Medical librarians adopted embedded librarianship decades ago, but until recently, they haven’t used the term embedded. The rich literature that documents their efforts uses terms such as clinical librarian, informationist, information specialist in context, and, perhaps borrowed from the higher education sector, liaison. But, until recently, hardly ever embedded. Still, based on the working definition of the embedded librarian as an “integral part to the whole,” a librarian who works closely with a group of information users, develops close working relationships, and delivers high-value, customized information to meet their needs—these medical librarians are indeed embedded. This chapter traces the development of embedded librarianship in the fields of healthcare education and delivery and explores the diversity of contexts and roles in which medical embedded librarians have been engaged. It reviews the lists of best practices that medical librarians have recommended to their colleagues and surveys the work done over decades to evaluate the contributions of embedded librarians in medical education and practice. The chapter closes with some thoughts on the future of embedded librarianship in the health sciences.
The germ of the idea for embedded librarianship in medicine can be traced back to the 1960s, when the practice of medicine began to be seen as a team activity, not the domain of a solo physician acting alone. Early in the decade, the report “Lifetime Learning for Physicians” (Dryer, 1962), also known as the Dryer Report, documented the gulf between what was known in the literature and what was being applied in medical practice. Also during this decade, the roles of pharmacists, social workers, nurses, nutritionists, and psychologists in clinical care began to receive recognition. In particular, there arose the idea of the clinical pharmacist, who accompanied the medical doctor in visiting patients and contributed specialized knowledge to treatment decisions.
Taking her cue from this trend, Dr. Gertrude Lamb, director of the library at the School of Medicine, University of Missouri–Kansas City, applied for and received a grant to initiate a “clinical medical librarian” program, which ran from 1972 to 1975. Dr. Lamb spoke about her initiative at the Medical Library Association conference in 1972. In the succeeding years, she and others replicated the model of clinical medical librarianship in other medical schools and hospitals (Cimpl, 1985).
The primary goals of these programs were to provide information to the healthcare teams quickly and to influence the healthcare professionals to make more and better use of the literature to inform their decision making and actions. The programs generally involved the librarian in rounding, or the process of doctors and other health professionals visiting patients, reviewing their statuses, and discussing diagnostic, treatment, and care questions. Given the technology of the time, the librarian harvested questions from these discussions and then returned to the library to conduct research and to identify and obtain relevant literature. The librarian then provided copies of key documents or summaries of findings to the appropriate members of the healthcare team. In addition, the clinical medical librarians sometimes provided information directly to patients. The program at McMaster University was one in which the clinical medical librarian devoted a significant proportion of time to supplying patient information (Marshall and Neufeld, 1981).
Clinical medical librarianship continued to develop into the 1990s, albeit somewhat slowly, and studies documented its beneficial effects. Still, the problem of applying scientific knowledge in the teaching of health sciences and in medical practice continued to grow. The term evidence-based medicine came into use to express the importance of applying evidence from the literature and was expanded into evidence-based healthcare to incorporate the recognition that nurses, pharmacists, and all healthcare professionals shared the need to use the literature effectively in their work. Ultimately, the term evidence-based practice emerged. Perry and Kronenfeld (2005) have identified these essential steps of the evidence-based model:
• Recognizing and formulating an answerable question
• Locating relevant “knowledge artifacts”
• Critically evaluating them
• Synthesizing an appropriate evidence-based answer from them
• Applying the answer in a collaborative way with the patient
The idea of evidence-based practice according to these steps was attractive, but it wasn’t so easy to follow in real life. Critical obstacles included the never-ending time pressures on medical professionals, the tendency of doctors (like most of us) to seek information from other people rather than from documents and databases, and lack of skill in using databases of medical literature (Coumou and Meijman, 2006). By the late 1990s, clinical librarianship had gained recognition at some institutions but was not in wide use. Librarians at institutions such as Vanderbilt University had gained considerable experience. They participated in rounds, capturing explicit questions and also listening for the implied questions that came up as the clinical team discussed the patients. The librarians recognized the importance of expertise and ongoing education in both the specialized subject matter of the clinicians and the use of information retrieval systems. They had conducted some evaluations that provided some validation of the worth of their efforts (Giuse, et al., 1998).
Meanwhile, the need to broaden and systemize the response to the knowledge gap in healthcare remained, and visionaries in the medical library community recognized that more needed to be done. Writing in 1997, Giuse articulated a vision of the clinical librarian as an “equal voice with other specialists who support clinical decision-making,” who “project themselves not as information ‘servers’ who trail the team in an auxiliary capacity, but as an integral part of the group with a specialized expertise that can contribute vitally to clinical situations” (p. 437). The key to achieving this status, she wrote, was to establish a trusted relationship with other members of the team, and the key to establishing trust was to be competent and well-prepared. Giuse advocated that the clinical librarian have a strong “medical knowledge base” in addition to mentored practice in “searching, retrieving, filtering, and summarizing information.” Soon, a new proposal would crystallize her recommendations.
The new proposal came in the form of an Annals of Internal Medicine editorial by Frank Davidoff and Valerie Florance (2000). They proposed a new kind of healthcare professional, with a new title: informationist.
Davidoff and Florance (2000) took as their starting point the ongoing gap between research and the practice of medicine. “The disappointing reality,” they said, “is that physicians still don’t regularly search the medical literature themselves, nor do they ask for professional help in searching nearly as often as they need to” (p. 996). Reviewing the 30-year history of clinical medical librarianship, they found that such programs are both efficient and effective, and they advocated that informationists become as widespread in medical care as “head nurses or office managers” (p. 997). The informationist would meet the information needs of every member of the clinical care team, and of patients as well.
Informationists would combine a knowledge of both information science and clinical work, and would be the product of an accredited education and professional certification. They would report to clinical management, and their services would be paid for directly, just as other specialists on the clinical team. Just as attending physicians rely on a range of technical specialists for chemical lab tests, computed tomography scans, and the like, they—and the rest of the clinical team—would rely on informationists to ensure that the best available findings from the literature were informing all aspects of care.
The informationist proposal has fundamentally influenced the field ever since its publication. In an early reaction, Plutchak (2000) recognized its radical nature: “This informationist is a true hybrid—still a librarian, but one steeped in the clinic in a substantially new way” (p. 392). In 2002, the Medical Library Association and the National Library of Medicine held a conference with the aim of developing a consensus definition and an action program to spread the implementation of the informationist concept (Shipman, et al., 2002). Participants wrestled with the issues of finding appropriate recruits; developing the necessary educational and training programs; funding informationist services; marketing the concept to medical administrators; finding opportunities to test and implement it; and assessing its value.
In the ensuing years, the concepts of clinical medical librarian and informationist continued to evolve, even as they spread. The Medical Library Association began to use the formulation information specialist in context for the new embedded-librarian role. Writing in 2005, Perry and Kronenfeld proposed a set of expanded roles for librarians in the areas of “engagement at the institutional level in setting the knowledge-based resource use agenda” (p. 14), and specifically the following:
• Building knowledge systems
• Training staff in evidence-based practice
• Building client-centric library services
• Delivering “highly customized, answer-focused mediated knowledge-based search services” (p. 14)
• Collaborating with information technology staff to integrate knowledge and institutional information systems
In a review of the literature, Rankin, Grefsheim, and Canto (2008) found reports from informationists at 15 different institutions, some represented by more than one article, and suggested that two distinct variations of the concept had emerged—the clinical informationist and the bioinformatics informationist. Significant differences between the two stem from their evolution. The clinical informationist began with an emphasis on customized service and the development of a team relationship, found opportunities to deliver credible information, and expanded the depth and sophistication of contributions to the team’s work. The bioinformatics informationist, on the other hand, emphasized strong technical skills at the outset and developed customized contributions and strong working relationships over time.
The Medical Library Association has continued to explore and develop the concept ever since, with a special program at its 2010 annual conference (the tenth anniversary of Davidoff and Florance’s editorial), the establishment of an interest group, and many other programs.
Throughout the history of embedded librarianship in healthcare and health sciences—from clinical medical librarian to informationist to information specialist in context—medical librarians take responsibility for providing information to patients, to clinical care teams, and to researchers. They also lead in information literacy instruction at all levels of education, from undergraduate programs in allied healthcare professions, to graduate and doctoral programs, to continuing education for medical professionals. While a few accounts focus exclusively on one of these roles, embedded librarians at most institutions wear a variety of hats.
Some of the earliest accounts of embedded medical librarians emphasize information delivery to patients. McMaster University in Toronto was a leader in this area. By 1975, Joanne Marshall and others were participating in hospital rounds—as prescribed in Lamb’s pioneering work—and preparing information packets to help patients understand the nature of their illness and the treatment they were receiving (Cimpl, 1985; Kates, 1978; Roseman, 1978). The clinical librarian served all members of the healthcare community; patient requests accounted for about a quarter (24 percent) of all requests fielded (Marshall and Neufeld, 1981).
Other accounts have focused on research and information analysis for the different professional specialists that make up clinical teams.
At a nonteaching hospital in Harrisburg, Pennsylvania, librarians established a role in the program for medical residents. They provided real-time research and answers to clinical questions during one “morning report” session per week. They found that the program improved the residents’ own use of the literature, increased demands for librarians to collaborate, and led to better, more-collegial relationships between residents and the clinical librarians (Coldsmith and Schwing, 2005). Studies of research services provided by clinical librarians and informationists to all medical staff in Brighton, England, and Dublin, Ireland, found good support for the value of the research and its impact on care, as well as a demand for closer collaboration of informationists with the clinical teams (Brookman, et al., 2006; Flynn and McGuinness, 2011).
Taking a different approach, librarians at the University of Sheffield, England, initiated a clinical research service aimed specifically at meeting the needs of the nursing staff. The service was intended to enable the nurses to make better use of evidence from the literature in their patient care. It was the practice for nurses to hold “handover meetings” every afternoon. At these meetings, nurses departing from the day shift would discuss issues with nurses coming in for the evening shift. The librarian arranged to attend these meetings once a week. By participating in these conversations, the librarian would be able to identify information needs and either provide relevant evidence from the literature immediately or follow up later (Tod, et al., 2007).
Institutions such as teaching hospitals, as well as undergraduate and graduate programs in health sciences, have teaching missions. In these institutions, clinical librarianship initiatives have emphasized the development of embedded information literacy instruction. At Virginia Commonwealth University in Richmond, Virginia, a professional health sciences librarian was embedded in the School of Medicine, given associate professor rank, and assigned as director of the school’s computer-based instruction lab, reporting to the dean, not the library director. The position included responsibility for embedded medical informatics instruction, which was carried out through collaboration with both subject faculty and library staff (Seago, 2004). Similarly, the Faculty of Medicine at Imperial College London established a Medicine Information Literacy Group, which embedded information literacy instruction in the college’s curriculum. Information literacy instruction was coordinated with the college’s Problem Based Learning curriculum and included such topics as plagiarism and critical appraisal of evidence, in addition to search and retrieval skills (Cousins and Perris, 2009).
In some cases, embedded information literacy instruction in the health sciences is similar to its counterpart in other subjects in higher education, as described in Chapter 3. Ferris State University, in Big Rapids, Michigan, offers online courses in its master’s of nursing and healthcare systems administration degrees, and the health sciences librarian is embedded in certain courses in these programs, participating in the courses through the course management system (Konieczny, 2010). At Southeastern Louisiana University, a pilot project in the Foundations of Advanced Nursing course expanded into embedded instruction for a master’s of science in nursing program offered by a consortium of three universities (Guillot, Stahr, and Meeker, 2010).
Published accounts of their work by several programs of clinical medical librarianship over a number of years provide some insights into the programs’ success and progress. The programs include those at the University of Florida Health Science Center, the U.S. National Institutes of Health (NIH), and the Vanderbilt University Eskind Biomedical Library.
The University of Florida Health Science Center refers to its program as a liaison librarian program. Begun in 1998, the program sought to establish closer communications and better services with six different colleges that made up the Health Science Center (Tennant, et al., 2001). The liaison plan identified seven areas of focus:
• Communication
• Collection development
• Education
• User services
• Information access
• Liaison development
• Program evaluation
These areas bear a strong resemblance to the more-innovative roles of liaison librarians articulated by academic librarians during this time, as discussed in Chapter 3.
By 2006, some, if not all, of the liaison librarians had significantly outperformed expectations. Two positions, one in nursing and one in a new Genetics Institute, had been cofunded by the respective information user groups, and the librarians were given office space colocated with the staff of these units, away from the library. By the time of an evaluation conducted in 2006, there were nine liaison librarians to the various colleges. The evaluation showed a wide variation in the levels of collaboration achieved by the liaisons: Three had published with subject faculty, two had given presentations at scholarly or professional meetings with subject faculty, one had been given a joint appointment in an academic department, and one had been asked to co-teach a course (Tennant, Cataldo, and Jesano, 2006). The article didn’t make clear whether it was the same liaison who was asked to co-teach, received a joint appointment, and was one of those who had co-presented and co-authored. The possibility exists that one or two of the liaisons were very successful and became strongly embedded or whether the successes reported were spread out among the liaison librarians.
In 2008, Tennant, the bioinformatics librarian, and Miyamoto, a professor in the Department of Zoology, published a case study documenting the collaboration of librarian and subject researcher on a specific research project (Tennant and Miyamoto, 2008). It provided a unique insight into the research process from the librarian and subject faculty perspectives, showed the nature of collaboration between them, and documented the contribution of the embedded librarian to the research outcome.
A 2009 update on the status of the University of Florida program documented both successes and challenges. There has been significant staff turnover, and vacant positions have been eliminated in response to deep budget reductions. However, it appears that highly embedded librarians remain in place in some areas. The update discusses the struggles of the library organization over the varying degrees of embeddedness. The solution has been to define four tiers of service, which establish a framework for discussions with information user groups and library management. At tier one, the roles appear very similar to those of traditional academic liaison librarians discussed in Chapter 3—collection development, communication, and little more. At tier two, there are elements of relationship building and collaborating with subject faculty on teaching and other activities. At tiers three and four, the duties are much more in line with the examples of strong embedded relationships discussed in Chapters 1 and 3, including engagement in instruction and the research process (Ferree, et al., 2009).
The NIH established its informationist program and hired its first informationist in 2001. The library director laid the groundwork by meeting with senior managers of different units within the organization. Senior staff in the information user groups were enlisted to provide visible endorsement, mentor the new informationists, and enable them to learn about the work of their groups. Training plans were developed to ensure that the informationists would have the level of scientific knowledge needed to be effective members of the clinical and research teams. Over time, the responsibilities of informationists evolved from traditional library tasks to “critical appraisal and summarizing and synthesizing the literature, … help with manuscript preparation, co-authoring articles and help with data analysis software” (Whitmore, Grefsheim, and Rankin, 2008, p. 138). By 2008, the program had grown to 14 informationists, working with more than 40 information user groups in 16 different organizational units.
A preliminary analysis of evaluations conducted in 2004 and 2006 indicated that the presence of an informationist was associated with greater use of the available literature and better performance in finding answers to clinical and research questions. The evaluation summarized user reaction as “strong acceptance and uptake …, growing demand for a wider variety of services, robust return on investment and unassailable satisfaction and loyalty ratings” (Whitmore, Grefsheim, and Rankin, 2008, p. 139).
Further documentation, published in 2010, noted that the NIH informationist program had increased to 15 staff members. Analyzing and comparing the 2004 and 2006 evaluations, the report found that collaboration with the informationists had become more widespread over time, as word of their ability and value spread among the researchers and clinicians. The presence of an informationist in a clinical or research group led to more-effective use of the literature and an increase in information user self-sufficiency in information retrieval. This doesn’t mean that the informationists worked themselves out of a job, however. Instead, it was found that they took on a variety of technical information management and informatics-related tasks (Grefsheim, et al., 2010).
Perhaps the most prolific and influential contributors to the literature over a long period of sustained success are the management and staff of the Eskind Biomedical Library at Vanderbilt University. As noted earlier in this chapter, Giuse, then deputy director of the Eskind Biomedical Library, anticipated Davidoff and Florance with her 1997 call for clinical medical librarians to assume a coequal role with other healthcare professionals (Giuse, 1997).
Clinical medical librarianship was initiated at Eskind in 1996 with one librarian participating in rounds in the Medical Intensive Care Unit. The initial focus was on retrieving, filtering, and summarizing articles in response to questions that arose during the rounds. Initially, the clinical librarians did not charge for their work, so as not to be under pressure to demonstrate an immediate return on investment. The librarians recorded each clinical question they worked on in a database so that they could demonstrate the work that they had performed. In 1997, a 10-question opinion survey of the professionals with whom the clinical librarians collaborated found strong support for the librarians’ work (Giuse, et al., 1998).
By 2004, the Eskind Biomedical Library had taken its model further. In 2001, the parent Medical Center implemented an electronic medical records system, called StarPanel. Recognizing the value of aligning its work with the new electronic workflow, the Library established in 2004 a capability to receive clinical questions via the secure messaging function of StarPanel. In response to these questions, a clinical librarian would prepare an information packet that retrieved, filtered, highlighted, and summarized the relevant literature. The final product would contain the librarian’s summary, highlighted copies of all articles summarized, and a disclaimer noting that the summary was not intended to substitute for clinical judgment. The service was further expanded by incorporating links to standard guidelines into the StarPanel system and by extending the service to questions related to inpatient care practices. Yet none of these innovations replaced or eliminated the ongoing relationship of the informationist with other members of the clinical team. The authors noted that “librarians make regular ‘drop-in’ visits to the clinic to maintain visibility and continue to develop trust as well as collegiality with the clinicians” (Giuse, et al., 2005, p. 252).
An update published in 2010 (Giuse, Williams, and Giuse, 2010) demonstrated that the Eskind Biomedical Library continued to develop its informationist model. The update highlighted the following aspects of the program’s integration and progression:
• Clinical Informatics Consult Service (CICS): Participate in rounds and provide evidence packets to clinical staff in response to specific patient-related needs.
• Patient Informatics Consult Service: Provide information packets to patients.
• StarPanel: Receive and respond to clinical questions using the Medical Center’s electronic patient record system, thus integrating the information specialists into the mainstream of clinical workflow.
• Maintain and update order sets (“point-of-care actions … usually grouped around a specific procedure [for example, asthma control]”) within the electronic system (p. 221).
• Participate in the design and content creation for a new patient web portal, designed to “promote patients as more proactive partners in their care management and increase patient-provider communication” (p. 222).
Giuse, Williams, and Giuse concluded by observing that during the decade-plus history of clinical medical librarianship at Eskind Biomedical Library, the librarians have been “fully integrated …, thus allowing a true value-added scalable approach to the provision of evidence” and “an integral partnership of skills and competencies” (p. 222).
In other words, we might say they have become very highly embedded in the clinical enterprise.
Education is an important, ongoing theme in the literature of embedded medical librarianship. Authors have held repeatedly that a substantial knowledge of medical science, in addition to a background in library and information science, is a requirement for success in the field. The emphasis on this point is much stronger in the medical library literature than in the literature of embedded librarianship in other sectors. Several articles have focused exclusively on the issue of education and training for clinical medical librarians and informationists.
Davidoff and Florance (2000) set the terms of this discussion in their manifesto, “The Informationist: A New Health Profession?” They called for informationists to have “a clear and solid understanding of both information science and the essentials of clinical work” (p. 997) and recognized the need for two tracks for the education of informationists: one for those who begin as librarians and the other for those who begin as clinical professionals.
Others soon elaborated on these recommendations. Detlefsen (2002) proposed five models for a comprehensive approach to the education of informationists:
• American Library Association accredited degree programs
• Training in universities and medical informatics centers not associated with American Library Association accredited programs
• Short courses and continuing education opportunities offered by academic health science centers
• Continuing education opportunities for librarianship, which might be offered by the Medical Library Association
• Distance education in both medical informatics and librarianship
Meanwhile, Hersh (2002), a faculty member at the Oregon Health and Science University, articulated a model of informationist training entirely outside librarianship. The model offered courses in medical informatics, health and medicine, computer science, and quantitative methods. Hersh argued that both library science and separate informatics educational programs were valid pathways for the development of informationists and could coexist in the healthcare field.
Vanderbilt University, continuing its leadership, developed an in-house training program for librarians without a background in biology, in order to equip them with bioinformatics knowledge (Lyon, et al., 2004). In the NIH informationist program (Robison, 2008), all have degrees in library and information science, while about half have prior education or experience in science or medicine, and all have worked in medical or science libraries. At NIH, the continuing education program includes the following elements:
• General courses in clinical research and biology
• Specialized courses in medicine and science relevant to the work of the group in which the informationist is embedded
• Learning by regular participation in lectures, conferences, rounds, and lab meetings
The American Medical Informatics Association website lists medical informatics programs both affiliated with schools of Library and Information Science and located in medical sciences institutes.
Beyond the issue of educational background, there are important questions regarding other competencies and organizational and management practices for initiating and sustaining the informationist or “information specialist in context” model.
Looking at the librarian’s instructional role, Seago (2004) and Konieczny (2010) reached similar conclusions. In embedding information literacy instruction in the medical school curriculum, Seago emphasized the importance of having the support of the course director and incorporating a graded informatics assignment in one or more courses. Konieczny elaborated on the importance of the librarian’s sharing with the subject instructor a clear understanding of their respective roles; having regular, reliable availability in course management systems; and meeting face-to-face with students if possible, or interacting with them via videoconference. The key is the quality of the relationships between librarian and instructor and between librarian and students.
Regarding relationship building and the development of trust in the clinical environment, Coldsmith and Schwing (2005) noted that success of an embedded clinical medical librarian in a residency program requires the commitment of all parties—the program director, the residency staff, and the librarian. The goal is that the residents “become comfortable” with librarians. In other words, they must develop a strong working relationship so that the residents will trust the librarians and rely on their information retrieval and analysis work products in making decisions. Brookman, et al. (2006) emphasized that trust is built over time through sustained visibility and ongoing commitment. They also noted that clinicians are eager for librarians to do more analysis—but only if the clinicians know and trust the librarians.
On the basis of an extensive review, Rankin, Grefsheim, and Canto (2008) recommended that to succeed, librarians must reverse a longstanding professional trend toward generalization and become specialists instead, with subject expertise relevant to the work of the information user groups in which they are embedded. Further, they found that an embedded relationship is necessary to “achieve credibility, acceptance, and sustainability,” whereas an “impersonal information service provided at a distance” is less likely to achieve success (p. 195). They also reported that programs funded as part of the library’s budget are more stable than those relying on funds from other sources.
Rankin, Grefsheim, and Canto (2008) also identified a range of enabling factors for embedded informationist programs. Enabling factors are organizational, programmatic, and service provider–related. Organizational factors include executive support and a supportive culture and environment. Programmatic factors include marketing; visibility; delivery of a high-quality, sophisticated work product; and evaluation or feedback. Service provider factors include characteristics such as a high level of professional competence and interpersonal skills to form strong, trusted working relationships. These required competencies are summarized in Figure 4.1.
Informationist Competencies
Personal
Communication
Professionalism
Lifelong learning
Quality assurance
Proactivity
Leadership
Customer service
Entrepreneurialism
Functional
Locating information
Critical appraisal
Information synthesis
Informaiton management & organization
Project management
Knowledge management
Education
Research
Applied informatics
Current awareness
Knowledge
Domain
Information environment
Research design & analysis
Technology
Organizational
Related disciplines
Health policy & regulations
Figure 4.1 Informationist Competencies [Source: Rankin, J. A., Grefsheim, S. F., & Canto, C. C. (2008). The emerging informationist specialty: A systematic review of the literature. Journal of the Medical Library Association, 96(3), 199. Used by permission.]
The experience of the Arizona Health Sciences Library illustrates the application of these practices. Seizing the opportunity created by construction of a new multidisciplinary research building, the library director won senior administrator support to provide space for embedded librarians. When the building opened, four librarians, including several with prior specialized subject education, were placed in it part-time. They continued to be funded by the library budget. Their former duties at the main library reference desk were covered by merging the reference and circulation desks and training paraprofessional employees to handle most types of requests.
Within a year, the library deployed an additional librarian to work with the College of Pharmacy and extended the program to the College of Nursing. Librarians were regular attendees at faculty meetings. Librarians were being written into grant applications, embedded information literacy instruction was up 40 percent, and librarians had come to be viewed as “partners who can contribute significantly to the mission of the colleges and the university” (Freiburger and Kramer, 2009, p. 141).
In reviewing the program’s success, Freiburger and Kramer echoed the factors highlighted by Rankin, et al.:
• Organizational support: Information user-group commitment in the form of support from a dean or associate dean of each college
• Programmatic support: Sustained effort on the part of library management, with provision of equipment, freeing embedded librarians from other duties, and ongoing promotion through visits to faculty meetings and class sessions, special events, and other outreach efforts
• Service provider skills: The combination of domain education, information competencies, and interpersonal skills of the embedded librarians
The literature on embedded librarianship in the health sciences is distinguished by its emphasis on evaluation. From the early days of clinical medical librarianship to the present, medical librarians have recognized the need to demonstrate that the embedded librarian can and does have a real impact on the behavior of clinical staff, the use of the medical literature, and ultimately the quality of medical care.
Adopting the methodology of the medical field, Marshall and Neufeld (1981) performed a randomized trial field experiment, assigning four healthcare teams to the study group and four to a control group. A clinical librarian was assigned to each team in the study group and worked with the team for 6 months. At that time, interviews were conducted with the members of all eight groups, and a second round of interviews was conducted 3 months later.
Marshall and Neufeld also collected data about the work of the clinical librarians. They found that the librarians interacted with virtually everyone in the clinical environment. They received almost 600 requests: 24 percent from patients, 40 percent from physicians, 21 percent from allied health professionals, and 15 percent from nurses. Of the 444 searches done for health professionals (not patients), more than 25 percent (113) were information needs recognized by the clinical librarians—not direct requests made by the healthcare team. In other words, the librarians often recognized needs not articulated by the other members of the team.
Additional data showed that the study group members made much greater use of library resources overall than the control group members did. For example, the study groups accounted for more than 15 times the number of online search requests by the control groups. This difference in library use was found to persist, but at lower levels, after the clinical librarians were withdrawn (Marshall and Neufeld, 1981).
Marshall and Neufeld also found that comments made in the interviews showed the impact and value of the clinical librarians:
• Some 70 percent of the study group members found the librarian “very helpful.”
• Many staff, especially nurses and allied health professionals, said they used library services for the first time after the clinical librarian program began.
• Physicians noted that they became aware of many questions that were raised during rounds but not followed up.
• And 83 percent of study group members said that the librarian had provided useful information that they would not have found otherwise.
Two-thirds of the study group members said that their information-seeking habits were changed by their experience with the clinical librarians. Additional responses showed that library resources became their highest-rated information sources, superseding colleagues and their own books and journals overall. The authors conclude that “clinical librarians serving health professionals, patients, and families can be successful” and “there are significant changes in information-seeking patterns among health professionals who receive the services of a clinical librarian” (p. 416).
Kay Cimpl Wagner (formerly Kay Cimpl) has done the field a great service by completing two reviews of the literature almost 20 years apart. Her first review (Cimpl, 1985) summarized eight programs, all at university medical centers in North America. She noted a wide range of documented benefits, including enhanced patient care, educational benefits for medical students and healthcare teams, time savings, greater exposure to the literature, and greater information sharing among colleagues.
Cimpl (1985) also noted a few objections. Some can be attributed to lack of skill and knowledge on the part of the medical librarian—misunderstanding of questions and delivery of unsolicited, irrelevant information. In one case, the service was discontinued because it seemed to be competing with traditional library reference service. She also found cost to be a major obstacle. Most of the programs she reviewed were funded by the library, not by the clinical programs. Where the costs were deemed prohibitive, she suggested that the calculation failed to take into account the various benefits of the program to the clinical staff or did not involve a comprehensive cost-effectiveness evaluation.
Kay Cimpl Wagner’s second review, co-authored with Gary Byrd (2004), identified 35 papers that reported a formal evaluation, by either quantitative or qualitative methods. It included papers from Nigeria and the U.K., in addition to Canada and the U.S. Wagner and Byrd noted that only four of the 35 studies used before-and-after evaluation, or comparison control groups. In the end, their detailed analysis found equivocal conclusions:
There is some relatively strong evidence that these programs have been well accepted and liked by most of the targeted clinicians and students served. However, the total amount of such evidence is not great, most of it is descriptive rather than comparative or analytically qualitative, and it does not rise to the level of the “best evidence” called for to support evidence-based medicine or librarianship. (pp. 30–31)
Medical librarians have continued to study and report the value of their embedded librarianship programs. A review by Weightman and Williamson (2005) examined the reported impacts of both traditional library services and clinical librarians on healthcare. Analyzing 28 studies, the authors found evidence that both traditional library services and clinical medical librarians have an impact on healthcare. “There is also evidence, notably from the clinical librarian studies, of time savings to healthcare professionals and cost-benefits” (p. 17). However, they qualified their findings with concerns over the size and quality of the research studies.
Among the more recent evaluations of embedded librarianship in medical settings, Schwing and Coldsmith (2005) found that, after maintaining their program for 2 years, 58 percent of the hospital residents surveyed said that the answers provided by the librarian influenced patient care “much” or “a great deal” (p. 36). However, only 40 percent said that they asked for the librarian’s help “frequently.”
In a British study, Brookman, et al. (2006) evaluated specific information searches that the clinical librarians conducted. The most common reason for requesting a search was patient care, and 50 percent of these searches resulted in a change in practice. Similarly, searches conducted for other purposes were highly rated, and overall, 85 percent of searches received a rating of 4 or 5 on a 5-point scale (with 5 being the most positive rating).
A team at Vanderbilt University conducted a randomized trial of its CICS in the intensive care unit during 19 months spanning 2004 to 2006 (Mulvaney, et al., 2008). During this period, all requests to the clinical informationists were randomly assigned to either receive a response or not. (Clinicians were able to opt out of the randomization process, and about 10 percent of the cases in the study were “opt-outs.”) The Vanderbilt team studied results of more than 200 requests for information and found that the CICS had a statistically significant, positive impact on the use of evidence in diagnosis, care, and treatment. Furthermore, clinicians who requested information from the CICS were also more likely to perform searches themselves, suggesting a synergistic effect between the role of the embedded informationist and information seeking by the clinical staff.
Throughout its development, embedded librarians in the health sciences have exemplified traits in common with embedded librarians in other sectors: a strong emphasis on building relationships with other medical professionals, becoming knowledgeable in the subject domain, and performing at a high level of sophistication and skill to add value to the mission of the parent organization. Health sciences librarians have recognized the importance of continuous change since the early days of the development of their clinical role, and the process of change continues today.
The idea that the medical library not only should change, but must change, has become a fixture in predictions about the role of libraries and librarians in healthcare. Already in 1995, Nina Matheson, director emerita of the Welch Medical Library at Johns Hopkins University, predicted that “[t]he time is coming … when libraries must choose between owning traditional resources and making them available; developing and owning new digital knowledge resources and making them accessible; or being an organization that knows where knowledge is and teaches how to mine it” (pp. 4–5). Just 2 years later, Giuse (1997) sharpened the point by adding, “We have no choice but to migrate into the clinical setting; to avoid doing so is to deny our future in the information age” (p. 437; italics in the original).
A few years later, Donald Lindberg and Betsy Humphreys (2005), director and deputy director, respectively, of the National Library of Medicine, laid out their own vision of the future in the New England Journal of Medicine. Avoiding Matheson’s stark choice, they offered inferences for the year 2015:
More librarians and information specialists are deployed “in context” outside the library to improve quality, to reduce the risks associated with inefficient or incomplete retrieval of the available evidence, and to do community outreach. Many librarians have advanced training in both subject-matter disciplines and information science. It is common to find librarians working as part of health care teams, writing grant proposals, serving on institutional review boards, working as bioinformatics database specialists within science departments, serving as faculty members in evidence-based medicine courses, and being involved in multilingual health-literacy programs and community partnerships. (para. 9)
In other words, embedded librarianship is a key element of their vision.
Circling back to Johns Hopkins, in 2010, Nancy Roderer, successor to Matheson as director of the Welch Medical Library, articulated a specific, near-term plan that fulfills Matheson’s vision. In accord with a clinician’s observation that “if the materials are going to be coming to me, then the librarians should too” (Kho, 2011, p. 1), the Welch Medical Library will distribute its librarians to the clinical and research teams as embedded informationists and reuse its building for other than traditional library purposes (Kolowich, 2010).
It’s clear that some 40 years after the first medical librarian joined clinicians on their rounds, embedded librarianship is not just alive and well, but a key element of the future of healthcare information.
This chapter showed how embedded librarianship in the healthcare field arose in the 1970s to meet a need. That need was to better utilize evidence from the literature in the diagnosis and treatment of disease and in the care of patients. Initial programs had some success, and early studies were able to document the fact that the programs resulted in more-extensive use of literature.
Along the way, the clinical medical librarians recognized that the problem was greater than simple access to articles—it also involved having the time to read the literature and synthesize the results. Accordingly, leaders in the field called for a new role, initially called informationist and later sometimes information specialist in context—a professional with both expertise in information retrieval and a sophisticated understanding of the clinical environment, able to identify, retrieve, read, and summarize the literature in order to provide direct input into clinical questions and to teach health sciences students how to use the available knowledge effectively for themselves. These services have become well established in many institutions, although we don’t know how many, as there is no comprehensive census.
Meanwhile, other developments created an even greater need for embedded medical librarians. Paradoxically, as more and more health science literature became available in digital form anytime, anywhere, through mobile communication devices, the sheer increase in its volume and the proliferation of interfaces, tools, and digital library resources increased the complexity of the research process. Greater emphasis was placed on information literacy skills, and at the same time the value of the traditional medical library with its emphasis on collecting paper copies of information resources was called into question.
American Medical Informatics Association. n.d. Academic informatics programs. Available from www.amia.org/education/programs-and-courses. (Accessed April 4, 2012.)
Brookman, A., A. Lovell, F. Henwood, and J. Lehmann. 2006. What do clinicians want from us? An evaluation of Brighton and Sussex University Hospitals NHS Trust Clinical Librarian Service and its implications for developing future working patterns. Health Information and Libraries Journal 23 (December): 10–21.
Cimpl, K. 1985. Clinical medical librarianship: A review of the literature. Bulletin of the Medical Library Association 73 (1): 21–28.
Coumou, H. C. H., and F. J. Meijman. 2006. How do primary care physicians seek answers to clinical questions. Journal of the Medical Library Association 94 (1): 55–60.
Cousins, J., and K. Perris. 2009. Supporting research at the faculty of medicine: The development of Imperial College London’s Medicine Information Literacy Group. Journal of Information Literacy 3 (1): 60–67.
Davidoff, F., and V. Florance. 2000. The informationist: A new health profession? Annals of Internal Medicine 132 (12): 996–998.
Detlefsen, E. G. 2002. The education of informationists, from the perspective of a library and information sciences educator. Journal of the American Medical Library Association 90 (1): 59–67.
Dryer, B. V. 1962. Lifetime learning for physicians: Principles, practices, proposals. Journal of Medical Education 37 (6): 1–134.
Ferree, N., N. Schaefer, L. C. Butson, and M. R. Tennant. 2009. Liaison librarian tiers: Levels of service. Journal of the Medical Library Association 97 (2): 145–148.
Flynn, M. G., and C. McGuinness. 2011. Hospital clinicians’ information behaviour and attitude towards the ‘clinical informationist’: An Irish survey. Health Information and Libraries Journal 28 (1): 23–32.
Freiburger, G., and S. Kramer. 2009. Embedded librarians: One library’s model for decentralized service. Journal of the Medical Library Association 97 (2): 139–142.
Giuse, N. B. 1997. Advancing the practice of clinical medical librarianship. Bulletin of the Medical Library Association 85 (4): 437–438.
Giuse, N. B., A. M. Williams, and D. A. Giuse. 2010. Integrating best evidence into patient care: A process facilitated by a seamless integration with informatics tools. Journal of the Medical Library Association 98 (3): 220–222.
Giuse, N. B., S. R. Kafantaris, M. D. Miller, K. S. Wilder, S. L. Martin, N. A. Sathe, and J. D. Campbell. 1998. Clinical medical librarianship: The Vanderbilt experience. Bulletin of the Medical Library Association 86 (3): 412–416.
Giuse, N. B., T. Y. Koonce, R. N. Jerome, M. Cahall, N. A. Sathe, and A. Williams. 2005. Evolution of a mature clinical informationist model. Journal of the American Medical Informatics Association 12 (3): 249–255.
Grefsheim, S. F., S. C. Whitmore, B. A. Rapp, J. A. Rankin, R. R. Robison, and C. C. Canto. 2010. The informationist: Building evidence for an emerging health profession. Journal of the Medical Library Association 98 (2): 147–156.
Guillot, L., B. Stahr, and B. J. Meeker. 2010. Nursing faculty collaborate with embedded librarians to serve online graduate students in a consortium setting. Journal of Library and Information Services in Distance Learning 1 (1/2): 53–62.
Hersh, W. 2002. Medical informatics education: An alternative pathway for training informationists. Journal of the Medical Information Association 90 (1): 76–79.
Kates, J. 1978. What’s obstetrics coming to? Ask the women at McMaster: At McMaster University, giving birth is a remarkable experience. Globe and Mail, Canada, September 13.
Kho, N. D. 2011. Embedded librarianship: Building relational roles. Information Today 28 (3): 1–36.
Kolowich, S. 2010. Embedded librarians: Johns Hopkins ahead of curve. USA Today, June 9. Available from www.usatoday.com/news/education/2010-06-09-IHE-embedded-librarians-Johns-Hopkins09_ST_N.htm. (Accessed April 4, 2012.)
Konieczny, A. 2010. Experiences as an embedded librarian in online courses. Medical Reference Services Quarterly 29 (1): 47–57.
Lindberg, D. A. B., and B. L. Humphreys. 2005. 2015—the future of medical libraries. New England Journal of Medicine 325 (11): 1067–1070.
Lyon, J., N. B. Giuse, A. Williams, T. Y. Koonce, and R. Walden. 2004. A model for training the new bioinformationist. Journal of the Medical Library Association 92 (2): 188–195.
Marshall, J. G., and V. R. Neufeld. 1981. A randomized trial of librarian educational participation in clinical settings. Journal of Medical Education 56: 409–416.
Matheson, N. W. 1995. The idea of the library in the twenty-first century. Bulletin of the Medical Library Association 83 (1): 1–7.
Mulvaney, S. A., L. Bickman, N. B. Giuse, E. W. Lambert, N. A. Sathe, and R. N. Jerome. 2008. A randomized effectiveness trial of a clinical informatics consult service: Impact on evidence-based decision-making and knowledge implementation. Journal of the American Medical Informatics Association 15 (2): 201–211.
Perry, G. J., and M. R. Kronenfeld. 2005. Evidence-based practice: A new paradigm brings new opportunities for health sciences librarians. Medical Reference Services Quarterly 24 (4): 1–16.
Plutchak, T. S. 2000. Informationists and librarians. Bulletin of the Medical Library Association 88 (4): 391–392.
Rankin, J. A., S. F. Grefsheim, and C. C. Canto. 2008. The emerging informationist specialty: A systematic review of the literature. Journal of the Medical Library Association 96 (3): 194–206.
Robison, R. R. 2008. Informationist education. Medical Reference Services Quarterly 27 (3): 339–347.
Roseman, E. 1978. The consumer game: Some hospitals listening to patients’ complaints. Globe and Mail, Canada, July 31.
Schwing, L. J., and E. E. Coldsmith. 2005. Librarians as hidden gems in a clinical team. Medical Reference Services Quarterly 24 (1): 29–39.
Seago, B. L. 2004. School of medicine CBIL librarian: An educational informationist model. Reference Services Review 32 (1): 35–39.
Shipman, J. P., D. J. Cunningham, R. Holst, and L. A. Watson. 2002. The informationist conference: Report. Journal of the Medical Librarians Association 90 (4): 458–464.
Tennant, M. R., and M. M. Miyamoto. 2008. The role of the medical librarian in the basic biological sciences: A case study in virology and evolution. Journal of the Medical Library Association 96 (4): 290–298.
Tennant, M. R., L. C. Butson, M. E. Rezeau, P. J. Tucker, M. E. Boyle, and G. Clayton. 2001. Customizing for clients: Developing a library liaison program from need to plan. Bulletin of the Medical Library Association 89 (1): 8–20.
Tennant, M. R., T. T. Cataldo, P. Sherwill-Navarro, and R. Jesano. 2006. Evaluation of a liaison librarian program: Client and liaison perspectives. Journal of the Medical Library Association 94 (4): 402–409.
Tod, A. M., B. Bond, N. Leonard, I. J. Gilsen, and S. Palfreyman. 2007. Exploring the contribution of the clinical librarian to facilitating evidence-based nursing. Journal of Clinical Nursing 16 (4): 621–629.
Wagner, K. C., and G. D. Byrd. 2004. Evaluating the effectiveness of clinical medical librarian programs: A systematic review of the literature. Journal of the Medical Library Association 92 (1): 14–33.
Weightman, A. L., and J. M. Williamson. 2005. The value and impact of information provided through library services for patient care: A systematic review. Health Information and Libraries Journal 22: 4–25.
Whitmore, S. C., S. F. Grefsheim, and J. A. Rankin. 2008. Informationist programme in support of biomedical research: A programme description and preliminary findings of an evaluation. Health Information and Libraries Journal 25: 135–141.