David Sackett

e- B a s e d
E v i d e n c
u ct io n t o a l
In tr od A p pr a is
C ri t ica l
c in e an d
Medi
u s Su b an ad a
Ida Bag h i l d H ealth, pi t a l , Denpasar
ent of C s
Departm a - S a nglah Ho
n
i v ers i t as Udaya
a l Sch ool of U n
Medic
 David Sackett

What is Evidence-Based
Medicice (EBM) ?
“EBM is the conscientious, explicit, and
judicious use of current best evidence in
making decisions about the care of individual
patients” David Sackett, McMaster University; 2001

ATAU
“kerangka untuk menerapkan bukti mutakhir
yang sahih dalam tata laksana pasien”
Sastroasmoro S. Evidence-based Medicine. Jakarta: Sagung Seto; 2014.
 David Sackett

What is Evidence-Based
Medicice (EBM) ?
“Evidence‐based medicine is the use of mathematical
estimates of the risk of benefit and harm, derived from
high‐quality research on population samples, to inform
clinical decision‐making in the diagnosis, investigation
or management of individual patients”

Greenhalgh T. How to Read a Paper. The Basic of Evidence-Based Medicine and

Healthcare. 6th ed. Oxford: Wiley Blackwell; 2019
 David Sackett

The term:
Clinical Epidemiology
‘Evidence-Based Medicine’

Referred to

Evidence-Based Practice (EBP)

 David Sackett

British epidemiologist ARCHIE COCHRANE:

The fact that the most treatment-related decisions
were not based on a systematic review of clinical
research.
But based on :
1. Variable quality scientific literature
2. Expert opinion
3. Trial and error
 David Sackett

Corticosteroid for preterm birth

1972
An RCT was published : improved outcomes for preterm babies when
mother given a short course of corticosteroid
1972-1989
6 more were published, all confirming the 1972 finding
But during this time, most obstetrician were still unaware did not treat
women who have a preterm birth
1989
The 1st Systematic review of corticosteroid treatment was published.
1989-1991
Seven more studies were published
 David Sackett

Conclusion
o Corticosteroid reduces the odds babies dying from
complications of immaturity by 30-50%.
o Thousands of babies have died or suffered
unnecessarily since 1972 because doctor did not
know about the effectiveness of the treatment in
1972 trial.
TEXTBOOKS ARE ALWAYS ABOUT
5-10 YEARS OUT OF DATE
 David Sackett
“Evidence-Based Medicine is the integration
of best research evidence with clinical
expertise and patient values”
David Sackett, McMaster University; 2001

Penggabungan dari:
1. Bukti-bukti yg sahih dari penelitian.
2. Kompetensi dokter sebagai pemberi jasa pelayanan
kesehatan.
3. Nilai-nilai dan pilihan pasien (ketersediaan obat,
fasilitas medis dan penunjang yg diperlukan, dan
biaya).
Sastroasmoro S. Evidence-based Medicine. Jakarta: Sagung Seto; 2014.
 David Sackett
“Evidence-Based Medicine is the integration
of best research evidence with clinical
expertise and patient values”

Clinical expertise

EBM
Best research evidence Patient values

Diagram Venn of EBM

David Sackett, McMaster University; 2001
 David Sackett

Why do we need EBM ?

There is a large though variable gap between what we

know from research and what we do in clinical
practice.

Because so much research is published – some valid

and some invalid – clinicians understandably are
unaware of most of it, or do not have the ‘tools’ to
assess its quality.

Glasziou P, et al. EBP workbook. 2nd ed. Oxford: Blackwell; 2007

 David Sackett

The amount of medical research

Glasziou P, et al. EBP workbook. 2nd ed. Oxford: Blackwell; 2007
 David Sackett
Review the World Literature Fortnightly*
5,000?
2500000
per
day
Medical Articles per Year

2000000

1500000 1,400
55 per
1000000
per day
500000 day

0
Trials MEDLINE BioMedical
Finding the high quality evidence is like trying to sip pure water
from a hose pumping dirty water or looking for rare pearl
 David Sackett

Doctor’s
knowledge Patient
of evidence, values,
Clinical decisions concerns,
skills,
attitude expectations

Health system
access rules Concern about
(PBS, Medicare, litigation
funding, etc)
 David Sackett

How did EBM help ?

1. EBM can help you find information you need.

2. EBM helped because the empirical data were easy

for the patient to understand and he could
participate in the clinical decision

3. EBM can help to reduce litigation

4. EBM helps doctors keep up to date across a very

wide spectrum of information
 David Sackett

Case study : dog bite

Case:
A patient came to the clinic with a fresh dog bite. It
looked clean. The GP & patient wondered
whether it was necessary to give prophylactic
antibiotic.
Searched MEDLINE
Meta-analysis:
Indicating that the average infection rate for dog bites
was 14%
 David Sackett

Empirical measures of outcome

Outcome measured as:

Absolute Risk Reduction (ARR)
Relative Risk (RR )
Number Needed to Treat (NNT)
 David Sackett

The risk of infection after dog bite without Antibiotics = 14%

(0,14)

The risk of infection after dog bite with Antibiotics =7% (0,07)
ARR for Antibiotics:
14% - 7%= 7%
(7 people in every 100 treated will be saved from infection)

NNT = 100/7 = 14
(Need to treat 14 dog bite patients with AB to prevent 1 infection)

RR of infection with AB compared to without AB

RR = 0,07 / 0,14 = 0.5
 David Sackett

STEPS IN EBM
1. Formulate an answerable question

2. Track down the best evidence of outcomes available

3. Critically appraise the evidence (find out how good it is

and what it means)

4. Apply the evidence : integrate the results with clinical

expertise and patient values

5. Evaluate our performance

 David Sackett

EBM step 1 : Formulate an answerable question

 David Sackett

Background and foreground questions

 David Sackett

Background questions

Ask for general knowledge about a condition, test, or

treatment

Have two essential components:

A question root (who, what, where, when, whay, and how) and
1
a verb

2 A disorder, test, treatment, or other aspect of health care

Example: a. How does pneumonia cause pleural effusions?

b. What causes COVID-19?
 David Sackett

Foreground questions

Ask for specific knowledge to inform clinical decisions

or actions

Have four essential components:

1 P: Patients, population, or problem

2 I: Intervention, exposure, tetst, or other agent

3 C: Comparison intervention, exposure, test, etc

4 O: Outcomes of clinical importance

 David Sackett

Example:

In patients with heart failure and reduce systolic function, would adding
implantation of an electronic resynchronization device to standard therapy
reduce morbidity or mortality over 3 to 5 years?
 David Sackett

Gap between “das Sollen” and “das Sein”

Problem

Problem solving: formulate an answerable question

The “PICO” (pronounced “pee-co”) principle

P : Patient/ population/ problem
I : Intervention/ indicator/ index text
C : Comparator/ control
O : Outcome
 Example
David Sackett
Julie is pregnant for the second time. She had her first baby when she was 33 and had
amniocentesis to find out if the baby had Down syndrome. The test was negative but it
was not a good experience, because she did not get the result until she was 18 weeks
pregnant. She is now 35 and 1 month pregnant, and ask if she can have a test that
would give her an earlier result. The local hospital offers serum biochemistry plus
nuchal translucency ultrasound screening as a first trimester test for Down syndrome.
You wonder if this combination of test is as reliable as conventional amniocentesis.

PICO principle
P Population/problem : pregnant women (first trimester)
I Index test : serum biochemistry plus nuchal translucency
ultrasound screening
C Comparator/control : conventional amniocentesis
O Outcome : accurate diagnosis (sensitivity and
specificity) of Down syndrome
Glasziou P, et al. EBP workbook. 2nd ed. Oxford: Blackwell; 2007
 David Sackett

Clinical question:
For pregnant women, is serum biochemistry plus
nuchal translucency ultrasound screening testing
in the first trimester as accurate (equal or better
sensitivity and specificity) as conventional
amniocentesis for diagnosing Down syndrome?

Glasziou P, et al. EBP workbook. 2nd ed. Oxford: Blackwell; 2007

 David Sackett

Question Best study designs

Intervention RCT
Aetiology and risk factors RCT, Cohort, Case-control
Frequency and rate Cohort, Cross-sectional
Diagnosis Cross-sectional with random
or consecutive sample

Prognosis and prediction Cohort/survival study

Straus SE, et al. EBM, 3 rd ed. London: Elsevier; 2005

 David Sackett

Question Question type Best Study design

What should I do about this INTERVENTION RCT
condition or problem?
What causes the problem? AETIOLOGY AND RISK RCT, Cohort study, Case-
FACTORS control study

How common is the FREQUENCY AND RATE Cohort study, Cross-

problem? sectional study

Does this person have the DIAGNOSIS Cross-sectional study

condition or problem with random or
consecutive sample

Who will get the condition PROGNOSIS AND Cohort/survival study

or problem PREDICTION
 David Sackett
Clinical question:
For pregnant women, is serum biochemistry plus nuchal
translucency ultrasound screening testing in the first
trimester as accurate (equal or better sensitivity and
specificity) as conventional amniocentesis for diagnosing
Down syndrome?

Step 2:Track down the best evidence (searching of

article)

Step 3 and 4: Critically appraise and apply the

evidence
Step 5: Evaluate our performance
 David Sackett

EBM step 2 : Track down the best evidence

o What study designs should you look for

(interventions, aetiology and risk factors, diagnosis,
prognosis, frequency, and rate)?

o How to recognize different study types [experimental,

observational (cohort, case-control, cross-sectional
studies, case report, case series), analytical,
descriptive]?
 David Sackett

o Where to search (PubMed, The Cochrane Library,

Medline, and others)?

o Searching tips and tactics [AND, OR, NEAR, NOT,

parentheses, * (truncation), ti, so]
 David Sackett

EBM Step 3 : Critically appraise the evidence

A careful look at the study or studies

and decide how good they are for
answering the clinical question.
 David Sackett

Principles of critical
appraisal – primary research

1. What is the PICO of the study, and is it close

enough to your PICO?
2. How well was the study done?
3. What do the results mean and could they have
been due to chance?
 David Sackett

1. Is the PICO of the study close enough to your

PICO?
o When you find a study that you think will help to
answer your clinical question, the first thing to look at
is whether the PICO of the study matches the PICO of
your question.

o Quickly decided if you want to appraise it further or

not.
 David Sackett

2. How well was the study done?

R Recruitment :
were the subjects representative of the target population?

A Allocation or Adjustment :
Was the treatment allocation concealed before randomization and were
the groups comparable at the start of the trial?

M Maintenance :
Was the comparable status of the study groups maintained through equal
management and adequate follow-up?

M Measurement :
Were the outcomes measured with :

b blinded subjects and assessor and/or

o objective outcomes?
 David Sackett
Study question Study design Aim Study methods Critical
Appraisal
Source
P Fair recruitment
Subjects
Recruit a large
enough sample R
representative of the PLUS recruit subject
Subjects target population randomly OR recruit
consecutive patient

I Fair allocation Allocate randomly to

groups A
Comparable study OR
groups Adjust for
confounding

C Fair Maintenance Manage groups

equally M
Follow up all subjects
and assess relevant
I outcomes in the
C starting groups
+
O Fair Measurement
Valid and unbiased
Measure outcomes
with: blinded subjects M
outcome measures and assessors and/or
- objective measures b
o
 David Sackett

3. What do the results mean?

Effect size and confidence intervals (CI ~ estimation)

CI is an estimate of the range of values that are likely to

include the real value
Usually CI are quoted as 95% CI
P-values (hypothesis testing)

Measure the probability that a result is purely due to

chance
P < 0,05 ~ statistically significant
 Outcome measures for binary outcomes
Measure
Relative Risk (RR)
= risk of outcome in the
treatment group/risk of
event in the control group
Absolute risk reduction
(ARR)
= risk of event in the control
group – risk of event in the
treatment group
Relative risk reduction
(RRR)
= ARR/risk of event in
control group
(or 1 – RR)
Number needed to treat
(NNT)
= 1/ARR
Meaning Example
David Sackett
RR tells us how many times more likely it is that an RR = 0,1/0,15 = 0,67
event will occur in the treatment group relative to the
control group Since this RR < 1 , the treatment
RR = 1 means that there is no difference between decreases the risk of death
the 2 groups
RR < 1 means that the treatment reduces the risk of
the event
RR> 1 means that the treatment increases the risk
of the event
ARR tells us the absolute difference in the rates of ARR = 0,15 – 0,10 = 0,05 (5%)
events between the two groups and given an
indication of the baseline risk and treatment effect The absolute benefit of treatment
ARR = O means that there is no difference between is a 5% reduction in the death
two groups rate
ARR positive means that the treatment is beneficial
ARR negative means that the treatment is harmful
RRR tells us the reduction in rate of the event in the RRR = 0,05/0,15 = 0,33 (33%)
treatment group relative to the rate in the control OR
group 1 – 0,67 = 0,33 (33%)
RRR is probably the most commonly reported
measure of treatment effects
NNT tells us the number of patients we need to treat NNT = 1/0,05 = 20
in order to prevent 1 bad event We would need to treat 20
people for 2 years in order to
prevent 1 death
 Statistically Clinically
significant important

= confidence interval

= point estimate OR
effect size

Minimum clinical
Important difference

Null hypothesis
(no effect)
(a) (b) (c) (d)
(a) Difference is clinically important and statistically significant
(b) Difference is clinically important but not statistically significant
(c) Difference is not clinically important but statistically significant
(d) Difference is not clinically important and not statistically significant
 David Sackett

RAMMbo
INTERVENTION PROGNOSTIC DIAGNOSTIC FREQUENCY

R R R R

A = allocation A = adjustment (-) (-)

M M M M

M M M M

b b b b

o o o o
 David Sackett

EBM Step 4 : Apply the evidence : integrate the results

with clinical expertise and patient values

The questions that you should ask before you decide

to apply the results of the study to your patient
are:
1. Is the instrument or test are feasible in my
setting?
2. What else do I need to apply this evidence?
3. What alternatives are available?
 David Sackett

EBM Step 4: Apply the evidence : integrate the results

with clinical expertise and patient values

4. Is my patient so different to those in the study

that the result can not apply at all?
5. Will the potential benefits of treatment outweigh
the potential harms of treatment for my patient?
6. What does my patient think about it?
(%)

B David Sackett
e Lower risk patients Higher risk patients
n
e
f 8
i
t
Trial patients Note:
Benefit
o 6
f Harm
t Threshold
r
e
4
a
t
m
2
e
n
t
0
0 5 10 15
Risk of outcome (%)
 David Sackett

The “f method” for estimating your

patient’s risk:
● If your patient is twice as susceptible as
those in trial, f = 2
● If your patient is half as susceptible as
those in the trial, f = 0.5

The NNT for your patient = NNT (trial)/f

 David Sackett

EBM Step 5: Evaluate our performance

1. Informal: evaluate the outcome of EBM practice in

each patient.
2. Formal : evaluate the input, process, and outcome
 David Sackett

Pasien dg
masalah

Terapkan Formulasikan
“evidence” permasalahan
klinis

Telaah kritis Cari

“evidence” “evidence”

Gambar. Siklus EBM

 David Sackett

Level of evidence
LEVEL INTERVENTION DIAGNOSIS PROGNOSIS AETIOLOGY
I Systematic review of Systematic review of Systematic Systematic
level II studies level II studies review of level II review of level II
studies studies

II RCT Cross-sectional - Inception cohort Prospective

consecutive study cohort study

III One of the following: One of the following: One of the One of the
•Non-randomized •Cross-sectional – following: following:
experimental study non-consecutive •Untreated •Retrospective
•Cohort study, case- •Diagnostic case- control – RCT cohort study
control study control •Retrospectively •Case-control
cohort study study

IV Case series Case series Case series, Cross-sectional

cohort study at
different stages
of disease
 Level of evidence
Level of Therapy/prevention/etiology/ harm
evidence
1a Systematic review of homogeneity of RCTs
1b Individual RCT with narrow CI
1c All or none
2a Systematic review (with homogeneity) of cohort studies
2b Individual cohort study (including low-quality RCT; e.g. <80%
f-up)
3a Systematic review (with homogeneity) of case-control study
3b Individual case-control study
4 Case series (and poor quality cohort and case-control study)
5 Expert opinion
 Grade of recommendation
A Consistent level 1 studies

B Consistent level 2 or 3 studies OR extrapolations from levels 1 studies

C Level 4 studies OR extrapolations from level 2 or 3 studies

D Level 5 evidence OR troublingly inconsistent OR inconclusive studies at any

level

Straus SE, et al. EBM, 3 rd ed. London: Elsevier; 2005

 David Sackett

References

1. Glasziou P, et al. EBP workbook. 2nd ed. Oxford: Blackwell; 2007

2. Sastroasmoro S. Evidence-based Medicine. Jakarta: Sagung Seto;

2014.

3. Greenhalgh T. How to Read a Paper. The Basic of Evidence-Based

Medicine and Healthcare. 6th ed. Oxford: Wiley Blackwell; 2019

4. Straus SE, et al. Evidence-Based Medicine. How to Practice and Teach

EBM. 5th ed. Edinburgh: Elsevier; 2019
David Sackett
David Sackett
 David Sackett

Latihan

Skenario 1

Kita menghadapi anak dg defisiensi growth hormone (GH),

apakah pemberian terapi GH eksogen dapat memacu
pertumbuhan anak tersebut?

Buat pertanyaan klinis berdasarkan prinsip PICO !!!!!

 David Sackett

Latihan

Skenario 2

Kita menghadapi perempuan muda yang telah didiagnosis

dengan lupus eritematosus sistemik. Pasien tersebut pernah
mengalami komplikasi gagal jantung 2 bulan yang lalu. Kita
ditanya oleh keluarga tentang harapan kesembuhan pasien
tersebut

Buat pertanyaan klinis berdasarkan prinsip PICO !!!!!

 David Sackett

Latihan

Skenario 3

Pada suatu studi kohort ditemukan insidens batuk kronik

berulang (BKB) pada pekerja pabrik tekstil sebesar 16%,
sedangkan pada pekerja pabrik minuman sebesar 6%. Risiko
relatif (RR) yang didapat adalah 2,67 (IK95% 1,8 sampai 4,4).

1. Bagaimana interpretasi dari nilai RR tersebut ?

2. Apakah nilai RR tersebut penting secara klinis ?
3. Bagaimana presisi estimasinya?
 David Sackett

Latihan

Skenario 4

Suatu uji klinis acak terandomisasi, dari 100 subyek yang diberi
obat standar sembuh 60%, sedangkan yang diberi obat baru
sembuh 75%.

1.Berapa nilai NNT yang didapat ?

2. Apa arti nilai NNT tersebut ?
David Sackett