Informationist and Expert Searcher Critical New Old Roles

Informationist and Expert Searcher: Critical New (Old) Roles for Health Sciences Librarianship? Gary D. Byrd, Ph. D. University at Buffalo (SUNY)

An Outline for this Talk ► Patient-Centered Librarianship: from CMLs to Informationists ► Expert Searching: from Intermediaries to Educators to Search Consultants ► Knowledge Management: from IAIMS to Decision Support ► Research Questions ► Training Implications

Patient-Centered Librarianship ►Evolving Paradigms: § Clinical Medical Librarians (Clinical Librarians) § Measuring the Impact of Hospital Library Search Services on Patient Care § Informationists

CML Services ► Integrate library resources and expertise into the clinical, patient care setting ► Overcome time, cost and expertise barriers ► Adjunct members of patient care teams ► Improve librarians’ understanding of patient care context of questions, information needs ► Potential to better anticipate information needs

CML Service Roles ► Providing research assistance for clinical faculty ► Providing requested articles & bibliographies ► Selecting, summarizing, abstracting articles based on observed (anticipated) needs ► Educating students, residents, others on team ► Providing information to patients & families ► Promoting use of traditional library services

Factors in CML Successes & Failures ► Acceptance of librarian as team member ► Medical, clinical knowledge of librarian ► Librarian’s willingness to assume CML role ► Frequency of team requests and service uses ► Costs of personnel and other resources ► Budget sources available to support services

CML Evaluative Studies ► Only 35 from 1974 to 2001 § Mostly descriptive (5 used controls) § Usually single, active programs (5 more general) § Most used actual program data (1 secondary, 1 simulated) § Data collection methods: use statistics (20), questionnaires (13), surveys (10), interviews (9) § Service aspects studied: effect on users (30), program functions (21), program development (12), costs (8), library effects (2), need (1) § Impacts studied: patient care (31), educational (21), library services (11), research (4)

Results of CML Studies ► Atypical studies: § Positive recollection of previous services (1977) § Programs discontinued due to budget and staff shortages (1980) § Review of reported benefits & problems (1974 -84) § Dept. chairs & librarians mildly receptive to CML concept (1990) ► Single, active program studies: § Weighted average--perceptions of positive impact : ►<600 users & <400 “uses” included in these studies ►Usefulness, quality of information (12 studies) 89% ►Patient care impact (12 studies) 65%

Hospital Libraries & Patient Care ► Studies of general impact of hospital library services on quality and costs of patient care § King (1986): random, unobtrusive survey of physicians & nurses in Chicago-area hospitals; 74% would handle case differently with library-provided information § Marshall (1990 -91): Rochester hospitals study; 80% would handle case differently § Klein (1989 -90): Search services impact on Detroit hospital charges and length of stay; “early” MEDLINE searches lowered costs and patient bed days

The “Informationist” Concept ► Davidoff & Florance (2000) Annals editorial § Proposed new, credentialed professional role § Combine knowledge, training and skills of librarian, biostatistician, computer scientist and medical professional § Evidence-based medicine mandate to bring best evidence to point of care § Institute of Medicine reports (1999, 2001) mandate to reduce medical errors

CML Redux, or Something New? ► Giuse (1997 -98) Vanderbilt Univ. CML experience § Prefigures the informationist movement § “Future of medical librarianship is in the clinical realm” § Lack of adequate CML preparation § Need to “assimilate the culture” § Ability to interact on rounds, search effectively, and interpret the literature

The Critical Informationist Difference (With thanks to Scott Plutchak, 2000 BMLA) ► More than helping role of librarian at bedside (ambivalence of librarians and physicians) ► Move information management to mainstream of clinical practice ► Clinic driven & funded, not library driven & funded ► Nationally recognized standard curriculum ► Still librarian, but hybrid steeped in the clinic ► Not concerned with survival of librarianship, but the survival of patients

Expert Searching ►Evolving Paradigms: § Search intermediary § Educator and trainer of end-user searchers § Expert search collaborator, consultant

Mediated Search Services ►Combination of knowledge (2003 MLA Policy Statement) ►Knowledge & skills areas: § Subject domain, discipline, or practice § Databases content, structure, & effective use § Retrieval system capabilities, limitations

Mediated Search Services (cont. ) ► Skills needed: § Clarify, refine & understand context of information needs § Find and effectively use information in all formats § Recognize personal and institutional limitations § Apply retrieval system logic, syntax and weighting § Be mindful and reflective (use iterative and heuristic methods) § Use deductive and inductive reasoning § Efficiently evaluate results to fit requestor’s needs & expectations § Expertly process and edit results to facilitate work of requestor § Thoroughly document search process

Evolution of Mediated Searching (with thanks to Catherine Smith, JMLA 2004) ► ► The impact of MEDLARS and MEDLINE Traditional reference, but with “machines” § Intermediary role “born of batch processing” § Separation of expert searcher and customer § Mediation between sophisticated inquirer and idiot computer NLM training to certify expertise & library school courses AAHSL mediated database search statistics, up to 1992 Time-shared, network access made search dialogs possible Systems to encourage “end-user” searching “Expert” concept shift (ca 1980): from mediation to just searching ► Implication: anyone can learn to be an expert searcher ► ► ►

Librarians as Search Educators ► NLM and Library Schools Trained Librarians ► More sophisticated and user-friendly search system interfaces ► NLM policy shift to encourage end-user searching ► AAHSL user education statistics, 1985 ff. ► Ubiquitous, easily accessible search systems ► Explosion in number of online searches ► Perception: librarian search expertise not needed

Renewed Needs for Librarian Searching Expertise ► Collaboration and consultation: § To conduct systematic reviews of the literature for best evidence § To confirm complex research designs (e. g. , for clinical trials) § To support basic research and grant proposals § To prevent medical errors (e. g, Johns Hopkins research volunteer death) § To overcome the illusion of the all-inclusive & comprehensive Web

Knowledge Management ►Evolving Paradigms: § Integrated information management systems § Network infrastructure for seamless access § Semantic and syntactic integration § Decision support

Integrated Information Management ► IAIMS (starting with Matheson Report, 1982) § Librarian leadership potential § Strategic planning for integrated advanced information management systems Evolution from infrastructure and organizational issues to binding knowledge to effective action ► Continuing barriers and problems: ► § Marking quality & authority in complex information spaces § Superficial integration (hyperlinks) vs. true integration (syntax and semantics) § Facilitating information flows across patient care, research and education missions

Knowledge Management ► The core mission of hospitals and academic health sciences centers ► Stewardship over the life cycle of health sciences knowledge ► Support for the creation, storage, manipulation, dissemination & use of data, information and knowledge ► An institution’s own knowledge store plus links to external knowledge

Binding Knowledge to Effective Action ► The end goal of knowledge management ► Support for effective decision-making § Decision support ► Make all relevant current, accurate, authoritative health information immediately useful and usable ► Support for care providers, researchers, administrators, patients, and public

The Decision Support Challenge ► Overcoming limitations human decision-making § We tend to be risk adverse § We make decisions to satisfy needs rather than maximize utility § Our decisions determined by opportunity, availability, uncertainty about consequences, personal preferences § We all have “cognitive limitations” § Our choices are constrained by stress, time pressure, limited resources

Information Characteristics Needed to Compensate for These Limitations ► Risk adverse § Authoritative ► Satisfy need (vs. maximize utility) § Relevance, timeliness, tied to context, optimized to work goals ► Determined by opportunity, availability, uncertainty about consequences, personal preferences § Accessible, current, known limitations, personalized ► Cognitive limitations § Clear, readable, succinct, non-redundant, focused ► Constrained by stress, time pressure, limited resources § Easily and quickly located (used, understood), inexpensive

The best strategies or tools to bind knowledge to effective action? ► Informationist expertise in decision making settings? ► Expert searching expertise? ► IAIMS infrastructures and resources? ► New information management technology applications? ► Others, or a combination?

Two world views. . . ► Health Sciences Librarianship (Librarians) § The service paradigm § CML, Informationist, Expert Searcher ► Medical Informatics (Mostly Physicians) § The information technology paradigm § IAIMS Leader, Decision Support Applications Developer

Medical Informatics is Strategically Focused ► An ambitious research agenda § Improving patient care via the electronic health record ►A growing number of doctoral training programs ► National leadership and funding § National Library of Medicine § Office of National Health Information Technology (David J. Brailer, MD, Ph. D)

The Challenge for Health Sciences Librarianship ►We need-- § A more focused research agenda ►Evidence-based librarianship § A clear vision for librarians’ role in the knowledge management process § A strategy for educating and training the next generation of librarian knowledge managers

Research Strategies ► Learn from and use the research results of others (mine the literature!) ► Collaborate with faculty and graduate students § Health sciences (clinical and basic sciences) § Medical informatics § Library science and informatics ► Study and report on the impact of our services and resources

Some Research Questions ► CMLs and Informationists § In what settings can they be most effective? ►In-patient, out-patient, research labs, clinical trials … § What information needs can be most effectively anticipated? § Best ratio of librarian to clinicians or researchers? § Time needed to best evaluate a program or service? § What is the cost-benefit ratio for these services?

More Research Questions ► Expert Searching § How important is recall and precision for meeting different information needs? § What characteristics of the search consultation communication process lead to successful results? § Can we measure the additional value provided by librarian search experts serving on research teams?

And Some More Questions ► Knowledge Management § Can we use metadata standards to effectively indicate the quality and authority of information resources in databases? § In what settings and situations can decision support be best provided by an information management expert (librarian) rather than being embedded in an IT application?

What is Our Vision? ► Univ. at Buffalo School of Informatics

What knowledge and skills are needed to achieve our vision? ► Patient-centered librarians? § CMLs, Informationists ► Expert searchers? § Intermediaries, educators/trainers, consultants ► Knowledge managers? § Information system integrators, decision supporters ► All the above?

Pharmacy Education as a Model ► Pharm. Ds now often serve as drug “informationists” in clinical settings ► Major transition in education and practice roles ► Deliberate process over past 30 years ► Intriguing parallels with evolution in our thinking about patient-centered librarianship

Pharmacists ► Professional training § Five-year baccalaureate degree ► Work environment § Drug store or hospital pharmacy ► Practice roles § Dispensing drugs and advice ► Philosophy § Provision of useful drug products and services

Doctors of Pharmacy ► Professional training § Six-year Pharm. D with clinical experience ► Work environment § Hospital wards, clinics, HMOs, clinical research teams ► Practice roles § Provide drug therapies for individuals & populations ► Philosophy § The provision of pharmaceutical care

Health Sciences Librarians ► Professional training § Masters degree, often with other training in health sciences ► Work environment § Hospital or academic health sciences library ► Practice roles § Develop collections, facilitate use, retrieval for individual requestors ► Philosophy § Provision of health information resources & services

Health Informationists? ► Professional training § Six-year clinical information doctorate with clinical experience ► Work environment § Hospital wards, clinics, HMOs, clinical research teams ► Practice roles § Provide knowledge management services for other health professionals and patients ► Philosophy § Provision of health information care

Other Potential Training Models ►Additional specialized graduate training beyond the professional degree ►MD/Ph. D or MD/JD models ►Certification of specialized knowledge and skills ►Fellowship training in clinical settings

Conclusions ► Growing recognition of the value of information and knowledge ► Many competing professions interested in facilitating knowledge management § Medical informatics § Pharmaceutical sciences § Management sciences ►A challenge & opportunity for us!

Thank you Questions?