Evidence-based Medicine

Overview

Why should we learn about

EBM?
 Objectives
After the lecture, students will be
able to
 Illustrate the concepts of EBM
 Formulate clinical questions that can
be researched
 Identify resources for finding the best
clinical evidence
 The Definition of EBM

 Evidence-based medicine (EBM) is the

process of systematically reviewing and
appraising clinical research findings with
the aim to identify and use the current
best evidence to aid in the optimal care of
individual patients.
Why learn about EBM?
 Medical research is continually discovering
improved treatment methods and therapies
 Research findings are often delayed in being
implemented into clinical practice
 Clinicians must stay current with changing
therapies
 Evidence-based practice keeps clinicians up
to date
 EBM solves clinical problems by examination
of clinical research, thus complementing the
practitioner’s clinical experience
 A Process of Life-long,
Problem-based Learning
 Converts information needs into focused
questions
 Efficiently tracks down the best evidence with
which to answer the question
 Critically appraises the evidence for validity
and clinical usefulness
 Applies the results in clinical practice
 Evaluates performance of the practitioner in
applying the evidence to the patient’s
treatment
The Steps in the EBM
Process
Formulate a clinical
question

Use EBM resources & identify the

current best evidence

Evaluate the evidence for its validity,

impact, and applicability

Integrate the critical appraisal with your

clinical expertise & the patient’s unique
circumstances
Evaluate your effectiveness &
efficiency in the EBM process; seek
ways to improve
 References
 Sackett D, Rosenberg W, Gray J, Haynes R, Richardson W.
Evidence based medicine: what it is and what it isn't. BMJ
1996;312:71-2.
 Greenhalgh T. How to read a paper. BMJ 1997;315:672-5.
 Guyatt Gordon; Drummond Rennie.
Users' guides to the medical literature : a manual for evide
nce-based clinical practice
. Chicago: AMA Press; 2002. 706 p.
 Grandage KK, Slawson DC, Shaughnessy AF.
When less is more: a practical approach to searching for ev
idence-based answers
. J Med Libr Assoc. 2002 July; 90 (3): 298-304.
 Strauss, Sharon E.; Richarson, W.; Glasziou, Paul, et
al.
Evidence-based medicine: How to practice and teac
Learning Evidence-based
Medicine

What are the principles of

EBM?
Principles
 Understand the difference between
“Background” and “Foreground” questions
 Identify the key elements of a clinical
question
 Understand the impact of a well-built clinical
question on EBM research
 Identify EBM searching strategies to find the
current best evidence
 Identify key issues for determining the
validity of a clinical study
 Apply the evidence to the individual patient
The Steps in the EBM
Process
Formulate a clinical
question

Use EBM resources & identify the

current best evidence

Evaluate the evidence for its validity,

impact, and applicability

Integrate the critical appraisal with your

clinical expertise & the patient’s unique
circumstances
Evaluate your effectiveness &
efficiency in the EBM process; seek
ways to improve
 The Clinical Question
 The 1st step: Formulate a
searchable clinical question
 Your clinical question will impact
the entire EBM research process

Search
Clinical Terms Literature EBM
Question (keywords Search Resource
)
 Background vs.
Foreground Questions
 Background questions ask for general
knowledge
 Examples
 How does heart failure cause emphysema?
 What causes sinus tachycardia?
Foreground Questions
 Foreground questions ask for specific
knowledge toPinform
atient clinical decisions
 They have four essential components:



I
Patient or problem
ntervention
Intervention or exposure

Comparison, if relevant
Clinical outcomes, including time
Comparison


Outcome
Example of a Foreground
Question
 In adults with heart failure who are
in sinus rhythm, would adding
warfarin to standard therapy
reduce morbidity or mortality from
thromboembolism enough over 3-5
years to be worth warfarin’s
harmful effects and
inconvenience?
 Identify the “PICO” elements in this
question
Effects of a well-developed
and answerable question
 Saves time when researching
 Keeps the focus directly on the
patient's need
 Suggests the appropriate form that
a useful answer may take
 Suggests the type of research
most likely to provide the best
answer
Important Reminder
 Your clinical situation may raise
more than one question. Do not try
to squeeze multiple topics into one
clinical research question!
Who is your PATIENT?
 A clinical question must identify a
patient or patient group.
 A clinical question should also
include any information that is
relevant to the treatment or
diagnosis or the patient:
 Sex, age or ethnicity
 Disease History
 Primary Complaint
What is the
INTERVENTION?
 The intervention is what you plan
to do for your patient or patient
group:
 Run tests
 Prescribe drug treatment
 Refer to a specialist
 Schedule surgery
What is the COMPARISON?
 In general most, but not all, clinical
questions have a comparison. A
comparison is the alternative to
compare to your intervention:
 Is aspirin as effective in preventing
strokes as warfarin?
 Is bed rest as effective as antibiotics
in treating Acute Otitis Media?
What is the OUTCOME?
 The outcome is the expected effect
of the intervention:
 If I prescribe ibuprofen for my patient
it will prevent pain. Outcome = Pain
Prevention.
 If I prescribe antibiotics for my patient
it will resolve the infection. Outcome
= Bacterial Clearance.
 P.I.C.O. Model for Clinical
Questions

Patient, Population, or How would I describe a group of patients

Problem similar to mine?
Intervention, Prognostic Which main intervention, prognostic factor,
Factor, or Exposure or exposure am I considering?
Comparison of Intervention What is the main alternative to compare
(if appropriate) with the intervention?
Outcome you would like to What can I hope to accomplish, measure,
measure or achieve improve or affect?
What type of question are Diagnosis, Etiology, Therapy, Prognosis,
you asking? Prevention, Harm
Type of study you want to What would be the best study design or
find methodology?
A Case
 Joseph Borden, an 8-year-old boy, complains of
ear ache. His mother reports that he has been
treated for ear infections three times in the
last 18 months. You diagnose acute otitis
media and prescribe amoxicillin. The mother
asks you whether there is anything can be
done to prevent the infection from recurring.
You remember a presentation during a
symposium on Xylitol.
 Examples of Clinical
Questions
 Here is a well-formulated clinical question that
a pediatrician might research based on the
case above:
 In children with a history of acute otitis media, will
Xylitol help prevent recurrent infections?
 You quickly perform a search of PubMed and
identify several articles on the use of Xylitol in
the prevention of acute otitis media. You select
one of these:
 Uhari M, Kontiokari T, Koskela M, Niemelä M. (1996) Xylitol
chewing gum in prevention of acute otitis media: double blind
randomised trial. BMJ 313:1180-4.
Another Case
 John Alden, a 68-year-old slightly man,
complains of profound tiredness and inability
to exercise without developing chest pain and
shortness of breath. He is 5’ 10” and weighs
190 lbs. He points out that his ankles get
swollen during the day. He smokes 1-2 packs
of cigarettes per day. The diagnosis is
congestive heart failure (CHF) with complete
left bundle branch block resulting from
ischemic heart disease. The initial left heart
failure has led to right heart failure. His
daughter plans to get married in 12 months
and he worries that he his health condition
might not allow him to attend the wedding.
 Examples of Clinical
Questions
 Here are two well-formulated clinical questions
that a physician might research based on the
case above:
 Among elderly patients with congestive heart failure
and complete left bundle branch block, does
insertion of a biventricular pacemaker combined with
ACE-inhibitors and beta-blockers reduce mortality
rates compared to treatment with ACE-inhibitors and
beta-blockers only?
 In elderly patients with congestive heart failure, will
the addition of an angiotensin II antagonist to
treatment with ACE-inhibitors reduce mortality over a
5-year period compared to treatment with ACE-
inhibitors alone?
Learning Evidence-based
Medicine

What are the resources for

clinical evidence?
Resources for Clinical
Evidence
 EBM resources provide:
 A current overview of a subject
 A critical appraisal of the literature
 A review and synthesis of the
research on a topic
 A comparison of emerging treatment
options to the gold standard
treatment
Three Major Categories of
EBM Resources
Filtered Unfiltered
Information Information
Resources Resources

Background Information
Resources
Note
 Background Information Resources
should not be confused with
resources to answer background
questions
 They consist of Practice Guidelines,
Consensus Statements, and
overviews of the current research
How to find Best Current
Evidence
 Only a small portion of the medical
literature is immediately useful in
answering clinical questions.
 Reports cover the whole spectrum
of the scientific research process -
from in-vitro studies to double-
blind randomized controlled trials.
How to find Best Current
Evidence
 Adding METHODOLOGY terms
and CLINICAL FILTERS to
SUBJECT terms will result in the
most efficient and optimal retrieval
of the highest level of evidence in
answering clinical questions.
 Optimal Retrieval of High
Levels of Evidence

SUBJECT TERMS
plus
METHODOLOGICAL TERMS
equal
OPTIMAL RETRIEVAL of HIGH LEVELS of
EVIDENCE
The Evidence Pyramid
 Levels of Clinical Evidence
in the Primary Literature
Type of
Type of Question MEDLINE Filters
Study/Methodology
Therapy Double-Blind Randomized Controlled Trial [PT],
Randomized Controlled Trial double [TW] and blind [TW]
Diagnosis Controlled Trial Sensitivity and Specificity [MH],
Diagnosis [SH]
Prognosis Cohort Studies, EXP Cohort Studies [MH],
Case Control, Prognosis [MH],
Case Series Survival Analysis [MH]
Etiology Cohort Studies EXP Cohort Studies [MH],
Risk [TW]
Prevention Randomized Controlled Trial, Randomized Controlled Trial [PT],
Cohort Studies Cohort Studies [MH],
Prevention and Control [SH]
Quality Randomized Controlled Randomized Controlled Trial [PT]
Improvement Trials Practice Guideline [PT]
Consensus Development Conference [PT]
 Issues
• EBM is one part of the clinical process. Clinical
decisions must combine clinical expertise and
experience, patient preferences, and solid
scientific that
• EBM is cook book medicine. It suggests evidence.
decisions are based solely on• research
The last step in the EBM process is to decide
evidence and negates sound clinical
whether or not the information and results are
judgement. applicable to your patient and to discuss the
• EBM is the mindless applicationresults with the patient.
of population
studies to the treatment of the individual.might
• Clinicians It consider the "evidence pyramid"
takes the results of studies of large groups
and look for of
the next best level of evidence.
people and tries to apply themClinicians
to individuals
need to understand that there may be
who may have unique circumstancesno good orevidence to support clinical judgement.
characteristics, not found in the study groups.
• Librarians can help identify the best resources
• Often there is no randomized controlled
and teachtrial or
clinicians effective searching skills.
"gold standard" in the literature to address the
clinical question.
• There is often great difficulty in getting access
to the evidence and in conducting effective
searches to identify the best evidence.