1. Research Rotation Part II
B Kanna MD, MPH, FACP
Associate Program Director of Internal Medicine
Associate Director of Graduate Medical Education for Research

2. Critique of A randomized controlled clinical trial
Section I
Critique of
A randomized controlled clinical trial

3. Validity
the best available approximation to the truth of a given proposition, inference, or conclusion
Accurate & reproducible
Person 2
Person 1

4. Checking Validity Conclusion Validity: Internal Validity:
In this study, is there a relationship between the two variables?
Internal Validity:
Assuming that there is a relationship in this study, is the relationship a causal one?
Construct Validity:
Assuming that there is a causal relationship in this study, did we implement the program we intended to implement and did we measure the outcome we wanted to measure?
External Validity:
Assuming that there is a causal relationship in this study between the constructs of the cause and the effect, can we generalize this effect to other persons, places or times?

5. Random Error Sample 2 Sample 1 CHANCE Sample 3 CHANCE
Universe of subjects
Sample 1
CHANCE
Sample 3
CHANCE

6. Checking Internal validity
Universe of subjects
Sample 2
Confounding Rx
Outcome
Sample 3
Bias

7. Applying this to a study/observation
Day 1
Red pill
Sx improves
Day 2
Red pill
Sx improves
Day 3
Red pill
Sx improves

8. THREE THINGS TO REMEMBER!
CHANCE – RANDOM ERROR
BIAS – SYSTEMATIC ERROR
CONFOUNDING – ADDITIONAL FACTORS

9. How to apply these concepts during the critique of RCT
Learn the general steps in a study design
Choose a topic
Define a hypothesis
Select patients
Sample size
Randomization
Treatment
Follow-up
Collection of data
Analysis
Interpretation
Conclusion
Generalize findings

10. Steps Statistical methods Principles Reading Choose a topic
Likelihood/Plausibility
Chance
Ethical/feasibility/cost/QOL etc
Definition a hypothesis
Hypothesis test
Null/Alternate hypothesis/
Superiority/non-inferiority designs
Selection of patients
Sampling
Bias/Confounder
Single/Multi-center/restriction criteria
Sample size
Power
Errors/Effect size
Randomization
Allocation
Definition/type and methods of randomization/Other study designs
Treatment
Intervention/controls
Types of controls/Intervention /observational study designs
Follow-up
ITT/Blinding/Comparability
Bias/Confounder/chance (power)
ITT/Blinding methods/accounting for censored data/changes in allocation/Kaplan Meier analysis
Collection of data
Measurements/Aggregate data
Bias
Type of variables/
Distribution of data /measures of central tendency
Analysis
Statistical testing/p values/CI
Tests of significance/p value/CI/
Interpretation
Checking for Bias/Confounders
Stratified analysis/Regression
Conclusion
Internal validity
Chance/Bias/Confounder
Validity ( internal Validity)
Generalizability
External Validity
External validity

11. Section II
Selected topics

12. Study designs Observational Interventional Case control Cohort
Cross sectional
Interventional
Randomized
Non-randomized

13. Number needed to Rx and harm
NNT = 1/ Absolute risk reduction
Rx = 0.9%
Placebo = 3.2%
ARR = = 2.3%
NNT = 1/ = 45
NNH = 1 / Absolute risk of adverse events

14. Sensitivity, Specificity & Post test probability

15. How to design & start a study
Step 1: Select a topic
 Step 2: Choose a mentor and guide
 Step 3: Define topic
 Step 4: Search literature
 Step 5: Define problem
 Step 6: Collect back ground information
 Step 7: Study Methods
Study design; Sample selection; Sample size; Inclusion & Exclusion criteria
Follow-up; Measurements; Endpoints;
Statistical methods;, P value, CI, Test of significance
Confounding factors, Stratification, Regression
Results, Discussion , Limitations 
Step 8: IRB application along with aggregate data request
Step 9: IRB approval
Step 10: Start project (Revise protocol if needed and apply for IRB modifications)

16. How to write an IRB protocol
Simplified Guide for residents 
1. Title page
Title; Authors; Contact information of primary authors; Department 
2. Summary of the protocol
3. Introduction
Problem definition; Magnitude; Need for study
4. Background
Physiologic principles; Cause and effect; Hypothesis
5. Methods
Study design; Sample selection; Inclusion & Exclusion criteria; Sample size
Data collection (tool); Follow-up( If applicable) ; Measurements ( if applicable); Endpoints 
6. Statistical methods
Testing hypothesis; P value; CI; Tests of significance; Others analyses (e.g. subgroups)
 7. Confounding & Bias
Confounding ; Stratification; Regression; Methods used to Limit Bias
 8. Confidentiality and Privacy
 9. Consent/Waiver ( see Lincoln IRB consent procedures)
 10. Reporting of results
 11. Significance or Implications of the study
 12. References
 13. Tables, Figures and legends