. These include medical societies, scientific organizations, industry associations, and others. Although most of these organizations do not generate technology assessment reports on a regular basis, they have memberships concerned with medical technologies, conduct relevant meetings and symposiums, publish conference proceedings and other documents of note, establish standards for technologies, or are active in government, regulatory, or other policy affairs relevant to the field. At the end of this part is a subject index to these organizations.

Part 5 includes an alphabetical listing with addresses and telephone numbers for all organizations described in the Directory, as well as a page index to the organizations. Among the appendices to the Directory are a description of the council, rosters of council and panel members, a listing of organizations that have contributed financial support to the council, and the survey form used to collect information for the 68 profiles in Part 1.

Scope and Limitations

This Directory covers many of the major national level assessment activities in the U.S. and certain significant ones from around the rest of the world. Its immediate utility as a hard-copy reference notwithstanding, the Directory is not a complete guide to the field or an all-encompassing information base. The set of 68 profiled programs and their 3,200 assessment report citations in Part 1, the listing of 2,500 of the assessment report citations by technology in Part 2 , the 73 information and data resources in Part 3, and the 72 organizational resources in Part 4 are illustrative, not exhaustive, of the scope of technology assessment activities in the U.S. and on the international scene. The Directory does represent an important step toward a more comprehensive resource, as a foundation and a template for building an online medical technology assessment information base and related services.

Types of assessment programs and other assessment resources are variously represented in the Directory. That an assessment program or other resource is not described in the Directory implies nothing about its value. Rather, the breadth of the sample of those included in this first edition helps to suggest how the Directory information base might be expanded. For instance, of the various bureaus, institutes, divisions, and other components of the National Institutes of Health that conduct clinical trials or other assessment activities, just three are profiled here. Seventeen of the 24 member organizations of the Council of Medical Specialty Societies (CMSS) provided information for the assessment program profiles or resource organization descriptions in the Directory; other CMSS members may also be appropriate for addition to the information base.

Eleven university based programs are profiled; there are others worthy of note. A recent publication of the American College of Radiology noted more than 50 U.S. and international societies in the field of radiology alone; certainly, some of these organizations generate technology assessment documents or engage in other activities relevant to technology assessment.

A survey of existing inventories of health care data sets conducted for the Methods Panel of the council by intern Norma Gavin identified more than 260 data sets that meet minimum provisional criteria for inclusion in a proposed inventory of data sources for health care technology assessment. Although the data sets were identified too late in 1987 to be categorically described in this volume, that survey suggests how the set of information and data resources in Part 3 of the Directory could be expanded and its contents more thoroughly elucidated.

Inclusion in the Directory of descriptions of any assessment programs, information or data resources, or other organizations does not imply any judgment on behalf of the council, Information Panel, or staff as to the quality of those assessment resources. Similarly, inclusion of any assessment report citations in the Directory implies no judgment as to the validity, utility, or other aspect of the quality of the assessment reports. The bulk of the content of the Directory was self reported by assessment programs and other resource organizations, and this information was not independently or systematically verified. Compilation of the Directory was largely limited by staff awareness of assessment activities, availability of information about these, and limitations of time, personnel, and other resources available for this edition.

The technology thesaurus in Part 2 was generated based upon the technologies appearing in the titles of the assessment report citations of the 68 profiled programs. Therefore, the 1,200-term thesaurus does not encompass all medical technologies, although it may be readily augmented with additional technology terminology. The technology heading under which an assessment report citation is listed in Part 2 does not imply that the technology has been found to be safe, effective, or otherwise appropriate for the use implied by that heading. That a report citation is listed under a particular heading implies only that the technology that is the subject of that citation may have been assessed for the application represented by that heading. For instance, a particular drug may be listed under Antineoplastics and under Antiarthritics. Assessed as an antineoplastic, the drug may have been determined by a regulatory or other organization to be safe, effective, cost-effective, etc. Assessed as an antiarthritic, the drug may or may not have been found to possess one or more of these qualities. Nevertheless, it would be listed under both headings because the report citations indicate that it was assessed for both applications, regardless of the assessment findings, not because the drug was determined to be appropriate for those uses.

Compilation of the Directory

Information for the Directory was compiled largely from three surveys. The assessment program profiles and report citation listings of Parts 1 and 2 are based primarily upon responses to a survey of organizations with ongoing medical technology assessment programs. Many of the descriptions in Part 3 were written from responses to a survey of organizations that compile or make available data or information relevant to medical technology assessment. The organizational resources descriptions in Part 4 were drawn largely from a request for brief descriptions of professional, industry, and other organizations having interests in technology assessment.

Council staff reviewed and edited all returned survey forms. No attempt was made to verify with third parties the information provided by the organizations. Staff did not assess the quality of assessment programs, information and data resources, organizational resources, or assessments listed in the Directory.

Organizations known or thought to have an ongoing medical technology assessment program were identified initially from council staff files in October 1986. Additional organizations were identified by those organizations surveyed initially and by further council staff work. More than 80 organizations were sent the survey instrument, which is shown in the appendices. This survey was not applicable to organizations that may have conducted medical technology assessments but did not have an ongoing program; that cataloged, indexed, or evaluated assessments but did not produce them; or that conducted assessments but generally did not make results available outside of the organization. Due to resource constraints and because this was not an effort to assemble a library, the survey requested bibliographic citations of assessment reports and sample reports, rather than complete sets of reports from all programs.

Most of the surveys were returned to council staff fully or partially completed. In several cases, council staff and program contacts determined that a program did not constitute an ongoing assessment program, but was appropriate for inclusion in Part 4 or another part of the Directory. Surveys were returned to several organizations with requests for additional information or more thorough responses to portions of the survey; often such requests were for more complete citations for assessment reports. Council staff also used telephone or face-to-face interviews with assessment program staff, as well as reports and other literature submitted by the programs, to complete the surveys. Following editing by council staff and further preparation by the contractor, these descriptions were forwarded to the contact persons at each assessment program with a request to verify their factual content or make other suggestions and to return these to council staff by a deadline date, or to expect no modifications if no response was received by that date.

Staff and consultants reviewed the lists of assessment report citations provided by the programs and deleted those that did not meet the criteria of a medical technology assessment report. For purposes of deciding whether or not a report should be listed in the Directory , the title had to contain or suggest one or more medical technologies, or imply the consideration of medical technologies, e.g., the management of cervical cancer. Further, it had to either state or imply technology assessment, or one or more properties or attributes of a technology that could be determined by an assessment, e.g., safety, efficacy, ethical implications, or state or imply an evaluation; or be or appear to be guidelines issued by a professional body or its equivalent known by staff to conduct assessments. Staff deleted, for example, assessments of the health effects of Agent Orange because it was neither intended to be nor has it been used as a medical technology. On the other hand, staff retained in the listing assessments of the detrimental consequences of medical technologies, such as the risk of birth defects among nurse anesthetists.

Staff tended to be inclusive in determining what constituted assessment reports for listing with program profiles in Part 1. The primary reasons for this leniency were that decisions were necessarily based on limited information, i.e. report titles rather than full reports; that as a first effort, this Directory ought not severely circumscribe this body of literature; and that users might gain a better understanding of programs' scope by viewing fuller portfolios of their reports.

Somewhat stricter criteria were applied for including reports in the list of assessment report citations by technology in Part 2. In instances where titles were ambiguous in this regard and where full reports were on file with council staff, these reports were reviewed to determine whether they constituted assessment reports. However, staff did not make special requests to acquire reports for this purpose.

Edited survey forms were sent to TRITON Corporation, a contractor retained by the council. Using the surveys, TRITON prepared uniform descriptions, or "profiles," of ongoing medical technology assessment programs and created a record for each of their assessment report citations. Each profile has approximately 20 descriptors such as program name, address, telephone, contact person, purpose, methods, reports, etc. Descriptions were edited for consistency and to ensure, as far as possible, uniformity in detail. In creating database records for assessment citations, TRITON followed Index Medicus style. The base of citation records was used for two purposes: for generating the coded list of citations following each profile in Part 1, and for generating the listing of assessment report citations by technology in Part 2.

TRITON indexed each assessment report citation using from one to five subject headings, including but not limited to technology terms. Of necessity, TRITON had to rely exclusively on titles to index assessment reports. The subject headings used for indexing are taken primarily though not entirely from the National Library of Medicine's controlled biomedical vocabulary MESH. The departure from MeSH terms occurred most often for terms related to technologies. The index to the citations appears at the end of Part 1, and is coded to citations appearing after their respective assessment programs.

The technology thesaurus that appears at the beginning of Part 2 was developed in order to arrange the listing of assessment report citations by technology that follows it. The technology thesaurus terms used as headings for Part 2 constitute, largely, a subset of the indexing terms used in Part 1, i.e., those that identify technologies. As noted above, technology terms accounted for most of the instances where indexing departed from MESH. If MeSH did not contain a subject heading appropriate for a technology used in an assessment report citation title, one was designated based on usage in the medical technology assessment literature as determined by staff, consultants, and TRITON. In some instances, MeSH did not include a term for a particular technology. In others, the closest MeSH term may have been too broad or too narrow to adequately identify a technology, or one not commonly used or perhaps not readily recognized by users who do not have a particular medical or scientific expertise. Linkages to MeSH are retained wherever possible with the use of synonyms, acronyms, cross-references, and broader and narrower terms that are used in MESH.

Information and data resources described in Part 3 were identified by council staff and by respondents who completed surveys for this part. About half of the descriptions of information and data resources in Part 3 were based entirely or in part on survey responses. The remainder of the descriptions was derived by staff from literature and samples received from the source organizations and telephone interviews. The copy for Part 3 that was submitted to TRITON was generated by staff.

The organizational resource descriptions in Part 4 were derived primarily from requests for narratives of 300 words or less made to organizations identified by council staff and suggested to staff by certain of these organizations. Following editing by council staff, drafts of these descriptions were forwarded to the contact persons at each organization with a request to verify their factual content or make other suggestions and to return these to council staff by a deadline date, or to expect no modifications if no response was received by that date. The copy for Part 4 that was forwarded to TRITON was generated by staff. TRITON submitted final, camera-ready copy of the Directory to council staff, who turned it over to the National Academy Press for printing.

Plans of the Information Panel

The Information Panel plans to develop technology assessment information products beyond this first edition of the Directory. Among the major proposed areas of activity are wide dissemination of the Directory , working with the NLM to augment the Library's MeSH vocabulary and DIRLINE (Directory of Information Resources Online) file with new information from the Directory, producing access guides to standard literature sources, enhancing and updating the Directory, instituting clearinghouse services, and establishing online access to a technology assessment information base.

Users of the Directory will be surveyed regarding its utility and format, and will be asked for suggestions for improvement. Additional responses will come in the form of number and distribution of sales, book reviews, and other user feedback. The Information Panel will take account of user responses in formulating its plans for further development of the Directory information base and related efforts.

The Information Panel plans a joint vocabulary project with the NLM to better index and thereby improve access to technology assessment information. The vocabulary effort began with the indexing of assessment report titles for the Directory and preparation of the technology thesaurus. From this project, the Information Panel may better determine the topical scope of the technology assessment literature, as well as the capacity of the NLM's MeSH vocabulary for indexing technology assessment topics. Instances in which current MeSH terms are found inappropriate for indexing technology assessment report titles in the Directory provide opportunities for augmenting MeSH with the intention of improving its utility for searching MEDLINE for assessment related information. Because the Directory indexes many report citations drawn from gray literature sources, it may provide a means for identifying newer terms that have not yet entered the lexicon of more traditional sources.

The vocabulary project would establish a framework for, and could begin mapping, certain existing terminology sets such as MESH, ECRI device codes, ICD-9-CM (International Classification of Diseases, 9th Revision, Clinical Modification) and CPT-4 (Current Procedural Terminology, 4th Edition) procedure codes, and DRG (diagnosis-related group) terms to the technology topics used in the Directory. The mapping of vocabulary terms is consistent with the mission of the Library's Unified Medical Language System project.

Information gathered in compiling the Directory will be used to add, update, and modify organizational descriptions in the NLM's DIRLINE file. A component of MEDLARS, DIRLINE is accessible for all users of MEDLARS and certain commercial databases. It contains descriptive information on many organizations and resource centers in the health field. Users interested in particular subjects may enter appropriate keywords to locate information about organizations active in those areas. Adding to DIRLINE descriptive information from the Directory about organizations involved in technology assessment will provide broad online access to it through MEDLARS, and any services that download DIRLINE from MEDLARS.

Whereas the Directory deals primarily with technology assessments generated by specific assessment organizations and found to a large extent in the gray literature, the Information Panel is planning access guides that would address retrieval of technology assessment information from existing citation bases and other sources. Some of these sources, such as BIOETHICSLINE, MEDLINE, and the National Technical Information Service, are noted in the information and data resources part of this Directory. The purpose of these guides would be to help users extract valuable medical technology assessment information from sources that are not specifically devoted to the field or are structured in ways that are not amenable to searching for this information. The guides would describe these sources in greater detail, explain what subfiles or other aspects of each may be useful for assessment purposes, indicate how to access them, and provide examples of assessment searches.

The Directory information base may be developed in several major ways. Profiles of programs entered in the first edition may be updated and their new assessments indexed. Additional assessment programs may be profiled and their assessments indexed, and the technology thesaurus updated accordingly. Based upon the information and data resources in Part 3, an expanded, more detailed inventory of data sets for technology assessment could be developed using more quantitative and qualitative descriptors. This inventory would be guided in part by the survey work on data sets described above.

Of particular interest to both the Evaluation Panel and Information Panel of the council is strengthening the information base to enable tracking of new and emerging medical technologies. Such information has value to different sectors of the health care enterprise. It can be used to anticipate assessment needs and prepare timely and well informed assessments, e.g., at the time of market entry of a new drug or device, or when a reimbursement claim is made for a new procedure. Further, it may be useful for remaining apprised of the state-of-the-art in medical practice, planning clinical services, making resource allocation decisions to foster development and diffusion of promising technologies, charting trends in health care product and practice innovation, and identifying investment opportunities. Sources that might be monitored for information concerning emerging and new technologies might include status reports of biomedical research projects, patent information, medical and bioengineering conference proceedings, medical and scientific bibliographic information bases, future-oriented surveys and monographs, market research reports, financial investment sources, product development databases, annual and quarterly reports of health care product companies, news services, FDA approval documents for new drugs and devices, and health insurance claims data. Although this edition of the Directory has descriptions of certain information sources that fall into these categories, development of the information base could emphasize addition of information sources, indexing, and retrieval means to help meet needs for intelligence about new and emerging technologies.

The Information Panel is considering implementing selected clearinghouse services with expanded staff or under contract. Initial clearinghouse services may be established based upon the Directory and continued updating of its information base, and access to MEDLINE and other online information sources. With these resources, the council could begin to respond to certain types of assessment related inquiries received via written correspondence, telephone, or electronic mail. Staff assigned to this activity could also conduct special searches or generate assessment report bibliographies on topics of current interest for broad distribution. These services would be initiated on an experimental basis, and the demand for them monitored carefully. Commitments to phone answering and conducting searches would have to be adjusted according to demand and availability of resources.

The Information Panel will continue to study the feasibility of providing computer access to technology assessment information. Computer access may entail use of central information bases and distribution of the Directory, its updates, and related products via optical or magnetic disks. The panel has determined that appending a technology assessment information base to an existing online system would be preferable to initiating a new stand-alone system. The panel will take account of what is learned from the response to the Directory and other current efforts, consult with the NLM and other information base authorities, and plan accordingly.

Providing computer access to the technology assessment information base will be furthered by implementing other of the Information Panel's plans. Together, the Directory and access guides will account for much of the gray literature and other online and traditional sources that could be integrated in an online system. Augmenting DIRLINE and MeSH will help to make NLM's MEDLARS more useful to users concerned with technology assessment. Response to the Directory, which represents a hard-copy analog of certain aspects of an online system, and access guides will inform the Information Panel about the desirable content and format of assessment information compilations. Clearinghouse services would enhance the council's ability to meet, and learn more about, user needs.

NLM's review of the Directory, described below, and other council efforts will provide insight into the extent to which an online technology assessment information base may be linked with the MEDLARS system. One venue for such a link would be to append the Directory information base to the HEALTH (HEALTH Planning and Administration) file of MEDLARS, which covers citations on nonclinical aspects of health care delivery such as facilities planning and administration, manpower, insurance, financial management, regulation, and quality assurance. HEALTH is compiled primarily by the American Hospital Association using selections from MEDLINE and augmented by a variety of sources not indexed by NLM. Thus, HEALTH stands as a model of an online base formed from published literature sources as well as certain gray literature sources.