Transcript Document 204020

Assessing Study Design
for Institutional Review
Committees/Boards
Monica Brown, PhD
Alternate Member
The Sutter Health Central Area
Institutional Review Committee
September 21, 2010
Presentation Objectives
Identify the 2 basic study designs.
What 2 elements drive clinical study
design?
Name the #1 reason why C.R.A.P.
detection in clinical study design is so
important?
Presentation Outline
Study Design Basics
Selecting the Appropriate Design to Meet the
Study’s Goal
Experimental Studies: Clinical Study Design
Issues in Poor Clinical Study Design
Conclusion
Introduction
Why must
IRC/IRB
members
understand
study
design?
Ethical codes and federal regulations
require that the IRB evaluate
study design and scientific quality
 Section 18 of the Declaration of Helsinki (2000):
 Medical research involving human subjects should only be
conducted if the importance of the objective outweighs the
inherent risks and burdens to subjects.
 Federal Regulation 45 CFR 46.111 (a):
 Criteria for IRB approval of research: Risks to subjects are
minimized by using procedures which are consistent with sound
research design, and which not, unnecessarily, expose subjects to
risk.
Study Design Basics
1. DEFINITION
2. PURPOSE
3. T YPES
A study design is a specific plan or
protocol for conducting a study,
which allows the investigator to
translate the conceptual
hypothesis into an operational one.
S TUDY
D ESIGN
Basic Study Design Types
Qualitative
• An inductive research
process involving collection
& analysis of non-numerical
data to search for patterns,
themes & holistic features
Quantitative
• An inductive research
process involving collection
& analysis of numerical data
to search for patterns,
themes & holistic features
Categories of Quantitative
Study Designs
Studies where there is no
manipulation of the study
factor or subjects
Observational
Experimental
Studies that entail manipulation
of the study factor (exposure) and
randomization of subjects to
exposure (treatment or
intervention)
Experimental studies are superior to
Observational studies
1. Provide stronger evidence of the effect
(outcome) compared to observational
designs
2. Yield more valid results, as variation is
minimized and bias controlled
3. Determine whether experimental
treatments are safe and effective under
“controlled environments” (as opposed to
“natural settings” in observational
designs), especially when the margin of
expected benefit is narrow (10 - 30%)
The primary purpose of a
clinical trial is to compare
the clinical efficacy, as
measured by clinical
endpoints,
of those receiving
experimental treatment, over
those receiving standard
treatment or placebo
C LINICAL
E FFICACY
Clinical trial design is driven by
endpoints and the sponsor’s
“labeling” goals for the
therapeutic agent
C LINICAL
T RIAL
D ESIGN
Clinical Endpoints
“Ultimate” clinical endpoint – survival
 Proximal Clinical Endpoints
 symptom benefit
 progression-free survival
 biomarkers
 response rates
Clinical endpoints define the
complexity of a trial - to recruit and
enroll participants; perform the
study; gather and analyze the data.
C LINICAL
T RIAL
D ESIGN
C OMPLEXITY
Basic Elements of a
Clinical Study Design
Patient Population
 Sample size
 Patient selection (diversity)
Clinical endpoint
 Survival
 Proxy
Desirable Treatment Differences
 How large is acceptable?
Control Treatment
 Standard treatment vs. Placebo
Design Options
 Controls: Use of a comparison or control group.
 Randomization: the random (by chance)
assignment of participants into exposure groups.
 Blinding:
Sponsor's
Clinician /
Analysis Team Investigator Patient
No
No
No
No
No
Yes
No
Yes
Yes
Yes
Yes
Yes
open/unblinded
single blinded
double blinded
triple blinded
Basic Clinical Design Strategy
CLINICAL ENDPOINT
Recruit / Enroll / Consent
R
Patient
Population
A
N
D
O
M
I
Z
A
T
I
O
N
outcome
Experimental
Treatment
Control
no outcome
outcome
(Std Tx or Placebo)
no outcome
baseline
future
TIME
Study begins here
Classic Drug Development Process
Phase I: Dose-finding & Safety
Phase II: Initial Efficacy Data
Phase III: Pivotal FDA Approval
Phase IV: Post-marketing Surveillance
(a.k.a., registries)
Comparison of Drug Development Phases
PHASE I
 Determine the metabolic
& pharmacological
actions
 Maximally tolerated dose
PHASE II
 Evaluate effectiveness
 Determine short-term side
effects
 Identify common risks
FACTORS
TO BE
IDENTIFIED:






DATA FOCUS:
 Vital signs
 Plasma /serum levels
 Adverse events










DESIGN
FEATURES:
 Single, ascending dose
tiers
 Unblinded
 Uncontrolled
DURATION:
Up to 1 month
POPULATION:
Healthy volunteers or
individuals with the target
disease (i.e., cancer or HIV)
SAMPLE SIZE:
EXAMPLE:
20 to 80
Study of a single dose of
Drug X in normal subjects
OBJECTIVES:
Bioavailability
Bioequivalence
Dose proportionality
Metabolism
Pharmacodynamics
Pharmacokinetics


Bioavailability
Drug-disease interactions
Drug-drug interactions
Efficacy at various doses
Pharmacodynamics
Pharmacokinetics
Patient safety
Dose response & tolerance
Adverse events
Efficacy





 Placebo controlled
comparisons
 Active controlled
comparisons
 Well-defined entry criteria
Several months




PHASE III
Obtain additional
information about
effectiveness on clinical
outcomes
Evaluate overall risk-benefit
ratio
Drug-disease interactions
Drug-drug interactions
Dosage intervals
Risk-benefit information
Efficacy & safety for
subgroups
 Laboratory data
 Efficacy
 Adverse events
Randomized
Controlled
2-3 treatment arms
Broader eligibility criteria
PHASE IV
 Monitor ongoing safety in
large populations
 Identify additional uses of
agent that might be
approved by the FDA
 Epidemiological data
 Efficacy & safety within
large, diverse populations
 Pharmacoeconomics






Efficacy
Pharmacoeconomics
Epidemiology
Adverse events
Uncontrolled
Observational
Several years
Ongoing
(following FDA approval)
Individuals with target disease
Individuals with target
disease , demographically
diverse sample
Individuals with target
disease, as well as new age
groups, genders, etc.
200 to 300
Double-blind study evaluating
safety & efficacy of Drug X vs.
placebo in patients with
hypertension
Hundreds to thousands
Drug X vs. standard treatment
for hypertension
Thousands
Economic benefit of newlyapproved Drug X vs. standard
treatment for hypertension
Source: Research Coordinator Orientation, University of Pittsburgh, 2002
Issues in Clinical Study
Design
Attributes of Well Designed Studies
The Protocol has …
a clear & concise research
question/statement of purpose/goal
/hypothesis
measurable aims or objectives – that directly
build to answer the research question
appropriate (validated) clinical endpoints
the participant’s steps, clinical measures,
data analysis, etc., laid out in an easy to
follow manner
Attributes of Well Designed Studies
The Investigator has …
 Local control




Over protocol implementation
Consenting participants
How treatments are administered
Manner in which clinical measures are taken, assessed &
recorded
 Data control – How data will be





Gathered
Who will have access
Where & how data will be stored
Analyses to be conducted
When destroyed
Well Designed Studies
Can be replicated
Bias is minimized
Valid – design & its elements can meet
endpoints
Feasible
 Low patient burden
 Time commitment
 Resources
Poorly Designed Studies
Affect recruitment & enrollment
Too difficult for participant to
understand, which results in poor
treatment/intervention compliance
Do not get published
Reflects poorly on the Investigator,
Institution & IRC/IRB
Issues in Clinical Study Design
Patient Selection
 Representative of the patient population
 Patient must be able to tolerate being in a trial
Sample Size
 Influences size of benefit anticipated
 Influences the amount of certainty we wish to have with which
to capture that treatment benefit (power)
 Can be dependant on resources rather than science
 Negative trial results could be due to small sample size or
insufficient power
Issues in Clinical Study Design
Blinding
 Randomization can minimize the influence of bias by balancing
groups for various characteristics
 Bias can still occur if personnel and/or patients know the
identity of the treatment – knowledge of treatment can lead to
preconceptions & subjective judgment in reporting, evaluating
and even data/statistical analysis
Study Design C.R.A.P.*
* Streiner/Norm/Monroe Blum,
“PDQ Epidemiology” 1989
C ONVOLUTED
R EASONING
OR A NTI I NTELLECTUAL
P OMPOSITY
Detecting C.R.A.P. is Important!
1. Patient Safety
2. Inability to meet clinical endpoints
3. Waste of resources
4. Reflects poorly on investigator &
institution
How I find C.R.A.P.
 Can this methodology be used to answer the
research question?
 Is there adequate time during the study period to
reach the clinical endpoints?
 Can study aims/objectives (which should be
measurable) be measured using this design?
 Is more data being requested/gathered than are
needed?
 What are the actual steps of the participants?
 When/where does he/she start the process?
 How long does each step take? Where will he/she wait for the next
step?
 Would I want the same done to me or a loved one?
Examples of Study Design C.R.A.P.
Participants/Patients
Data
F OR
R ESEARCHERS
Recommendations
F OR
IRC/IRB
M EMBERS
Recommendations for Researchers:
In Your Protocol
Name the type of study
 i.e., “data-only, retrospective, case-control study”
State the research
question/purpose/goal/hypothesis CONCISELY
Aims & objectives should be QUANTIFIABLE
Clinical endpoints should be DEFINED (not just
named) – give rationale for its/their use
S HOW THAT THE CLINICAL TRIAL EXPERIENCE OF THE
PATIENT HAS BEEN CONSIDERED ( NOT JUST THAT IT IS
SAFE )
Recommendations for Researchers: …
List study design limitations & respond with
your solution – don’t wait for the IRC/IRB to
point it out
Not sure what will fly with your IRC/IRB? Ask
for an informal review by their member with
the greatest study design expertise
Recommendations for IRC/IRB Members
Assessing Study Design for Patient Safety
Assign protocols to reviewers who have the
appropriate expertise to evaluate study design
Encourage informal discussion of design
issues between members & investigators
Unless you’ve been trained in study design,
don’t second guess an investigator – ask the
investigator to explain why a study has been
designed a certain way
 i.e., multiple control groups, run-in phase
In Conclusion
W HY
Well-designed protocols are
important for conducting research
studies safely & in a cost-effective
manner.
THE
FOCUS
ON
STUDY
DESIGN ?
Resources
 www.ICH.org Efficacy Guidelines
 E8: General Considerations for Clinical Trials
 E9: Statistical Principals for Clinical Trials
 E10: Choice of Control Group and Related Issues in Clinical
Trials
 http://clinicaltrials.gov/ct2/info/understand
Understanding Clinical Trials
 http://www.hhs.gov/ohrp/irb/irb_chapter 4.htm
Institutional Review Board Guidebook, CHAPTER IV,
CONSIDERATIONS OF RESEARCH DESIGN
 http://humansubjects.stanford.edu/research/docume
nts/eval_study_designGUI03017.pdf Evaluating Sound
Study Design
Thank you