|
THE COMMITTEE ON HUMAN RESEARCH
UCSF GUIDANCE ON RESEARCH TOPICS AND ISSUES
USE OF
INACTIVE INTERVENTIONS (e.g., PLACEBOS) IN PLACE OF POTENTIALLY EFFECTIVE
THERAPY (Posted October 2004)
 • Brief
Overview
• Definitions
• Purpose
of this Guideline
• Limitations
of this Guideline
• Ethical
Concern
• Inactive
Controls
• Washout
Periods
• Issues
Investigators Need to Address
• The
CHR Application Form
• The
Consent Form
• Sample
Wording for Consent Forms
Brief Overview
| • |
Definitions: Inactive interventions
include placebos, treatment interruptions (“washout periods”),
treatment withdrawals, and sham procedures. |
|
| • |
Purpose
of this Guideline: The
purpose of this guideline is to explain when the use of an inactive
intervention (e.g., placebo, sham procedure) poses a serious
ethical concern for the protection of human subjects, and to
help investigators anticipate and address those concerns in the
initial CHR Application and consent form. |
|
•
|
Limitations
of this Guideline:
The guideline focuses on specific and recurring issues the CHR
encounters when reviewing studies involving inactive control
arms used in place of potentially effective treatments. This
guideline is not intended as a comprehensive overview of controlled
clinical trials. The CHR has no policy for or against particular
study designs and raises concerns about study design when the
risks to subjects do not appear to be reasonable in relation
to anticipated benefits. |
Ethical Concern
| • |
The
use of an inactive control in place of potentially effective treatment is unacceptable
if harm will come to subjects as a result. When an inactive
intervention
is proposed to be used in place of an effective therapy, or
any time a research protocol calls for the absence, withdrawal
or
interruption of active treatment, the rationale for the study
design must be carefully explained, and the risks and the steps
taken to minimize those risks must be carefully articulated
in the CHR application. Serious delays in the CHR review process
result when these issues are inadequately explained, or unexplained.
|
| |
• |
Unfortunately, too often these
issues are not addressed in the multicenter protocols and sample
consent forms provided by study sponsors, even when the protocol
and sample consent form have been reviewed by the FDA. This
leaves local investigators in the position of having to provide
the additional information.
|
|
|
|
IMPORTANT NOTE: The FDA conducts
its scientific review of research with the understanding that
there will be local review focused on the protection of human
subjects. |
| |
• |
Note that the CHR may only approve
studies in which the risks to subjects appear reasonable in relation
to the expected benefits of the research, and in which the risks
appear to be minimized (45 Code of Federal Regulations 46.111).
This is why the CHR Application Form requires a clear description
of the study’s risks and benefits and the steps the investigators
have taken to minimize the risks. |
| |
Example of an acceptable use
of an inactive intervention: Investigators propose a short-term
placebo-controlled trial in patients with a chronic condition.
Subjects in the control arm will forego standard therapy for
a short and clinically insignificant period of time. Subjects
in the control arm may experience discomfort from their condition
but are very unlikely to be harmed. They will be closely monitored
to ensure they are withdrawn from the study and placed on active
therapy if their condition deteriorates.
Example of an unacceptable use
of an inactive intervention:
Investigators propose a long-term placebo-controlled trial
in patients with a chronic condition. Subjects in the control
arm would forego standard therapy for a relatively long and
clinically significant period of time. There is a high likelihood
that subjects in the control will be harmed if they do not
receive active therapy regardless of how closely investigators
monitor subjects. |
| • |
The
use of washout periods raises
many of the same issues discussed above, whether the washout
period involves tapering off of previous treatment, withholding
treatment, switching treatment, or providing a placebo. The CHR
Application Form and the Consent Form should address the issues
raised in the section below (“Issues Investigators Need
to Address”) taking into account the following differences: |
| |
• |
Duration. Although washout periods
may be of shorter duration than the risks of being assigned to
an inactive control group, |
| |
|
• |
For some conditions withholding treatment
for even a brief period of time may cause serious risks or discomforts, |
| |
|
• |
For other conditions, a brief period off
treatment may involve negligible risk. |
| |
|
• |
Some medications may be less effective following
a period of withdrawal and readministration. |
| |
• |
Blinded Washout Periods.
Some study designs require withholding treatment from all subjects
for a period of time at the beginning of the study. Subjects
need to be told that they must go without effective treatment
for the specified period. If it is necessary that subjects be
blinded to the initial washout period, the Procedures section
of the consent form should explain that at some period in the
study, all subjects will receive inactive “treatment” for
the specified duration, but that they will not be told when the
inactive “treatment” is happening. The risks of going
without active treatment for that period of time must be described
in the consent form. |
Top of Page
Issues Investigators
Need to Address
| • |
The
CHR Application Form:
When an inactive intervention is used in place of potentially
effective treatment (e.g., giving a placebo instead of a
standard therapy), the CHR Application Form should address
the following questions:
|
| |
• |
What are the risks of being
assigned to the inactive intervention (e.g., the control group
or washout period)? That is, |
| |
|
• |
How is the disease or condition likely
to progress over the course of the study or doing the period
without active treatment? |
| |
|
• |
How long will subjects continue to receive
the inactive intervention without receiving active treatment? |
|
|
• |
What are the consequences
of delaying treatment? |
| |
• |
What is standard care for
the patients who are eligible for the study? |
| |
|
• |
In answering this question, the CHR application
should include information on treatments which may be commonly
used at UCSF even if not accepted as “standard.” |
| |
• |
How effective are available
treatments, especially in relation to placebo? |
| |
|
• |
Are there significant side effects associated
with available treatments? |
|
|
•
|
Is there a significant placebo effect
associated with the available treatments? |
| |
• |
What criteria are used to
exclude patients who may be harmed if not given treatment immediately? |
| |
• |
How are subjects monitored during
the study to ensure their safety? |
| |
• |
What are the specific withdrawal
criteria, stopping rules and unblinding rules? |
| |
• |
What is the plan for the use
of rescue medication (if applicable)? |
| • |
The
Consent Form: If the committee
finds that the use of an inactive intervention in place of
potentially effective treatment is safe and ethical, the procedures
and risks related to the control group must be fully described
in the consent form. In particular, the consent form must address
the following issues: |
| |
• |
The Purpose and Background
section should include: |
| |
|
• |
A statement that the study involves an
inactive intervention, |
| |
|
• |
A statement about the chances of receiving
an inactive intervention, and |
| |
|
• |
If a placebo is used, an explanation of
what a placebo is. |
| |
|
Note: These three points can
be addressed in a single sentence (see example
below). |
| |
• |
The Procedures section should
include (as applicable): |
| |
|
• |
A description of how subjects are assigned
to treatment and control groups, |
| |
|
• |
A statement of the chances of being assigned
to each group, |
| |
|
• |
A statement of how long subjects will continue
to receive the inactive intervention, |
| |
|
• |
An explanation of blinding (see example
below), if applicable, and |
|
|
• |
A description of how subjects will be monitored
and followed during the study. |
| |
|
Note: Please review the sample
biomedical consent form format for suggested wording. |
| |
|
• |
The Risks section should
include a straightforward description of the risks associated
with the control group. |
| |
|
• |
Note that statements such as “you
will not receive active treatment for your condition” do
not describe the risk. |
| |
|
• |
Subjects should be provided an assessment
of the consequences of not receiving active treatment. |
| |
• |
The Benefits section should
state that subjects assigned to the inactive intervention will
not benefit from being in the study. |
Top of Page
Sample Wording for Consent Forms
The following are examples of statements like those
from approved consent forms. The statements should be edited as appropriate
for individual studies.
| • |
Purpose and Background Section
Half the subjects in this study will receive
a placebo (an inactive pill).
|
| • |
Procedures Section
You will be randomly assigned to one of two groups. This
means that you have a 50/50 chance (like flipping a coin)
of being in either group and that
neither the researchers nor you will make the choice of which group you
are in. The two groups are Group A (Drug XXX) or Group B (placebo, an
inactive substance). You will not be told whether you are given the study drug
or placebo unless you experience a serious side effect and
the study doctors need to find out this information to recommend
treatment. |
| • |
Risks Section
If you are assigned to Group B, you will be receiving a placebo in place of active
treatment for your condition for 4 weeks. Your symptoms may get worse or become
more frequent during this time. You will be closely monitored during the study
and if you appear to be getting worse, or even if you think you are getting
worse, you will be withdrawn from the study immediately and be given active
treatment. |
•
|
Benefits Section
If you are assigned to Group B (placebo), you will not benefit from being in
the study. |
|
