Study Guide of Evidence-based Medical Practice Block
CURRICULUM
EVIDENCE-BASED MEDICAL PRACTICE
Aims:
1. To develop basic information skills and their integration with the evidence-based
practice in the primary care setting
2. To develop skills to obtain, appraise and use valid and reliable new information using
on-line resources
3. To develop skills to convey electronic and oral communication
Learning Outcomes:
1. Possess skills to gain access to on-line resources
2. Able to critically appraise medical literatures
3. Able to keep patient’s medical records and comprehend ethical and legal
imperatives
4. Able to communicate with colleagues and co-workers using oral, written,
electronic means
and
Curriculum Contents:
1. Internet searching
2. Association and causation
3. Principles and applications of statistical analysis
4. Effect size, hypothesis testing and confidence interval
5. Principle of critical appraisal (diagnostic test, clinical trial, prognosis study)
6. Record keeping and Clinical Practice
7. Presentation at the meeting.
Udayana University Faculty of Medicine, MEU
1
Study Guide of Evidence-based Medical Practice Block
PLANNERS TEAM
No
1
2
Name
Department
Phone
Prof. Dr. dr. I Gde Raka Widiana,
SpPD-KGH (Coordinator)
Internal Medicine
0816297956
Ida Bagus Nyoman Putra Dwija,
Microbiology
08179747502
S.Si.,M.Biotech
3
Dr. dr. I P. G. Adiatmika, Mkes
Physiology
08123811019
4
dr. Lanang Sidiartha, SpA
Pediatrics
08123966008
5
dr. I.B. Subanada, SpA
Pediatrics
08123995933
Department
Phone
Internal Medicine
0816297956
LECTURERS
No
Name
1.
Prof. Dr. dr. I Gde Raka Widiana,
SpPD-KGH
2.
dr. Dewi Sutriani Mahalini, SpA
Pediatrics
08123641466
3.
dr. Eka Gunawijaya, Sp A
Pediatrics
081338599801
4.
dr. Lanang Sidiartha, SpA
Pediatrics
08123966008
5.
dr. I.B. Subanada, SpA
Pediatrics
08123995933
6.
Dr. dr. I P. G. Adiatmika, Mkes
Physiology
08123811019
7.
dr. I Wyn. Sudhana, SpPD-KGH
Internal Medicine
08123914095
8.
Dr.dr. Ketut Suega,Sp.PD-KHOM
Internal Medicine
081338728421
Udayana University Faculty of Medicine, MEU
2
Study Guide of Evidence-based Medical Practice Block
FACILITATORS
Regular Class (Class A)
Veue
No
Name
Group
Departement
Phone
(3rd floor)
1
dr. I Putu Adiartha Griadhi,M.Fis
1
Physiology
081999636899
3nd floor:
R.3.09
2
dr. I Gusti Ayu Eka
Pratiwi,M.Kes.,Sp.A
2
Pediatric
08123920750
3nd floor:
R.3.10
3
dr. Gede Kambayana,Sp.PD-KR
3
Interna
08124683416
3nd floor:
R.3.11
4
dr. Ni Nyoman Margiani, Sp.Rad
4
Radiology
081337401240
3nd floor:
R.3.12
5
dr.Ni Putu Sri Widnyani,Sp.PA
5
Pathology
Anatomy
081337115012
3nd floor:
R.3.13
6
dr. A.A.Bagus Ngurah Nuartha,
SpS.(K)
6
Neurology
08123687288
3nd floor:
R.3.14
7
dr. Anak Agung Wiradewi
Lestari,Sp.PK
7
Clinical
Pathology
08155237937
3nd floor:
R.3.15
8
dr. Nyoman Paramita Ayu, Sp.PD
8
Interna
08123837372
3nd floor:
R.3.16
9
dr. Nyoman
Suryawati,M.Kes.,Sp.KK
9
Dermatology
0817447279
3nd floor:
R.3.17
10
dr. Kunthi Yulianti, SpKF
10
Forensic
081338472005
3nd floor:
R.3.19
Udayana University Faculty of Medicine, MEU
3
Study Guide of Evidence-based Medical Practice Block
English Class (Class B)
Veue
No
Name
Group
Departement
Phone
(3rd floor)
1
dr. A.A. Ngurah Subawa,M.Si
1
Clinical
Pathology
2
Dr.dr.Bagus Komang
Satriyasa,M.Repro
2
Pharmacology
3
dr. Pontisomaya Parami,Sp.An
3
Anesthesiology
4
dr. Pratihiwi
Primadarsini,M.Biomed,Sp.PD
4
Interna
5
dr. Putri Ariani Sp.KJ
5
Psychiatry
6
IBN. Putra
Dwija,S.Si.,M.Biotech
6
Microbiology
7
dr. I Made Agus Kresna
Sucandra,Sp.An
7
Anesthesiology
8
dr. Putu Budiastra,Sp.M (K)
8
Ophthalmology
9
dr. Putu
Patriawan,M.Sc.,Sp.Rad
9
Radiology
10
Desak Gde Diah Dharma
Santhi, S.Si, Apt, M.Kes
10
Clinical
Pathology
Udayana University Faculty of Medicine, MEU
08155735034
3nd floor:
R.3.09
087777790064
3nd floor:
R.3.10
08123661312
3nd floor:
R.3.11
081805530196
3nd floor:
R.3.12
082237817384
3nd floor:
R.3.13
08179747502
3nd floor:
R.3.14
08123621422
3nd floor:
R.3.15
085238238999
3nd floor:
R.3.16
08123956636
3nd floor:
R.3.17
0817569021
3nd floor:
R.3.19
4
Study Guide of Evidence-based Medical Practice Block
TIME TABLE
REGULAR CLASS
Day/Date
1
Time
Activity
Person-incharge
09.00 – 10.00
Introduction
Class
Room
Prof. Raka
10.00 - 12.30
Scenarios : Problems
identifications with clinical
questions
Class
Room
dr. Dewi
12.30 – 13.00
Break
13.00 – 16.00
Practical Work 1.
Computer
room
Dr.
Adiatmika
Tuesday
May,19
Venue
2015
Record Keeping
Computer
room
09.00 – 10.00
Lecture 1. Principal of critical
appraisal
Class
Room
dr.Eka
10.00 – 11.00
Lecture 2. Association and
Causation
Class
Room
dr.
Subanada
May,20
11.00 – 12.00
Individual Learning
2015
12.00 – 13.00
Student Project 1
Class
Room
dr. Lanang
13.00 – 13.30
Break
13.30 – 15.00
SGD
Discussion
Room
Facilitator
15.00 – 16.00
Plenary Session
Class
Room
dr. Eka,
dr.Subanada
09.00 – 16.00
Practical Work 2.
Computer
room
Dr.
Adiatmika
Class
Room
dr.
Subanada
Class
Room
dr. Lanang
2
Wednesday
3
Thursday
Searching articles (address
will be given)
May,
21.2015
09.00 – 10.00
Lecture 3: Effect size,
Hypothesis Testing and
Confidence Interval
10.00 – 11.00
Individual Learning
11.00 – 12.00
Lecture 4. Principles and
Application of Statistical
Analysis
4
Udayana University Faculty of Medicine, MEU
5
Study Guide of Evidence-based Medical Practice Block
Friday
12.00 – 13.00
Student Project 2
May,22.
13.00 – 13.30
Break
2015
13.30 – 15.00
SGD
15.00 – 16.00
Discussion
Room
Fasilitator
Plenary Lecture 3,4
Class
Room
Lecture
09.00 – 10.00
Lecture 5: Methodological
and Statistical Principles and
Application in Descriptive
Studies
Class
Room
dr. Eka
10.00 – 11.30
Student Project 3
11.30 – 12.00
Break
12.00 – 13.30
Individual Learning
13.30 - 15.00
SGD
Discussion
Room
Facilitators
15.00 – 16.00
Plenary Lecture
Class
Room
dr.Eka
09.00 -10.00
Lecture 6. Methodological
and Statistical Principles and
Application in Analytical
Studies
Class
Room
dr. Lanang
10.00 -11.30
Student Project 4
11.30 -12.00
Break
12.00 – 13.30
Individual Learning
13.30 – 15.00
SGD
Discussion
Room
Facilitators
15.00 – 16.00
Plenary session
Class
Room
dr. Lanang
09.00 – 10.00
Lecture 7. Diagnostic Test
7
10.00 – 11.30
Student Project 5
Wednesday
11.30 -12.00
Break
May,27.
12.00 – 13.30
Individual Learning
2015
13.30 - 15.00
SGD
15.00 – 16.00
Plenary session
5
Monday
May,25
2015
6
Tuesday
May,26
2015
Udayana University Faculty of Medicine, MEU
Dr. Sudhana
Discussion
Room
Facilitators
Class
Room
Dr. Sudhana
6
Study Guide of Evidence-based Medical Practice Block
09.00 – 10.00
Lecture 8. Clinical Trial
10.00 – 11.30
Student Project 6
11.30 – 12.00
Break
12.00 - 13.30
Individual Learning
13.30 – 15.00
SGD
15.00 – 16.00
Class
Room
Prof Raka
Discussion
Room
Facilitator
Plenary session
Class
Room
Prof Raka
09.00 – 10.00
Lecture : 9. Study about
Prognosis
Class
Room
dr. Eka
10.00 – 11.30
Student Project 7
11.30 – 12.00
Break
12.00 – 13.30
Individual Learning
13.30 – 15.00
SGD
Discussion
room
Facilitators
15.00 – 16.00
Plenary session
Class
Room
dr. Eka
09.00 – 10.00
Lecture: 10. How to Write a
Paper and Present at a
Meeting
Class
Dr. Suega
8
Thursday
May,28
2015
9
Friday
May,29
2015
10
Monday
10.00 – 11.30
Student Project 8
June,1.
11.30 – 12.00
Break
2015
12.00 – 13.30
Individual Learning
13.30 – 15.00
SGD
15.00 – 16.00
Plenary session
Room
Discussion
room
Facilitator
Class
Room
Dr. Suega
11.
Wednesday
Examination
June,3
2015
Udayana University Faculty of Medicine, MEU
7
Study Guide of Evidence-based Medical Practice Block
English Class
Day/Date
Time
08.00 – 09.00
Activity
Introduction
1
Practical Work 1.
Tuesday,
May 19,
09.00 – 12.00
Record Keeping
Person-incharge
Class
Room
Prof. Raka
Computer
room
Dr.
Adiatmika
Computer
room
2015
2
Venue
12.00 – 12.30
Break
12.30 - 15.00
Scenarios : Problems
identifications with clinical
questions
08.00 – 15.00
Practical Work 2.
Wednesday,
May 20,
Searching articles (address
will be given)
Class
Room
dr. Dewi
Computer
room
Dr.
Adiatmika
2015
08.00 – 09.00
Lecture 1. Principal of critical
appraisal
Class
Room
dr.Eka
09.00 – 10.00
Lecture 2. Association and
Causation
Class
Room
dr.
Subanada
10.00 – 11.00
Individual Learning
11.00 – 12.30
SGD
Discussion
Room
Facilitator
12.30 – 13.00
Break
13.00 – 14.00
Student Project 1
Class
Room
dr. Lanang
14.00 – 15.00
Plenary Session
Class
Room
dr. Eka,
dr.Subanada
08.00 – 09.00
Lecture 3: Effect size,
Hypothesis Testing and
Confidence Interval
Class
Room
dr.
Subanada
09.00 – 10.00
Individual Learning
10.00 – 11.00
Lecture 4. Principles and
Application of Statistical
Analysis
Class
Room
dr. Lanang
11.00 – 12.30
SGD
Discussion
Facillitator
3
Thursday
May 21,
2015
4
Friday
May 22,
2015
Udayana University Faculty of Medicine, MEU
8
Study Guide of Evidence-based Medical Practice Block
Room
12.30 – 13.00
Break
13.00 – 14.00
Student Project 2
14.00 – 15.00
Plenary Lecture 3,4
Class
Room
dr.Lanang;
dr.Subanada
08.00 – 09.00
Lecture 5: Methodological
and Statistical Principles and
Application in Descriptive
Studies
Class
Room
dr. Eka
09.00 – 10.30
Individual Learning
10.30 – 12.00
SGD
Discussion
Room
Facilitators
12.00 – 12.30
Break
12.30 – 14.00
Student Project 3
14.00 – 15.00
Plenary Lecture
Class
Room
dr.Eka
08.00 – 09.00
Lecture 6. Methodological
and Statistical Principles and
Application in Analytical
Studies
Class
Room
dr. Lanang
09.00 – 10.30
Individual Learning
10.30 – 12.00
SGD
Discussion
Room
Facilitators
12.00 – 12.30
Break
12.30 – 14.00
Student Project 4
14.00 – 15.00
Plenary session
Class
Room
dr. Lanang
7
08.00 – 09.00
Lecture 7. Diagnostic Test
Wednesday
09.00 – 10.30
Individual Learning
May,27
10.30 – 12.00
SGD
12.00 – 12.30
Break
12.30 – 14.00
Student Project 5
14.00 – 15.00
Plenary session
5
Monday
May 25
2015
6
Tuesday
May 26
2015
2015
Udayana University Faculty of Medicine, MEU
Dr. Sudhana
Discussion
Room
Facilitators
Class
Room
Dr. Sudhana
9
Study Guide of Evidence-based Medical Practice Block
08.00 – 09.00
Lecture 8. Clinical Trial
09.00 – 10.30
Individual Learning
10.30 – 12.00
SGD
12.00 – 12.30
Break
12.30 – 14.00
Student Project 6
14.00 – 15.00
Class
Room
Prof Raka
Discussion
Room
Facilitator
Plenary session
Class
Room
Prof Raka
08.00 – 09.00
Lecture : 9. Study about
Prognosis
Class
Room
dr. Eka
09.00 – 10.30
Individual Learning
10.30 – 12.00
SGD
Discussion
room
Facilitators
12.00 – 12.30
Break
12.30 – 14.00
Student Project 7
14.00 – 15.00
Plenary session
Class
Room
dr. Eka
08.00 – 09.00
Lecture: 10. How to Write a
Paper and Present at a
Meeting
Class
Dr. Suega
8
Thursday
May 28
2015
9
Friday
May 29
2015
10
Monday
09.00 – 10.30
Individual Learning
June 1
10.30 – 12.00
SGD
2015
12.00 – 12.30
Break
12.30 – 14.00
Student Project 8
14.00 – 15.00
Plenary session
Room
Discussion
room
Facilitator
Class
Room
Dr. Suega
11.
Wednesday
EXAMINATION
June 3
2015
Udayana University Faculty of Medicine, MEU
10
Study Guide of Evidence-based Medical Practice Block
Note :
: Class Room 3.01, 3rd floor
1. Lecture
2. Small Group Discussion : Discussion Room 3rd floor,west wing (Room 3.01-3.08)
and Room 3.21-3.24 beside the IT Room at 3rd Floor
3. Examination
: Class Room 1, 4th floor, east wing (R 401) & Discussion
Room ,3rd floor,west wing & Multi-function Laboratory,4th
floor.
4. For activity of practical work record keeping and searching articles students have to
bring their own laptop, because the number of the personal computers in the
computer room is not enough.
MEETING
Meeting of Student Representatives
Meeting of the planners team with the student representatives (Regular and English class)
will be held on Monday,May 25.2015, from 10.00 - 11.00 in the class room. It is hoped that
the planners’ team will get some inputs and suggestions from the student representatives to
improve the next implementation of the program. For the meeting each discussion group
must choose one of his members as their representative.
Meeting of Facilitators
All facilitators will be invited to discuss all the block activities with the planners team on
Monday, May.25.2015 from 11.00 – 12.00 in the class room.
ASSESSMENT METHOD
The student final assessment (CBT) will be held on Wednesday, June,3.2015. The time of
examination will be informed letter. The minimal passing level is 70.
Udayana University Faculty of Medicine, MEU
11
Study Guide of Evidence-based Medical Practice Block
LEARNING PROGRAMS
CURRICULUM
EVIDENCE-BASED MEDICAL PRACTICE
INTRODUCTION TO THE BLOCK:
An Introduction to Evidence-Based Medical Practice
Curriculum and Evidence- Based Medicine
Raka Widiana
Aims:
1. To develop basic information skills and their integration with the evidence-based
practice in the primary care setting
2. To develop skills to obtain, appraise and use valid and reliable new information using
on-line resources
3. To develop skills to convey electronic and oral communication
Learning Outcomes:
1. Possess skills to gain access to on-line resources
2. Able to critically appraise medical literatures
3. Able to keep patient’s medical records and comprehend ethical and legal
imperatives
4. Able to communicate with colleagues and co-workers using oral, written, and
electronic means
Curriculum Contents:
1. Internet searching
2. Association and causation
3. Principles and applications of statistical analysis
4. Effect size, hypothesis testing and confidence interval
5. Principle of critical appraisal (diagnostic test, clinical trial, prognosis study)
6. Record keeping and Clinical Practice
7. Presentation at the meeting.
Udayana University Faculty of Medicine, MEU
12
Study Guide of Evidence-based Medical Practice Block
ABSTRACTS (I Gde Raka Widiana)
Since the early 1980s, medicine has been undergoing a continuing revolution. It has led to e
fertile development of EBM (evidence based medicine). The practice of EBM (evidence
based medical practice) may be perceived as a meta-field because it involves concepts and
tools from many disciplines, including statistics and bio-statistics, research design, computer
programming, database management, and mathematical modeling. EBMP applies medical
informatics which is used for the purposes of generating, organizing, and making accessible
and intelligible huge amounts of information. Learning the skills to manage information is of
paramount importance to modern physicians. The computer is an important tool to facilitate
this process. Those who learn the skills of using computers to manage information will have
a greater advantage over those who do not. In daily practice, EBMP is important to the
medical students because can help them to deal with problems which include how to find,
appraise, procure, apply and store the best evidence to diagnose, treat, and determine the
prognosis of patients. Skills in this area may help them to pursue life long, students
centered and problem based education. The teaching of EBMP consists of lecture,
searching articles, critical appraisal and application to problems that may be introduced in
clinical scenarios
SELF DIRECTING LEARNING
Basic knowledge that must be known:
1. Search related articles to the patients problems (diagnose, treat, and prognosis) in
the internet,
2. Critically appraise the related articles and procure from the internet
3. Apply the articles to patents problems
4. Store the best evidence to of patients in your personal electronic library.
SCENARIO
A 25 year old woman was consulted to a nephrologist with lupus GN. The patient had
been treated with methyl prednisolon for 3 months  failed to get remission.The patient
has not married  hope to get pregnant in the future (contraindicated for CYP).The
doctor knew MMF, a promising drug for LGN available in the market. However the
doctor was not sure the drug is save and effective  he would like to find out best
evidence in the internet, but he was not so familiar with EBM and searching in the
internet. Please, help the doctor to find the best evidence to answer his problem
Learning Task:
1. Comprehend above scenario
2. Make clinical (foreground question) using acronym PICO
3. Search articles about therapy (randomized clinical trial) in internet (using
Highwire or other addresses)
4. Procure 3 related articles
Udayana University Faculty of Medicine, MEU
13
Study Guide of Evidence-based Medical Practice Block
5. Appraise and select one best article
6. Apply whenever you find valid and important best evident article
Self Assessment:

How to make clinical (foreground question) using acronym PICO using above
clinical scenario

How to search articles about therapy (randomized clinical trial) in internet
(using Highwire or other address)

How to procure related articles

How to appraise and select the best article

How to apply the valid and important best evident article to your patient’s
problems
LEARNING OBJECTIVE
Comprehend and skillful making clinical (foreground question) using acronym PICO,
searching articles in internet, procure related articles, appraise, select the best article and
apply the valid and important best evident article to patient’s problems
Scenarios: Problems Identifications with Clinical Questions
Dewi Sutriani Mahalini
:to Search for The Best Evidence
Evidence based medical practice (EBMP) is the use of the best scientific evidence to
support the clinical decision making. The identification of the best evidence requires the
CLINICAL SCENARIOS TO SEARCH FOR THE BEST
construction of an appropriate research question and review of the literature. Many
EVIDENCE
questions about patient care arise
at the patient bedside. You cannot simply enter your
question directly into a database and expect to get an answer. There are 4 steps in EBMP:
1. Formulate an answerable question; 2. Track down the best evidence outcome available;
3. Critically appraise the evidence; 4. Apply the evidence. The first step of EBMP is to
convert an information need into a focused question. This part of the EBMP process is often
overlooked but is essential if a search is to be conducted efficiently. Questions often spring
to mind in a form that makes finding answers in the medical literature a challenge.
Dissecting the question into its component parts and restructuring it so that it is easy to find
the answers is an essential first step in EBMP.
Udayana University Faculty of Medicine, MEU
EBMP process starts with a clinical
14
Study Guide of Evidence-based Medical Practice Block
scenario that needs the best answer. One way of defining a focused question is to use the
PICO or PECO framework. "PICO" is the acronym for this 4 part question which consists of
the first letters of Patients, Intervention/Exposure, Comparison, Outcome. PICO doesn't
necessarily work perfectly for all kinds of questions, the main thing is the break down your
question into separate concepts, regardless of the headings you put them under. You can
usually identify three of the four PICO elements. By far the most common type of clinical
question is about how to treat a disease or condition. In EBMP, treatments and therapies
are called ‘interventions’ and such questions are questions of INTERVENTION. However,
not all research questions are about interventions. Other types of questions that may arise
are: ETIOLOGY and RISK FACTOR, FREQUENCY, DIAGNOSIS, PROGNOSIS and
PREDICTION. In each case the PICO method can be used to formulate the question.
SCENARIO
Scenario 1:
A medical student, 19 years old, female came to a doctor with chief complaint of mass in her
left breast with 2 cm in diameter. Based on physical examination, the doctor didn’t sure if the
mass was a malignancy or not. The doctor told her to do mammography but she was
worried that mammography will expose her to x ray. She asked for ultrasonography. She
thought that ultrasonography was safer compared to mammography. The problem was the
doctor didn’t know if ultrasonography can accurately diagnosed breast cancer compared to
mammography.
Can you find the answer for scenario above by search for the best evidence from the
internet?
Scenario 2:
A 40 years old male came to a doctor. He had diabetes mellitus since 5 years ago but he
didn’t attend medical visit regularly. This morning, the patient came to laboratory; he wanted
to check for microalbuminuria, based on his friend’s advice. His friend also had diabetes
mellitus. Apparently, he was positive for microalbuminuria. He was told by a nephrologists
that irbesartan can prevent renal failure in diabetes mellitus patients with microalbuminuria.
He asked this doctor whether it was true or not, and if it was true what was the estimation
for the preventive effect. The doctor didn’t have data to answer that question. Could you
help this doctor to search for best evidence from the internet?
Udayana University Faculty of Medicine, MEU
15
Study Guide of Evidence-based Medical Practice Block
Scenario 3:
A marketing staff from a well-known laboratory came to a doctor to offer a homocysteine
serum test for diabetes mellitus patients. This staff said that this new marker, homocysteine,
can predict the mortality rate in patients if the level in serum was high. The doctor asked for
an evidence for that statement. The staff didn’t have the answer for the doctor question.
Could you help the marketing staff to search for the evidence from the internet?
Scenario 4:
A 40 years old man who work as a teacher in department of Agriculture, had diabetes
mellitus for 6 years, came to a doctor. The doctor said that he had mild decreased of renal
function (secondary to diabetic nephropathy). The doctor advised him to have low protein
diet and go to a dietitian to ask for a menu that he needed. He asked if the low protein diet
really necessary because other doctor advised him to have a low calories and normal
protein intake to maintain his nutritional status. This doctor told him that low protein diet was
needed to prevent the progression of renal failure. He was confused which doctor was right
and asked for an evidence. The doctor didn’t have an evidence to show his patient. Could
you help this doctor to find evidence from the internet?
Learning task:
Please fill the worksheet as defined in next page.
1. Identify what type of question of the above scenario?
2. Please, build up a clinical research question using PICO !
3. Formulate a clinical research question from scenario above !
Self assessment
1. Please describe the steps in EBMP !
2. Please describe the components of a good clinical questions !
3. Two additional elements of the well built clinical question are the type of question
and the type of study. This information can be helpful in focusing the question and
determining the most apppropriate type of evidence. Please describe the type of
questions and the typeof the study !
Udayana University Faculty of Medicine, MEU
16
Study Guide of Evidence-based Medical Practice Block
WORKSHEET TO BUILT UP CLINICAL RESEARCH QUESTIONS
Scenario # ..................
1. Type of question: Choose one of the term below:
a. Diagnosis
b. Therapy/ intervention
c. Prognosis/ Prediction
d. Etiology/Risk factors
e. Rate/Frequency
2. Built up a research question using PICO
P Population/problem=...........................................................................................
I Intervention
=............................................................................................
C Comparator/control =...........................................................................................
O Outcome
= ...........................................................................................
3. Clinical research question:
.......................................................................................................................................
.......................................................................................................................................
.......................................................................................................................................
Udayana University Faculty of Medicine, MEU
17
Study Guide of Evidence-based Medical Practice Block
LECTURE 1: Principles of Critical Appraisal
Eka Gunawijaya
What exactly is critical appraisal and what is the difference between "appraising" an article
and simply reading it? If you have been conscientious enough to organize a literature
search, go to the library and copy a promising article, why can't we leave you alone to read
it over a coffee? Why do we ask you to put it through some complex process called critical
appraisal?
Critical appraisal is the process of systematically examining research evidence to assess its
validity, results and relevance before using it to inform a decision. Critical appraisal is an
essential part of evidence-based clinical practice that includes the process of systematically
finding, appraising and acting on evidence of effectiveness. Critical appraisal allows us to
make sense of research evidence and thus begins to close the gap between research and
practice. The aim of critical appraisal is to identify the quality of an article.
Appraisal is a technique which offers a discipline for increasing the effectiveness of your
reading, by enabling you to quickly exclude papers that are of too poor a quality to inform
practice, and to systematically evaluate those that pass muster to extract their salient points.
Critical appraisal is usually applied to quantitative studies (e.g. "randomized" or "blinded"
controlled trials, crossover trials, meta-analyses or systematic reviews) of the effectiveness
of different health and medical interventions. However, the skills can also be applied to the
assessment of qualitative studies of psychosocial and behavioral interventions: e.g.
observational or interview data obtained from case or cohort studies.
Furthermore, studies on the effectiveness of cognitive, behavioral and other psychosocial
interventions are also being conducted using quantitative research methodologies (e.g.
randomized controlled trials); especially where these interventions are used as part of a
combination therapy, which includes medication.
The medical literature is vast and rapidly expanding. Forays into the library can be
exhausting, as the reader is overwhelmed by the huge number of papers offered. When
reading, someone will cite interesting references, which spur the reader into a lengthy paper
haze. A major hazard of reading is to pursue a subject in too much depth. Instead of
following this haphazard course, the process of reading should be carefully planned to
provide a worthwhile return on the time invested. Establishing control over your reading
means following a number of steps: clarify your reasons for reading; specify your
information need; identify the relevant reports; critically appraise the papers
Udayana University Faculty of Medicine, MEU
18
Study Guide of Evidence-based Medical Practice Block
LECTURE 2: Association and Causation
I B Subanada
When reading a medical literature we sometimes encounter such phrases like “association
with”, “linked to”, and “related to”. The authors have avoided dogmatic statement of “causes”
or “produces”, statement like ”smoking causes lung cancer ” or “birth control pills produce
vein thrombosis”. Association is not necessarily a causality. Here is an analogy: a town has
a large number of unemployed people and a very high crime rate. It does not necessarily
follow that the unemployed are committing the crime. In other word the presence of
unemployment and crime tells us nothing at all about either the presence or direction of
causality. On the other hand, even if we are convince that we have pinpointed the
responsible etiology, we still think that some unsuspected risk factor is actually causing the
disease or an unappreciated co-intervention is responsible for the treatment effect.
Concept of causation
There are two concepts of causation: First, contributory cause. In most medical
phenomena, there is no single specific cause of a disease. In physics we consider that
metal will expand because of heat and shrink when cold, therefore expansion in a metal is
mathematically caused by heat only as a single cause. Generally, however, most of
diseases are caused by multi-factorial etiology. A risk factor may be a contributory cause.
The second concept of cause is often called necessary cause. In the 19th century, Robert
Koch developed a series of conditions that must be met before a microorganism can be
considered the cause of a disease. Such conditions are related to what is known as Koch’s
Postulates, which include a requirement that organism is always found with the disease. In
the real medical world most of medical phenomena can only be explained by contributory
cause as cause and effect relationship. For instance, even though cigarettes have been well
established as a risk factor, it is a contributory cause for the development of lung cancer, but
cigarette smoking is not necessarily the condition for development of lung cancer, since not
every one with lung cancer has smoked cigarettes. In order to describe the model of cause
and effect relationship, we can construct it into two perspectives:
Udayana University Faculty of Medicine, MEU
19
Study Guide of Evidence-based Medical Practice Block
LECTURE 3: Effect Size, Hypothesis Testing
and Confidence Interval
I.B. Subanada
In analytical studies, investigators seek to determine effect size of the outcome of interest
between variables or between groups and to determine whether there is a statistical
significant of those effect sizes (hypothesis testing).
Most investigations are conducted on only sample or subset of larger group of individuals or
subset of population. Researcher, therefore, are confronted with the question of whether
results of the investigation in the sample would be similar if the investigation included the
entire population or whether chance in the selection of samples (by random sample)
produced unusual results in their sample. Hypothesis testing is a method, which is used to
answer this question. Hypothesis testing is based on null hypothesis, assuming there is no
difference between groups being compared or there is no relationship between variables. Pvalue, therefore, is the probability that observes data, or outcome, that would have occurred
by chance (just due to sampling variation) when the null hypothesis is true. If P- value is
small, probability of chance would be also small and one may doubt about the null
hypothesis, which thus can be rejected. If P-value is large, the chance may be great and the
data are plausibly consistent with the null hypothesis, which thus cannot be rejected.
LECTURE 4: Principles and Application
of Statistical Analysis
Lanang Sidiartha
Statistical analysis is basically a method to assist us to answer a question under study
(research question). The research question is commonly formulated in a study hypothesis.
A study hypothesis in an analytical study mostly contains one or more independent
variables and one dependent variable. The relationship between independent and
dependent variable is an important issue in statistical analysis. Regarding this issue, it is not
our intention to describe basic statistics, since this subject will be applied for nonstatisticians. Many computer programs packed with statistical soft ware are available and
easy to operate. However, this subject aims to guide the student to understanding the
principle of selecting statistical analysis and interpreting the results and the meaning
parameters in statistics.
Statistics have three purposes in the analysis of health research studies
1. To make estimates of the strength of relationship or magnitude of differences
2. To be used in hypothesis testing, which is allows us to draw inferences about
population from samples, obtained from the same population.
Udayana University Faculty of Medicine, MEU
20
Study Guide of Evidence-based Medical Practice Block
3. To adjust to the influence of confounding variables on these estimates and
inferences.
When selecting a specific statistical method, we must think about variables. A variable
expresses or represents data in the mathematical procedures that are part of statistics.
About variable, we should identify:
1. What the function of each variable, and
2. What type of data is represented by each variable
With regard to the function of variable, we have to distinguish dependent variable from
independent variable. Dependent variable can be identified as the outcome or end-point of
a study. On the other hand, there may be no, one, or several independent variables that
may be identified as risk factors or treatment of interest. The third is confounding variable,
which needs to be taken into account when hypotheses are to be tested and estimates are
to be made.
With regard to the type of data, we have to categorize them as continuous and discrete.
Continuous data are defined as data that provide the possibility of observing any of an
infinite number of equal spaced numerical values between any two points in its range of
measurement, for examples, blood pressure, serum cholesterol, age, and weight. Discrete
data can only a finite or limited number of values in their range of measurement, for
examples, number of pregnancies, stage of disease, and gender. For each of these
variables, we can select two values between which it is not possible to imagine other value.
For instance, there is no number of hypertension stages between stage 1 and stage 2. Then
data can be defined further by their scale of measurement. Continuous data are measured
on scales, called ratio or interval scales. Discrete data, on the other hand, can be
measured as nominal (such as treatment, gender, race, and eye color) and ordinal scales
(such as stage of the disease and levels of education).
For the purpose of selecting a statistical procedure or interpreting the result of such
procedure, it is important to distinguish between three categories: 1) continuous (contain a
great number of possible values), 2) ordinal (data are ordered one higher than the next and
with at least three), and 3) nominal (only two possible values, such as alive or dead)
variables. Continuous variable can be rescaled to ordinal or nominal variable. For example,
data of blood pressure (in continuous) can be categorized to stage of hypertension (stage 1,
stage 2, and stage 3) as ordinal variable or can be categorized to normal and hypertension
as ordinal variable. Continuous variables contain more information than ordinal variable and
nominal variables. Thus, continuous variables are considered to be at higher level than
nominal variables.
Thus in selecting statistical procedure, the initial steps are:
1. Identify one dependent variable and all independent variables, if present, on the
basis of the research question.
2. Determine for each variable whether it represents continuous, ordinal, or nominal
data.
The three basic statistical procedures: univariable, bivariable and multivariable analysis
Udayana University Faculty of Medicine, MEU
21
Study Guide of Evidence-based Medical Practice Block
LECTURE 5. Methodological and Statistical Principles and
Applications in Descriptive Studies
Eka Gunawijaya
A typical sequence for studying a topic begins with observational studies of a type that is
often called descriptive. This type of study often explore distribution of diseases and healthrelated characteristics in the population (How common is TB patients in Bali) or the
sensitivity and specificity of a diagnostic test. Descriptive studies are usually followed by
analytical study that to evaluate associations to discover cause and effect relationship
(Does TB vaccination lower the incidence of lung TB in Bali). Descriptive studies is
characterized by a set of measurements contains one dependent variable and no
independent variable.
A univariate analysis is therefore commonly used to test the data. There are three
application of this method:
1. Descriptive studies (e.g. case series), only one sample might be presented
2. To determine mean or percentage, point estimation and confidence interval of
particular groups.
3. Comparing two measurements of same characteristic on the same or very similar
individuals (using paired data).
In continuous dependent variable, data are usually assumed to come from population with a
Gaussian distribution. Population mean is an estimate of primary interest and dispersion is
measured by the standard deviation or variance.
For ordinal dependent variable, we do not assume a particular distribution of population
data (distribution free or non-parametric). Estimate of population is median defined as midpoint of a collection of data.
For nominal dependent variable, we determine only the presence or absence of the
condition, and we can estimate the frequency of condition occurs in the population. The
data is assumed to have either a binomial or Poisson distribution.
Estimating the sample size for descriptive study including studies of diagnostic test is based
to that investigator is aiming to calculate descriptive statistics, such as means and
proportions (prevalence, incidence, mortality rate) of particular disease in a single
population.
Descriptive studies commonly report confidence intervals, a range of values about the
sample mean or proportion. A confidence interval is a measure of the precision of the
sample estimate. The investigator sets the confidence level, such as 95% or 99%. An
Udayana University Faculty of Medicine, MEU
22
Study Guide of Evidence-based Medical Practice Block
interval with a greater confidence level (say 99%) is wider, and therefore more likely to
include the true population value, than an interval with a lower confidence level (say 95% or
90%). The width of confidence interval depends on sample size. The more sample size the
.narrower the width.
LECTURE 6. Methodological and Statistical Principles and Applications in
Analytical Studies
Lanang Sidiartha
Based on variable, statistical analysis was divided to bivariate analysis if consist of one
independent variable and one dependent variable and multivariate analysis if consist of one
dependent variable and two or more independent variable or two or more dependent
variable and one independent variable. There are 5 steps to choose appropriate statistical
analysis. First, identified the study hypothesis or research question; second, identified how
many variable was compared; third, identified the variable is it related or unrelated variable;
fourth, identified scale of measurement of data; and fifth, identified the criteria for parametric
test and non-parametric test. Study hypothesis was classified as comparative hypothesis,
associative hypothesis and correlative hypothesis, each of them has different statistical
analysis. Variable or set of data was classified as related group if it was take from the same
subject.
LECTURE 7: Diagnostic Test
Principles In Critical Appraisal
Sudhana
The fundamental principle of diagnosis testing rests on the belief that individual with a
disease are different from individual without disease and that diagnostic test can distinguish
between these two groups. Ideally, diagnostic test have the following features: (1) all
individuals without the disease under study have one uniform value on the test, (2) all
individuals with the disease have a different but uniform value for the test, thus, (3) all test
results would coincide with the results of diseased or those of the disease-free group. If this
was the situation in reality, then one perfect test could distinguish disease from health.
However, none of these three conditions is usually present. Variation exist is due to same
factors coming from subjects being studied, instrument being used and the observer.
Subject variation is condition of the individual subject being tested may vary from
performance to performance, resulting in changes in phenomenon being assessed.
Instrument variation may occur as a result of technical methods used to perform the test.
Errors may occur because of variations when using the same testing instrument (intrainstrument error) or when using different instrument (inter-instrument error). Observer
Udayana University Faculty of Medicine, MEU
23
Study Guide of Evidence-based Medical Practice Block
variation may occur as a result of the observer who assesses the results. Errors may occur
because of the variation in measurement by the same observer (intra-observer variation) or
error using different observers (inter-observer variation).
The test or criterion used to unequivocally define the disease is known as a gold standard.
The gold standard may be a biopsy, an angiogram, an autopsy, or any established test. The
use of a gold standard tests that is possible to be 100% correct in making a diagnosis.
There might be a cheaper or more convenient test. In diagnostic test, we can ask whether
the test measures up to the gold standard. The investigator classifies each patient as either
having the disease or being disease-free according to the gold standard test, and as
positive or negative by the test being evaluated. They then can calculate the number of
individuals for whom the test and the gold standard test agree and the number for whom
they disagree and display their result in2X2 table as follows:
(+)
Gold standard
(-)
Total
Predictive
Test
(+)
(-)
Total
A
C
a+c

Sensitivity
b
d
b+d
a+ b
c+d
a +b+c+d

Value
Positive
 negative
Specificity
Table 1. Table 2 x 2 table
We can determine, a = number of individuals with the disease and test positive: true
positive; b = number of individuals diseased free and test positive: false positive; c =
number of individuals with the disease and test negative: false negative; d = number of
individuals disease-free and test negative: true negative. Then, a + c = total number of
individuals with the disease and b + d = total number of disease-free individuals. From these
components we can calculate sensitivity = a/(a+c) and specificity = d/(b + d) of the test.
If we want to critically appraised an article about diagnosis, we are confronted with V I A,
that stands for valid, important and applicable which appraises the validity, importance and
applicability of the report.
1. The validity will questions whether evidence about the accuracy of a diagnostic test
valid
2. The importance query of whether this (valid) evidence demonstrate an important
ability to accurately distinguish patients who do and don’t have specific disorder
3. The applicability may ask of whether we can apply this valid important diagnostic test
to specific (our) patients.
Udayana University Faculty of Medicine, MEU
24
Study Guide of Evidence-based Medical Practice Block
To step on a critical appraisal, an important issue about the test being studied that we need
to define is what is normal and what is abnormal. Both parameters will be applied to the test
being studied and the gold standard.
We can define the normalcy from six approaches:
1. Gaussian: mean ± 2standard deviations – this one assumes a normal distribution for
all tests and results in all “abnormalities” having the same frequency
2. Percentile 5-95%- has the same basic defect as the Gaussian definition
3. Culturally desirable: when “normal” is that which preferred by society, the role of
medicine gets confused
4. Risk factor: changing risk factor necessarily changing risk
5. Diagnostic: range of result beyond which target disorder became highly probable
6. Therapeutic: range of result beyond which treatment does more good than harm.
Means we have to keep up with the advances in therapy.
Problem may arise with parameters in continuous variables. With such tests, several
Fig 1. Receiver operating characteristic (ROC) curve for some cutoff points of a continuous
variable.
Values of sensitivity and specificity are possible, depending on the cutoff point chosen to
define a positive test. This trade- off between sensitivity and specificity can be displayed
using a graphic technique. This technique originally developed in electronic equipment
called receiver operating characteristic (ROC) curves. We select several cutoff points and
determine sensitivity and specificity at each point. We then graph the sensitivity (or truepositive rate) on Y-axis as a function of 1-specificity (false-positive rate) on X-axis. An ideal
Udayana University Faculty of Medicine, MEU
25
Study Guide of Evidence-based Medical Practice Block
test is one that reaches the upper left corner of the graph (100% true positives and no false
positives). A worthless test follows the diagonal from the lower left to the upper right
corner (see fig.1). The area under curve (AUC), which does ranges from 0.5 for useless test
to 1.0 for prefect test, is useful summary of the overall accuracy of a test and can be used to
compare the accuracy of two or more tests.
LECTURE 8: Clinical Trial
Principle in Critical Appraisal
Raka Widiana
Some treatments are so clearly advantageous that they require no formal assessment; this
is true of antibiotics for pneumonia and surgery for serious trauma. However, this situation is
relatively rare in clinical medicine. Usually the effects of treatment are much less obvious
and most interventions require research to establish their value. Not only must specific
interventions be shown to do more good than harm among patients who use them (i.e. they
are theoretically effective or efficacious), but they should also do more good than harm in
patients to whom they are offered (i.e. they should be practically effective). In studies of
efficacy it is advantageous to include who are likely to be compliant. Practical effectiveness
is determined by studying outcome in a group of people offered treatment, only some of
who will be compliant. The most accurate method for measuring effectiveness is a clinical
trial.
Clinical trials are a special kind of cohort study in which the conditions of study – selection of
treatment groups, nature of interventions, management during follow up, and measurement
of outcomes – are specified by the investigator for the purpose of making unbiased
comparisons. Clinical trials are more highly controlled and managed than are cohort studies
(observational study). The investigators are conducting an experiment, analogous to those
done in the laboratory. They have taken it upon themselves (of course, with their patient’s
permission) to isolate for study the unique contribution of one factor by holding constant, as
much as possible, all other determinants of the outcome. Hence, other names for clinical
trials are experimental and intervention studies.
We need to define exactly what is meant by ‘clinical trial’; briefly the term may be applied to
any form of planned experiments which involves patients and is designed to elucidate the
most appropriate treatment of future patients with a given medical conditions. Perhaps the
essential characteristic of a clinical trial is that one uses results based on a limited sample of
patients to make inferences about how treatment should be conducted in the general
population of patients who will require treatment in the future. Randomized controlled trials
(RCT) are the standard of excellence for scientific studies of the effects of treatment in
clinical trial.
Udayana University Faculty of Medicine, MEU
26
Study Guide of Evidence-based Medical Practice Block
RANDOMIZED CONTROLLED TRIAL
The structure of an RCT is shown in Figure-1. The patients to be studied are first selected
from a larger number of patients with the condition of interest. They are then divided, using
randomization into two groups of comparable prognosis. One group, called the experimental
or treated group, is exposed to an intervention that is believed to be helpful. The other
group, called a control or comparison group, is treated the same in all ways except that its
members are not exposed to the intervention. The clinical course of both groups is then
observed and any differences in outcome are attributed to the intervention.
The main reason for structuring RCT in this way is to avoid bias (systematic error) when
comparing the respective value of the two or more kinds of treatments. The validity of RCT
depends on how well they result in an equal distribution of all determinants of prognosis,
other than the being tested, in treated and control patients.
Experimental
Intervention
NOT IMPROVED
Population of
patients with
IMPROVED
SAMPLE
TIME
OUTCOMES
the condition
IMPROVED
Comparison
NOT IMPROVED
Intervention
Figure 1: The structure of a RCT
SAMPLING
Any RCT requires a precise definition of which patients are eligible for inclusion. The early
stages of protocol development may proceed with only a rough outline of the intended type
of patient, but before the RCT gets underway this must be transformed into detailed
specification. The main objective is to ensure that patients in the RCT may be identified as
representative of some future class of patients to whom the RCT finding may be applied as
shown in Figure 2.
The kinds of patients that are included in an RCT determine the extent to which conclusions
can be generalized to other patients. Of the many reasons why patients with the condition of
interest may not be a part of an RCT, three account for most of the losses. They do not
meet specific entry criteria, they refuse to participate or they do not cooperate with the
conduct of the trial.
The first, entry criteria is intended to restrict the heterogeneity of patients in the trial.
Common exclusion criteria are atypical disease, the presence of other diseases, an
Udayana University Faculty of Medicine, MEU
27
Study Guide of Evidence-based Medical Practice Block
unusually poor prognosis (which may cause patients to drop out of the assigned treatment
group), and evidence of unreliability. Patients with contraindications to one of the treatments
are also excluded, for obvious reasons. As heterogeneity is restricted in this way, the
internal validity of the study is improved; there is less opportunity for differences in
outcomes that are not related to treatment itself. Also, generalizing the results is more
precise because one knows exactly to whom the results apply. But exclusions come at the
price of diminished scope of generalizability, because characteristics that exclude patients
occur commonly among those ordinarily seen in clinical practice, limiting generalize ability to
these patients, the very ones for whom the information is needed.
Second, patients can refuse to participate in the trial. They may not want a particular type of
treatment or to have their medical care decided by a flip of a coin or by someone other than
their own physician. Patients who refuse to participate are usually systematically different –
in socioeconomic class, severity of disease, other health-related problems, and other ways
– from those who agree to enter the trial.
Third, patients who are found to be unreliable during the early stages of the trial are
excluded. This avoids wasted effort and the reduction in internal validity that would occur if
patients moved in and out of treatment groups or out of the trial altogether. For these
reasons, patients in an RCT are usually highly selected, biased sample of all patients with
the condition of interest. Because of the high degree of selection in trials, it often requires
considerable faith to generalize the results of RCT to ordinary practice settings.
Treated
Outcome
Allocation
Sample
Control
Outcome
Population
sampled
OTHER
POPULATION
Statistical
analysis
Figure 2: Generalizability in an RCT
Udayana University Faculty of Medicine, MEU
28
Study Guide of Evidence-based Medical Practice Block
CRITICAL APPRAISAL
The critical appraisal of an RCT is one of the basic elements of medical informatics and
evidence-based medicine curricula. Various frameworks for critical appraisal have been
proposed. This paper proposes a framework for evaluating an RCT that consists of 12
questions grouped under 3 headings (table 1,2 and 3). These questions are very helpful to
do critical appraisal on RCT, and enclosed at the end of this paper.
• Are the results of the RCT valid?
• What were the results?
• Will the results help me in caring for my patients?
The paper put together a flow diagram of an RCT and the point’s at which bias can creep in;
this flow diagram serves as a memory aid and can act as a framework on which to "hang"
whatever critical appraisal guide the user is most comfortable with.
Each of the 5 numbered steps in the RCT flow diagram can act as the focal point for a
discussion of methods.
4
5
1
2
3
Figure 3: Flow diagram for an RCT
Figure 3 above shows the flow of an RCT and can be drawn as the starting point for a
critical appraisal exercise. Each of the 5 numbered steps in the RCT flow diagram can act
as the focal point for a discussion of methods.
Step 1: selection and sampling issues
When drawing the larger circle, one can ask such questions as "What sorts of patients were
recruited, and where were they recruited from (i.e. primary care or a referral centre)?" When
drawing the arrow that indicates which patients are recruited into the study, one can ask
whether the inclusion and exclusion criteria make sense and whether consecutive patients
are being recruited; exclusions can be represented graphically by drawing a second arrow
peeling off from the main one. This arrow also serves to indicate the total number of patients
Udayana University Faculty of Medicine, MEU
29
Study Guide of Evidence-based Medical Practice Block
screened to obtain the enrolled sample. All these considerations affect the external validity
and generalizability of the study, whereas subsequent steps affect internal validity. The
graphical representation of this concept is the box drawn around the remainder of the study
(fig 4).
Step 2: randomization
The arrows representing the allocation of participants to the 2 groups serve to highlight
graphically the process of randomization, which evenly (or at least randomly) distributes
known and unknown confounders. One can then discuss the principle of randomization,
including concealment (the person enrolling the patient in the study must not be able to
predict to which group the patient will be randomly allocated, and less secure processes,
such as quasi-randomization. One can then ask how to check the success of randomization.
Are the baseline characteristics of the study groups similar? Any imbalance in those
characteristics can signal 1 of 2 things: randomization has not been done properly (the
participants were not truly randomly allocated) or randomization was done properly but a
discrepancy has arisen by chance (which is more likely when the number of patients is
small). In the latter case, one can check to see whether the study authors statistically
adjusted for the differences
Internal
validity
External
validity
4
5
1
2
3
Exclusion
Figure 4: Internal validity of an RCT
Step 3: follow up
A catchy mnemonic that is helpful in highlighting the sources of bias at this step are the "5
Cs": contamination, crossover, compliance, co-intervention, and count (i.e. loss to follow
up).
Contamination often occurs in a trial of an educational intervention where, for example, the
control group may adopt such lifestyle changes as dietary modification or exercise intended
Udayana University Faculty of Medicine, MEU
30
Study Guide of Evidence-based Medical Practice Block
for the intervention group only. Contamination can also happen when control and treated
patients share their medications, as in certain AIDS trials.
Crossover occurs most often in open trials in which patients or clinicians know what
medication the patient is receiving and also what the alternative treatment is: for example, in
trials where there are medical and surgical arms or a conservative and aggressive treatment
protocol, patients in the less aggressive arm who worsen may cross over to the more
aggressive treatment arm. Crossovers may represent graphically as arrows that cross
between the two arms of the trial (Fig 5).
Internal validity
External
validity
Cross-over
4
1
5
Contamination
2
3
Exclusion
Co-intervention
Loss to follow-up
Figure 5: Sources of bias in RCT
Lack of compliance with the intervention can introduce bias, as can count or loss to follow
up, if patients who drop out differ in their characteristics from those who remain. Both
sources of bias may be represented by arrows that peel off from the 2 arms (fig 5).
Co-interventions also introduce bias if they are applied differentially to the 2 groups. Cointerventions can be represented by arrows that join the two arms (Fig 5). Contamination,
crossover, and lack of compliance will bias results toward the null (attenuate any effect);
whereas co-intervention and count may bias results in either direction.
Step 4: outcomes
The boxes in the figures represent the outcomes in the 2 groups. Discussion can focus on
whether the chosen outcomes were reasonable and whether all-important outcomes were
considered. Who judges or counts the outcome is of concern. Outcome assessment leads
to a discussion of the importance of blinding, which can include blinding of the patient, the
caregiver, the outcome assessor (particularly important when such patient-reported
subjective measures as quality of life are used), and even the statistician doing the analysis.
Udayana University Faculty of Medicine, MEU
31
Study Guide of Evidence-based Medical Practice Block
Having the assessors and patients guess their assignments at the end of the trial can give
an indication of whether un-blinding occurred.
Step 5: analysis
This step includes a discussion of intention to treat (ITT) analysis. ITT analysis is important
because it preserves the benefit of randomization. If those patients who dropped out, were
non-compliant, or had intolerable adverse effects are not considered in the analysis, it is the
equivalent of allowing participants to self select or opt out of the study. If these patients'
characteristics differ from those of the rest of the group, the even distribution of confounders
obtained through randomization is lost.
Other issues of analysis include the magnitude of the effect (measured by relative and
absolute risk differences, odds ratios, and numbers needed to treat), precision (confidence
intervals), and subgroup analyses (were these preplanned or derived from a "fishing
expedition" once the data became available?).
LECTURE 9: Study about Prognosis
Principle in Critical Appraisal
Eka Gunawijaya
As clinicians, we consider questions about prognosis all the time. Sometimes the questions
are posed by patients and are quite direct: “How long have I got?” At other times, we pose
these questions ourselves, and they may be less direct, as when deciding whether to treat
at all (e.g. an elderly man with chronic lymphocytic leukemia who feels well – would his
prognosis be importantly altered if he were left alone until becomes symptomatic?) or
deciding whether to screen (e.g. for abdominal aortic aneurysms – what is the fate of the
undetected 4 cm aneurysm?). These questions share three elements: a qualitative aspect
(Which outcomes could happen?), a quantitative aspect (How likely are they to happen?)
and a temporal aspect (Over what time period?).
Prognosis is a prediction of the future course of disease following its onset and is expressed
as the probability that a particular event will occur in the future. Doctor and patients think
about prognosis in several different ways. First, they want to know the general course of the
illness the patient has. Second, they usually want to know, as much as possible, the
prognosis in the particular case. Third, patients especially are interested to know how an
illness is likely to affect their lives, not only whether it will or will not kill them, but how it will
change their ability to work, to walk, to talk, how it will alter their relationships with family and
friends, how much pain and discomfort they will have to endure.
Udayana University Faculty of Medicine, MEU
32
Study Guide of Evidence-based Medical Practice Block
NATURAL HISTORY AND PROGNOSIS
Prognosis is closely associated with the natural history or clinical course of the disease. The
term natural history refers to the stages of disease, which include:
1. Pathological onset
2. The pre-symptomatic stage from onset of pathological changes to the first
appearance of symptoms and signs.
3. The stage when the disease is clinically obvious and may be subject to remissions
and relapse, regress spontaneously, or progress to death
Detection and treatment at any stage can alter clinical course of a disease, but the effects of
treatment can only be determined if clinical course of the disease in the absence of
treatment is known. The prognosis of disease without medical intervention is termed the
natural history of disease. Natural history describes how patients will fare if nothing is done
for their disease. A great many medical conditions, even in countries with advanced medical
care systems, often do not come under medical care. They remain unrecognized, perhaps
because they are asymptomatic or are considered among the ordinary discomforts of daily
living. Examples include mild depression, anemia, and cancers that are occult and slow
growing (e.g. some cancers of the thyroid and prostate).
Patients usually come under medical care at some time in the course of their illness when
they have diseases that cause symptoms such as pain, failure to thrive, disfigurement, or
unusual behavior. Once disease is recognized, it is also likely to be treated.
PROGNOSIS STUDIES
Studies of prognosis deal with these above clinical questions in ways similar to cohort
studies of risk. A group of patients having something in common (a particular medical
disease or condition, in the case of prognostic studies) are assembled and followed forward
in time, and clinical outcomes are measured, Often, conditions that are associated with a
given outcome of the disease, i.e. prognostic factors, are sought.
Epidemiological information is necessary to provide sound prediction on prognosis and
outcome. Clinical experience alone is inadequate for this purpose since it is often based on
a limited set of patients and inadequate follow-up. For example, patients who are seen by a
doctor are not necessarily representative of all patients with particular disease. Patients may
be selected according to severity or other features of their disease, or by demographic,
social or personal characteristics of the patients themselves. Further more, since many
doctors do not systematically follow up their patients, they have a limited, and often
excessively pessimistic, view of the prognosis of disease. For these reasons
epidemiological studies are required to describe accurately the natural history and
prognosis of disease.
Ideally, the assessment of prognosis should include measurement of all clinically relevant
outcomes, not just death, since patients are usually as interested in the quality of life as they
are in its duration. In studies to determine natural history and prognosis, the group of
patients should be randomly selected; otherwise selection bias may severely affect the
information obtained. For example, the prognosis of patients with chest pain admitted to
Udayana University Faculty of Medicine, MEU
33
Study Guide of Evidence-based Medical Practice Block
hospital is likely to be worse that that of patients with chest pain seen by health workers in
the community.
Prognosis in term of mortality is measured as case-fatality rate or probability of survival.
Both the date of onset and the duration of follow-up must be clearly specified. Survival
analysis is a simple method of measuring prognosis. Life-table analysis is a more
sophisticated method that attempts to predict the onset of events over time from previous
patterns for all patients at risk. In the follow-up of cohorts of patients to determine prognosis,
bias can arise from the method of assembling the cohort and from incomplete follow-up.
PROGNOSTIC FACTORS
Although most patients are interested in the course of their disease in general, they are
even more interested in a prediction for their given case. Prognostic factors help identify
groups of patients with the same disease who have different prognosis.
DIFFERENT BETWEEN PROGNOSTIC FACTORS AND RISK FACTORS
Studies of risk factors usually deal with healthy people, whereas prognostic factors –
conditions that are associated with an outcome of the disease – are, by definition, studied in
sick people. There are other important differences as well, outlined below.
Different Factors
Factors associated with an increased risk are not necessarily the same as those marking a
worse prognosis and are often considerably different for a given disease.
Onset of Acute Myocardial Infarction
Well
RISK
PROGNOSIS
Outcomes
Death
Re-infarction
Risk Factors
Prognostic Factors
Age ↑
Age ↑
Male
Female
Smoking
Smoking
Other
Hypertension
Hypotension
Figure 1: Differences between risk and prognostic factors for AMI
LDL ↑ & HDL ↓
Anterior Infarction
For example, low blood pressure decreases one’s chance of having an acute myocardial
Inactivity
Heartpresent
Failure during an acute event (figure 1).
infarction,
but it is a bad prognostic sign, when
Similarly, intake of exogenous estrogens during
menopause increases women’s risk of
Ventricular arrhythmias
endometrial cancer, but the associated cancers are found at an earlier stage and seem to
Udayana University Faculty of Medicine, MEU
34
Study Guide of Evidence-based Medical Practice Block
have a better-than-average prognosis. Some factors do have a similar effect on both risk
and prognosis. For example, both of the risk of experiencing an acute myocardial infarction
and the risk of dying of it increase with age.
Different outcomes
Risk and prognosis describe different phenomena. For risk, the event being counted is the
onset of disease. For prognosis, a variety of consequences of disease are counted,
including death, complications, disability, and suffering.
Different rates
Risk factors generally predict low probability events. Yearly rates for the onset of various
diseases are on the order of 1/100 to 1/10,000. As a result, relationships between exposure
and risk usually elude even astute clinicians unless they rely on carefully executed studies,
often involving a large number of people over extended periods of time. Prognosis, on the
other hand, describes relatively frequent events. Clinicians often can form good estimates of
prognosis on their own, from their personal experience. For example, they know that few
patients with lung or pancreatic cancer survive as long as 5 years, whereas most with
chronic lymphocytic leukemia survive much longer. A combination of factors may give a
more precise prognosis than each of the same factors taken one at a time. Clinical
prediction rules estimates the probability of outcomes according to a set of patient
characteristics. In this paper, we’ll present a framework for appraising the validity,
importance and applicability of evidence about prognosis. When we are scanning articles
bout prognosis, we can use the first two validity guides for screening, selecting only those
that pass both guides to spend more time on.
LECTURE 10: How to Write a Paper and Present at a Meeting
Suega
Writing, even writing poorly, is not easy to most researchers. Some lack the initial energy to
get started. Others may even have half dozen manuscripts in various stages of completion
but never manage to finish. Whatever the problem, you ran improve your writing skill.
First, decide whether it is worth your time and effort to write the manuscript. If you are
having a hard time motivating yourself to start, perhaps it is because you do not have much
to say. Does the world really need the manuscript? If not, cut your losses and move on to
another project.
Next, be sure to use word-processing software that is well suited to medical writing. If you
do not already have software you are comfortable with, choose a program that is commonly
used at your institution so you can ask your colleagues questions about it. Take advantage
of any spell-checking capabilities; make sure that you spell medical words correctly when
you add them to the dictionary. Use a bibliography that will automatically number and format
Udayana University Faculty of Medicine, MEU
35
Study Guide of Evidence-based Medical Practice Block
your references. Learn how to download references electronically into the program so you
do not have to key them in manually. Use a laser printer, if one is available. The clean look
of a laser-printed manuscript increases the motivation to write.
Finally, you cannot write well if you do not write at all. Do not fuss over the first draft of a
manuscript, just write, without worrying about style, brevity, or clarity. Prepare a draft that
includes a title page, each section of the manuscript, and mock tables and figures. Put your
name, a running title, the page number, and the date at the top of each page. The more the
draft looks like a manuscript, the prouder you will feel, and the more likely it will that you and
your coauthors will be willing to invest additional time in improving it. This works even if
some of the sections consist of nothing more than the word pending.
You cannot write well if you do not read what you have written. Print a copy of the first draft.
Read it to make sure that you have not left anything out. Make the necessary additions to
the next draft, incorporate them into the word-processed version, and print the revised
manuscript. Each version will be better than its predecessor, and the incremental
improvements will encourage you to proceed.
When you think you have a nearly complete draft containing all of the scientific material,
print a copy and sharpen a pencil. Cross out every paragraph that is not necessary. (If
trimming your writing pains you, substitute the word critical for necessary). Delete every
extra sentence within the remaining paragraphs, and finally, the extraneous words within the
remaining sentences. Underline everything that is not crystal clear. Circle words that do not
make sense. Put an arrow before every paragraph that does not flow logically from its
predecessor and write: “Segue?” Sharpen another pencil, and begin clarifying the
ambiguities, replacing the awkward words and phrases, and filling in the missing links.
Style will be the objective for the next two or three drafts of the manuscript. But if you never
learned how to write well, or are someone for whom English is a second language, do not
fuss over this aspect. Hire a copy editor to do it for you.
When you think you are finished, punt another version, and read it carefully for typographic
and spelling errors. Make sure that the numbers make sense. Are they consistent between
the methods and results?, Between the text and tables?, Between the results and the
discussion?
If you are a non-English-speaking author, do not submit a manuscript to an Englishlanguage journal until it has been read and edited by a native English speaker who
understands the scientific content. Editors do not enjoy the thought of sending
reviewers a manuscript that is difficult to understand. Reviewers do not like having to
read them because they take more time. Thus, manuscripts describing good or even
excellent work may be rejected because the English is poor.
HOW TO PRESENT AT A MEETING
Presentation is a one method of communication, but has a limited understanding.
Communication by definition is, a two-way process of interaction, while presentation tends
to be one way only.
Udayana University Faculty of Medicine, MEU
36
Study Guide of Evidence-based Medical Practice Block
Assuming that the intention of your presentation is to inform your audience, so that some
thing is learned from you, what do we know in general about how people learn?
People learn best when:
1. they are motivated
2. they recognize their need to learn
3. the learning is relevant in context and matches their needs
4. the aims of the learning are clear
5. they are actively involved
6. a variety of learning methods is used
7. it is enjoyable.
SMALL GROUP DISCUSSIONS (SGD)
SGD 1:
Association and Causation
Learning task

What is the basic difference between association and causation?

What are the differentiation between contributory cause and necessarily cause?

What are the features of contributory cause?

What are the source of spurious association between cause and effect in a study?
SGD 2.
Effect Size, Hypothesis Testing and Confidenc Interval
1.
2.
3.
4.
What is the meaning of effect size?
What is the principle of hypothesis testing?
What is the principle of confidence interval?
What is the relationship between effect size, hypothesis testing, and confidence
interval?
5. How to report an abstract showing effect size, confidence interval, and effect size?
Principal and application of statistical analysis
Severe acute malnutrition affects 13 million children worldwide and causes 1–2 million
deaths every year. Our aim was to assess the clinical and nutritional efficacy of a probiotic
and prebiotic functional food for the treatment of severe acute malnutrition in a HIVprevalent setting. We recruited 795 Malawian children (age range 5 to 168 months) from
July 12, 2006, to March 7, 2007, into a double-blind, randomized, placebo-controlled
efficacy trial. Children were randomly assigned to ready-to-use therapeutic food either with
Udayana University Faculty of Medicine, MEU
37
Study Guide of Evidence-based Medical Practice Block
(n=399) or without (n=396) Synbiotic2000 Forte. Primary outcome was nutritional cure
(weight-for-height >80% of National Center for Health Statistics median on two consecutive
outpatient visits). Nutritional cure was similar in both Symbiotic and control groups (53.9%
[215 of 399] and 51.3% [203 of 396]; p=0.40). The characteristic of subjects were presented
in table 1 below.
Table 1. Characteristic of subject
Variable
Symbiotic (n=399)
Placebo (n=396)
22 (15 – 32)
21 (15 – 31)
Gender, boys, n(%)
214 (53.6)
216 (55.0)
Girls, n(%)
185 (46.4)
177 (45.0)
142 (35.6)
146 (36.9)
257 (64.4)
250 (63.1)
170 (42.6)
192 (48.5)
203 (51.0)
190 (48.0)
26 (6.4)
14 (3.5)
208 (53.9)
217 (56.8)
132 (46.1)
117 (43.2)
215 (53.9)
203 (51.3)
Age in months
Nutritional status, Marasmic, n(%)
Kwashiorkor, n(%)
Status HIV, Seropositive, n(%)
Seronegative, n(%)
Unknown, n(%)
Household water source,piped,
n(%) Borehold, n(%)
Primary outcome, cure, n(%)*
*) p = 0.40
Question
1. Explain the type of data of each variable based on scale of measurement: age,
gender, nutritional status, status HIV, household water source, and primary outcome!
2. Explain the statistic testing for primary outcome between symbiotic and placebo
group!
3. To adjusted compounding factors, explain the statistic testing that was used!
Udayana University Faculty of Medicine, MEU
38
Study Guide of Evidence-based Medical Practice Block
SGD 3:
Methodological and Statistical Principles and Applications in Descriptive Studies
1.
2.
3.
4.
Explain the difference between descriptive and analytic studies.
Explain why must undergo a descriptive study, give an example for describe it.
Explain the types of descriptive studies.
Write a 1 study title for an example of each type of studies.
SGD 4 :
Methodological and Statistical Principles and Application of Analytical Studies
Researcher wants to study effectiveness of probiotic as additional treatment of acute
diarrhea in children. The research question: in children with acute diarrhea, is the probiotic
as additional therapy compare with placebo give faster recovery. Based on simple
randomized, 50 subjects receive probiotic and 50 subjects receive placebo. The
characteristic of subject was presented in table 1 below.
Table 1. Characteristic of subject
Variable
Probiotic
Placebo
(N = 50)
(N = 50)
Gender, boys, n(%)
24 (48)
25 (50)
girls, n(%)
26 (52)
25 (50)
24 (1,5)
25 (1,4)
Obese
0 (0)
1 (2)
Normal
30 (60)
32 (64)
Malnourished
20 (40)
17 (34)
Body temperature, oC
37,5 (36,1-39,5)
37,8 (36,5-39,4)
Length of stay (days),
mean(SD)
3,5 (0,2)
5,5 (0,3)
Age (month), mean (SD)
Nutritional status, n (%)
Udayana University Faculty of Medicine, MEU
39
Study Guide of Evidence-based Medical Practice Block
Question
1. What kind of statistical analysis is used to compare the length of stay between
probiotic and placebo group?
2. What kind of statistical analysis is used to compare the proportion of gender
between probiotic and placebo group?
3. What kind of statistical analysis is used to compare the different of nutritional status
between probiotic and placebo group?
4. What kind of statistical analysis is used to compare the mean body temperature
between probiotic and placebo group?
SGD 5: Critical Appraisal of Diagnostic Test. Theory
To step on a critical appraisal, an important issue about the test being studied that we need
to define is what is normal and what is abnormal.
Please define six approaches of normalcy.
Please explain:
Are the results of this diagnostic study valid?
1. Independent, blind comparison with a reference (gold standard of diagnosis. What
this way is to avoid?
2. Appropriate spectrum of patients.
3. Reference standard applied regardless of the diagnostic test result.
What is work-up bias?
4. Test validated in second independent group of patients. Does the study show an
important ability?
Look at some parameters, what do they mean:




Sensitivity and specificity?
Positive and negative predictive value?
Likelihood ratio (multilevel)
Pre-test and post-test probability
Critical Appraisal of Diagnostic Test . Article
An article will be given two days before the discussion
Learning task
Please fill in the worksheet after discussing an article about diagnostic test that has been
given to you.
Udayana University Faculty of Medicine, MEU
40
Study Guide of Evidence-based Medical Practice Block
Diagnosis Worksheet
Citation:
Are the results of this diagnostic study valid?
Was there an independent, blind
comparison with a reference (“gold”)
standard of diagnosis?
Was the diagnostic test evaluated in an
appropriate spectrum of patients (like
those in whom it would be used in
practice)?
Was the reference standard applied
regardless of the diagnostic test result?
Was the test (or cluster of tests)
validated in a second, independent
group of patients?
Are the valid results of this diagnostic study important?
SAMPLE CALCULATIONS
Target disorder
Totals
(iron deficiency anemia)
Present
Absent
731
270
1001
(<65 mmol/L)
A
B
a+b
Negative
78
1500
1578
(65 mmol/L)
C
D
c+d
809
1770
2579
a+c
b+d
a+b+c+d
Positive
Diagnostic
test result
(serum
ferritin)
Totals
Udayana University Faculty of Medicine, MEU
41
Study Guide of Evidence-based Medical Practice Block
Sensitivity=a/(a+c)=731/809=90%
Specificity=d/(b+d)=1500/1770=85%
Likelihood ratio for a positive test result=LR+=sensitivity/(1 – specificity)=90%/15%=6
Likelihood ratio for a negative test result=LR–=(1 – sensitivity)/specificity=10%/85%=0.12
Positive predictive value=a/(a+b)=731/1001=73%
Negative predictive value=d/(c+d)=1500/1578=95%
Pre-test probability (prevalence)=(a+c)/(a+b+c+d)=809/2579=32%
Pre-test odds=prevalence/(1 – prevalence)=31%/69%=0.45
Post-test odds=(pre-test odds)LR
Post-test probability=(post-test odds)/(post-test odds+1)
YOUR CALCULATIONS
Target disorder
Diagnostic
test result
Totals
Present
Absent
Positive
A
B
a+b
Negative
C
D
c+d
a+c
b+d
a+b+c+d
Totals
Can you apply this valid, important evidence about a diagnostic test in caring for
your patient?
Is the diagnostic test available, affordable,
accurate, and precise in your setting?
Can you generate a clinically sensible
estimate of your patient’s pre-test probability
(from personal experience, prevalence
statistics, practice databases, or primary
studies)?
 Are the study patients similar to your
own?
 Is it unlikely that the disease possibilities
or probabilities have changed since the
Udayana University Faculty of Medicine, MEU
42
Study Guide of Evidence-based Medical Practice Block
evidence was gathered?
Will the resulting post-test probabilities
affect your management and help your
patient?
 Could it move you across a test–
treatment threshold?
 Would your patient be a willing partner in
carrying it out?
Would the consequences of the test help
your patient?
Additional notes:
SGD 6. Critical Appraisal of Clinical Trial. Theory
Please explain:
Are the results of this RCT valid?
1. Was the assignment of patients to treatment randomized?
a. Was the randomization list concealed?
2. Was follow-up of patients sufficiently long and complete?
3. Were all patients analyzed in the groups to which they were randomized? Some less
important points:
4. Were patients and clinicians kept blinded to treatment?
5. Were groups treated equally, apart from the experimental therapy?
6. Were the groups similar at the start of the RCT?
Udayana University Faculty of Medicine, MEU
43
Study Guide of Evidence-based Medical Practice Block
Are the valid results of this RCT important?
1. What is the magnitude of the treatment effect?
2. How precise is this estimate of the treatment effect?
Critical Appraisal of Clinical Trial. Article 1.
An article will be given two days before the discussion
Please fill in the worksheet after discussing an article about clinical trial that has been given
to you.
THERAPY WORKSHEET
Citation:
Are the results of this single preventive or therapeutic trial valid?
Was the assignment of patients to
treatments randomized?
Was the randomization list concealed?
Was follow-up of patients sufficiently long
and complete?
Were all patients analyzed in the groups
to which they were randomized?
Were patients, clinicians, and study
personnel kept “blind” to treatment?
Were the groups treated equally, apart
from the experimental treatment?
Were the groups similar at the start of the
trial apart from the experimental therapy?
Udayana University Faculty of Medicine, MEU
44
Study Guide of Evidence-based Medical Practice Block
Are the valid results of this randomized trial important?
What is the magnitude of the treatment
effect?
How precise is the estimate of the
treatment effect?
SAMPLE CALCULATIONS
Occurrence of diabetic
neuropathy at 5 years among
insulin-dependent diabetics in
the DCCT trial
Relative risk
reduction
(RRR)
Absolute risk
reduction
(ARR)
Number
needed to
treat (NNT)
CER – EER
1/ARR
Usual insulin
regimen
control event
rate (CER)
Intensive insulin
regimen
experimental
event rate (EER)
CER – EER
9.6%
2.8%
9.6% – 2.8%
9.6% – 2.8%
1/6.8%
9.6%
=6.8%
=15 patients
4.4% to 9.2%
11 to 23
CER
=71%
95% CI
a95% confidence interval (CI) on an NNT
=1/(limits on the CI of its ARR)
CER  1  CER 
EER  1  EER 

 

 1.96 


 number of control patients   number of experimental patients 
 0.96  0.904   0.028  0.972 
 1.96 


730
711

 

 2.4%
Udayana University Faculty of Medicine, MEU
45
Study Guide of Evidence-based Medical Practice Block
YOUR CALCULATIONS
Relative risk
reduction
(RRR)
CER
EER
CER – EER
Absolute risk
reduction
(ARR)
Number
needed to
treat (NNT)
CER – EER
1/ARR
CER
95% CI
Can you apply this valid, important evidence about therapy in caring your patient?
Do these results apply to our patient?
Is our patient so different from those in
the study that its results cannot apply?
Is the treatment feasible in our setting?
What are our patient’s potential benefits and harms from the therapy?
Method I: f
Risk of the outcome in our patient, relative
to patients in the trial.
Expressed as a decimal:______
NNT/f=______/______=______
(NNT for patients like ours)
Method II: 1/(PEERRRR)
Our patient’s expected event rate if they
received the control treatment (PEER)
=______
1/(PEERRRR)=1/________=______
(NNT for patients like ours)
Are our patient’s values and preferences satisfied by the regimen and its
consequences?
Do we and our patient have a clear
assessment of their values and
preferences?
Udayana University Faculty of Medicine, MEU
46
Study Guide of Evidence-based Medical Practice Block
Are they met by this regimen and its
consequences?
Additional notes:
HARM WORKSHEET
Citation:
Are the results of this harm study valid?
Were there clearly defined groups of
patients, similar in all important ways other
than exposure to the treatment or other
cause?
Were treatments/exposures and clinical
outcomes measured in the same ways in
both groups (was the assessment of
outcomes either objective or blinded to
exposure)?
Was the follow-up of study patients
sufficiently long and complete?
Do the results satisfy some “diagnostic tests for causation”?
Is it clear that the exposure preceded the
onset of the outcome?
Is there a dose–response gradient?
Is there positive evidence from a
“dechallenge–rechallenge” study?
Is the association consistent from study to
study?
Does the association make biological
sense?
Udayana University Faculty of Medicine, MEU
47
Study Guide of Evidence-based Medical Practice Block
Are the valid results from this harm study important?
What is the magnitude of the association
between the exposure and outcome?
What is the precision of the estimate of the
association between exposure and
outcome?
Adverse outcome
Exposed to
the
treatment
Totals
Present (case)
Absent (control)
Yes
(cohort)
a
b
a+b
No
(cohort)
c
d
c+d
a+c
b+d
a+b+c+d
Totals
In a randomized trial or cohort study: relative risk=RR=[a/(a+b)]/[c/(c+d)].
In a case–control study: odds ratio (or relative odds)=OR=ad/bc.
Should these valid, potentially important results change the treatment of your
patient?
Do the results apply to our patient?
Is our patient so different from those in
the study that its results don’t apply?
What are our patient’s risks of the
adverse event?
To calculate the NNH (number of patients
we need to treat to harm one of them) for
any odds ratio (OR) and our patient’s
expected event rate for this adverse
event if they were not exposed to this
treatment (PEER):
NNH 
PEER  OR  1  1
PEER  OR  1  1  PEER 
Udayana University Faculty of Medicine, MEU
48
Study Guide of Evidence-based Medical Practice Block
What are our patient’s preferences,
concerns and expectations from this
treatment?
What alternative treatments are
available?
Additional notes:
SGD 8. Critical Appraisal of Study about Prognosis. Theory
Please explain:
Is this evidence about prognosis valid?
1. Was a defined, representative sample of patients assembled at a common (usually
early) point in the course of their disease?
2. Was patient follow-up sufficiently long and complete?
3. Were objective outcome criteria applied in a ‘blind’ fashion?
4. If subgroups with different prognosis are identified:
a. Was there adjustment for important prognostic factors?
b. Was there validation in an independent group of ‘test-set’ patients?
Is this valid evidence about prognosis important?
1. How likely are the outcomes over time?
2. How precise are the prognostic estimated?
Critical Appraisal of Study about Prognosis. Article
An article will be given two days before the discussion
Please fill in the worksheet after discussing an article about prognosis that has been given
to you.
PROGNOSIS WORKSHEET
Citation:
Udayana University Faculty of Medicine, MEU
49
Study Guide of Evidence-based Medical Practice Block
Are the results of this prognosis study valid?
Was a defined, representative sample of
patients assembled at a common (usually
early) point in the course of their disease?
Was patient follow-up sufficiently long
and complete?
Were objective outcome criteria applied in
a “blind” fashion?
If subgroups with different prognoses are
identified:
 Was there adjustment for important
prognostic factors?
 Was there validation in an independent
group (‘test set’) of patients?
Are the valid results of this prognosis study important?
How likely are the outcomes over time?
How precise are the prognostic
estimates?
IF YOU WANT TO CALCULATE A CONFIDENCE INTERVAL AROUND THE MEASURE
OF PROGNOSIS
Clinical measure
Proportion (as in the rate
of some prognostic event,
etc.) where:
n=the number of patients
Standard error (SE)
 p  1  p  / n 
where p is proportion
and n is number of
patients
p=the proportion of these
patients who experience
the event
Udayana University Faculty of Medicine, MEU
Typical calculation of CI
If p=24/60=0.4 (or 40%) and
n=60:
SE  0.4  1  0.4  / 60
=0.063 (or 6.3%)
95% CI is 40%±1.96×6.3% or
27.6% to 52.4%
50
Study Guide of Evidence-based Medical Practice Block
n from your evidence:___
Your calculation:
p from your evidence:___
SE:____________
95% CI:________
Can you apply this valid, important evidence about prognosis in caring your patient?
Do the results apply to our patient?
Is our patient so different from those in
the study that its results cannot apply?
Will this evidence make a clinically
important impact on our conclusions
about what to offer or tell our patient?
Additional notes:
SGD 8. How To Write a paper and Present at a Meeting
Learning Task will be given before SGD
PRACTICAL WORK
PRACTICAL WORK 1: RECORDS KEEPING
IP.G. Adiatmika
Compiling and keeping an accurate and clinically meaningful patient’s record is an important
basic clinical skill in primary care practice. The use of personal computer is now becoming
increasingly common in keeping the patient’s record. There are many approaches to
keeping the patient’s record for easy access and retrieval. In this module we will learn how
to store patient’s data in Microsoft Excel for ready access and retrieval, and use the data in
merge file for referral letter for a second opinion or further investigation.
Storing patient’s data in one table on a worksheet or in separate worksheets can be done by
using the Microsoft Excel. The table may contain information such as registration number,
patient’s personal identity, history, results of physical examination, laboratory investigations,
imaging study, treatment and management. Data may be put in one column in order to
make it easier for storing and retrieval. Each patient’ data is put in one row as one data
base. This data may be used and copied for various purposes, such as the compiling of
referral letter. Electronic letters can be designed in Microsoft Word by using the mail merge.
By using the templates on the mail merge menu, we can design the letter and put the
database in the letter.
Udayana University Faculty of Medicine, MEU
51
Study Guide of Evidence-based Medical Practice Block
PRACTICAL WORK 2: SEARCHING ARTICLES
(Track down the best evidence)
Dewi Sutriani Mahalini
Learning task:
Step 1. Create a search strategy
Step 2. Convert the questions to search strategy
1. Underline the key terms
2. Number in order of importance from 1 to 4
3. Think of alternative spellings, synonyms and truncations
Scenario #...................
Clinical research question :
Question part
Question term
Synonyms
P Population/problem
OR
I Intervention
OR
C Comparator/control
OR
O Outcome
OR
Udayana University Faculty of Medicine, MEU
Number in
order of
importance
52
Study Guide of Evidence-based Medical Practice Block
Result of search (Journal):
1. ..............................................................................................
..............................................................................................
.............................................................................................
2. ..............................................................................................
..............................................................................................
..............................................................................................
3. ..............................................................................................
..............................................................................................
..............................................................................................
STUDENT PROJECTS
Scenario 1
Mastitis is one of the most common problems experienced by women who are
breastfeeding. The aims of this paper are to found the prevalence rates and compare of
mastitis in primiparous women receiving public hospital care (standard or birth centre) and
care in a co-located private hospital. The RCT (Attachment to the Breast and Family
Attitudes to Breastfeeding, ABFAB) which was designed to test whether breastfeeding
education in mid-pregnancy could increase breastfeeding duration recruited public patients
at the Royal Women's Hospital at 18–20 weeks gestation. A concurrent survey recruited
women planning to give birth in the Family Birth Centre/FBC (at 36 weeks gestation) and
women in the postnatal wards of Frances Perry House/FPH (private hospital). All women
were followed up by telephone at 6 months postpartum. Mastitis was defined as at least 2
breast symptoms (pain, redness or lump) AND at least one of fever or flu-like symptoms.
Prevalence of mastitis in ABFAB, FBC and FPH was 17%, 23% and 24%, respectively. The
characteristic of data was presented in table 1 below.
Udayana University Faculty of Medicine, MEU
53
Study Guide of Evidence-based Medical Practice Block
Question:
1. Which of the characteristic above was classified data as Nominal, Ordinal dan
Numeric (ratio)?
2. How many groups were compared in this study? Explain each groups are there
paired groups or unpaired groups?
3. How did you compare the prevalence of mastitis between groups?
Scenario 2
In this study the effects of supplementation of iron and zinc, alone or combined, on iron
status, zinc status and growth in Indonesian infants is investigated. In a randomized,
double-blind, placebo controlled supplementation trial, 478 infants, 4 mo of age, were
supplemented for 6 mo with iron (10 mg/d), zinc (10 mg/d), iron 1 zinc (10 mg of each/d) or
placebo. Anthropometry was assessed monthly, and micronutrient status was assessed at
the end of supplementation. The results were presented in table 2 below
Udayana University Faculty of Medicine, MEU
54
Study Guide of Evidence-based Medical Practice Block
Question
1. Explain the type of data of ‘hemoglobin, g/L’; hemoglobin that classified as anemia
and moderate/severe anemia’; ‘plasma ferritin’ and ‘plasma zinc’ !!
2. How many groups were compared in this study? Are there paired or unpaired
groups?
3. Why the hemoglobin level was reported as mean ± SD, on the other hand the
plasma ferritin level as median (minimum-maximum)?
4. What kind of hypothesis testing or statistic analysis was used to compared :
a. Hemoglobin level (g/L) between groups?
b. Plasma ferritin level (µg/L) between groups?
c. Plasma zinc level (µmol/L) between Placebo and Iron group?
d. Plasma ferritin level (µg/L) between Placebo and Zinc group?
e. Hemoglobin that classified as anemia and moderate/severe anemia between
groups?
Scenario 3
Based on scenario 2 the researcher wants to compare the proportion of anemia between
placebo and iron group. The results were presented in table 3 below.
Table 3. Proportion of anemia between placebo and iron group
Supplementation group
Placebo
Iron
n = 87
n = 90
Anemia
72
29
101
Non-anemia
15
61
76
Total
87
90
177
Udayana University Faculty of Medicine, MEU
Total
55
Study Guide of Evidence-based Medical Practice Block
Question
1. What kind of hypothesis testing or statistic analysis was used to compare the
proportion of anemia between placebo and iron group? And why did you use this
test?
2. Calculate the relative risk and confidence interval of this study!
3. How we conclude this study if p value < 0,05?
Scenario 4
An increase in the incidence of necrotizing fasciitis (NF) occurring in previously healthy
children with primary varicella was noted in the Washington State area between December
1993 and June 1995. Our objective was to investigate ibuprofen use and other risk factors
for NF in the setting of primary varicella. Case-control study was done. Demographic
information, clinical parameters, and potential risk factors for NF were compared for cases
and controls. A case was defined as a child with NF hospitalized within 3 weeks of primary
Varicella (n 5 19). Controls were children hospitalized with a soft tissue infection other than
NF within 3 weeks of primary Varicella (n 5 29). Odds ratios (ORs) of ibuprofen, as well as
other potential risk factors were evaluated. The results were presented in table below.
Question
1. How did you calculate the odds ratio and confidence interval of this study, especially
for ibuprofen treatment?
2. And how did it mean?
Udayana University Faculty of Medicine, MEU
56
Study Guide of Evidence-based Medical Practice Block
Scenario 5
Correlations among measures of quality in HIV care in the United States. Cross sectional
study of data abstracted during an evaluation of an initiative to improve quality of care for
people with HIV. Data sources were medical records of 9020 patients. Main outcome
measures were adjusted performance rates at site level for eight measures of quality of care
specific to HIV and a site level summary performance score (the number of measures for
which the site was in the top quarter of the distribution). The results were presented in table
2 and 3 below.
Question: how did your comment about this correlation study?
SELF-ASSESSMENTS
Introduction
1. What is Evidence Based Medical Practise (EBMP) ?
2. Why do we need EBMP?
3. Please describe the steps in EBMP !
4. Please describe the components of a good clinical questions !
Udayana University Faculty of Medicine, MEU
57
Study Guide of Evidence-based Medical Practice Block
5. Two additional elements of the well built clinical question are the type of question
and the type of study. This information can be helpful in focusing the question and
determining the most apppropriate type of evidence. Please describe the type of
questions and the typeof the study !
Searching articles
1. Demonstrate how you get access to the internet/on line resources !
2. Select the appropriate resources and conduct the search !
3. Select & download the relevant articles/journals to the clinical question !
Lecture 1: Principles of critical appraisal
1. Describe the principles of critical appraisal and explain the meaning of critical
appraisal.
2. Explain the process of critical appraisal.
3. Describe the basic principles for assessing validity, results and relevance of an
article before using it to inform a decision.
4. Show how to do a medical literature research.
Lecture 2: Association and causation
1. Even though cigarettes have been well established as a risk factor for development
of lung cancer, apparently not every one with lung cancer has smoked cigarette. In
this case, the role of cigarette smoking is a:
 Necessary cause of lung cancer
 Contributory cause of lung cancer
2. Mycobacterium tuberculosis was established as a cause of tuberculous disease
because the organism is always found in tuberculous disease. In this case, the role
of Mycobacterium tuberculosis is a:
 Contributory cause
 Necessary cause
Lecture 3: Effect size, hypothesis testing and CI
A result of one study found that there are association between cigarette smoke and
coronary heart disease with OR 3.5 (95% CI 2.9 to 4.6), P=0.001.
1. Which one of them is effect size and what does it mean?
2. Which one of them is confidence interval and what does it mean?
3. Which one of them is the end result of hypothesis testing and what does it mean?
Udayana University Faculty of Medicine, MEU
58
Study Guide of Evidence-based Medical Practice Block
Lecture 4: Principles and applications of statistical analysis
1. Explain the type of data based on scale of measurement and give the example,
respectively!
2. Explain the method and the example of bivariable analysis!
3. Explain the method and the example of multivariable analysis!
Lecture 5: Methodological and Statistical Principles and Applications in Descriptive
Studies
1. Describe a dichotomous and categorical variables, give an example for each.
2. Describe a continuous variables, with a preference of using mean (standard
deviation, SD), median (range) and 95% confidence interval for the mean in the
report of descriptive study.
3. Try to make a descriptive data of study sample into a bar-chart, pie-chart, linegraphic.
4. Try to make a descriptive data of a study sample into a table of base characteristics
of the sample.
Lecture 6: Methodological and Statistical Principles and Applications in Analytical
tudies
1. Explain the 5 steps of statistical analysis to know the relation between independent
and dependent variable!
2. Explain the criteria of parametric test and criteria of chi square test!
3. Explain the alternative test if there is not appropriate for the parametric and chi
square test!
4. Explain the interpretation of each statistical analysis if p value < 0.05!
Lecture 7: Diagnostic test. Principles of critical appraisal
1. Please, explain the aim of “blind and independent evaluation” of the new test
2. Give reason, why the follow-up of subjects who did not undergo gold standard is
important
3. What parameters are needed to determine post-test probability
4. Please, explain why the likelihood ratio is more applicable than sensitivity or
specificity in looking at the importance of the new test
5. How to set the optimal cut-off value using ROC curve
Lecture 8: Clinical trial. Principles of critical appraisal
1. What is the purpose of randomization in clinical trial
2. Why it is important that the randomization list concealed?
3. What is the basic concept if intention-to-treat analysis
Udayana University Faculty of Medicine, MEU
59
Study Guide of Evidence-based Medical Practice Block
4. What is the purpose of double blind treatment in clinical trial
5. Please, mention the definition of “number need to treat”
6. Please describe in words, “relative risk”
Lecture 9: Study about Prognosis. Principles of critical appraisal
1. When you look at a study about prognosis, how can you know whether the patients
were followed-up sufficiently long and complete
2. Please explain “inception cohort”
3. How can we control the confounding factors in cohort study
4. What the basic differences between cohort and case-control study
5. Explain, why the objective outcome criteria should be applied in a ‘blind’ fashion
6. The is the basic understanding of “risk
Lecture 10: How to write a paper
1. Mention and describe the basic structure of a paper as summarized by the acronym
IMRAD.
2. Mention four elements that should be included in every manuscript introduction
3. Mention three elements that should be included in the methods of manuscript
4. Describe the major contents in the results of manuscript
5. Mention six elements that should be included in the discussion of manuscript
How to present at meeting
1. Mention and describe the five elements of preparation of presentation
2. Describe the contents of a presentation
3. Describe the delivery technique of your talk
4. Describe about the principle of making slides using Power Point
Udayana University Faculty of Medicine, MEU
60
Study Guide of Evidence-based Medical Practice Block
~ CURRICULUM MAP ~
Smstr
Program or curriculum blocks
10
Senior Clerkship
9
Senior Clerkship
8
Senior clerksh
ip
7
Medical
Emergency
(3 weeks)
Special Topic:
-Travel medicine
(2 weeks)
Elective Study III
(6 weeks)
Clinic Orientation
(Clerkship)
(6 weeks)
6
BCS (1 weeks)
The Respiratory
System and
Disorders
(4 weeks)
The
Cardiovascular
System and
Disorders
(4 weeks)
The Urinary
System and
Disorders
(3 weeks)
The Reproductive
System and
Disorders
(3 weeks)
BCS (1 weeks)
Alimentary
& hepatobiliary systems
& disorders
(4 Weeks)
BCS (1 weeks)
The Endocrine
System,
Metabolism and
Disorders
(4 weeks)
BCS (1 weeks)
Clinical Nutrition
and Disorders
(2 weeks)
BCS (1 weeks)
BCS (1 weeks)
Musculoskeletal
system &
connective
tissue disorders
(4 weeks)
Neuroscience
and
neurological
disorders
(4 weeks)
Behavior Change
and disorders
(4 weeks)
BCS (1 weeks)
Hematologic
system & disorders & clinical
oncology
(4 weeks)
BCS (1 weeks)
Immune
system &
disorders
(2 weeks)
BCS(1 weeks)
Infection
& infectious
diseases
(5 weeks)
BCS
(1 weeks)
The skin & hearing
system
& disorders
(3 weeks)
BCS (1 weeks)
Medical
Professionalism
(2 weeks)
BCS(1 weeks)
Evidence-based
Medical Practice
(2 weeks)
BCS (1 weeks)
Health Systembased Practice
(3 weeks)
BCS(1 weeks)
Community-based
practice
(4 weeks)
-
BCS (1 weeks)
Studium
Generale and
Humaniora
(3 weeks)
Medical
communication
(3 weeks)
BCS (1 weeks)
The cell
as biochemical machinery
(3 weeks)
Growth
&
development
(4 weeks)
BCS (1 weeks)
BCS(1 weeks)
BCS: (1 weeks)
BCS (1 weeks)
Elective Study
II
(1 weeks)
5
4
3
2
1
BCS (1 weeks)
Special Topic :
- Palliative
medicine
-Compleme
ntary &
Alternative
Medicine
- Forensic
(3 weeks)
Elective
Study II
(1 weeks)
Special Topic
- Ergonomi
- Geriatri
(2 weeks)
Elective
Study I
(2 weeks)
The Visual
system &
disorders
(2 weeks)
Pendidikan Pancasila & Kewarganegaraan (3 weeks)
Udayana University Faculty of Medicine, MEU
61
Study Guide of Evidence-based Medical Practice Block
REFERENCES
Student Standard References
1. Udayana Medical Informatics Team: Medical informatics (Guide Book), 2003
Additional Reference
1. Strauss SE: Evidence Based Medicine: How to Teach and Practice EBM 2005
Udayana University Faculty of Medicine, MEU
62