Evidence based

medicine
By dr. Balgis MSc.
CM FM.

The science of medicine

is continuously changing
if we do not adapt to
this change, the
following might happen;
a. We may have not give
our patient effective
treatments
b. We may have wasted
precious resources.

Old model of decision making

The traditional model of medical education
depended on learning collection facts,
principles and rules of thumbs that would
serve individual for a professional life.
The method; passive dissemination of
information (didactic lecture in large
session halls)
The Professor or the Expert become
the standard source of information for
decision making
Problem can be answered by asking
colleagues and local expert, reading text
books ( out of date by the time they are
published), or browsing journals (but which
journal should you look at)

New model of medical decision making

There is now a strong trend towards
the view that medical education should
be lifelong, oriented towards problem
solving and based on the principle that;
1. Clinical experience is important but
observations must be recorded
systematically and without bias
2. Regular reference must be made to
original literature
3. The results of studies must be
critically examined using rules of
evidence before applying it to practice

Meaning of EBM
according to others
Experience-based medicine- the
oldest ones are the most valid
Eloquenc-based medicine- the
loudestvoice is the most heard

Evidence-Based Medicine
Definition
the
the conscientious,
conscientious,

explicit
explicit and
and judicious
judicious
use
use of
of current
current best
best
evidence
evidence in
in making
making
decisions
decisions about
about the
the
care
care of
of the
the individual
individual
patient
patient

The
The practice
practice of
of EBM
EBM requires
requires the
the integration
integration of
of

individual
individual clinical
clinical expertise
expertise and
and patient
patient values
values
with
with the
the

best
best available
available external
external clinical
clinical evidence
evidence from
from
systematic
systematic research
research..
Best Clinical Evidence

Clinical
Experience

Patients Needs/
Preferences

What evidence-based
medicine is:
Good doctors use both individual clinical
expertise and the best available external
evidence, and neither alone is enough.
Without the former, practice risks becoming
evidence-tyrannised, for even excellent
external evidence may be inapplicable or
inappropriate for an individual patient.
Without the latter, practice risks becoming
rapidly out of date, to the detriment of
patients and patient-care.

Evidence based medicine

EBM was developed in canada to
teach medical students but was
latter applied to solving clinical
problem

 The aim of EBM is to encourage the use of

diagnostic test and therapeutic
interventions that are proven to be
accurate, effective and safe in clinical
research.
EBM fits with recommendations to make
undergraduate and postgraduate medical
education curriculum more problem
centered, directed toward lifelong learning
and less about memorizing a static body of
knowledge.

Evidence-based medicine is
not cook-book medicine:
Patients cannot go through a treatment
tunnel assuming the same management is
appropriate and the same outcomes
important to each
External clinical evidence can inform, but
can never replace, individual clinical
expertise
Your clinical accumen decides whether the
external evidence applies to the individual
patient at all and, if so, how it should be
integrated into a clinical decision.

Evidence-based medicine is
not cost-cutting medicine:
The aim is to give most benefit to
each individual patient
To apply the most efficacious
interventions which will maximise
their function, quality, and quantity of
life
may raise rather than lower the cost
of their care.

EBM is neither old-hat nor

impossible to practice:
The former argument falls before the evidence:
of striking variations in the integration of
patient values into our clinical behaviour
of striking variations in the rates with which
clinicians provide interventions of
established benefit and uselessness to their
patients.
in the inability of clinicians to keep abreast
of important medical advances reported in
primary journals

The patient

1. Start with the patient -- a clinical problem or question

arises out of the care of the patient

The question

2. Construct a well built clinical question derived from the

case and classify it into one category

The resource

3. Select the appropriate resource(s) and conduct a search

The evaluation

4. Appraise that evidence for its validity (closeness to the

truth) and applicability (usefulness in clinical practice)

The patient

5. Return to the patient Make a clinical dicesion to apply

or not to apply the result of the study /integrate that
evidence with clinical expertise, patient factors to carry out
the dicesion

Self-evaluation

6. Evaluate your performance with this patient

What is Evidence-based
Practice
Clinical Skills

Keeping
up to date

Clinical question
Audit
Find the Evidence
Apply to Practice
Critical Appraisal

Types of Questions
Therapy
/prevention Question

Diagnosis Question

A question concerning the effectiveness

of a treatment or preventative measure

A question concerning outcome of

A patient with a particular condition

Prognosis Question

A question concerning
the ability of a test to
predict the likelihood of a disease

Harm Question

A question concerning the likelihood

of a therapeutic intervention to cause harm

Type of Question

Suggested best type of Study

Therapy

RCT>cohort > case control > case series

Diagnosis

Prospective, blind comparison to a gold standard

Etiology/Harm

RCT > cohort > case control > case series

Prognosis

Cohort study > case control > case series

Prevention

RCT>cohort study > case control > case series

Clinical Exam

Prospective, blind comparison to gold standard

Cost

Economic analysis

Types of Research Studies

Certain study designs are superior to others
when answering particular questions.
RCT are considered the best for addressing
questions about therapy.
Etiology questions may be addressed by case-control
& cohort studies.

Other aspects relevant to study design include

placebo comparison group and follow-up.

Evidence Pyramid
Meta-Analysis
Systematic Review
Randomized Controlled Trial
Cohort studies
Case Control studies
Case Series/Case Reports
Animal research

This is often referred to as the "evidence pyramid". It is

used to illustrate the evolution of the literature.
The base of the pyramid is where information usually
starts with an idea or laboratory research. As these ideas
turn into drugs and diagnostic tools they are tested in
laboratories models, then in animals, and finally in
humans. The human testing may begin with volunteers
and go through several phases of clinical trials before
the drug or diagnostic tool can be authorized for use
within the general population. Randomized controlled
trials are then done to further test the effectiveness and
efficacy of a drug or therapy. As you move up the
pyramid the amount of available literature
decreases, but increases in its relevance to the
clinical setting.

Meta-Analysis takes the systematic review (see

below) a step further by using statistical
techniques to combine the results of several
studies as if they were one large study.
Systematic Reviews usually focus on a clinical
topic and answer a specific question. Extensive
literature searches are conducted to identify
studies with sound methodology. The studies are
reviewed, assessed, and summarized according
to the predetermined criteria of the review
question.

Randomized Controlled Trial (RCT)

Randomized Controlled Trials are carefully
planned projects that study the effect of a
therapy or test on real patients. They include
methodologies that reduce the potential for
bias and that allow for comparison between
intervention groups and control groups (no
intervention). Evidence for questions of
diagnosis is found in prospective trials which
compare tests with a reference or "gold
standard" test.

Cohort study
Cohort Studies take a large population and follow patients
who have a specific condition or receive a particular
treatment over time and compare them with another group
that is similar but has not been affected by the condition
being studied.
Answers prognosis, etiology, prevention questions
Cohorts are defined populations that, as a whole, are
followed in an attempt to determine distinguishing
subgroup characteristics.
Researchers identify and compare two groups over a
period of time.
At the start of the study, one of the groups has a
particular condition or receives a particular treatment,
and the other does not.
At the end of a certain amount of time, researchers
compare the two groups to see how they did.

Case Control study

Answers prognosis, etiology, prevention
questions
Identifies patients who already have the
outcome of interest and control patients without
that outcome, and look back to see if they had
the exposure of interest or not.
These types of studies are often less reliable
than randomized controlled trials and cohort
studies because showing a statistical
relationship does not mean than one factor
necessarily caused the other.

Case series & Case Reports

Case Series consist of collections of reports
on the treatment of individual patients.
Case Report is a report on a single patient.
Because they are reports of cases and use
no control groups with which to compare
outcomes, they have no statistical validity .
Answers prognosis, etiology, prevention
questions

Practice Guidelines
Practice Guidelines are systematically developed
statements to assist practitioner and patient make
decisions about appropriate health care for specific
clinical circumstances. Guidelines review and evaluate
the evidence and then make explicit recommendations
for practice.
The pyramid serves as a guideline to the hierarchy of
evidence available. You may not always find the best
level of evidence to answer your question. In the
absence of the best evidence, you then need to consider
moving down the pyramid to other types of studies.

Levels of Evidence
Level 1: Randomized Clinical Trials
Level 2: Head to Head Trial or
Systematic Review of Cohort Studies
Level 3: Case-Control Studies
Level 4: Case-series
Level 5: Expert Opinion

Conducting an efficient
computer-aided literature
search

 A computer datebase can be likened to a file

composed of several index cards
Each index cards may represent a published
article
Each index cards contain structured
information such as its TITLE, the
AUTHORS, the SOURCE, the ABSTRACT,
KEYWORDS and other important
information
These index cards can then be put together
in certain sequence that will facilitate
access to information ( according to years of
publication, keywords)

 In computer, each index card is

referred to as a record
Each piece of structured information is
referred to as a FIELD
All of index cards is referred to as a
DATA BASE

The advantages of computer

database
The very compact
format
Efficient access to
mountains of information

 Physicians as primer clinical decision makers have

significant impact on the cost and quality of care.
The current information explosion in medicine makes
it progressively more difficult for physicians to
access the information necessary to make intelligent
and cost effective clinical decisions
What is needed is a new model of medical knowledge
acquisition that emphasizes the important of
knowledge seeking skills and the ability of widely
available computer technology to augment the
physician cognitive potential.
As example the use of computer technology to
support the physicians need for medical knowledge is
the retrieval of information from a wide variety of on
line medical database such as MEDLINE

MEDLINE
Is the largest database of publication in healthrelated topics and most widely used medical
bibliographic available in the market
It is being maintained by US National Library of
Medicine
It covers more than 3500 journal derived from 70
countries around the world and 75% of the articles
are in english.
It also contains 7 million references since 1966 to
1994.

From the guru of EBM MEDLINE is the best general source of

current best evidence at present because of

its breadth and constant maintenance. Thus,
it is important to evidence-based practice
that clinicians develop and hone MEDLINE
searching skills and acquire local access.
Evidence-based Medicine: how to practice & teach
EBM - Sackett D L et al 1998

Clinical scenario
Patients usually come in to the clinic for
problems. Unfortunately these problem
are vague and sometimes not clearly
stated.
To state the problem clearly, you must
bear in mind that there are only three
important elements that the patient want
to know

1. What their disease are

2. What treatment they should
be given
3. What is the result to expect

 Before
Before attempting
attempting to
to

search
search for
for the
the
evidence
evidence in
in the
the
literature,
literature, construct
construct
aa well-built
well-built or
or clearly
clearly
structured
structured question
question
that
that could
could be
be broken
broken
into
into the
the following
following
four
four parts
parts

Three important elements in

clinical research are baically

PIO analysis):

P - Patient/Population
I - Intervention
O - Outcome

PICO analysis):

P - Patient/Population
I - Intervention
C Comparison
O - Outcome

Some time the researcher

can add Method

(1) Patient / Population

What is the primary problem,
disease or co-existing conditions?
On what groups do you want
information?
How would you describe a group of
patients similar to the one in
question?
Sometimes age or sex of a patient
may be relevant and should be
included.

Example: post-menopausal women

with hypertension

(2) Intervention

What medical event do you want to study

the effect of?
Which main intervention are you
considering, prescribing a drug, ordering
a test, ordering surgery?. . . .

(3) Comparison

Compared to what?
Better or worse than no intervention at all
or than another intervention?
What is the main alternative to compare
with the intervention, are you trying to
decide between two drugs, a drug and a
placebo, or two diagnostic tests?
Sometimes there is no comparison.

(4) Outcome

What is the effect of the intervention?

What do you hope to accomplish,
measure, improve, or affect with this
intervention?
What are you trying to do for the patient,
relieve or eliminate the symptoms, reduce
side effects, reduce cost? .. .
Example: improve quality of life

What makes a clinical question

well built?

Question should be directly

relevant to the problem at hand.

Question should be phrased to

facilitate searching for a precise
answer.

Asking a well-built question, like

many clinical skills, needs practice.

Categories of EBM
Questions
After

constructing the PICO

analysis, determine the category
of the question.
There are the four categories of
EBM questions

1) Therapy

Solves
Solves questions
questions about
about which
which

treatment
treatment to
to administer,
administer, and
and what
what
might
might be
be the
the outcome
outcome of
of different
different
treatment
treatment options.
options.

For
For most
most therapy
therapy questions
questions one
one may
may
want
want to
to look
look for
for the
the best
best evidence
evidence
namely
namely a
a randomized
randomized controlled
controlled study,
study,
and
and if
if the
the study
study can
can be
be a
a double
double blind,
blind,
so
so much
much the
the better.
better.
-- Example:
Example: find
find the
the evidence
evidence for
for the
the use
use of
of
anticoagulants
anticoagulants in
in patients
patients with
with atrial
atrial
fibrillation
fibrillation

2) Diagnosis

Solves questions about degree to which a

test is reliable and clinically useful, to
decide whether the patient would benefit
from it

Most articles compare result of a certain

diagnostic test with that of a standard
test regarded as being a "gold standard".
- Example: search for the best diagnostic
test for Alzheimer's disease

3) Etiology

Solves
Solves problems
problems

about
about the
the relationship
relationship
between
between a
a disease
disease
and
and a
a possible
possible cause
cause
-- Example:
Example: find
find out
out if
if a
a diet
diet
rich
rich in
in saturated
saturated fats
fats
increases
increases the
the risk
risk of
of heart
heart
disease,
disease, and
and if
if so
so by
by how
how
much
much

4) Prognosis

Answers questions about a patient's

future health, life span and quality of
life in the event one chooses a
particular treatment option

- Example: find how would the quality of

life change for a patient who undergoes
surgery for prostate cancer
The next step would be to determine the
best study design needed to answer the
clinical question.

The Boolean Principle

The venn diagram depicts 3 sets
Which are designated as setA, set B, set C
Each set has contents which are referred
to elements
The element consist of some numbers.
Note that the different sets
haveoverlapping elements

Boolean logic can be used to express the

relationships between the various sets in
this diagram, or to identify specific
elements.
This achieved using connectives AND
or OR. The union between set A and
Set B can be expressed in Boolean logic
as
A OR B
in the diagram, this would be
represented by elements 1,2,3,4,5 and 6

BOOLEAN LOGIC
AND / OR / NOT

OR

HEART DISEASES

HYPERTENSION

OR is MORE
The union between set
A and Set B can be
expressed in Boolean
logic as
A OR B
in the diagram, this
would be represented
by elements 1,2,3,4,5
and 6

BOOLEAN LOGIC
AND

HEART
AND
DISEASES

HYPER
TENSION

2. The Intersect of sets A and B can

be expreseed as
A AND B
in the diagram, this would be
repsented by elements 2 and 3.
Note that unions / OR tend to
incease the number of elements
included, while Intersects / AND
tend to do the opposite

 To make things more exciting, brackets

() can help us use unions and intersects in
the same expression for example:
( A AND B) OR C
would refer to elements 2,3,4,6 and 7 but
the similar expressions
A AND (B OR C)
would refer to elements 2,3, and 4.

BOOLEAN LOGIC
NOT

HEART
DISEASES

Not
CONGENITAL
HEART
DISEASES

Limits

Local holdings (for library only)

Age groups
Publication types
Language
Year of publication
Human/animal
etc

Still too much?

Use the most specific subject
headings
Use focus
Use subheadings
Use AND
Use limits
Restrict textword searching to title

Or too little?

Use OR
Use explode
Use All Subheadings
Dont use focus
Consider synonyms, alternative
spellings or truncation

Or too little?

Use OR
Use explode
Use All Subheadings
Dont use focus
Consider synonyms, alternative
spellings or truncation

Clinical Scenario
Supposing a patient with
hypecholesterolemia came into your
clinic and asks for an anti-cholesterol
drug because he wants to avoid
sudden death.
How you formulate the problem

Types of
Questions
If

you stated your

problem as among
patients with
hypercholesterolem
ia will anticholesterol drug
prevent sudden
death

The key concepts

1. hypercholesterolemia
2. Anti-cholesterol OR statin
3 sudden death

Type these word in the MEDLINE

1.
2.
3.

Term
Hypercholesterolemia
Anti-cholesterol OR statin
Sudden death

Articles
34.089
12.343
102.367

Naturally, I dont have time to read thousands of articles! So I

limit my serach further to
4. #1 AND #2 AND #3
1.835
Still a lot of articles to read, so I added some limiters like
focusing only on randomized controlled trials
5. Randomized controlled trial
6.098
6. #4 AND #5
12
I scanned the titles and come up with one study to read and
confident that this study is the most recent, up todate
and relevant

In searching through medline

there are several general steps
STEP I- develop an initial search strategy using
a single disc
Step I A. phrase the question as precisely as
possible
1. The kind of patient will be perform
( specified as a disease entity)
2. Intervention being evaluated
( a drug, a diagnostic test, an operative
procedure)
3. The outcomes of interest to the researcher (
mortality, morbidity, quality of life)
4. If needed we can add methodological filter

 STEP I B Identify key concepts in

the question phrase
At least one of key concept should be
specified in sufficient detail for a
meaningful search to be conducted

STEP I C- Broaden each key concepts , to

account for differences in use of
terminology
In preparing to search a database for the
key concepts that one identifies, a major
difficulty to be addresed is the problem of
non-uniform medical terminology.
This problem can be solved in two ways:
1. Using a free text search ( cancer or new
growth or neoplasm or tumor)
2. Using a medical subject heading ( MeSH)
search.

MeSH
Removes the uncertainty

Have I thought of all the synonyms?

Have I tried all the spellings?
Have I included all the variants?
Did the author use my term(s) at all?

From broad to specific

Start off by being
inclusive - keep it
broad
THEN narrow it
down by using limits
Narrowing too soon
will make you miss
things

THE MeSH
VOCABULARY

Is a controlled
Database
dictionary of almost
17000 terms maintained AIDSLINE
by NLM
MeSH is used to index
citation in a number of
biomedical databases
BIOETHICS
Table 1 list a sample of
this databases that use
MeSH vocabulary

Content
Citations to
the aids
literature
Citations to
biomedical
ethics
literature

The Scope of MeSH

The MeSH vocabulary
reflects the scope of the
biomedical literature in
that NLM adds terms as
new concepts change.
MeSH is updated on a
yearly basis to reflect
these changes.
Each of categories is the
root a complex hierarchical
arrangement of
increasingly specific MeSH
terms
These categories provide
an overview of the general
concept areas covered by
MeSH

Analytical, Diagnostic and Therapeutic

Technique ( Including Anesthesia)
Anatomical Terms, Body Regions, Organs &
Systems, Cylology and Embryology
Anthropology, Education, Human Activities and
Social Sciences
Biological Phenomena, Genetis, Physiology,
Occupations and Public Health
Human & Animal Diseases, Symptoms and
General Pathology
Health Care, Demography, Organizations and
Population Characteristics

Special MeSH Terms

MeSH contains some special types of terms
that are never designated as Major Concept
headings but can be used when searching.
These special MeSH terms are:
Publication Types :

these terms provide an additional

clasification dimension for citations in MEDLINE

Check Tags :

this group of MeSH terms designates very broad

attributes of the content of journal articles

Geographic Terms:

includes terms identifying individual

geographic regions, contiments, countries, states and selected
cities

MeSH Subheadings
MeSh contains a group of 80 terms
called MeSH subheadings.
Subheadings are used to qualify the
use of MeSH terms and allow the
searcher to limit retrieval citations
that deal with a spesific aspect of a
biomedical concept
The apropriate use of MeSH
subheadings can help us create highly
spesific MEDLINE queries that may
improve search precision

Abnormalitios
Administration &
Dosage
Adverse Effects
Analogs & Derivatives
Biosynthesis
Blood
Cerebrospinal fluid

The MeSH Tree Structures

MeSH terms are organized into a set
of 15 hierarchies called the MeSH
Tree Structures.
In this hierarchy the MeSH terms
are arranged into a set of branching,
treelike structures of increasing
specificity.

MeSH Diseases
Neoplasms
Neoplasms by site
Digestive System Neoplasms
Gastrointestinal Neoplasms
Intestinal Neoplasms
Colorectal Neoplasms

Intestinal
Neoplasms

Cecal Neoplasms
Colonic Neoplasms

Appendiceal Neoplasms
Colonic Polyps

Duodenal Neoplasms

Colorectal Neoplasms

Ileal Neoplasms

Sigmoid Neoplasms

Intestinal Polyps

Colonic Polyps
Gardner Syndrome

Jejunal Neoplasms
Rectal Neoplasms

Anus Neoplasms

Anal Gland
Neoplasms

MeSH searches

Using a standarized dictionary of medical terms

( if it existed) ex: dictionary form national
library of medicine (NLM)
The advantage is obviates the need for long
search strategies marked by an endless list of
synonyms.
The disadvantage is many concepts have not
found their way into the least of MESH term
( ex new disease or new drug)

 Experienced users are sometimes

wary of possible retrieval errors with
MESH since data entry is subject to
human error
To solved this problem expert users
sometimes combine MESH and FREE
TEXT search

STEP ID- NARROW DOWN THE YIELD OF

THE SEARCH BY USING THE INTERSECT
OF THE EXPANDED SETS
In step 1c we are assuring the retrieval of all
relevant articles but we face the danger most of
articles we might not even need.
The final step in developing an initial strategy, is
to narrow the yield by searching for the intersect
of the concepts that were previously expanded.
Ex: [carcinoma or neoplasm or tumor or
lnewgrowth ] AND [ lung or pulmonary or
bronchogenic] AND Randomized Controlled Trial
[ MESH search]

 STEP 2-RUN THE SEARCH AND MODIFY IF

NECESSARY TO OBTAIN AN OPTIMAL
STRATEGY
STEP 3 SAVE THE SEARCH STRATEGY AND
RUN IT THROUGH THE OTHER DISC
STEP 4 DOWNLOAD THE YIELDED ARTICLES
ON A FLOPPY DISC AND SCAN THEM IN
ANOTHER COMPUTER