Therapy Study

Sitti Wahyuni
Evidence Based Medicine

Same slides modified from P.Trihono’s

CEEBM Unit, Cipto Mangunkusumo Hospital Faculty of Medicine
Universitas Indonesia
 Clinical trial
• Is the specific form of EXPERIMENTAL STUDY (INTERVENTIONAL
STUDY)
• To determine cause – effect relationship
• The investigator assigns who should receive which treatment, and
analyzes the outcome of the intervention
• A cohort- prospective- interventional- experiment- controlled study

21/03/2024 Sitti Wahyuni/ EBM 2

21/03/2024 Sitti Wahyuni/ EBM 3
 Level of evidence in Therapeutic Study

21/03/2024 Sitti Wahyuni/ EBM 4

Structure of a Simple Trial
Comparing Two Treatments

Population of patients
with condition

21/03/2024 Sitti Wahyuni/ EBM 5

Classification of clinical trial
1. Type of intervention
• Majority form: Concern with the evaluation of drug therapy (drug trials)
• Other forms: Surgical procedures, radiotherapy for cancer, medical advice
(diet, psychological approach, health education, etc)
2. Objective of the study
• Pragmatic trial
• Explanatory trial

21/03/2024 Sitti Wahyuni/ EBM 6

Pragmatic trial
• Attempt: To determine the
Y
cause-effect relationship c a
• Assuming the results
will be applied in
actual clinical
practice
Exp N
• Preferably: binomial
outcome (Yes/ No) R b
• Analysis: Intention to treat
analysis where all
Ctrl Y
randomized subjects are
accounted for the final
calculation according to
their original allocation
N
21/03/2024
a, b, & c are accounted
Sitti Wahyuni/ EBM
as failure of Exp arm 7
Explanatory trial
• Attempt to explain the cause-effect relationship
• Usually in laboratory investigations (pharmacology,
pharmacodynamics, etc)
• Analysis is on treatment analysis:
• only subjects who completed the trial are accounted in the
analysis
• Only minimal dropout is allowed, or replacement for dropouts

21/03/2024 Sitti Wahyuni/ EBM 8

 Key elements of RCT

Control
Randomization
Blinding

• Randomized, Double-Blind, Placebo-Controlled, Clinical Trial

• Randomized Controlled Clinical Trial
21/03/2024 Sitti Wahyuni/ EBM 9
1. Control
• The essence of any research is a comparison
• Compare the experience of a group of patients on new tx with a
control group of similar patients receiving standard tx
• Results of clinical trial based on different effects between control and
experiment groups

21/03/2024 Sitti Wahyuni/ EBM 10

2. Randomization
• Random allocation/ assignment
• The process of assigning trial subjects to treatment or control group
using an element of chance to determine the assignments in order
to reduce bias
• Random allocation:
• Aimed to create comparable groups
• Balances the groups for prognostic factors
• If it is not distributed evenly, it may exaggerate, cancel or
counteract the effects of treatment

21/03/2024 Sitti Wahyuni/ EBM 11

Randomization

SINE QUA NON --- TRUE EXPERIMENTAL

To exclude CONFOUNDING BIAS
Usually = start of trial
21/03/2024 Sitti Wahyuni/ EBM 12
Confounder
• Variable that explains (partially or wholly) the association between
independent and dependent variables, or masks the existing
association
• Might be unevenly distributed between experimental and control
groups

Example
Maternal Low birth weight
nutritional status infant

Low
social status
21/03/2024 Sitti Wahyuni/ EBM 13
 Another example of confounding

Drinking Drinking
Coffee Coffee

Smoking
Cigarettes
Pancreatic
Cancer
Pancreatic
Cancer

The relationship between coffee drinking and pancreatic cancer is confounded by

cigarette smoking.
21/03/2024 Sitti Wahyuni/ EBM 14
3. Blinding= Masking

• A procedure in which one or more parties to the trial are kept unaware of the
treatment assignment
• Aim
• To reduce or eliminate bias, exp. measurement bias
• To avoid patients’ reporting symptoms or clinician’s interpretation being
affected by hunches about whether the treatment is effective
• Highly recommended whenever applicable
• May single or double-blind (the patient, the treatment team, the evaluator, and
data analysts)
• Double blind prevent patient and clinician from adding additional treatment to
just one of the groups
21/03/2024 Sitti Wahyuni/ EBM 15
Is it really effective?

INTERVENTION Outcome
Drug A 70% cure rate
STUDY
PARTICIPANTS R
CONTROL Outcome
Drug B 50% cure rate

21/03/2024 Sitti Wahyuni/ EBM 16

Influential factors for treatment outcome

If a patient is treated with drug E and gets the OutCome, the outcome
(OC) may be influenced by one or more of the following:
⚫ TE→ The treatment effect of the d r u g
⚫ NC→ The natural course of the disease (including the risk profile of the
patients)
⚫ EF→ The extraneous factors are given besides d r u g E
• Attention of the physician,etc
• Psychology of patient OC = TE + NC + EF + IE
• Adjustment lifestyle
• Co-medication
⚫ IE→ The measurement/information about the outcome
21/03/2024 Sitti Wahyuni/ EBM 17
 Are they really different?

NC1 + EF1 + IE1 + TE1 = TO1

NC2 + EF2 + IE2 + TE2 = TO2

NC = Natural Course: age, stage of disease, etc

EF = Extraneous Factors: concurrent drugs, more attention
IE = Information Error = measurement of outcome
TE = Treatment Effect
TO = Treatment Outcome

21/03/2024 Sitti Wahyuni/ EBM 18

Are they really different?

NC1 + EF1 + IE1 + TE1 = TO1

NC2 + EF2 + IE2 + TE2 = TO2

NC= Natural Course: age, stage of disease, etc

EF= Extraneous Factors: concurrent drugs,
Blinding more attention
IE= Information Error = measurement of outcome
Masking/blinding TE= Treatment Effect
TO= Treatment Outcome
Randomization
21/03/2024 Sitti Wahyuni/ EBM 19
 Critical appraisal of clinical trial

- Valid ➢ Methods
- Important ➢ Results
- Applicable ➢ Discussion

21/03/2024 Sitti Wahyuni/ EBM 20

 GATE (GraphicApproachTo Epidemiology):
identifying where errors occur in Epidemiology studies
RAMBOMAN
Recruitment
Allocation
Maintenance
Blind
Objective
Measurements
ANalyses

21/03/2024 Sitti Wahyuni/ EBM 21

R-A-M-B-OM-AN
• R→ has recruited participants relevant to the study objectives?→ who are the
findings applicable to?
• A→ was allocation to EG & CG successful?→ RCT: Allocated by randomization
• M→ how well were participants maintained in the groups they were allocated to
(i.e. to EG & CG) throughout the study? Completeness of follow-up, compliance,
contamination co-interventions
• B→ Were outcomes measured blind to whether the participant ( both ?)
• OM→ Were Outcomes Measured objectively?
• AN→ Were the ANalysis done appropriately?→ Intention to treat?

21/03/2024 Sitti Wahyuni/ EBM 22

A. Validity
Primary guides
1. Was the assignment of patientst treatment randomized?→ Was the
randomization list concealed?
2. Was the follow-up sufficiently long and complete?
3. Were all patients analyzed in the groups to which they were randomized?

Secondary guides: Some less important points

1. Were patients, health workers, and study personnel “blind” to
treatment?
2. Were the groups treated equally apart from the experimental therapy?
3. Were the groups similar at the start of the trial?

21/03/2024 Sitti Wahyuni/ EBM 23

B. Importance

1. To measure the magnitude of the treatment effect

• P- value
• Relative Risk Reduction (RRR)
• Absolute Risk Reduction (ARR)
• Relative Risk Increase (RRI)
• Absolute Risk Increase (ARI)
• Number Needed to Treat (NNT)
• Number needed to harm (NNH)

2. To measure the precise of the treatment effect

• Construct Confidence Interval (CI)

21/03/2024 Sitti Wahyuni/ EBM 24

Exercise: Klissen TP, et al. Nebulized budesonide children with mild
to a moderate group. N Eng J Med 1994;331(5): 285-9.

21/03/2024 Sitti Wahyuni/ EBM 25

Exercise: Klissen TP, et al. Nebulized budesonide children with
mild to a moderate group. N Eng J Med 1994;331(5): 285-9.

• A randomized double-blind control trial on 3 mo – 5 yr old children with mild to

moderate group (laryngotracheobronchitis)
• The experimental group: 2mg (4 ml) nebulized budesonide
• The control group: 4 ml nebulized normal saline.
• The event being prevented: hospital admission due to upper-airway obstruction
• Study protocol 54 children
• 27 (1 hospitalized, 26 non-hospitalized)→ received Budesonide
• 27 (7 hospitalized, 20 non-hospitalized)→ received NACL

21/03/2024 Sitti Wahyuni/ EBM 26

 Exercise
Upper airway obstruction
No Yes (Hospitalized)
Budesonide→ E 26 1 27
NaCl →C 20 7 27

X2= ; df=1 p=0.04

Experiment Event Rate-EER = 1/27=0.04
Control Event Rate-CER = 7/27 = 0.26
Absolut Risk Reduction =EER-CER == 0.04-0.26=0.22
Relative Risk Reduction RRR = (CER-EER)/CER = (0.26-0.04)/0.26 = 85%
• Absolute Risk Reduction (ARR)
• is the difference in risk between the control group and the treatment group
• Relative Risk Reduction (RRR)
• Is the percent reduction in risk in the treated group compared to the control group
• Is a measure of how much the treatment studied has reduced the frequency of an adverse event
21/03/2024 Sitti Wahyuni/ EBM 27
Exercise→ NNT
Upper airway
obstruction • EER = 1/27=0.04
No Yes • CER = 7/27 = 0.26
• ARR = [CER-EER] = 0.26-0.04 = 0.22
Budesonide→ E 26 1 27
• NNT = 1/ARR = 1/0.22 = 5
NaCl →C 20 7 27

• NNT= number needed to treat • NNT= 5

• The average number of patients that need to be • 5 Number of patients should be treated to
treated for a benefit to occur to one person. give benefit in one person
• NNT = 1 / ARR
• The smaller the better

21/03/2024 Sitti Wahyuni/ EBM 28

Exercise→ NNH
MF= adverse effect
Mild Fever
No Yes
Budesonide→ E 8 19 27
NaCl →C 21 6 27
CER = 6/27 = 0.2 • NNH= number needed to harm
EER = 19/27=0.7 • The average number of patients that need to be
exposed for a bad outcome in one person.
ARI = [CER-EER] = 0.2-0.7 = 0.5 • NNH = 2→ 2 patients need to be exposed to
Budesome to get mild fever
NNH = 1/ARI = 1/0.5 = 2 • The higher, the better

21/03/2024 Sitti Wahyuni/ EBM 29

 Summary of risks and benefits of therapy
Outcome CER EER RRR ARR NNT
Upper-airway 26% 4% 85% 22% 5
obstruction

• NNT: We would need to treat average 5 children to prevent one additional child
from being hospitalized due to upper-airway obstruction

Outcome CER EER RRI ARI NNH

Mild fever 20% 70% 250% 50% 2
70%
• NNH: We would only treat 2 children 250%
to harm/ cause one more50%
child being a 2
mild fever

21/03/2024 Sitti Wahyuni/ EBM 30

Confidence interval (CI)
NNT
is to precise the treatment effect CER = 6/27 = 0.26
EER = 1/27=0.04
CI for NNT (1/ARR) ARR = (CER-EER) = 0.26-0.04 = 0.22
NNT = 1/ARR = 1/0.22 = 5
• Calculate CI for ARR (diff
between proportion) 95%CI ARR
• Calculate 1/(upper CL of ARR) = ARR +/- 1.96 V(p1q1/n1 + p2q2/n2)
and 1/(lower CL of ARR) = 0.22 +/- 1.96 V(0.26x0.74)/27+(0.04x0.96/27)
= 0.22 +/- 1.96 x 0.092
= 0.22 +/- 0.18 = 0.04 ; 0.40

95%CI NNT
= 1/0,40 - 1/0.04
21/03/2024 Sitti Wahyuni/ EBM 31
= 3 - 25
Applicability
Can you apply this valid, important evidence about a treatment in caring for
your patient?
• Do these results apply to your patient?
• Is your patient so different from those in the trial that its results
can not help you?
• Is the treatment feasible in your setting
• Are your patient’s values and satisfied by the regimen? And its
preferences consequences?
• Do your patient and you have a clear assessment of their values
and preferences?
• Are they met by this regimen and its consequences?
21/03/2024 Sitti Wahyuni/ EBM 32