2. Key Definitions
2
Research
Research is a systematized effort
to gain new knowledge
or to test previously established facts
Methods to convert
a research problem into
an operational one to get conclusions
Systematic analysis of data
to draw accurate conclusions
Research Methods
Statistical Analysis
8. Decision Tree
Observational Study
No Yes
Experimental Study
Comparison Group
Descriptive Study Analytic Study
Cohort Study
Cross-Sectional Study
Case-Control Study
Direction of Study
E O
No Yes
E O
E = O
Randomization
Blinding
Control
Intervention Done
9. Epidemiological Study Design
Observational Studies
Descriptive Studies
Analytic
Cross-Sectional
Case-Control
Cohort
Experimental / Interventional studies
As per Control: RCT/NRCT
As per Blinding: Single /Double Blind/Triple
Blind/Open Label
As per Design: Simple/Cross-over
11. Other Name: Case-Series/Population
Unit of Study: Case/Individuals
Study Question: What is happening
Direction Of Inquiry: at present
Study Design:
☻☻☻☻☻☻ desired information
☻☻☻☻☻☻ about cases/individuals is
collected
Descriptive type of Observational Studies
12. Describe the research Output
Descriptive Study
Where?
What type of places ?
What are the factors responsible?
Outcome Analysis & Inferences
Qualitative data: Percentage and Proportions
Quantitative data: Mean and SD
Ordinal data: Median and Percentiles
13. Data collected at a single point in time
Describes associations
Prevalence
A “Snapshot”
Cross-sectional Studies
14. Other Name: Prevalence Study
Unit of Study: Individual
Study Question: What is happening
Direction of Inquiry: at present
Study Design
Population
Diseased
Not
Exposed
to Factor
Exposed
to Factor
Non-
Disease
Exposed to
Factor
Not
Exposed to
Factor
Cross-sectional Studies
15. Significance Tests
Chi-square Test in Qualitative data
Unpaired ‘t’ Test in Quantitative data
Mann Whitny Test in Ordinal data
To find out association
Significance of difference of exposure
in with and without diseased
Analysis
&
Inference
Cross-sectional Studies
16. Start with people who have disease(Cases)
Match them with controls that do not have
disease (Match Confounding)
Look back and assess exposures
Case-Control Studies
17. Other Name Retrospective Study
Unit of Study Cases/Control
Study Question What has happened
Direction of Inquiry= F O
Study Design
Cases
Not
Exposed
Exposed
Control
Exposed
Not
Exposed
Case-Control Studies
18. Significance Tests
Chi-square Test in Qualitative data
Unpaired ‘t’ Test in Quantitative data
Mann Whitny Test in Ordinal data
To find out association
1. Significance of difference of exposure
in with and without diseased
2. Odd’s Ratio (Risk in times)
Analysis
&
Inference
Case-control Studies
19. Begin with disease-free individuals
Classify patients as exposed/unexposed
Record outcomes in both groups
Compare outcomes using relative risk
Cohort Studies
20. Other Name Prospective / Follow-up Study/Incidence Study
Unit of Study Individual
Study Question What will happen
Direction of Inquiry F O
Study Design
Cohort
Exposed to
Factor
Not Non
Diseased
Not
Exposed to
Factor
Diseased
Diseased
Non-Diseased
Cohort Studies
21. Significance Tests
Chi-square Test in Qualitative data
Unpaired ‘t’ Test in Quantitative data
Mann Whitny Test in Ordinal data
To find out association
1. Significance of difference of exposure
in with and without diseased
2. Odd’s Ratio (Risk in times)
3. Relative Risk
4. Attributed RiskAnalysis
&
Inference
Cohort Studies
22. Clinical trials provide the “gold standard” of
determining the relationship between factor
and the event
Experimental Studies
A clinical trial is a comparative, prospective
experiment conducted in human subjects
24. Controls
Controls give better comparison for
better estimates of effect
Concurrent Parallel Design (RCT)
Sequential RCT Design
RCT with External Control
25. Randomization
Randomization: Random distribution of subjects in
experimental groups is done by giving equal chance to
every subject to be included in either of experimental
group.
Simple random: Subjects were distributed in various
experimental group either by lottery method or chit box
method or by computer generated random numbers.
Systemic random: Subjects were distributed in various
experimental group by systematic like alternate
allocation, every 3rd etc.
First group may be chosen by either by lottery method
or chit box method then with a system.
26. Blinding
Good practice: Factors that can affect the evaluation of
outcome should not be permitted to influence the
evaluation process
Single-blind: Patient does not know the type of Rx given to
him/her
Double-blind design: Neither patient nor doctor knows
which Rx is given to which patient
Triple-blind design: Neither patient nor doctor nor outcome
evaluator knows which Rx is given to which patient
Open label design: Everyone i.e. patient, doctor, outcome
evaluator knows which Rx is given to which patient
27. Steps of Experimental Study
Drawing up a Protocol
Reference Population
Sample Population
Randomization
Experimental Group Control Group
Manipulation/Intervention
Follow - up
Assessment of Outcome
Inclusion Criteria
Exclusion Criteria Excluded
Inferences
28. Experimental Study
Other Name Intervention Study
Objective To know the effect of intervention
Unit of Study Individual meeting entry criteria
Study Question What will happen after
intervention in groups
Direction of Inquiry I E
Study Design:
1. Before and After Comparison
2. Comparison with Placebo
3. Comparison Of two medicine/procedure/tests
4. Comparison Of > two medicine/procedure/tests
29. Intervention Study - Design 1
(Comparison of Effect before and after )
Cases
Meeting
Entry criteria
Pre Intervention
Assessment
Outcomes
Follow ups
Intervention
Post Intervention
Outcomes
Outcome Analysis
Inferences
30. Intervention Study - Design 2
(Comparison of Effect with Placebo )
Cases
Meeting
Entry criteria
Group -2Group - 1
Outcome
Intervention PlaceboIntervention
Outcome
Randomization
Follow up
Outcome Analysis
Inferences
Pre-I
A-Outcomes
Pre-I
A-Outcomes
31. Intervention Study - Design 3
(Comparison of Effect of Two Interventions)
Cases
Meeting
Entry criteria
Group -2Group - 1
Outcome
Intervention 1 Intervention 2Intervention
Outcome
Randomization
Follow up
Outcome Analysis
Inferences
Pre-I
A-Outcomes
Pre-I
A-Outcomes
32. Intervention Study – Design 4
(Comparison of Effect of >Two Interventions)
Cases
Meeting
Entry criteria
Group -2Group - 1
Outcome
Intervention 1 Intervention 2
Outcome
Randomization
Follow up
Outcome Analysis
Inferences
Group -3
Outcome
Intervention 3
Follow up
33. Cross over Intervention Study
Cases
Meeting
Entry criteria
Group -2Group - 1
Intervention 1 Intervention 2
Wash out period
Randomization
Cross Over
Outcome Analysis
Outcome Outcome
Group -1Group - 2
Intervention 1 Intervention 2
Outcome Outcome
Pre-I
A-Outcomes
Pre-I
A-Outcomes
34. Dr. Kusum Gaur
Type of Research Problems :
to Compare 2 Independent Groups
Cases
Control
Change in
Outcomes
Change in
Outcomes
Analysis & Inference:
Chi-square Test in Qualitative data
Unpaired ‘t’ Test in Quantitative data
Mann Whitny Test in Ordinal data
Intervention
35. Dr. Kusum Gaur
Type of Research Problems :
to Compare >2 Independent Groups
study
Gp ‘A’
Control
Change in
Outcomes
Change in
Outcomes
Analysis & Inference:
Chi-square Test in Qualitative data
ANOVA Test in Quantitative data
Kruskal Wallis Test in Ordinal data
Study
Gp ‘B’
Change in
Outcomes
Intervention
36. Dr. Kusum Gaur
Type of Research Problems :
to Compare 2 Dependent Observations
Study
Gp
Pre-
intervention
Outcomes
Post-
intervention
Outcomes
Analysis & Inference:
Mec Nemar Test in Qualitative data
Paired ‘t’ Test in Quantitative data
Wilcoxon Test in Ordinal data
Intervention
37. Dr. Kusum Gaur
Type of Research Problems :
to Compare >2 Dependent Observations
Study
Gp
Pre-
intervention
Outcomes
1st Follow-up
Outcomes
Analysis & Inference:
Cohrane Test in Qualitative data
Repeated ANOVA Test in Quantitative
Friedmen Test in Ordinal data
2nd Follow-
up Outcomes
Intervention
38. STUDY QUESTIONS AND APPROPRIATE DESIGNS
Type of Question Appropriate Study Design
Burden of illness Field Surveys
- Prevalence Cross Sectional Survey
- Incidence Longitudinal survey
Causation, Risk & Prognosis Case Control Study, Cohort
study, Experimental study
Treatment Efficacy Experimental study
Diagnostic Test Evaluation Experimental study
Cost Effectiveness Experimental study
Programme Evaluation Quasi Experimental study
39. Hierarchy of Epidemiological Study Design
Generate Hypothesis
Establish Causality
Case Report
Case Series
Cross-Sectional
Case Control
Cohort
Experimental Study
Find Association
Test Hypothesis
Reviews Systemic Reviews Meta-analysis
40. Web sites related to Statistics
• http://stattrek.com
• http://vassarstat.net
• http://www.scribd.com
• http://www.statistixl.com
• http://statistics calculators.com
• http://stat.ubc.ca/~rollin/stats/ssize/
• ………………………………………………………
……
41. Computer Softwares in Statistics
• Microsoft Excel
• SPSS
• Epi info
• Epi tab
• Mini tab
• Graph Pad
• Primer
• Medcal
• ……………..
42. Further Reference from Presenter’s Own
A book entitled '' Community Medicine: Practical
Guide '' authored by me.
My Slide Share presentation on
Biostatics and research methodology,
Research methodology,
Sampling techniques,
Data analysis
My (M) 9414036057
My (e-mail) drkusumgaur@gmail.com
Launching Soon
For Sample size & Analysis
P Value
(A mobile App )