Introduction to

EVIDENCE-BASED
MEDICINE

DEPARTMENT OF FAMILY & COMMUNITY MEDICINE

UP-PHILIPPINE GENERAL HOSPITAL
 WHAT IS EVIDENCE BASED MEDICINE?

The conscientious, explicit,

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

It means integrating individual

clinical expertise with the best
Dr. David Sackett, 1996 available external clinical evidence
from systematic research.
EVIDENCE-BASED
MEDICINE
 STEPS IN EBM PROCESS

The patient 1. Start with the patient – a clinical problem or

question arises out of the care of the patient

The question 2. Construct a well built clincial question from the case

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

The evaluation 4. Appraise the evidence for its validity and applicability

The patient 5. Return to the patient – intergrate the evidence with the
clinical expertise, patient preference and apply it to practice

Self evaluation 6. Evaluate your performance with this patient

 STEP 1: CLINICAL SCENARIO

 Maria, a 67 year old

 History of congestive
heart failure brought on
by several myocardial
infarctions.

 Hospitalized 2x in the last

6 months for worsening of
heart failure

Duke University Medical Center and Health Sciences Library, 2006

 STEP 1: CLINICAL SCENARIO

 Presently in normal sinus rhythm

 She is extremely compliant about taking her medications

(enalapril, aspirin and simvastatin)

 She desperately wants to stay out of the hospital

 You think she should also be taking digoxin, but you are not
certain if this will help keep her out of the hospital.

 You decide to research this question before her next visit.

Duke University Medical Center and Health Sciences Library, 2006

STEP 2. CONSTRUCT A WELL BUILT CLINICAL QUESTION
ANATOMY OF A CLINICAL QUESTION

POPULATION • primary problem, disease, or co-existing conditions

• sex, age or race of a patient

INTERVENTION • main intervention, prognostic factor, or exposure

OR TREATMENT • prescribe a drug, Order a test? Order surgery?

COMPARISON OR • main alternative to compare with the intervention

CONTROL • Another drug? Another test? Placebo?
• Your clinical question does not always need a specific
comparison

OUTCOME • What do you hope to accomplish/measure/improve/affect?

• What are you trying to do for the patient?
• Symptoms? Number of adverse events? Functionality? Test
scores?

METHODOLOGY • Type of Evidence/Study

 GOING BACK TO MARIA

Patient / Problem Elderly with congestive heart

failure
Intervention Digoxin

Comparison Placebo

Outcome Primary: reduce number of

hospitalizations

Secondary: reduce mortality

Methodology Randomized Controlled Trial
Meta-analysis
 TYPE OF QUESTION: WHAT TYPE OF STUDY?

Type of Question Suggested best type of Study

Therapy RCT>cohort > case control > case series
Cross sectional>prospective, blind comparison
Diagnosis
to a gold standard
Etiology/Harm RCT > cohort > case control > case series
Prognosis cohort study > case control > case series
Prevention RCT>cohort study > case control > case series
Cost economic analysis

What type of study is best suited for Maria’s case?

 STEP 3: SELECT THE APPROPRIATE RESOURCE AND
CONDUCT THE SEARCH

SEARCH STRATEGIES
 Identify the concepts: KEY TERMS
 Phrase search: “quotation marks”
 Boolean Principle: OR, AND
 MeSH
 Truncation and Wild Card
 Limits
 Related Articles
 Clinical Queries
 BOOLEAN PRINCIPLE

By using OR &
Congestive
Heart Digoxin AND…
Failure A
How do we get
D
B C •A?
•B?
•C?
Hosptialization
•D?
 OTHER WEBSITES

NATIONAL GUIDELINE CLEARINGHOUSE

THE COCHRANE COLLABORATION
NEW ENGLAND JOURNAL OF MEDICINE
SPECIALTY JOURNALS
Once you find an article,
What will you do with it?
 Do the Critical Appraisal!

USE
 The Evidence-Based Family and Community Practice
(EBFCP) and Quality Improvement in Health Care Manual
(FMRG, 2003)

Other source to Understand:

 Painless EBM by Dr. Dans
 World Wide Web

 BUT the guide questions to be answered must come from

the EBFCP Manual
 FORMAT OF JOURNAL REPORT

 Case Scenario
 Research Question
 Search
 Title
 Source
 Authors
 Appraisal
 EVIDENCE-BASED FAMILY AND
COMMUNITY PRACTICE

I. Is it relevant?
II. Is it valid?
III. What are the results?
IV. Is it applicable to my patient?
Critical Appraisal of An Article
on Diagnostics
 Decision analysis

 appraisal of an article on diagnostics start with the

decision analysis line

0 10%(DT) 50% 60% 80% (TT) 100%

Diagnostic Pre-test Treatment
Treshold Probability Treshold
 Decision Analysis Line

 Diagnostic Threshold (DT): arbitrary point at which

you rule out the disease below this point
 After this point, you will request for a diagnostic test
 Therapeutic Threshold: arbitrary point at which you
decide to treat beyond this point
 Below this point, you will request for a diagnostic test
 Pre-Test Probability: probability that the patient has
the disease
 Post-Test probability: probability that the patient has
the disease after doing the diagnostic exam
 Case Scenario

 24 year old male

 Presenting with 1 month history of RUQ pain
 Started out as epigastric pain then localizing in the
RUQ
 Usually associated with intake of fatty food
 No other associated symptom
 PE: (-) Murphy’s

 What is our pre test probability that this patient has

Acute Cholelithiasis?
 Will we request for an HBT ULTRASOUND?
 Determine the POST TEST PROBABILITY

APPRAISE

SEARCH

0 10%(DT) 60% 90% (TT) 100%

pre-test probability
 Is it relevant?

 Is the objective of the study similar to your clinical

dilemma?
CLINICAL STUDY
DILEMMA
Population

? ?
Intervention and
comparative
intervention
Methodology

Outcome
 Is it valid?

PRIMARY VALIDITY GUIDES

1. Was there an independent comparison with a reference
standard?
2. Did the patient sample include an appropriate spectrum of
patients to whom the test will be done?

SECONDARY VALIDITY GUIDES

1. Was the reference standard done regardless of the result of
the of the diagnostic test being evaluated?
2. Were the methods for performing the test described in
sufficient detail to permit replication?
 PRIMARY VALIDITY GUIDES

1. Was there an independent comparison with a

reference standard?

Reference standard.
Accuracy.
Precision.

* The reference standard should not be part of the

diagnostic procedure in question.
 PRIMARY VALIDITY GUIDES

2. Did the patient sample include an

appropriate spectrum of patients to whom the
test will be done?
 Inclusion and Exclusion Criteria.
 Baseline Characteristics (table 1).
 Representativeness includes subjects with the
whole spectrum of the disease

 The accuracy of a diagnostic test among patients

with low risk for the disease is different from
patients with high risk for the disease.
 SECONDARY VALIDITY GUIDES

1. Was the reference standard done regardless

of the result of the of the diagnostic test
being evaluated?
 SECONDARY VALIDITY GUIDES

2. Were the methods for performing the test

described in sufficient detail to permit
replication?
 2. Were the methods for performing the test
described in sufficient detail to permit replication?

 Look at the results and methodology section.

 Clear procedure including the preparation of subjects.
 Diets, drugs to avoid.

 Precautions.

 Step by step descriptions.

 Be able to duplicate the exam in your setting and still

get the same results.
 II. Overall is the study valid?

 If you want to be strict about it, you should answer

yes in all 4 questions.

 A simple rule might be to answer yes to at least, one

primary guide and two secondary guides.
 III. What are the results?

What are the likelihood ratios for the different test

results?

Likelihood ratios indicates by how much a given test

result increases the pre-test probability of the disease.
LR of 1 means that the pre-test probability is same after
doing the test
LR >1 increases the probability that the disease is
present
 Sensitivity & Specificity

Disease

Present Absent
Test Positive TP FP
Negative FN TN

SENSITIVITY: probability/likelihood that the

diseased patient will test positive: TP / TP + FN

SPECIFICITY: probability/likelihood that those

without the disease will test negative: TN/ TN + FP
Remember? SpPIN & SnNOUT
 LIKELIHOOD RATIO

 Likelihood ratio of a positive test

 Probability that the test is + among diseased person

Probability that the test is + among non-diseased person

 LR (+) = Sensitivity/1-Specificity

 Likelihood ratio of a negative test

 Probability that the test is - among diseased person

Probability that the test is - among non-diseased person

 LR (-) = 1-Sensitivity/Specificity
 What do Likelihood ratios do?

 It can change the probability of the disease after a

diagnostic test is done.
 Remember our Case Scenario

0 10%(DT) 60% 90% (TT) 100%

pre-test probability
Will the likelihood ratios
shift the pre-test
probability
 Use a NOMOGRAM
 Given a pre test
probability of 60%
 The HBT UTZ yielded a
LR of 2.5
 PLOT
 Post test Probability is
93%

 What does this mean?

 Determine the POST TEST PROBABILITY

APPRAISE

SEARCH

0 10%(DT) 60% 90% (TT) 100%

pre-test probability
IV. Can the results help me in caring for my
patients?

1. Will the reproducibility of the test result and its

interpretation be satisfactory in my setting?
 Is the interpretation simple enough?
 Is the basis of the interpretation clear and specific?

2. Are the results applicable to my patient?

 setting is somewhat similar
 inclusion criteria include characteristics of your patient
 clinical judgment is required

3. Will the results change my management?

Critical appraisal of an article
on therapeutics
 Is it relevant?

 Is the objective of the study similar to your clinical

dilemma?
CLINICAL STUDY
DILEMMA
Population

? ?
Intervention and
comparative
intervention
Methodology

Outcome
 I. Is it valid?

Primary Validity Guides

1. Was the assignment of patients to treatment randomized?
2. Were all the patient who entered the trial properly
accounted for and attributed at its conclusion?
Secondary Validity Guides
1. Were patient, their clinicians and study personnel
“blinded”?
2. Were the groups similar at the start of the trial?
3. Aside from the experimental intervention, were the
groups treated equally?
 1. Was the assignment of patients to treatment
randomized?
 Look at the abstract or methodology.
 Define randomization.
 Random sampling?

 Randomization assures that both known and unknown

determinants of outcome are even distributed between the
treatment and control

 Avoids selection bias

 More severe disease = worse results
 less severe disease = better results
2. Were all the patient who entered the trial properly
accounted for and attributed at its conclusion?

b. Were the patients analyzed in the group to which

they were randomized?

 patients, including drop outs or withdrawn, were

analyzed to their original grouping at the start of
the study
 This preserves the value of randomization
 No crossing over
2. Were all the patient who entered the trial properly
accounted for and attributed at its conclusion?

a. Was follow up complete?

 Methodology and Results Section
 Look at the number of patients enrolled at the outset and
compare this with the number of patient reported at the
results table.
 What is the drop out rate?
 20 % or more is considered substantial
 If less than 20% dropped out, check if and intention to treat
analysis was done

 What is intention to treat analysis?

 I. Is it valid?

Primary Validity Guides

1. Was the assignment of patients to treatment randomized?
2. Were all the patient who entered the trial properly
accounted for and attributed at its conclusion?
Secondary Validity Guides
1. Were patient, their clinicians and study personnel
“blinded”?
2. Were the groups similar at the start of the trial?
3. Aside from the experimental intervention, were the
groups treated equally?
 1. Were patients, their clinicians, and study
personnel "blind" to treatment?

 Look at the abstract or methodology.

 Define blinding.
 What is its importance?
 SINGLE blinding?

 DOUBLE blinding?

 TRIPLE blinding?
 2. Were the groups similar at the start of the trial?

 Baseline characteristics usually labeled

Table no. 1

 The greater the similarity between known prognostic

factors for the control and experimental group, the
more likely that the results can be attributed to the
intervention, rather than due to the differences in
these factors
 3. Aside from the experimental intervention,
were the groups treated equally?

 Look for interventions other than the treatment under

study
 “co-interventions” distort the results

 Where outcomes measurements clearly described and

determined in the same way between interventions?
 DEFINE
 Clinical Outcomes
 Surrogate Outcomes
 II. Overall is the study?

 If you want to be strict about it, you should answer

yes in all 5 questions.

 A simple rule might be to answer yes to at least, one

primary guide and two secondary guides.
 III. WHAT ARE THE RESULTS?

a. How large was the treatment effect?

RISK IN TREATMENT (Rt)

No. of patients who did not get well in the treatment group
Total no. of patients in the treatment group

RISK IN CONTROL (Rc)

No. of patients who did not get well in the control group
Total no. of patients in the control group
 a. What are the results?

 Absolute Risk Reduction (ARR) = Rc – Rt

 Relative Risk (RR) = Rt/Rc

RR of 1 : No difference between Treatment and

Control
RR of >1: Treatment is more harmful
RR of <1: Treatment is more effective

 Relative Risk Reduction (RRR) = 1 – RR

 B. How precise was the treatment effect?

 What is the confidence interval?

 What is the p-value?
IV. Are the results applicable to my patients?

 Are the medical, social and economic resources needed

to administer the treatment available in your setting?

 Consider the acceptability by your patient, his family and

community.

 Are the likely treatment benefits worth the potential

harm and costs?

Number needed to treat – 1/ARR

Cost effectiveness Formula: NNT x price/unit x
dose x duration
THANK YoU!
The doctor of the future
will give no medicine, but
will interest her or his
patients in the care of the
human frame, in a proper
diet, and in the cause and
prevention of disease.
Thomas A. Edison
US inventor (1847 - 1931)