Evidence Based Information Retrieval (Martin)

Existing knowledge can prevent…
•Poor quality clinical care
•Poor patient experience
•Adoption of interventions of low value
•Failure to adopt interventions of high value
Source: Sir Muir Gray, Chief Knowledge Officer of Britain’s National Health
Service. Quoted on http://www.nks.nhs.uk/.
Learning Objectives
At the end of the presentation, you will be able to:
• Define evidence-based medicine (EBM)
• Understand the Five Steps to practice EBM
• Use the 4S approach to organizing clinical research
• Conduct an efficient online search to track down best
• Access online and print tools to critically appraise the
• Use the Five Steps in future clinical encounters
What is EBM?
“Evidence-based medicine is the integration of best
research evidence with clinical expertise and patient
Best research
evidence Expertise
Sackett DL, Rosenberg WMC, Gray JAM, Haynes RB, Richardson WS: Evidence based medicine: what it is and what it
isn’t. BMJ 1996;312:71-2.
Evolution of EBM in the Literature
Term first appeared in the literature in a 1991 editorial in
ACP Journal Club Volume 114, Mar-April 1991, pp A-16
Seminal article by the Evidence-Based Medicine
Working Group published in JAMA Volume 268, No. 17,
1992, pp 2420-2425
Fundamentally new approach
becomes widely recognized
JAMA published a series of Users’ Guides to the Medical
Literature that served as the first learning tools
Courses were developed in residency training and
medical school curricula
The first handbook, Evidence-Based Medicine: How to
practice and teach EBM, by Sackett, et al, was published
in 1996. Fourth edition published in 2010.
New York Times listed EBM as one of its ideas of the
year in 2001
BMJ listed EBM as one of the 15 greatest medical
milestones since 1840
New Approach Required New Skills
Clinical question formulation
Search and retrieval of best evidence
Critical appraisal of study methods to ascertain validity of
Integration of EBM into medical school
curricula patient-doctor courses
Information Retrieval for Evidence
Based Patient Care
Using research findings versus conducting research
Retrieving and evaluating information that has direct
application to specific patient care problems
Selecting resources that are current, valid, and
available at point of care
Developing search strategies that are feasible within
time constraints of clinical practice
Key developments that streamlined the
practice of EBM
Advances in ease of accessing and understanding
Development of preprocessed (preappraised) tools
Improvements in search interfaces to MEDLINE
Collaboration between EBM Working Group and
National Library of Medicine in development of hedges,
“clinical queries” tool, that filters search results to
specific study types and levels of evidence
Dissemination of systematic reviews of primary studies
and growth of the Cochrane Collaboration
Ongoing Developments…
Continuing development of point of care (POC),
evidence based summaries, e.g., Dynamed, ACP
PIER, UptoDate, Clinical Evidence, FirstConsult
None are integrated with individual patient EMRs
Can be run through same computers as EMRs
Some have a direct link from within the EMR via the “Info
Ongoing Developments…
“Evolving” decision support systems that will integrate
and summarize relevant evidence that answer clinical
questions and automatically link, through the EMR, to a
specific patient problem
Production models exist for parts of decision support systems
Cover limited range of clinical problems
Ongoing Developments
Uptake of Evidence Based Medicine process by health policy, nursing,
allied health, and psychosocial fields
Terminology is important when searching the literature
Subject Heading is Evidence Based Medicine
Evidence Based Optometry = Evidence Based Medicine and Optometry
Evidence Based Practice is a broad subject heading that is used for
Evidence Based Health Care in general
EBM Process – 5 Steps
ASSESS: Recognize and prioritize important patient
ASK: Construct clinical questions that facilitate an
efficient search
ACQUIRE: Track down the best evidence to answer the
APPRAISE: Systematically evaluate best available
evidence for validity, importance, and usefulness
APPLY: Interpret the applicability of evidence to
specific problems, given patient preferences and values
Step 1
ASSESS the clinical problem
Begin with the patient encounter
Select question that
Is most important to the patient’s well being
Fills gaps in your clinical knowledge
Is feasible to answer in the time available
Step 2
ASK focused clinical questions
Four common types of clinical questions:
Well Built Clinical Questions
Deal with patient management issues
Contain elements of PICO format
Comparison Intervention (if useful)
Facilitate an efficient search
Example – Therapy/Prevention
In patients with POAG or OHT, does
topical pharmacological treatment
compared to no treatment prevent
progression or onset of VF defects?
Step 3
ACQUIRE: Track down the evidence to
answer the question
Use the 4S approach to select the most
likely resource
Start with the highest level resource
4S Hierarchy
6S Hierarchy
Point of Care Summaries:
Uptodate, Dynamed,
DARE (synopses of syntheses)
ACP Journal Club (synopses of
Cochrane and other Systematic
MEDLINE Searches limited to
Clinical Queries
SOURCE: Haynes, R. B. (2001). Of studies, syntheses, synopses, and systems: the “4S” evolution of services for finding
current best evidence. Evidence-Based Medicine, 6 (2), 36-38. Retrieved 2-07-07 from
Critically Appraised Content
Evidence Based Retrieval
1. Find the answer that is supported by valid
studies appropriate to the type of
question and that is available in a timely
2. Requires search terms plus best study design
for question plus highest level of evidence
Extract search terms from question
Therapy/Prevention Question in PICO
In patients with primary open angle glaucoma
or ocular hypertension [Patient/Population],
does topical pharmacological treatment
[Intervention] versus no treatment
[Comparison Intervention], prevent
progression or onset of visual field defects
Possible Search Terms
Ocular hypertension, OHT, intraocular
pressure, IOP, primary open angle
glaucoma, POAG, medical treatment,
medical intervention, visual fields, VF
Best Study Design for Type of Question
Type of Question
Study Design
Randomized controlled
Prospective cohort, blind
comparison to a gold
Cohort, Case Control, Case
Cohort, Case Control, Case
As you move up the pyramid the amount of available literature decreases, but it increases in its
relevance to the clinical setting.
Source: Sackett, D.L., Richardson, W.S., Rosenberg, W.M.C., & Haynes, R.B. (1996). Evidence-Based Medicine: How to practice and teach
EBM. London: Churchill-Livingstone.
Find critically appraised content under “Evidence Based Medicine”
• DynaMed
– Summaries for more than 3,000 topics
– Monitors >500 medical journals and
systematic review databases
– Updated daily
– Each article evaluated for clinical relevance
and scientific validity
– Includes “graded evidence”
Glaucoma Summary
Evidence-based answer found in 1 minute, 39 seconds
– Over 400 evidence-based summaries for primary care
published by the American College of Physicians
– Recommendations based on citations from medical
literature and ranked by experts at McMaster
– Updated monthly
– Reference citations rated by type and quality of study
• UptoDate
– Evidence based summaries of over 9,500
topics in over 20 specialties
– Ophthalmology not one of the specialties
– Good for information on systemic conditions
– Available through individual subscription.
Online access plus Mobile app for iPhone and
iPad. Cost: $199 per year in training; $499
per year in practice
Available in the Optometry Clinic at Hastings Hospital. Use Resource tab and scroll
down to the UptoDate link
Sample summary at http://www.uptodate.com/contents/management-of-persistent-hyperglycemia-in-type-2diabetes-mellitus
• FirstConsult
– Available through NSU subscription to MD
Consult for iPhone or iPad only
– Create a personal account in MD Consult
– Download the app from the Apple app store
– Login with your username and password
– Summaries are detailed and include sections
on Differential Diagnosis
– Eyes and Vision topics well covered
Synopses of Syntheses
• DARE (Database of Abstracts of Reviews
of Effects)
– contains structured abstracts of systematic
reviews from around the world
– covers topics such as diagnosis, prevention,
rehabilitation, screening, and treatment
– available on Ovid
Synopses of Studies
• ACP Journal Club
– consists of two journals, ACP Journal Club, a
publication of the American College of
Physicians, and Evidence-Based Medicine, a
joint publication of the American College of
Physicians and the British Medical Journal
– includes studies which ACP's editors have
selected as methodologically sound and
clinically relevant
• Cochrane Database of Systematic
Reviews (DSR)
– Part of the Cochrane Library (1996)
– 916 completed reviews, 1905 protocols
– Among the highest level of evidence upon
which to base treatment decisions
– Includes Dx since 2008
– Eyes & Vision Research Group
• Contains over 165 reviews
Systematic Review
Analyzes data from several primary studies to
answer a specific clinical question
Provides search strategies and resources used
to locate studies
Includes specific inclusion and exclusion criteria
(results in less bias)
Meta-Analysis (subclass) statistically
summarizes results of several individual studies
Access full text of Cochrane reviews in OVID
Cochrane DSR
Review found in 15 seconds
Copyright: The Cochrane Library, Copyright 2009, The Cochrane Collaboration
Levels of Evidence
Grade the quality of evidence based on the
design of the clinical study
Variety of hierarchies in use
GRADE Grading of Recommendations
Assessment, Development and Evaluation
High: High confidence that the evidence supports this finding. Further
research is unlikely to change our confidence in the relationship between
this clinical action and the reported patient outcome(s)
Moderate: Moderate confidence that the evidence supports this finding.
Further research may change our confidence in the relationship between
this clinical action and the reported patient outcome(s)
Low: Low confidence that the evidence supports this finding. Further
research is likely to change our confidence in the relationship between
this clinical action and the reported patient outcome(s)
Very Low (Insufficient): Very low confidence that the evidence supports
this finding. There is either no evidence available, or what is available
does not allow evaluation of the relationship between this clinical action
and the reported patient outcome(s)
FirstConsult in transition from SORT to GRADE. See JAMA 300(15) October
2008,”Progress in Evidence-Based Medicine”, for details on GRADE
Appraise the Filter (pre-appraised
Speed of updating
Scope and number of summaries
Summaries with graded evidence
Authors’ credentials, affiliations listed
Developing and using a rubric for evaluating evidence-based medicine point-of-care
tools. Journal of the Medical Library Association, Volume 99, No. 3, July 2011
Speed of updating online evidence based point of care summaries: prospective cohort
analysis. BMJ 2011;343:doi:10.1136/bmj.d5856 (Published 23 September 2011)
Appraisal Required by User
Step 4
Appraise best available evidence from
original studies
If the other “S’s” don’t provide the answer,
search for original studies
Use “clinical queries” limit in Ovid
MEDLINE to speed retrieval
Least efficient (in terms of time) to answer
clinical questions
Primary (Original) Studies
Articles that report results of original
research investigations
Conclusions supported by data and
reproducible methodology
Require time to acquire and appraise
Good Source: MEDLINE (OVID)
• Premier biomedical database from the National Library
of Medicine (NLM)
• Covers 1946-present
• Full text available from subscribed NSU journals plus
Ovid Deposit Account journals available only for
• Ovid interface includes tools to quickly filter search
results to specific study types and levels of evidence
• Access from
To Use Clinical Queries Tool, enter search terms and then click on “Additional Limits”.
Ovid MEDLINE Clinical Queries Tool
Levels of Evidence in Ovid based on AAFP SORT
Level A = “Specificity” in Ovid Clinical Queries
 Systematic reviews of randomized controlled trials including metaanalyses
 Good-quality randomized controlled trials
Level B = “Sensitivity” in Ovid Clinical Queries
 Good-quality nonrandomized clinical trials
 Systematic reviews not in Level A
 Lower-quality randomized controlled trials not in Level A
 Other types of study: case control studies, clinical cohort studies,
cross sectional studies, retrospective studies, and uncontrolled
Level C
 Evidence-based consensus statements and expert guidelines
Step 5
APPLY the evidence to patient care
Practice the EBM process in daily patient
Access databases and tools from
UptoDate available at Hastings Hospital
Use Resource tab and scroll down to UptoDate
Take Home Points
Focused clinical question (PICO) reveals
your search terms
Start your search at top of 4S hierarchy
and work down
Be aware of the filter, i.e., levels of
evidence, speed of updating
Look at more than one resource in the
hierarchy. Findings may differ.
Practice makes perfect
Evidence Based Medicine Lecture
NSUOCO Residents Journal Club
Sandra A. Martin, M.L.I.S.
Health Sciences Resource Coordinator
Instructor of Library Services
John Vaughan Library Room 305B
[email protected] – 918-444-3263

similar documents