3. Types
Types of trials
Randomised clinical trials (RCTs)
Non- Randomised clinical trials-Allocation to the
different treatment groups of a trial is not done
through proper randomization process
4. Level of evidenceFive levels of evidence
Level Research Design Description
Level 1
Randomized controlled
trial (RCT)
Randomized controlled trial, PEDro score 6. Includes within
subjects comparison with randomized conditions and cross-over
designs
Level 2
RCT Randomized controlled trial, PEDro score < 6.
Prospective controlled trial Prospective controlled trial (not randomized)
Cohort
Prospective longitudinal study using at least 2 similar groups with
one exposed to a particular condition.
Level 3 Case control
A retrospective study comparing conditions, including historical
controls
Level 4
Pre-post
A prospective trial with a baseline measure, intervention, and a
post-test using a single group of subjects.
Post-test
A prospective post-test with two or more groups (intervention
followed by post-test and no re-test or baseline measurement) using
a single group of subjects.
Case Series
A retrospective study usually collecting variables from a chart
review.
Level 5
Observational Study using cross-sectional analysis to interpret relations.
Clinical Consensus
Expert opinion without explicit critical appraisal, or based on
physiology, biomechanics or "first principles"
Case Report Pre-post or case series involving one subject
Editors: Riegelman, Richard K.Title: Studying a Study & Testing a Test: How to
Read the Medical Evidence, 5th Edition, 2005, Lippincott Williams & Wilkins
5. Randamization
Methods of Randomization
1. Fixed Randomization Methods
Simple randomization-tossing coin, random number
tables or computer generated random numbers
Block randomization
Stratified randomization
Randomized consent method
2.Adaptive Randomization Methods
Play the winner
Minimization
6. Non randomization
Alternate allocation
Geographical distribution
Cluster allocation
Odd/even number allocation
Based on date of birth/surnames /alphabetical orders
7. Designs
RCT designs
1. Factorial design
2. Cross- over design
3. Cluster random design
4. n-of-1 trials
5. Open trials
9. Statistical tests used
Nonparametric Tests Parametric Tests
Chi-Square test- For frequency data
Fisher’s exact test.
For N x N design and very small
sample size Fisher's exact test should be applied
McNemar test
McNemar test can be used with two
dichotomous measures on the same subjects
(repeated measurements). It is used to measure
change.
Mann-Whitney test -Ordinal data
independent groups.
Wilcoxon signed rank test- Ordinal data
dependent (paired samples)groups
Kruskal-Wallis test -Ordinal data
independent groups (more than 2 groups)
Friedman test- Ordinal data repeated
measures.
Measures of association.
Spearman correlation coefficient- for
correlation of non -parametric data.
One group t-test - Comparison of
sample mean with a population mean.
Independent groups t-test- Comparison
of means from two unrelated groups
Dependent or paired t-test- Comparison
of means from two related samples
k unrelated group, one -way ANOVA
test, two way ANOVA test- Comparison
of means from k unrelated
groups(more than 2 groups)
Repeated measures ANOVA- Interval or
ratio data repeated measures
(Comparison of means from k related
groups)
Measures of Association.
Pearson correlation coefficient- for
correlation of parametric data
10. Parametric tests and analogous nonparametric procedures
Type of analysis Example Parametric tests Nonparametric tests
Compare means between
two independent groups
Is the mean Quality of life score of
incontinence patients (at baseline)
for patients assigned to placebo
different from the mean for
patients assigned to the treatment
(Interferential therapy) group?
Two-sample t-
test
Wilcoxon rank-sum
test
Compare two quantitative
repeated measurements
taken from the same
sample
Was there a significant change in
Quality of life score of incontinence
patients between baseline and the
Four-month follow-up
measurement in the treatment
group?
Paired t-test Wilcoxon signed-rank
test
Compare means between
three or more
independent groups
If the study have three groups (e.g.,
placebo, Interferential therapy,
kegal’s exercises),to know the mean
difference among the three groups?
Analysis of
variance
(ANOVA)
Kruskal-Wallis test
Association between two
quantitative variables
Is Quality of life score associated
with the patient’s age?
Pearson
coefficient of
correlation
Spearman’s rank
correlation
11. Advanced statistical tests
Relative risk
Relative risk reduction
Relative risk increase
Absolute risk reduction
Absolute risk increase
Number needed to treat
Number needed to harm
Risk-benefit analysis
12. Critical appraisal
Strength and weakness/issues of
RCT designs
Sample size calculation stated in the
study
Selection methods- randomization
in allocation -handles the issue of
confounding factors of the study
Selection bias –inclusion and
exclusion criteria
Homogenous groups are compared-
no distribution bias
Blinding – reduces the bias of
experimenter observation
Performance bias- groups treated
equally apart from the treatments.
Standardized procedures followed
in administering the treatment
protocols
Reliability and validity of outcome
measures used.
Duration of the study – adequate
follow up logically needed to find a
difference.
Intention to treat analysis
performed- whether subjects
received or completed the
treatment.
Drop outs management.
Worst case scenario method.
Last data carried forward method
Appropriate statistical tests
13. Critical appraisal tools
Appraisal of a meta-analysis or systematic review
Appraisal of a controlled study
Appraisal of a cohort or panel study
Appraisal of a case control study
Appraisal of a cross-sectional study (survey)
Appraisal of a qualitative study
Appraisal of a case study
Checklist for Analytical Cross Sectional Studies
Checklist for Quasi-Experimental Studies (non-
randomized experimental studies)
Checklist for Prevalence Studies
14. Critical appraisal tools
AGREE - http://www.agreecollaboration.org/
Alberta University Evidence Based Medicine Toolkit - http://www.ebm.med.ualberta.ca/
CASP - http://www.sph.nhs.uk/what-we-do/public-health-workforce/resources/critical-
appraisals-skills-programme
CATwalk - http://www.library.ualberta.ca/subject/healthsciences/catwalk/index.cfm
CEBMH - http://cebmh.warne.ox.ac.uk/cebmh/education_critical_appraisal.htm
Centre for Evidence Based Medicine - http://www.cebm.net
CLIST Resources for Critical Appraisal - http://www.londonlinks.nhs.uk/groups/clinical-
librarians-informationskills-trainers-group/trainers-toolkit/resources-for-critical-
appraisal
Scottish Intercollegiate Guidelines Network (SIGN): • http://www.sign.ac.uk/
AMSTAR • Shea, B.J., Grimshaw, J.M., Wells, G.A., Boers, M., Andersson, N., Hamel, C., &
Bouter, L.M. (2007). Development of AMSTAR: A measurement tool to assess the
methodological quality of systematic reviews. BMC Medical Research Methodology,
7(10). doi: 10.1186/1471-2288-7-10 • Article link: http://www.biomedcentral. com/1471-
2288/7/10
15. Clinical trial design issues
Appropriate statistical tests (e.g.;-pre-post design
analysis ‘T’ test Vs Anova)
Surrogate end points –Information bias
Hawthorne effect( handled by keeping a silent group)
Interim analyses( to avoid abundant sample coverage)
Drop outs management.
Worst case scenario method.
Last data carried forward method