SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Clinical librarianship (CL), currently receiving renewed interest world-wide, seeks to provide quality-filtered information to health professionals at the point of need to support clinical decision-making. This review builds upon the work of Cimpl (Bulletin of the Medical Library Association 1985, 73, 21–8) and attempts to establish the evidence base for CL. The objectives were to determine, from the literature, whether CL services are used by clinicians, have an effect on patient care, and/or clinicians’ use of literature in practice and/or are cost-effective. The methodology used was a systematic review of the literature, following, where possible, the NHS Centre for Reviews and Dissemination (CRD) framework. Modifications to this methodology included the resources searched, and the critical appraisal checklist (CriSTAL) used. Two hundred and eighty-four unique references were retrieved. Seventeen (16 unique) evaluative and a further 33 descriptive studies met the inclusion criteria. The quality of reporting of the literature was generally poor. CL programmes appear to be well-used and received by clinicians. However, there is insufficient evidence available on their effect on patient care, clinicians’ use of literature in practice, and their cost-effectiveness, thus highlighting the need for further high-quality research.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Background

Clinical librarianship (CL), also referred to as clinical medical librarianship (CML), seeks to integrate information professionals into health care teams, mainly through their attendance at ward rounds and/or meetings in the clinical setting. CL programmes therefore aim to support clinical decision-making and/or education by providing timely, quality-filtered information to clinicians at the point of need. Such activities attempt to promote evidence-based health care, with the ultimate goal of improving patient care, as well as enhancing clinicians’ use of research literature and knowledge of library and information resources.

Literature Attached to the Chart (LATCH) programmes1,2 are often regarded as the predecessors of CL. The concept of CL was originally conceived by Dr Gertrude Lamb in 1971 at the University of Missouri, Kansas City, in response to an unmet need for clinical questions to be answered from the research literature on ward rounds.3 Lamb recognized librarians’ expertise in accessing information, and hence the concept of CL was born.4,5 According to Cimpl, clinical pharmacists were perhaps the role models for the first CLs.6 Lamb, and subsequently Algermissen, were instrumental in obtaining support, in particular from the National Library for Medicine, for several CL initiatives in the US in the 1970s.7,8

Since then a number of such programmes emerged with many having been documented in a narrative review by Kay Cimpl in 1985.6 From this review, a common definition of the purpose of CL emerged:

to provide information quickly to physicians and other members of the healthcare team; to influence the information seeking behaviour of clinicians and to improve their library skills; and to establish the medical librarian's role as a valid member of the health care team.6

The recurrent themes of CL as intermediary, educator and disseminator are evident in other definitions, such as that of Veenstra and Gluck:9

… to provide quality filtered case specific information to the physician in support of clinical decision making.

Lamb10 readdressed CL as a system with a number of elements. One of these elements involved defining the goals of CL programmes: to improve patient care by providing answers to patient care questions and by teaching health professionals to interact with literature, as well as to improve continuing education to health professionals.

Cimpl6 selectively listed and described 23 CL programmes involving both hospital and academic libraries in North America. Hospital librarians generally provided CL services in addition to their usual library duties. They spent, on average, 3 hours a week on rounds and 4 hours retrieving relevant literature. Dissemination was via topical bibliographies and distribution of pertinent articles. Additional roles for CLs included the development of a Patient Care Related Reading Program (PCRRP) which involved the routine delivery of articles on a pre-selected subject.8

In the 1980s, widespread financial constraints, particularly in the US, resulted in the demise of many CL initiatives.11 However, support for CL has remained high.12 Twenty-nine US-based CL programmes were identified in 1993,13 and more recently extensive programmes have been evaluated at Vanderbilt University.14,15 Following early attempts in the 1970s at Guy's Hospital, London,16,17 the concept has also received renewed interest in the UK.18−21 Here, CL initiatives are at varying stages of development.20−23 March 2002 saw the first CL conference held in Leicester, England.24

Possible reasons for this increased interest include the agendas for evidence-based health care and clinical governance that have undoubtedly promoted the information professionals’ role in supporting clinical practice. By contrast in the US, Giuse warns that librarians may become obsolete if, as a profession, they do not ‘migrate into the clinical setting’.14 The literature suggests a number of benefits from CL programmes. CLs have been found to be time saving for clinicians,11,25 to influence patient care and practitioners’ knowledge,9,26 and to increase efficiency in evaluating recent literature.13 However CL services have also been purported to be expensive and labour intensive,1,25 thus increasing an imperative for such programmes to demonstrate a tangible impact.25

Cimpl recognized the importance of evaluating CL and LATCH programmes: to determine the quality of the service, to assess information delivery methods, to measure costs and to gain user feedback. Cimpl's review remains an important piece of secondary research, providing a clear overview of the development of CL in North America from the initiation of LATCH programmes in 1967 to CL programmes in the early 1980s. However, the methods used for this review are inadequately described and, by Cimpl's own admission, it was by no means comprehensive in its coverage of the literature. Furthermore the review was published almost 20 years ago. It is appropriate therefore that in this current climate of evidence-based librarianship27−28 a systematic review, described below, was undertaken.

Aims and objectives of this review

The overall aim of this review was to build upon Cimpl's (1985) review6 and to establish an evidence base for CL programmes. The specific objectives of this review were therefore to determine, from the literature, whether CL programmes:

  • 1
    Are used by clinicians.
  • 2
    Have an effect on patient care.
  • 3
    Have an impact on clinicians’ use of literature in practice.
  • 4
    Are cost-effective.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

A systematic review was undertaken over a 12-month period from May 2001 to May 2002. Where possible, the authors followed the review framework described in the NHS Centre for Reviews and Dissemination (CRD) Report Number 4, Undertaking Systematic Reviews of Research on Effectiveness.29

Search strategy

The search strategy employed multiple methods in an attempt to identify all the relevant literature in the field. Sensitive search strategies were employed in the major health (e.g. medline, embase), health-related (e.g. cinahl, British Nursing Index, amed, HealthSTAR), science (e.g. Science Citation Index), social science (e.g. Social Sciences Citation Index, assia), and information science (e.g. lisa, inspec) electronic bibliographic databases. Several ‘grey literature’ sources were searched, including Health Management Information Consortium (HMIC), Index to Theses, the National Research Register, Current Research in Britain (CRiB), and copac. In addition, the websites of relevant bodies were consulted, supplemented by a general Internet search using a meta-search engine (Copernic). Citation searches were conducted on key papers and authors (e.g. Cimpl6), and reference lists of relevant papers and several prominent bibliographies30−32 were checked to identify additional studies missed by the electronic searches. Finally, key journals in the field (e.g. the Bulletin of the Medical Library Association and Health Information and Libraries Journal, formerly Health Libraries Review) were hand searched. Searches were initially conducted in May 2001 with the major database searches being repeated in December 2001.

A combined free-text and thesaurus approach was adopted. Search terms included: clinical (medical/support) librarian*, clinical information librarian*/professional*/specialist*, informationist*, etc. No date or language limits were applied at the search stage.

A list of all sources searched, together with the search strategies used, is available on request from the authors.

Inclusion and exclusion criteria

Retrieved references were assessed against explicit pre-determined inclusion and exclusion criteria, as described below.

Types of study. As this review aimed to build upon Cimpl's review,6 studies were only included if they were published after 1982. In addition, due to time and resource constraints, only English language papers were considered for inclusion. The review aimed to identify all primary CL studies which included an evaluative research element, such as a survey. The investigators also identified additional ‘descriptive’ papers, i.e. studies that simply described CL initiatives.

Types of participants. For a study to be included in the review, information professionals had to provide an identifiable CL initiative to health professionals (or students). Health science librarians providing a general hospital library service, outreach librarians involved in education of remote health care professionals, similar initiatives aimed at patients, or those utilizing a health care (as opposed to an information) professional as the information provider, were excluded.

Types of intervention. The following definition of CL was adopted, ‘the provision of quality-filtered case-specific information directly to health professionals to support clinical decision-making’.9 This required the CL to attend clinical rounds and/or meetings and become a member of the clinical team (as opposed to the librarian providing a remote service via a hospital library, for example).

Types of outcome measure. Four types of outcome measures were considered:

  • 1
    General outcomes, particularly in terms of service usage (e.g. the number of literature searches conducted by the CL during a specified time period).
  • 2
    Patient care outcomes (e.g. whether the information provided by the CL aided treatment, diagnosis, etc.).
  • 3
    Clinicians’ use of the literature in practice, i.e. how the clinician used the information provided by the CL (e.g. whether the information was distributed to colleagues, etc.).
  • 4
    Cost-effectiveness (e.g. whether a cost-benefit analysis was undertaken).

Quality assessment strategy

Evaluative research studies were critically appraised using the CriSTAL (Critical Skills Training in Appraisal for Librarians) checklist for user studies.33 This checklist follows the format used by the Critical Appraisal Skills Programme (CASP)34 and comprises three components (validity, reliability and applicability) common to many appraisal tools.

Data extraction strategy

Key data were elicited from both the evaluative and descriptive literature using pre-determined data extraction forms. The data extraction procedure was performed independently by two reviewers, and discrepancies were resolved by discussion.

Assumptions, limitations and uncertainties

Although every attempt was made to identify all the relevant literature in the field, it is possible that local reports of CL programmes have been overlooked. The investigators are also aware of a number of ongoing CL programmes, particularly within the UK, which have yet to be published.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Quantity of research available

Figure 1 details the number of references at each stage in the review. Two hundred and eighty-four unique references were retrieved, of which 191 were ordered for a more detailed examination. Seventeen evaluative (including one duplicate study) and a further 33 descriptive studies met the inclusion criteria. Several programmes were reported more than once. A list of all the included studies is provided in Table 1. A list of excluded references, together with reasons for their exclusion, is available on request from the authors.

image

Figure 1. The number of references at each stage in the review. *This includes two references containing data for the same study, i.e. 16 unique studies. $None of these references appeared directly relevant from a title assessment

Download figure to PowerPoint

Table 1.  The 50 studies included in this review with evaluative studies shown in bold.
1Barbour, G. L. & Young, M. N. Morning report—role of the clinical librarian. Journal of the American Medical Association 1986, 255(14), 1921–2
2Bellamy, L. M. & Selig, S. A. Application of a general database manager in a clinical medical librarian program. Medical Reference Services Quarterly 1987, 6(1), 59–71
3Billick, P. & Bresien, P. The clinical librarian program at Saint Luke's Hospital, Cleveland. Clinical Librarian Quarterly 1983, 2(1), 16–8
4Corcoran, V. Clinical librarianship in nursing. Clinical Librarian Quarterly 1983, 2(2), 8–12
5DePres, K. A. & Bloom, V. D. CML programs: a deterrent to library use? Clinical Librarian Quarterly 1983, 1(3), 11–2
6Dodson, S. A. (2001) A clinical medical librarian program into the next millennium. Available from: http://healthlinks.washington.edu/hsl/liaisons/dodson/cml.html (accessed 30 October 2002)
7Eaton, E. K. Evaluation and model of a clinical librarian program [Dissertation]. University of Texas Medical Branch at Galveston, 1986
8Ekstrand, N. Clinical librarianship: alternatives and new directions. Clinical Librarian Quarterly 1985, 3(3), 5–12
9Ekstrand, N. L, Maynard, C. D. & Sprinkle, M. D. A comprehensive information service for an academic radiology department. American Journal of Roentgenology 1983, 141(5), 1077–80
10Faust, J. A CML program which was spoiled by its own success. Clinical Librarian Quarterly 1983, 1(3), 5–6
11Gilbert, C. M. Adapting clinical librarianship. Medical Reference Services Quarterly 1999, 18(1), 69–72
12Giuse, N. B., Kafantaris, S. R., Miller, M. D., Wilder, K. S., Martin, S. L., Sathe, N. A et al. Clinical medical librarianship: the Vanderbilt experience. Bulletin of the Medical Library Association 1998, 86(3), 412–6
13Glassington, L. (2001) The library and information support for clinical effectiveness (LISCE) project—eight months on. Available from: http://www.ucl.ac.uk/kmc/kmc2/News/ACKM/ackm4/glassington.html (accessed 30 October 2002)
14Glick, J. & Sullivan, M. CML in a satellite library. Clinical Librarian Quarterly 1984, 3(1/2), 5–9
15Halbrook, B. Clinical librarian programs reflections on successes and failures. Clinical Librarian Quarterly 1983, 2(1), 9–12
16Hamberg, C. An unexpected success: grand rounds bibliography. Clinical Librarian Quarterly 1985, 3(4), 10–2
17Hamberg, C. The librarian who knew too much. Clinical Librarian Quarterly 1985, 4(1/2), 8–9
18Haskell, D.A. The role of the clinical medical librarian in medical education. Clinical Librarian Quarterly 1984, 2(4), 6–9
19Hayden, R. A clinical librarian program for oncology nursing at Roswell Park Memorial Institute. Clinical Librarian Quarterly 1983, 2(1), 13–6
20Honeybourne, C. & Ward, L. Report of a six-month pilot project at Leicester General Hospital. Leicester: Leicester General Hospital, 2000
21Iruoje, R. The role of the clinical librarian in patient management. Nigerian Quarterly Journal of Hospital of Hospital Medicine 1999, 9(3), 215–7
22Jerome, R. N., Giuse, N. B., Wilder, K. S., Sathe, N. A. & Dietrich, M. S. Information needs of clinical teams: analysis of questions received by the Clinical Informatics Consult Service. Bulletin of the Medical Library Association 2001, 89(2), 177–84
23Kuller, A. B., Wessel, C. B., Ginn, D. S. & Martin, T. P. Quality filtering of the clinical literature by librarians and physicians. Bulletin of the Medical Library Association 1993, 81(1), 38–43
24Landau, L. The evolution of a clinical librarian. Clinical Librarian Quarterly 1984, 2(3), 7–8
25Lusher, A. Getting evidence to the bedside: the role of the clinical librarian. In: Bakker, S. (ed.) Libraries Without Walls: Changing Needs—Changing Roles, Proceedings of the 6th European Conference of Medical and Health Libraries, pp. 66–70, Dordrecht: Kluwer Academic Publishers, 1998
26Millard, S. K. Clinical librarian program for dental faculty. Clinical Librarian Quarterly 1984, 2(3), 9–11
27Miller, N. & Kaye, D. The experience of a department of medicine with a clinical medical library service. Journal of Medical Education 1985, 60(5), 367–73
28Miller, N. Effects of cost-sharing and end-user searching on a clinical medical librarian program. Bulletin of the Medical Library Association 1989, 77(1), 71–3
29Miller, N. Keeping the clinical librarian up to date. Clinical Librarian Quarterly 1983, 2(2), 1–4
30Miller, N. Teaching the clinical librarian program's users. Clinical Librarian Quarterly 1984, 2(3), 1–3
31Morley, S. K. & Buchanan, H. S. Clinical medical librarians: extending library resources to the clinical setting. Journal of Hospital Librarianship 2001, 1(3), 15–30
32Nichols, M. & Watts, A. On being new clinical librarians. Clinical Librarian Quarterly 1986, 4(3/4), 12–3
33Panzarella, M. A CML program for the nursing department. Clinical Librarian Quarterly 1985, 3(3), 1–3
34Pasquinelli, L. M., Buescher, E. S. & Gowen, C. W. Report of a survey: impact of a clinical medical librarian on resident education. Journal of Investigative Medicine 1998, 46(1), 2 A
35Reid, L. The impact of clinical governance on the library and information service: clinical librarian case study. IFMH Inform 2001, 12(1), 1–3
36Royal, M., Grizzle, W. E., Algermissen, V. & Mowry, R. W. The success of a clinical librarian program in an academic autopsy pathology service. American Journal of Clinical Pathology 1993, 99(5), 576–81
37Sagers, V. & Seeger, M. Medical clinical librarian: a special resource person. Nursing Management 1983, 9, 9–12
38Selig, S. A. & Graves, K. J. The clinical librarian program as an integral component of graduate medical education. Clinical Librarian Quarterly 1983, 1(4), 7–10
39Sullivan, M. B. & Sarkis, J. M. The clinical medical librarian program as perceived by the CML. Bulletin of the Medical Library Association 1987, 75(2), 169–71
40Tobia, R. C., Kronick, D. A. & Harris, G. D. A clinical information consultation service at a teaching hospital. Bulletin of the Medical Library Association 1983, 71(4), 396–9
41Tobia, R. C. Clinical librarianship at the University of Texas Health Science Center at San Antonio library. Clinical Librarian Quarterly 1984, 3(1/2), 1–4
42Turman, L. U., Koste, J. L., Horne, A. S. & Hoffman, C. E. A new role for the clinical librarian as educator. Medical Reference Services Quarterly 1997, 16(1), 15–23
43University of Missouri-Kansas City. (2001) Clinical medical librarian program. University of Missouri-Kansas City. Available from: http://www.umkc.edu/lib/hsl/cmlservices.pdf (accessed 30 October 2002)
44University of Vanderbilt (1997) Clinical librarianship at Eskind. Available from: http://www.mc.vanderbilt.edu/biolib/catalist/may97/librarianship.html (accessed 30 October 2002)
45Veenstra, R. J. & Gluck, E. H. A clinical librarian program in the intensive-care unit. Critical Care Medicine 1992, 20(7), 1038–42
46Veenstra, R. J. Clinical medical librarian impact on patient care—a one-year analysis. Bulletin of the Medical Library Association 1992, 80(1), 19–22
47Ward, L. M., Honeybourne, C. & Harrison, J. A clinical librarian can support clinical governance. British Journal of linical Governance 2001, 6(4), 248–51
48Waterman, E. Evidence-based medicine and the clinical medical librarian. National Network 1998, 22(4), 16
49Watson, J. A. & Weist, A. The Forest Healthcare clinical support librarian: 6 months on. Health Libraries Review 2000, (174), 219–21
50Yates-Imah, C., Goldschmidt, R. H. & Johnson, M. A. The clinical librarian: new team member for a Family Practice Inpatient Service. Family Medicine 1985, 17(6), 262–4

Quality of evaluative studies

Sixteen unique references met the criteria for inclusion in this review as research studies, and underwent the quality assessment described above. The majority of studies employed a survey design. On the whole, the quality of reporting of the individual studies was poor, making it difficult to make a valued assessment of the research undertaken. In particular, the sample size, sampling method and response rate(s) were rarely stated. Where the sample size and response rate were reported, these were often very small, thus casting doubt on the generalisability of findings. Other concerns about the methodology relate to the reliability and validity of the approach adopted; for example, reference was rarely made to tools used in other studies, but instead in-house questionnaires were developed. Copies of these questionnaires were generally not provided. In the majority of cases, the CLs were involved in evaluating their own programmes with authors failing to acknowledge the potential biases associated with doing so. It was decided that none of the studies should be excluded on the basis of their quality. However, the poor quality of reporting should be taken into consideration when interpreting the results of this review.

Description of included studies

Eighteen of the included studies were published within the last 10 years (1992–2002); 14 of these within the past 5 years (1997–2002). The majority of studies were published in journals. However, two studies were reported on websites, two as articles in newsletters, one as a conference proceeding, and one as a PhD thesis.

The majority of CL programmes were based in hospital settings within the US. The five examples from the UK covered both hospital18−21 and community settings.35 One programme was based in Nigeria.36 Most programmes served medical specialities, although five examples served multidisciplinary teams,19,21,35,37,38 four served nursing and/or allied health professionals,39−42 and one specifically served dentists.43

Descriptions of the CL programmes were very similar: librarians attended relevant meetings (e.g. morning report) and/or rounds, conducted literature searches and presented the results to the clinicians, usually in the form of a bibliography. In some programmes, the librarians were also responsible for selecting and obtaining the full-text of the most relevant articles. One service produced digests of the evidence.19 Three programmes delivered training or instruction to clinical staff35,44,45 and one developed a current awareness service.46

Of the 16 unique evaluative studies, 15 reported general results (e.g. relating to service usage), 11 reported results relating to patient care, 15 reported results relating to clinicians’ use of the literature in practice, and three reported results concerning the costs of CL programmes.

Results relating to general outcomes

The reporting of the CL programmes and the associated measures of service usage were inadequate, leading to confusion concerning the number of the CLs involved, the duration of the programmes, the samples studied, the different methods of evaluation and the level of service usage. This meant that it was not possible to pool the results quantitatively in a meaningful manner.

All the evaluated CL programmes served the medical profession, in particular house officers and attending physicians. Veenstra26 found that senior residents were most likely to use the service, junior residents the least. Only three studies discussed the reasons for clinicians accessing the CL service.13,47,48 The most frequently cited reason was patient related.47,48 Other reasons included: for presentations, continuing education, publication, research,47 and to provide general information on a subject.13,48

Only estimates of the average time taken by CLs to respond to requests are provided. The majority of programmes aimed for a same day turnaround time, with a range of 47 minutes9 to several days.19 Tobia et al.48 found that the average turnaround time was 0.9 h, with all respondents (n = 37) stating that this was adequate. The study found that the CL spent an average of 1.35 h per request.48 Three studies45,48,49 reported time savings associated with having a CL service. Tobia et al.48 estimated the average time saved to be 2.2 h.

Two papers9,15 indicated that the CL became an integrated member of the clinical team, but only Miller and Kaye47 and Eaton45 measured this as an outcome. Sixty-nine per cent (n = 61) of respondents thought that the CL had been accepted very well.47 However, Pasquinelli et al.50 noted that their project would have been enhanced had there been increased communication between the clinicians and the CL at the onset of the project to define goals, expectations and clinicians’ patient care needs.

Several studies examined clinicians’ attitudes towards the CL service.11,15,35,45,46 Eaton45 and Giuse15 for example, both found a positive effect. In addition, several studies reported comments made by CLs regarding their services. Turman et al.11 for example, found that they had to be flexible to accommodate the teams’ schedule and patient load. Ward et al.19 noted that the rapid rotation of clinical staff, especially junior doctors, meant that knowledge of the CL service was not sustained.

Four studies13,26,48,50 indicated the success of their respective CL programmes by reporting the number of clinicians who thought the service should be continued, or planned to continue using the service in the future. All respondents to the Tobia et al.48 (n = 37) and Royal et al.13 (n = 31) studies thought that the CL programme should be continued. Seventy-seven per cent (n = 30) of respondents to the Veenstra26 survey planned to use the CL service more often in the future.

Results relating to patient care

Five studies reported that the literature provided by the CL had an effect on patient care or management.15,19,26,45,50 Methods of reporting varied between studies and hence the results are not directly comparable. Sample sizes were also very small. The second most frequently reported outcome was the effect of information on diagnosis, with four studies reporting that the information had or would have an impact on diagnosis.9,26,49,51 Between 37%9 and 95%49 of respondents reported this as an outcome.

Three studies9,26,51 stated that information had had an impact on the choice of treatment or therapy with over half the respondents highlighting this outcome. Two studies26,51 reported that the information had had an impact on choice of tests. In addition, the information provided by the CL played a role in avoiding additional tests51 and/or cancelling unnecessary tests,9 clinical decision-making, choice of drugs, the length of stay, post-hospital care, patient advice, avoidance of mortality, surgery and/or infection.51

The information provided by the CL also had an impact on the knowledge of the recipients, in that it provided new information, gave a new insight, and/or updated previous knowledge,13 led to a better informed clinical team,45 and helped clinicians to gain a better understanding.9

Results relating to clinicians’ use of the literature in practice

Seven studies reported that the literature identified by the CL was relevant to the clinical situation and met the clinicians’ needs.9,15,36,48−51 Figures for the usefulness and relevance of the material identified ranged from 67% (n = 6)19−100% (n = 31).51 The presentation of the literature by the CL was rated highly in the Giuse et al.15 study, represented by 9.4 on a 1–10 Likert scale, where 10 was high. The majority of clinicians in the Tobia et al.48 study thought that the CL accurately pinpointed the subject in the literature, whilst other respondents stated that the CL expedited the information retrieved.45

Only six studies explored subsequent use of the literature identified via the CL service by the clinicians. Ward et al.19 provide the most comprehensive overview, with 64% (n = 11) of respondents using the information for teaching, 60% (n = 10) for case presentations, 45% (n = 11) for research, 30% (n = 10) for management and 27% (n = 11) for publications. A large proportion of clinicians26,48 were found to share the information with other members of their ward team.

From the literature it is also evident that the CL service had an impact on clinicians’ subsequent use of library resources. One study found that clinicians did not attend the library as frequently after the implementation of a CL service.45 This contrasts with the findings of Glick and Sullivan46 who found library usage to increase in a number of areas: reference requests (up 21%), delivery service (up 36%) and computer search requests (up 178%). Finally, one study found that clinicians demonstrated an increased awareness of the librarians’ role as an information specialist and educator and an enhanced awareness of information resources.11

Results relating to cost-effectiveness

Only three papers reported costs associated with a CL service.9,45,48 However, none of these studies attempted to undertake a cost-benefit analysis, and all have limited applicability; for example, failing to take into account recent advances in online searching, etc. Tobia et al.48 simply reported the medline charges accrued during the course of a 3-month evaluation, whereas Veenstra and Gluck9 and Eaton45 attempted to calculate the total costs involved in their CL programmes. Eaton45 calculated the total yearly costs of a CL service to be $57 383.70. This is based on the salary for two CLs and one clerical assistant ($46 000), the cost of photocopying materials and purchasing books ($4238), along with databases searches ($3000) and associated costs ($4147.50).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Are CL programmes used by clinicians?

The findings of this review certainly indicate that CL services are utilized by health care professionals within a clinical setting. In addition, individual programmes are well received by clinicians.

Setting and population. CL programmes were traditionally medically orientated, and this was reflected in this review. Notable exceptions to the setting for such programmes include emergency care and specialties (e.g. infectious diseases). More recently, programmes spanning primary and community care have emerged, particularly in the UK, and there will undoubtedly be further growth in these areas, particularly as a result of the latest reconfiguration of the NHS.52 In the current climate of joint working and collaboration,53 CL programmes will need to adopt a multi-professional approach. Indeed, in the UK, nursing staff represent the largest proportion of the clinical workforce, and hence it would be beneficial to extend CL services and develop allegiances with this group. Such diversification will require librarians to become familiar with a greater range of information sources, covering the allied health, social science and educational literature.

Purpose. The reasons for clinicians utilizing CL services is sparsely reported. This may be because researchers believe that such reasons are well recognized. However, this is a dangerous assumption to make. In order to develop responsive programmes, CL services must be based on the expressed needs of their users.54 It is already evident from the literature that not all enquiries to CL services were patient related. Many requests for information were for research, publication and continuing education. Not only is this likely to have an effect on the type of service provided by the librarian, but it also may determine sources of funding for such programmes.

Integration into the clinical team. An original aim of CL programmes was to integrate librarians into the clinical team through their attendance at rounds. Unfortunately, this aspect of CL programmes has been under researched, with only anecdotal evidence presented. For CL programmes to be seen as successful, librarians need to demonstrate their importance in enhancing patient care. Librarians need to overcome cultural barriers by advocating their role in supporting evidence-based practice in terms of their expert searching skills and ability to quality filter and present literature in a manageable format. It is evident from the literature that the use of ‘champion’ clinicians can help to foster interest and support for CL programmes.19

Various models of CL exist, ranging from librarians simply attending key meetings, to participation in all ward rounds and clinical meetings. Honeybourne and Ward22 found, for example, that clinicians preferred CLs to be present at audit and other meetings. It was thought that attendance at ward rounds was not time effective, yet regular contact in the clinical setting was essential as it raised awareness of the service and stimulated requests for information. Further research is required to establish the optimal approach, although it should be noted that this will depend on the needs and resources of individual organizations.

How can researchers measure librarians’ integration into the clinical team? Miller and Kaye47 specifically asked clinicians whether they thought that the librarian had been integrated into the team. However, such questioning does not really explore this issue in sufficient depth. One possibility would be for an external researcher to undertake some observational research, supplemented by in-depth interviews and/or focus groups with both clinicians and librarians. The feasibility of doing this within a busy clinical setting may prove problematic, but, where possible, this should be pursued.

Performance. All services should attempt to evaluate their performance in relation to pre-determined aims and objectives. The time taken for librarians to respond to requests is an important factor, particularly as many CL programmes advocate information being available at the point of need. However, the fact that a search takes two days to complete does not necessarily mean the value or quality of the service is reduced or that the results are obsolete. Instead it may reflect a more complex query.

The relevance of the literature identified by the CL and the extent to which it meets the clinicians’ needs is a key outcome. A librarian may be serving a diverse range of health care professionals, be fully integrated into the clinical team, and be rapidly responding to queries, but this is all superfluous if the information retrieved is irrelevant.

Determining time savings for clinicians is more problematic. This will inevitably depend on the skills and knowledge of individual clinicians and librarians. To date, only estimates of time savings have been reported by clinicians. However, is the primary aim of a CL programme to save clinicians’ time?

In order for services to evolve, it is important for librarians to receive and respond to feedback, both positive and negative, from their users.55 Although a number of studies included in this review specifically asked clinicians whether they thought that the CL service should be continued, this is only a superficial success indicator. There is the potential for the aim of such questions to be misinterpreted and for false-positive responses to be recorded. For example, a CL may be well liked and users may express a wish for the service to be continued, yet the service itself may be ineffective when analysed against other criteria.

Do CL programmes have an effect on patient care?

Clear instances of the librarian's positive impact in the clinical setting exist in the literature. However, the CL's impact on patient care is generally indirect, with many studies not reporting this as a main outcome measure. It is therefore difficult to draw firm generalisable conclusions. The majority of studies simply asked clinicians whether the information provided had an impact on patient care. Interestingly, the three studies that explored specific areas of impact on patient care in more detail all reported positive results.9,26,51 However, the design and reporting of all these studies were flawed; for example, by unclear reporting of the methods and results, small sample sizes, etc.

There is considerable variation in the terminology used in the literature; for example, several studies reported that the CL service had a positive effect on or improved patient care or management. ‘Positive effect’ and ‘improvement’, and ‘patient care’ and ‘patient management’ have been interpreted in this review as being synonymous. However, this may not be the case. The exact meaning of ‘impact on patient care’ also remains elusive. Impact may be interpreted as affecting a range of factors, such as diagnosis, course of treatment, length of stay, complication and infection rates, etc. A clear definition is required to enable researchers to obtain measurable, comparable outcome data.

The lack of firm evidence on this issue raises a number of questions, in particular, what are CL programmes really trying to achieve? How does the CL relate to Davidoff and Florance's proposed new health ‘informationist’?56 Is the ultimate aim of a CL programme to improve patient care? If so, why has this issue not been adequately explored in the literature to date? There is probably a simple answer to this: measuring a direct impact on patient care is extremely difficult, if not impossible, to do. However, if new CL programmes are to be funded and existing programmes to continue, tangible evidence will need to be sought.

Do CL programmes have an impact on clinicians’ use of the literature in practice?

A small body of evidence indicates that CLs do provide relevant and useful literature to clinicians. Librarians have generally presented this material to clinicians in the form of bibliographies, although there has been limited evaluation of this approach. It would be advantageous to determine whether it is effective, in terms of both time and cost, for librarians to provide a value-added service, for example, in terms of summaries or digests of the best available evidence.18

What is not clear from the literature is how clinicians use material located by librarians. A handful of studies suggest that clinicians use the material in teaching, publications, case presentations and management, but the extent to which the information is used to inform practice and is disseminated to other health care professionals (and patients?) is unclear. It is well documented57−59 that the implementation stage of the evidence-based practice cycle is perhaps the most problematic. How can we be sure that the librarian is not identifying all this literature only for the results to be filed away and never read, let alone implemented? Future evaluations of CL programmes should therefore attempt to track requests from the first initial contact with the service to delivery of literature to clinicians and subsequent implementation of its results.

Are CL programmes cost-effective?

It is not possible to ascertain from the existing literature whether CL programmes are cost-effective. In fact, the costs of such programmes are not even considered in the majority of published studies. This aspect of CL has clearly not advanced since Cimpl's review in 1985.6 With ever-increasing demands being placed on health budgets, CLs need to demonstrate their benefits quantitatively. A cost-benefit analysis of several CL services is required to establish whether the benefits outweigh the costs of setting up and running such a programme. As the majority of the benefits associated with a CL service are indirect, such as effects on patient care, it is important to involve health economists in this exercise.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Although it is widely accepted that CLs are effective, this review has identified little evidence to support this. CL programmes are generally well used and received by clinicians but there are only a few isolated instances of the beneficial impact CL services can have on patient care or clinicians’ use of the literature in practice. The review has also revealed that no studies have considered the cost-effectiveness of such programmes. However a lack of evidence does not necessarily equate to ineffectiveness. Instead, this review indicates the need for further high quality research in this area.

Implications for practice

The scope for the development of CL programmes is vast, but will inevitably be limited by financial, organizational and cultural constraints. The challenge for both new and existing services is for librarians, clinicians, managers and, possibly even patients, to demonstrate the value of such programmes in terms of clinical effectiveness and cost-effectiveness, and, in doing so, to expand the evidence base for this practice. In particular,

  • • 
    Programmes should be designed around the expressed needs of health care professionals. The whole ethos of CL is to improve clinical practice and care, and therefore such programmes need to meet the needs of their users.
  • • 
    The design and project management arrangements for CL programmes should be carefully considered at the start. Clear project aims and objectives must be set, and data collected and analysed accordingly.
  • • 
    Greater attention should be paid to demonstrating the unique impact CLs can have, particularly on patient care; for example, by tracking the progress of individual requests made to the CL service.
  • • 
    Finally, librarians should be encouraged to share their experiences within the professional arena; for example, by disseminating details of CL programmes in journal articles and at national and international conferences.24

Implications for research

Perhaps the most significant finding of this review is the need for further research, as indicated by gaps in the existing literature. Future research priorities should include:

  • • 
    Independent evaluation of CL programmes undertaken by external researchers.
  • • 
    Further exploration of the impact CL services have on patient care, involving the identification of tangible measures of success.
  • • 
    Investigation of the preferred format of information provided to clinicians by the CLs; for example, bibliographies, evidence-based digests, etc.
  • • 
    Determining the most effective role the librarian can play in the process, such as the extent to which CLs should become members of the clinical team and become involved in the implementation of research findings.
  • • 
    Establishing the cost-effectiveness of CL programmes.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

We would like to thank prominent authors in the field of clinical librarianship for exchanging their reference lists with us, namely Kay Cimpl, Gary Byrd and Linda Ward, as well as acknowledge the help and support provided by the Information Resources Section at ScHARR, in particular Andrew Booth.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  • 1
    Babish, J. M. & Warner, E. R. LATCH: initiating a program and addressing legal concerns. Medical Reference Services Quarterly 1983, 2(2), 1720.
  • 2
    Nippert, C. C. Online LATCH. Medical Reference Services Quarterly 1985, 4(1), 239.
  • 3
    Lamb, G. And now clinical librarians on rounds. Journal of the American Medical Association 1974, 230, 521.
  • 4
    Algermissen, V. Biomedical librarians in a patient care setting at the University of Missouri-Kansas City School of Medicine. Bulletin of the Medical Library Association 1974, 62, 3548.
  • 5
    Lamb, G. Clinical librarians in patient care/teaching settings: terminal progress report. Hartford, CT: Hartford Hospital, 1976.
  • 6
    Cimpl, K. Clinical medical librarianship—a review of the literature. Bulletin of the Medical Library Association 1985, 73, 218.
  • 7
    Greenberg, B., Battison, S., Kolisch, M. & Leredu, M. Evaluation of a clinical medical librarian program at the Yale Medical Library. Bulletin of the Medical Library Association 1978, 66, 31926.
  • 8
    Hutchinson, S., Malamud, J., Stearns, N. S. & Moulton, B. Preselecting literature for routine delivery to physicians in a community hospital-based patient care related reading program. Bulletin of the Medical Library Association 1981, 69, 2369.
  • 9
    Veenstra, R. J. & Gluck, E. H. A clinical librarian program in the intensive-care unit. Critical Care Medicine 1992, 20, 103842.
  • 10
    Lamb, G. Clinical librarianship. Clinical Librarian Quarterly 1984, 3(1/ 2), 1021.
  • 11
    Turman, L. U., Koste, J. L., Horne, A. S. & Hoffman, C. E. A new role for the clinical librarian as educator. Medical Reference Services Quarterly 1997, 16(1), 1523.
  • 12
    Schacher, L. F. Clinical librarianship: its value in medical care. Annals of Internal Medicine 2001, 134, 71720.
  • 13
    Royal, M., Grizzle, W. E., Algermissen, V. & Mowry, R. W. The success of a clinical librarian program in an academic autopsy pathology service. American Journal of Clinical Pathology 1993, 99, 57681.
  • 14
    Giuse, N. B. Advancing the practice of clinical medical librarianship. Bulletin of the Medical Library Association 1997, 85, 4378.
  • 15
    Giuse, N. B., Kafantaris, S. R., Miller, M. D., Wilder, K. S., Martin, S. L., Sathe, N. A., et al. Clinical medical librarianship: the Vanderbilt experience. Bulletin of the Medical Library Association 1998, 86, 4126.
  • 16
    Moore, A. The Clinical Librarian in the Department of Surgery: A Report on 2 Years’ Experience. London: Guy's Hospital Medical School, Wills Library, 1980.
  • 17
    Moore, A. The Clinical Librarian in the Department of Surgery. British Library Research and Development Report No. 5686. London: British Library, 1980.
  • 18
    Lusher, A. Getting evidence to the bedside: the role of the clinical librarian. In: Bakker, S., ed. Libraries Without Walls: Changing Needs—Changing Roles, Proceedings of the 6th European Conference of Medical and Health Libraries, pp. 6670. Dordrecht: Kluwer Academic Publishers, 1998.
  • 19
    Ward, L. M., Honeybourne, C. & Harrison, J. A clinical librarian can support clinical governance. British Journal of Clinical Governance 2001, 6, 24851.
  • 20
    Watson, J. A. & Weist, A. The Forest Healthcare clinical support librarian: 6 months on. Health Information & Libraries Journal 2000, 17, 21921.
  • 21
    Glassington, L. The library and information support for clinical effectiveness (LISCE) project—eight months on. 2001. Available from: http://www.ucl.ac.uk/kmc/kmc2/News/ACKM/ackm4/glassingotn.html (accessed 30 October 2002).
  • 22
    Honeybourne, C. & Ward, L. Report of a Six-Month Pilot Project at Leicester General Hospital. Leicester: Leicester General Hospital, 2000.
  • 23
    Reid, L., Ikkos, G. & Hopkins, W. Clinical librarians at Barnet Primary Care NHS Trust: addressing the information requirements of clinical governance. Health Information and Libraries Journal 2002, 19, 525.
  • 24
    University Hospitals of Leciester, NHS Trust. Clinical librarian conference. 2002. Available from: http://www.le.ac.uk/li/lgh/library/confer.htm (accessed 30 October 2002).
  • 25
    Makowski, G. Clinical medical librarianship: a role for the future. Bibliotheca Medica Canadiana 1994, 16(1), 713.
  • 26
    Veenstra, R. J. Clinical medical librarian impact on patient care—a one-year analysis. Bulletin of the Medical Library Association 1992, 80, 1922.
  • 27
    Booth, A. ‘Librarian heal thyself?’: evidence based librarianship, useful, practicable, desirable? Proceedings of the 8th International Congress on Medical Librarianship, 2–5 July 2000, London, UK. Available from: http://www.icml.org/tuesday/themes/booth.htm (accessed 30 October 2002).
  • 28
    Eldredge, J. D. Evidence-based librarianship: an overview. Bulletin of the Medical Library Association 2000, 88, 289302.
  • 29
    NHS Centre for Reviews and Dissemination (CRD). Undertaking Systematic Reviews of Research on Effectiveness. CRD’s. Guidance for Carrying Out or Commissioning Reviews. York: NHS Centre for Reviews and Dissemination, University of York, 2001.
  • 30
    Williams, M. D. A bibliography on clinical medical librarianship. Clinical Librarian Quarterly 1985, 4(1/ 2), 127.
  • 31
    Brown, H. A selected bibliography on clinical librarianship. Clinical Librarian Quarterly 1986, 4(3/ 4), 46.
  • 32
    Pfannenstiel, B. R. (1999) Clinical medicine librarian bibliography. Available from: http://research.med.umkc.edu/teams/cml/CMLbib.html (accessed 30 October 2002).
  • 33
    Booth, A. Research. Health Information & Libraries Journal 2000, 17, 2325.
  • 34
    Critical Appraisal Skills Programme (CASP). 2001. Available from: http://www.public-health.org.uk/casp/ (accessed 30 October 2002).
  • 35
    Reid, L. The impact of clinical governance on the library and information service: clinical librarian case study. IFMH Information 2001, 12(1), 13.
  • 36
    Iruoje, R. The role of the clincial librarian in patient management. Nigerian Quarterly Journal of Hospital of Hospital Medicine 1999, 9, 2157.
  • 37
    Sullivan, M. B. & Sarkis, J. M. The clinical medical librarian program as perceived by the CML. Bulletin of the Medical Library Association 1987, 75, 16971.
  • 38
    White, C. E. Clinical medical librarian skills applied in a non-patient care setting. Clinical Librarian Quarterly 1985, 3(4), 14.
  • 39
    Corcoran, V. Clinical librarianship in nursing. Clinical Librarian Quarterly 1983, 2(2), 812.
  • 40
    Hayden, R. A clinical librarian program for oncology nursing at Roswell Park Memorial Institute. Clinical Librarian Quarterly 1983, 2(1), 136.
  • 41
    Ekstrand, N. L., Ladner, B. H., Olds, W. W. & Thodeson, K. Cancer information for librarians: medical-subject-based continuing education. Bulletin of the Medical Library Association 1990, 78, 30711.
  • 42
    Gilbert, C. M. Adapting clinical librarianship. Medical Reference Services Quarterly 1999, 18(1), 6972.
  • 43
    Millard, S. K. Clinical librarian program for dental faculty. Clinical Librarian Quarterly 1984, 2(3), 911.
  • 44
    Yates-Imah, C., Goldschmidt, R. H. & Johnson, M. A. The clinical librarian: new team member for a family practice inpatient service. Family Medicine 1985, 17, 2624.
  • 45
    Eaton, E. K. Evaluation and Model of a Clinical Librarian Program. Dissertation. Galveston, TX: University of Texas Medical Branch at Galveston, 1986, .
  • 46
    Glick, J. & Sullivan, M. CML in a satellite library. Clinical Librarian Quarterly 1984, 3(1/ 2), 59.
  • 47
    Miller, N. & Kaye, D. The experience of a department of medicine with a clinical medical library service. Journal of Medical Education 1985, 60, 36773.
  • 48
    Tobia, R. C., Kronick, D. A. & Harris, G. D. A clinical information consultation service at a teaching hospital. Bulletin of the Medical Library Association 1983, 71, 3969.
  • 49
    Barbour, G. L. & Young, M. N. Morning report—role of the clinical librarian. Journal of the American Medical Association 1986, 255, 19212.
  • 50
    Pasquinelli, L. M., Buescher, E. S. & Gowen, C. W. Report of a survey: impact of a clinical medical librarian on resident education. Journal of Investigative Medicine 1998, 46(1), 2A.
  • 51
    Dodson, S. A. A clinical medical librarian program into the next millennium. 2001. Available from: http:// healthlinks.washington.edu/hsl/liaisons/dodson/cml.html (accessed 30 October 2002).
  • 52
    Department of Health. Shifting the Balance of Power. London: Department of Health, 2002. Available from: http://www.doh.gov.uk/shiftingthebalance/ (accessed 30 October 2002).
  • 53
    Department of Health. Partnership Working. National Service Frameworks: A Practical Aid to Implementation in Primary Care. London: Department of Health, 2002. Available from: http://www.doh.gov.uk/pricare/pdfs/nsf_partnershipworking.pdf (accessed 30 October 2002).
  • 54
    Booth, A. Identifying users’ needs. In: Booth, A. & Walton, , G., eds. Managing Knowledge in Health Services, pp. 10111, London: Library Association, 2002.
  • 55
    Banwell, L. Evaluating information services. In: Booth, A., & Walton, , G., eds. Managing Knowledge in Health Services., pp. 17381, London: Library Association, 2002.
  • 56
    Davidoff, F. & Florance, V. The informationist—a new heath profession? Annals of Internal Medicine 2001, 132, 9968.
  • 57
    Cranney, M., Warren, E., Barton, S., Gardner, K. & Walley, T. Why do GPs not implement evidence-based guidelines? A descriptive study. Family Practice 2001, 18, 35963.
  • 58
    Gross, P. A. Implementing evidence-based recommendations for health care: a roundtable comparing European and American experiences. Joint Commission Journal on Quality Improvement 2000, 26, 54753.
  • 59
    McInnes, E., Harvey, G., Duff, L., Fennessy, G., Seers, K. & Clark, E. Implementing evidence-based practice in clinical situations. Nursing Standard 2001, 15(41), 404.