Evidence-based librarianship

1
Evidence-based librarianship

• Jessie McGowan
• Senior Information Scientist, Institute of
  Population Health / Ottawa Health Research
  Institute Adjunct Professor, Departments of
  Medicine and Family Medicine, University of
  Ottawa Associate Editor, Journal of Clinical
  Epidemiology

2
Overview

• What is evidence-based stuff?
• What is evidence-based librarianship (EBL)?
• EBL examples
• Knowledge translation (KT) and EBL
• Question to ponder Do we give users what they
  want or what they need? How does the quality of
  resource effect collection development policies?

3
Evidence-based history

• Evidence-based philosophies integrate a health
  professionals experience and knowledge with the
  best currently available clinical evidence.
• It was introduced as EBM by ACP Journal Club in
  1991 and this editorial was the impetus for a a
  series called the Users Guides to the Medical
  Literature in order to help clinician decide how
  to incorporate these philosophies in to their
  daily practice.
• Guyatt GH. Evidence-based medicine (editorial).
  ACP Journal Club suppl 19912A-16.
• Guyatt GH. Users/ guides to the medical
  literature (editorial). JAMA 1993270(17)2096-7.

4
Evidence-based evolution
Evidence-based medicine
Evidence-based health care, practice, dentistry,
nursing etc.
Evidence-based medical / health librarianship
Evidence-based librarianship
5
Evidence-based approach

• 1. Formation of a clinical question (s)
  (usually from an existing clinical scenario). It
  is formed using the following formula of PICO(T)
• Patient/population
• Intervention
• Comparison
• Outcome
• Time

6
PICO example

• In a 41 year-old women with osteoporosis
  (P), does the use of etidronate (I) versus
  exercies (C) improve bone density and reduce the
  risk of fractures (O)?

7
Evidence-based approach

• 2. A literature search to identify the
  evidence
• Biomedical database searches, textbook searches,
  grey literature, hand searching etc.
• 3. Critical Appraisal of the literature search
  results
• 4. Application of the results to the patient
• 5. Quality assurance of the previous steps

8
Evidence-based health librarianship

• EBL began with health librarians helping health
  professionals integrate EBP in their daily
  practice.

Andrew Booth
9
Evidence-based health librarianship

• Many different roles evolved
• Educators / teachers (involved in curriculum)
• Trainers (in use of library applications)
• Researchers (methods research)
• Collaborators (in grants, systematic reviews)
• Information providers (providing expert
  searching)

10
Evidence-based for all librarianship

• Now its almost mainstream! There is a CLA
  Evidence Based Librarianship Interest Group
  (EBLIG) and a journal Evidence Based Library and
  Information Practice and an international
  conference help every two years.

11

• "Improve library practice by utilizing the best
  available evidence combined with a pragmatic
  perspective developed from working experience in
  librarianship." Madge, B. Evidence Based
  Librarianship Tools we all can use. MLA 2000.

12
Librarianship domains

• The following six domains and their definitions
  were developed based upon the major areas
  librarians deal with in their daily practice
• Reference/Enquiriesproviding service and
  access to information that meets the needs of
  library users.
• Educationfinding teaching methods and
  strategies to educate users about library
  resources and how to improve their research
  skills.
• Collectionsbuilding a high-quality collection
  of print and electronic materials that is useful,
  cost-effective and meets the users needs.
• Managementmanaging people and resources within
  an organization.
• Information access and retrievalcreating
  better systems and methods for information
  retrieval and access.
• Marketing/Promotionpromoting the profession,
  the library and its services to both usersand
  non-users.
• Crumly E, Koufogiannakid D. Developing
  evidence-based librarianship practical steps for
  implementation. Health Information and Libraries
  Journal 2002196170.

13
Evidence-based librarianship

• This is about getting the science in
  information science.
• Steps in EBL
• 1. Define the problem
• Formulate a question arising from an issue or
  problem in a library or in the use of a library
  based philosophy
• 2. Find the evidence
• 3. Critically appraise the evidence
• 4. Apply the appraised evidence to the problem
• 5. Quality assurance evaluate your plan

14
Formulate a question

• Lets think of a few scenarios and questions!
• Scenario The quality of bibliographies in
  the master level theses of music students was
  noted to be poorer than in other faculties
  (including poor referencing, low referencing,
  mistakes, etc.). The Dean of Music raised this
  issue with the Chief Librarian. The Chief
  Librarian suggested that perhaps database
  training could help.

15
Formulate a question

• The following question was proposed Does
  database search training improve the quality of
  bibilographies in Masters thesis in music
  students.
• Population Master level music students
• Intervention - Database search training
• Comparison Students with no training (maybe
  use another faculty for comparison)
• Outcomes To improve the quality of
  bibliographies

16
Search the evidence

• What are the sources of information available to
  librarians? What are specific resources?
• Databases
• Textbooks
• Internet
• Unpublished (grey) literature
• Expert opinion

17

• EVIDENCE PYRAMID
• http//servers.medlib.hscbklyn.edu/ebm/2100.htm

18
(No Transcript)
19
Quality and types of librarian literature

• Systematic reviews or Meta-analyses
• Randomized controlled trials (RCTs)
• Cohort studies - describe possible causal links
  and pose probabilities of risk
• Comparative studies
• Descriptive articles (including narrative
  reviews)
• Surveys
• Case studies
• Qualitative research (such as focus groups,
  bibliometrics)

20

• Case study a description of a process, project,
  program, technology implementation, organization,
  library service etc.
• Cohort study a study that tracks over time a
  defined population (the cohort). These groups
  may or may not be exposed to factors hypothesized
  to influence the probability of the occurrence of
  a particular disease or other outcome. Cohorts
  are defined populations which, as a whole, are
  followed in an attempt to determine
  distinguishing subgroup characteristics.
• Comparative study a study that uses a
  systematic effort to find similarities and
  differences between two or more observed
  phenomena. An example could include comparing
  and contrasting OVID MEDLINE and Pubmed.

21

• Data-mining/ biblio-mining a method that
  involves the discovery of meaningful patterns
  from data retrieved from automated methods.
  Bibliomining uses the combination of data mining
  with bibliometrics, statistics, and reporting
  tools look at patterns in library systems.
• Descriptive survey a survey that describes the
  respondent perspectives or experiences on the
  questions that were asked in a predefined manner.
  Citation analysis represents a variation of the
  descriptive survey method.
• Focus Group a method that generate data or
  information from a meeting of individuals. This
  is also sometimes used to add to information
  gathered from other research methods such as
  questionnaires.

22

• Gap analysis a method involves surveys to
  identify discrepancies or gaps between individual
  expectations.
• Narrative review (review article) a review or
  overview of a subject that has been created from
  the results of an expert literature search.
  These can provide concise introductions to
  subjects.
• Program evaluation a method that assesses the
  operation or outcomes of a program. This can be
  very valuable in evaluating a program or policy
  at different levels in their development or
  implementation.

23

• Randomized controlled trial (RCT) a type of
  primary research that tests control and treatment
  groups, where the group participants have been
  randomly assigned. These trials can use
  individuals or groups. For example, a study
  could randomize different libraries to receive an
  intervention (such as access to certain
  resource).
• Systematic review (Meta-analysis) a review that
  uses systematic and explicit methods to identify,
  select and critically appraise relevant research,
  and to collect and analyze data from the studies
  that are included in the review. Statistical
  methods (meta-analysis) may or may not be used to
  analyze and pool the results of the included
  studies. Systematic reviews (also known as
  systematic overviews, evidence summaries and
  integrative reviews) use recently developed
  scientific methods to summarize results from
  multiple research studies.

24
Critical Appraisal

• Less is written about critical appraisal in
  librarianship. It can be difficult for
  librarians to interpret the evidence.
• Journals clubs can be used to improve skills.
• Pearce-Smith N. A journal club is an
  effective tool for assisting librarians in the
  practice of evidence-based librarianship a case
  study. Health Information and Libraries Journal.
  2006 Mar 23(1) 32-40.

25
Application to practice

• Now that you asked your question, gathered your
  evidence, reviewed your evidence, what do you
  do!!!!

26
Challenges for EBL

• Quality of the evidence
• Dispersion of evidence sources (e.g., education,
  management, computer science)
• Skills in conducting research
• Skills in disseminating research
• Skills in interpreting research
• Time!
• Andrew Booth Anne Brice (http//www.chl.wales.nh
  s.uk/cpd/librarians/e-based.pdf)

27
EBL Resources

• Evidence-based Practice for Information
  Professionals A Handbook (Hardcover) by Andrew
  Booth (Editor), Anne Brice (Editor). 2004
• ISBN-10 1856044718

28
http//www.cla.ca/about/igroups/evidence_based.htm
29
http//ejournals.library.ualberta.ca/index.php/EBL
IP/about/submissions
30
EBL Example - PRESS

• PRESS Peer Review Electronic Search Strategy
• 1) Develop a quality assessment checklist for
  electronic search strategies used in HTAs/SRs
• 2) Create a web-based peer review forum to these
  searches.
• This project was graciously funded through a
  grant by CADTH Canadian Agency for Drugs and
  Technologies in Health (formally CCOHTA)
• The final report to CADTH is complete and an HTA
  Report is due Dec 2007/Jan 2008.

31
Why?

• The quality of Health Technology Assessment (HTA)
  reports depends on many factors. One factor is
  the evidence base, the literature and other
  information, upon which the HTA is based.
• Performing a high quality search of information
  resources ensures accuracy and completeness of
  the evidence base used in HTA reports. A current
  review does not exist to tell us what elements of
  the search process have the most impact in the
  overall quality of the resulting evidence base.

32
Research Components

1. a systematic review of the literature
2. a consensus of expert opinion
3. a peer review forum to evaluate electronic search
   strategies during the initial stage of a
   systematic review or health technology assessment

33
Systematic review

• Systematic searches were performed to
  identify evidence on the importance of checklist
  items to the validity of the search.

• 7 bibliographic databases
• MEDLINE
• LISA
• CINAHL
• The Cochrane Librarys Methodology Register and
  Methodology Reviews
• PsycINFO
• HAPI
• HealthSTAR
• 3 in-house databases

34
Systematic review results

• A total of almost 9,700 citations were retrieved.
  Relevancy screening was done by 2 reviewers with
  the use of SRS, a web-based systematic reviewing
  platform by TrialStat.
• Over 100 relevant articles were obtained.
• Although 26 tools were identified that could
  potentially be used as checklists, none were
  validated for assessing electronic search
  strategies.

35
Consensus building - survey

• A survey of HTAs/SRs searchers was conducted
  using Survey Monkey, a web-based survey platform.
  
• Fifty-eight respondents completed the survey and
  allowed the elements to be ranked into three
  tiers of importance based on assessment of the
  potential impact of the elements on recall and
  precision.

36
Peer review forum

• Before programming the interface for a fully
  operational version of the peer review forum, 10
  survey respondents participated in a pilot of a
  web-based peer review forum in which 10 sample
  searches.
• This was also used for validation testing of the
  checklist and use of the web-site forum.

37
(No Transcript)
38
Implications for Decision Making

• Errors in the electronic search strategy have
  been demonstrated to reduce the effectiveness of
  electronic search strategies used in systematic
  reviews and (HTA) reports. Without assurance of a
  bias-free and complete evidence-base, the true
  outcomes of a systematic review cannot be tested.
  A checklist has been developed to assess the
  electronic search strategy for errors.
• An evidence-based peer review process has been
  developed to evaluate electronic search
  strategies for use in systematic reviews and
  health technology assessment reports.

39
Implications for Decision Making

• As the evidence base for systematic reviews and
  HTA reports is created in large part through the
  electronic search strategy, that strategy should
  be peer reviewed before the evidence base is
  created and reviewed.
• The evidence-based peer review of search
  strategies requires the same body of expert
  knowledge needed to create search strategies
  therefore the peer review should be undertaken by
  librarians or other suitably qualified and
  experienced individuals.

40
Evidence the ultimate goal is KT
41

• Knowledge translation is the exchange,
  synthesis and ethically-sound application of
  researcher findings within a complex system of
  relationships among researchers and knowledge
  users.

42
KT

• Biomedical research is consistently producing new
  findings that may contribute to effective and
  efficient patient care.
• The findings of such research will not change
  population outcomes unless health services and
  health care professionals adopt them in practice.
  
• Grimshaw, Ward, Eccles. Oxford Handbook of
  Public Health.

43
Audiences for KT in health research
Audience Basic Clinical Health Services Population Health
Researchers
Professionals
Patients
Administrators
Policy Makers
Industry
..
44
What is the librarians role?

• This is the step beyond dissemination
• Current examples include
• Evidence-based actionable messages (EBAMS)
• Consumer summaries for Cochrane reviews
• Summaries of newspaper articles
• Clinical question and answering service

45

• Be it Resolved that Evidence Based
  Librarianship is NOT Hooey

Commentary. Evidence Based Library and
Information Practice 2007, 22
46

• It is interesting that librarians, just like
  physicians, have felt offended by being asked
  to embrace this philosophy.
• A few years after the introduction of evidence
  based medicine in the literature, Sackett et al.
  replied to all criticisms and verifying that the
  historical origins date back to mid-19th century
  Paris.
• This article is now over 20 years old and
  evidence based medicine is now standard in most
  medical schools we need get on with things and
  catch up! Our question to proponents is, What is
  the alternative? Just make do with what feels
  good or see what Google has to say? We think not.
• Sackett, David L., William M. C. Rosenberg, J.
  A. Muir Gray, R. Brian Haynes, and W. Scott
  Richardson. Evidence based medicine what it is
  and what it isnt. BMJ 3127023 (1996) 71-2.

47
Question revisit

• Question to ponder Do we give users what they
  want or what they need? How does the quality of
  resource effect collection development policies?

48
(No Transcript)
49
Thanks jmcgowan_at_uottawa.ca