5. Consider how you would go about evaluating
the following interventions
– Surgical versus medical termination of pregnancy
– Referral guidelines for radiographic examination
– Paracetamol and/or ibuprofen for treating children with
fever
– Nurse counsellors as an alternative to clinical geneticists
for genetic counselling
– Single dose of chemotherapy versus radiotherapy treating
testicular cancer Cervical cancer vaccine
6. The need to evaluate health care
• Variations in health care
• Unproven treatments
• Inadequacies in care
• Inaccurate medical models
• Limitation of resources
• New innovations
• …
Crombie (1996)
7. Simple Definition
• A study in which people are allocated at
random to receive one of several
interventions
(simple but powerful research tool)
9. • Random allocation to intervention groups
• All participants have equal chance of being
allocated to each intervention group
why RCTs are referred to as randomised
controlled trials
10. • One intervention is regarded as control
treatment (the group of participants who
receive this are the control group)
• NOTE: Contemporaneous (not historical
controls)
why RCTs are referred to as randomised
controlled trials
11. Experimental Group
• receive new intervention
• (also called treatment
group or intervention
group interchangeably)
Control Group
• can be
– conventional practice
– no intervention (this
may be conventional
practice)
– placebo
12. • RCTs are
– Experiments: investigators can influence
number, type, regime of interventions
– Quantitative: measure events rather than try to
interpret them in their natural settings
– Comparative: compare two or more
interventions
13. Inclusion/exclusion criteria
• Decision rules applied to potential trial
participants to judge eligibility for inclusion in
trial
• See CONSORT statement
www.consort-statement.org
• Important that they are applied identically to
all groups in a trial!
14. Randomization
• Randomisation is the process of random allocation
• Allocation is not determined by investigators,
clinicians or participants
• Equal chance of being assigned to each intervention
group
• Individual people
– patients
– caregivers (physicians, nurses etc)
• Groups of people, ‘cluster randomisation’
15. Pseudo-randomization
• Other allocation methods include
– according to date of birth
– the number on hospital records
– date of invitation etc.
• These are NOT regarded as random
• These are called pseudo- or quasi-random
16. • Controlled clinical trials (CCTs) are not the
same as randomised controlled trials
• Controlled clinical trials include non-
randomised controlled trials and
randomised controlled trials
17. Why randomization?
• Characteristics similar across groups at baseline
• Can isolate and quantify impact of interventions
with effects from other factors minimised
• Risk of imbalance not abolished completely even
if perfect randomisation
• To combat selection bias
18. Why do we need a control group?
• Don’t need a control group if completely
predictable results
– Parachutes when jumping from plane
– New drug cures a few rabies cases
• But
– No intervention has 100% efficacy
– Many diseases recover spontaneously
19. Classification of Randomised Trials
• Parallel Group
– Randomly assigned to 2 groups
– Treatment vs non treatment
• Crossover
– Over time random sequence of intervention
• Split Body
– eg Right vs Left Random treatment vs No Treatment
• Cluster
– Pre-existing Groups eg Villages/Schools
• Factorial
– Combination of interventions / non interventions in Random groups
20. Classification of Randomised Trials
By outcome of interest (efficacy vs. effectiveness)
• RCTs - "explanatory"
– Test efficacy
– Highly selected participants
– Highly Controlled conditions
• RCTs – “pragmatic.”
– Test effectiveness in every day practice
– Unselected Participants
– Flexible Conditions
– Give informed decisions about practice
21. Classification of Randomised Trials
By hypothesis (superiority vs. noninferiority vs. equivalence)
• "superiority trials" (most)
– statistically significant
• "noninferiority trials”
– new treatment no worse than existing Rx
• "equivalence trials"
- x2 Rx indistiguishable
22. Classification of Randomised Trials
• By blinding
• Open trials
• Single blind trials
• Double blind trails
• Triple blind trials
23.
24. Randomisation - Goals
• Equal Group Sizes for Adequate Statistical Power
(Especially Subgroup Analyses)
• Low selection bias
(investigator cannot predict the next subject's group
assignment by examining which group has been
assigned the fewest subjects up to that point)
• Low probability of confounding
(i.e., a low probability of "accidental bias"),
25. Methods of carrying out
randomization
• Coin tossing
• Random number table
• Computer based randomization
• Sealed envelopes
• Telephone randomization
• Pharmacy randomization
• Web based randomization
• Attendance sheets
26. Six essential steps
• Drawing up a protocol
• Selecting reference and experimental
populations
• Randomization
• Manipulation or intervention
• Follow up
• Assessment of outcome
27. Drawing up a protocol
• Aims and objectives of the study
• Questions to be answered
• Sample size
• Criteria and procedure for selection of both groups
• Procedure for allocation in both groups
• Treatments to be applied
• Responsibilities of parties
• Once a protocol ahs been made its should be strictly
adhered
• Prevents bias and errors
• Pilot studies’
– Feasibility, acceptability, operational efficiency
28. Selecting reference and experimental
population
Reference (Target)population
• Populations to which the
findings may be applicable
• As broad as mankind or
Limited to geographical area,
age, sex, occupation, social
groups
• Whole city, school children,
industrial workers, obstetric
population
Experimental population
• Derived from reference
population
• Actual population that
participates in study
• Ideally randomly chosen
• Choose a stable population
• 3 criteria's
– Informed consent
– Representative
– Eligible
29. Randomization
• Definition
• Attempt to eliminate bias and allow
comparability- not matching
• Like can be compared to like
• No selection bias
• Best done using random number tables
30. Manipulation
• Intervene or manipulate the study
• Delibrate application or withdrawal
• Reduction of a suspected factor
• Independent variable ( drug, vaccine, habit)
and dependent variable ( incidence of disease,
survival time, recovery period)
31. Follow up
• At definite intervals of time
• In a standard measure
• Equal intensity
• Same given circumstances
• Duration of trial= outcome(start- end)
• Losses to follow up(attrition): death,
migration, loss of interest
32. Assessment
• Positive results
– Reduced incidence of disease
– Reduced cost to HCS
• Negative results
– Side effects
– Complications
– Death
• Bias may arise at three sources
– Participants : Subject variation
– Observer bias
– Bias in evaluation
– Randomization cannot guard against this bias so a technique called blinding is
used.
34. Types of Randomized controlled trials
• Clinical trail
• Preventive trial
• Risk factor trial
• Cessation experiments
• Trial of etiological agents
• Evaluation of health services
35. Clinical trial
• Evaluation of beta-blockers in decreasing CVD
mortality
• Trails of folate supplementation before
conception to prevent NTD.
• Trials of Aspirin in CVD mortality
• RCT of coronary bypass surgery in MI.
37. Risk factor trials
• Study of risk factors
• Major risk factor for CAD is elevated cholesterol,
smoking, HTN, sedentary lifestyle.
• So reducing cholesterol, quitting smoking, control
of BP and regualr physical exercise prevents CAD.
• The Stanford Three community trial
• North Karelia Project
• MRFIT trial
40. Evaluation of Health services
• Effectiveness and efficiency of health services
• Domicillary treatement vs sanitoriaum
treatment of TB
41. Non-randomized trials
• Uncontrolled trials (trials with no comparison groups).
Eg. Pap smear test reduces mortality from Cx CA
• Natural experiments: cigerrete smokig: smokers vs non
smokers
• Before and after comparison studies
• (eg. James Lind and scurvy)
– Using a control group
– Without a control group
Randomization is a statistical procedure by which participants are allocated into groups called “study” and “control” groups, to receive or not to receive AN EXPERIMENTAL PREVENTIVE OR THERAPEUTIC PROCEDURE, MANOEVRE, OR INTERN=VENTION.