Study Design Tridandapani 2016

41
Anatomy and Physiology of Clinical Research Srini Tridandapani, PhD MD Department of Radiology and Imaging Sciences Emory University

Transcript of Study Design Tridandapani 2016

Page 1: Study Design Tridandapani 2016

Anatomy and Physiology of Clinical Research

Srini Tridandapani, PhD MDDepartment of Radiology and Imaging

SciencesEmory University

Page 2: Study Design Tridandapani 2016

Goal

• To understand the anatomy and physiology of clinical research, i.e., understand the principles of study design, analysis, reporting and bias

Page 3: Study Design Tridandapani 2016

Learning Objectives

At the end of this lecture you should be able• To enumerate the various elements of a

research study• To describe how the elements of the a

research study interact• To name the two major classes of error in

research studies

Page 4: Study Design Tridandapani 2016

Clinical Research

• Anatomy: What it’s made of…– Research question– Design– Subjects– Measurements– Sample size

calculation

• Physiology: How it works…– Make valid inferences– What happened in the

study sample– How does it generalize

to the world outside– How to errors,

random and systematic

Page 5: Study Design Tridandapani 2016

Anatomy of Research: What it’s Made of

Page 6: Study Design Tridandapani 2016

Anatomy: Outline of Study Protocol

Element PurposeResearch Questions What questions will the study address?

Background and Significance Why are these questions important?

Design

(Time frame, epidemiologic approach)

How is the study structured?

Subjects

(Selection criteria, sampling design)

Who are the subjects and how will they be selected?

Variables

(Predictor variables, confounding variables, outcome variables)

What measurements will be made?

Statistical issues

(Hypotheses, sample size, analytic approach)

How large is the study and how will it be analyzed?

Page 7: Study Design Tridandapani 2016

Anatomy: Outline of Study Protocol

Element PurposeResearch Questions What questions will the study

address?Background and Significance Why are these questions important?

Design

(Time frame, epidemiologic approach)

How is the study structured?

Subjects

(Selection criteria, sampling design)

Who are the subjects and how will they be selected?

Variables

(Predictor variables, confounding variables, outcome variables)

What measurements will be made?

Statistical issues

(Hypotheses, sample size, analytic approach)

How large is the study and how will it be analyzed?

Page 8: Study Design Tridandapani 2016

Research Question

• Objective of the study• Starts as a general question, but must be

narrowed down to a concrete, researchable issue

• Example:– Should people eat more fish?

Page 9: Study Design Tridandapani 2016

Research Question• Break this question into more specific

components:– How often do Americans eat fish?– Does eating fish the risk of CV disease?– Is there a risk of mercury toxicity from fish

intake in older adults?– Do fish oil supplements have the same effects

on CV disease as dietary fish?– Which fish oil supplements don’t make people

smell like fish?

Page 10: Study Design Tridandapani 2016

A well-formatted research question:The four PICO elements

P patients

I intervention (independent variable)

C comparison

O outcome (dependent variable)

• S subjects

• E exposure

• C comparison

• O outcome

Page 11: Study Design Tridandapani 2016

Using PICO to develop a question

Includes each of the following:

• Intended patients/subjects to study

• The independent variable of interest– Intervention, exposure, pt characteristic– The item of primary interest of the study

• Comparison group or procedure

• Outcome measure

Combines them together as a question

Page 12: Study Design Tridandapani 2016

Question example• Issues to address:

– Define CHD– Define study population sample source– Define fish intake

Final result:Among a sample of patients seen in the investigator’s

clinic who have a previous CHD diagnosis and respond to a mailed questionnaire, what proportion report taking fish oil supplements?

Page 13: Study Design Tridandapani 2016

PICO in practice

Essentially all proposed research questions should include each of the

PICO elements

Page 14: Study Design Tridandapani 2016

Research question comments

• It has been proposed that if your entire question can be stated in one breath, it is too short

• That is extreme. It need not include your entire inclusion and exclusion criteria

• The research question is analogous to the pt chief complaint. It is not the entire history and exam.

Page 15: Study Design Tridandapani 2016

Refine the Research Question: Make it a FINER question

• Feasible = can it be done with the resources that are available?

• Interesting = do you have passion for it?• Novel = is it a new research idea?• Ethical = does it comply with research ethics

and codes of conduct?• Relevant = does it pass the “so what” test?

Page 16: Study Design Tridandapani 2016

Applying the FINER criteria

• Carefully review your proposed research question– Research in progress meetings can be helpful for this

• Consider how it meets each of the FINER criteria……or not

Then decide:• Do you still want to proceed with your proposed

project? Is it feasible, etc.?

Page 17: Study Design Tridandapani 2016

The FINER review

This is where most good research questions die an early death

(or should have died)

Page 18: Study Design Tridandapani 2016

Anatomy: Outline of Study Protocol

Element PurposeResearch Questions What questions will the study address?

Background and Significance Why are these questions important?Design

(Time frame, epidemiologic approach)

How is the study structured?

Subjects

(Selection criteria, sampling design)

Who are the subjects and how will they be selected?

Variables

(Predictor variables, confounding variables, outcome variables)

What measurements will be made?

Statistical issues

(Hypotheses, sample size, analytic approach)

How large is the study and how will it be analyzed?

Page 19: Study Design Tridandapani 2016

Background and Significance

• Provide context and Rationale• References, including your own work (toot

your horn!...but not too loudly)• Be succinct! Need to pare this down with

new NIH format

Page 20: Study Design Tridandapani 2016

Anatomy: Outline of Study Protocol

Element PurposeResearch Questions What questions will the study address?

Background and Significance Why are these questions important?

Design (Time frame, epidemiologic approach)

How is the study structured?

Subjects

(Selection criteria, sampling design)

Who are the subjects and how will they be selected?

Variables(Predictor variables, confounding variables, outcome variables)

What measurements will be made?

Statistical issues

(Hypotheses, sample size, analytic approach)

How large is the study and how will it be analyzed?

Page 21: Study Design Tridandapani 2016

Design of Study

Observational Study• Cohort study

– Retrospective– Prospective

• Cross-sectional study• Case-control study

Clinical Trial• Randomized blinded trial

– Can also have nonrandomized or un-blinded designs

Page 22: Study Design Tridandapani 2016

Anatomy: Observational Designs

Study Design Key Feature ExampleCohort Study A group followed over

timeMeasure fish intake at baseline and periodically examine subjects at fu visits to see if eating more fish CHD events

Cross sectional study A group examined at one point in time

Interview subjects about current and past history of fish intake and correlate with h/o CHD and current coronary calcium score

Case-control study Two groups selected based on the presence or absence of an outcome

Examine a group of patient w/ CHD (“cases”) and compare them with group who did not have CHD (“controls”), asking about past fish intake

Page 23: Study Design Tridandapani 2016

Anatomy: Clinical Trial Design

Study Design Key Feature ExampleRandomized Blinded Trial

Two groups created by a random process, and a blinded intervention

Randomly assign subjects to receive fish oil supplements or placebo, then follow both treatment groups for several years to observe the incidence of CHD

Page 24: Study Design Tridandapani 2016

Anatomy: Outline of Study Protocol

Element PurposeResearch Questions What questions will the study address?

Background and Significance Why are these questions important?

Design

(Time frame, epidemiologic approach)

How is the study structured?

Subjects

(Selection criteria, sampling design)

Who are the subjects and how will they be selected?

Variables

(Predictor variables, confounding variables, outcome variables)

What measurements will be made?

Statistical issues

(Hypotheses, sample size, analytic approach)

How large is the study and how will it be analyzed?

Page 25: Study Design Tridandapani 2016

Anatomy: Study Subjects

• Inclusion and Exclusion Criteria defining target population

• How best to recruit

Page 26: Study Design Tridandapani 2016

Anatomy: Study Subjects

• In the Fish Study example, how is CHD defined?– Could be subjects seen in the clinic with

diagnosis codes for MI, angioplasty, or CABG in their EMR

• Recruitment Trade-off– Using random sample of CHD patients in the

entire country would enhance generalizability but more expensive

Page 27: Study Design Tridandapani 2016

Anatomy: Outline of Study Protocol

Element PurposeResearch Questions What questions will the study address?

Background and Significance Why are these questions important?

Design

(Time frame, epidemiologic approach)

How is the study structured?

Subjects

(Selection criteria, sampling design)

Who are the subjects and how will they be selected?

Variables

(Predictor variables, confounding variables, outcome variables)

What measurements will be made?

Statistical issues

(Hypotheses, sample size, analytic approach)

How large is the study and how will it be analyzed?

Page 28: Study Design Tridandapani 2016

Variables• Predictor variables

– Age, race, sex, smoking history, fish and fish oil supplement intake

• Outcome variables– Heart attacks, strokes, quality of life,

unpleasant odor

• Confounder variables– Other predictors of outcome such as intake of

red meat or income level that could be related to dietary fish and confuse interpretation

Page 29: Study Design Tridandapani 2016

Anatomy: Outline of Study Protocol

Element PurposeResearch Questions What questions will the study address?

Background and Significance Why are these questions important?

Design

(Time frame, epidemiologic approach)

How is the study structured?

Subjects

(Selection criteria, sampling design)

Who are the subjects and how will they be selected?

Variables

(Predictor variables, confounding variables, outcome variables)

What measurements will be made?

Statistical issues

(Hypotheses, sample size, analytic approach)

How large is the study and how will it be analyzed?

Page 30: Study Design Tridandapani 2016

Anatomy: Statistical Issues

• Consult a Statistician• Develop a sound statistical hypothesis

(different from a scientific hypothesis)• Reframe research question to test for

statistical significance of findings• Hypothesis allows calculation of sample

size

Page 31: Study Design Tridandapani 2016

Anatomy: Statistical Issues

• Reframe scientific into a statistical hypothesis

• Hypothesis: 50- to 69-year-old women with CHD who take fish oil supplements will have a lower risk of myocardial infarction than those who do not

Page 32: Study Design Tridandapani 2016

Physiology of Research: How it Works

Page 33: Study Design Tridandapani 2016

Inferences

Findings in the Study

Truth in the Universe

INFER

Page 34: Study Design Tridandapani 2016

Inferences

Truth in the Study

Study Plan

Truth in the Universe

Research Question

Infer Findings in the Study

ActualStudy

Infer

Design Implement

EXTERNALVALIDITY

INTERNALVALIDITY

DESIGNING AND IMPLEMENTING

DRAWING CONCLUSIONS

Page 35: Study Design Tridandapani 2016

Designing the Study

• What is the prevalence of regular use of fish oil supplements among people with CHD?

• Among a sample of patients seen in the investigator’s clinic who have a previous CHD diagnosis and respond to a mailed questionnaire, what proportion report taking fish oil supplements?

Page 36: Study Design Tridandapani 2016

Design ErrorsTruth in the Study

Study Plan

Truth in the Universe

Research Question

EXTERNALVALIDITY

Target Population:People with CHD

Intended Sample:All patients with a

history of CHD seen in clinic last year

Phenomenon of Interest:

Proportion who take fish oil supplements

Intended variables:Self-reported use of fish oil supplements

Infer

Design

Errors

Page 37: Study Design Tridandapani 2016

Implementation ErrorsFindings in the Study

Actual study

Truth in the Study

Study plan

Infer

Implement

INTERNALVALIDITY

Intended Sample:All 215 patients with a h/o CHD seen in clinic

in last year

Actual Subjects:104 patients w/ CHD diagnosis in chart

last year completing questionnaire

Errors

Intended Variables:Self-reported use of fish oil supplements

Actual Measurement:Responses to

questions on fish oil supplements

Page 38: Study Design Tridandapani 2016

Types of Errors

• Systematic Errors (Bias)• Random Errors (Chance) – this can be

reduced by increasing the sample size

Page 39: Study Design Tridandapani 2016

Summary of the Physiology of Clinical Research: How it Works

Truth in the Universe

Research question

Infer

Design

INTERNALVALIDITY

Target population

Phenomenon of Interest

Truth in the Study

StudyPlan

Intended Sample

Intended Variables

Findings in the Study

ActualPlan

Actual Subjects

Actual Measurements

Infer

Design

EXTERNALVALIDITY

Random and

Systematic Errors

Random and

Systematic Errors

Page 40: Study Design Tridandapani 2016

Summary

• Anatomy of research: what it’s made of

• Physiology of research: how it works

• Research ? study plan actual study

Page 41: Study Design Tridandapani 2016

Thank You