Thursday, September 12, 2002
Session 2: Regulation 6: Institutional Review Boards
Ezekiel J. Emanuel, M.D., Ph.D.
Chief, Center for Clinical Bioethics, National Institutes
CHAIRMAN KASS: In this session the Council
returns to the general topic of oversight and regulation of
development and uses of biotechnologies, a subject of interest
to us in general and also, I remind you, in relation to our
work on cloning and stem cells, in which many of the people
on both sides of the aisle on the cloning debate were interested
in exploring the prospects for and the principles of regulation.
In previous meetings we have been looking both at what other
countries are doing and also learning about the current regulatory
activities in the United States. So far, we've had an overview
of what is happening around the world with Lori Knowles and
a report on the Canadian practices in reproductive technologies
with Patricia Baird. We've had three sessions domestically
on the patenting of living organisms.
Tomorrow morning we will hear from Professor Simitis on Germany's
practices and policies, and today we will be looking at the
Institutional Review Boards, those local bodies that are now
charged with oversight and approval of all federally-funded
research involving human subjects.
We are very fortunate to have as our presenter this morning
Dr. Ezekiel Emanuel, physician, oncologist, Doctor of Political
Philosophy, Chair of the Department of Clinical Bioethics
at the NIH. His paper he wrote with co-workers, Ann Wood and
Christine Grady, entitled, "The Crisis in Human Participants
Research: Identifying the Problems and Proposing Solutions,"
was distributed to Council members in advance.
I remind everyone present that the opinions expressed in
this paper and in this presentation are those of the authors.
They do not reflect any official position or policy of NIH,
the Public Health Service, or the Department of Health and
It's a special privilege for me to welcome Zeke Emanuel here
who I first met when — I shudder to think when — he was
still a student, a senior at Amherst, and I was a much younger
Zeke, welcome and thanks for being with us.
DR. EMANUEL: My speech coach always tells
me to stand and never sit. So I hope the Council will indulge
me in that.
It's very wonderful to be invited here, and many of the Council
members have been my mentors, many are my current colleagues,
and many are people whose writings and work and research I
have admired over the years.
I think it's fair to say that everyone seems dissatisfied
with the current system of protecting human research participants.
Many of our researchers find the system onerous and more of
a hurdle to get over than something that is value-added.
IRB members who serve on the oversight bodies — IRB stands
for Institutional Review Boards — find the regulations quite
opaque, feel in a bind between federal oversight and what
they're supposed to do for their institution. The pharmaceutical
and biotechnology industry finds the process very time-consuming,
very inefficient, and very resistant to innovative and novel
approaches in research.
Regulators feel frustrated. They get criticized by the federal
government and the public for not doing enough. They get criticized
by institutions for being too intrusive.
And the public is afraid, uncertain. They believe greatly
in biomedical research, but recent polls have certainly shown
them to be much more concerned than they ever were. I think
the cover of Time magazine over the summer that showed a human
subject in a guinea pig cage attests to this concern.
I don't think it is misplaced. Since 1999, the federal government
oversight body, OHRP — that is the Office for Human Research
Protection — has stopped or suspended research at 13 major
research institutions. This includes two of the top ten medical
centers in this country, and a third of the top ten medical
centers in this country had a major scandal involving the
death of a patient, but was not actually closed.
Now that dissatisfaction has led to tremendous activity to
respond to the problems. You can see all those acronyms up
there. You know you're in Washington; it's alphabet soup time.
Recently, there was the creation of this Office for Human
Research Protections during the end days of the Clinton Administration
that reports directly to the Secretary of the Department of
Health and Human Services.
At the FDA there was the creation of the Office for Good
Clinical Practice to oversee all their ethical issues. The
NIH, my employer, has issued requirements on training and
education of investigators and one-time grants to improve
the function of IRBs. There are new initiatives for accrediting
IRBs, which I'll talk about in a moment. And there's introduction
at least of two bills by Senator Kennedy and Representatives
DeGette and Greenwood in Congress for reforming the system.
This attests to a huge amount of activity.
I think there's a problem, though, and a lot of this activity
really tends to address specific problems without stepping
back and assessing the whole system. In the next few minutes
I really want to sort of systematically delineate what I take
to be the problems of the system, evaluate how the various
proposed reforms address or don't address those problems,
and then I want to float for you a more comprehensive solution.
It is at this point which I want to just reiterate Dr. Kass'
disclaimer for me, which being a federal government employee
I always do at every talk, which is that these are my opinions
and the opinions of my colleagues, Ann Wood and Christine
Grady, and our Department, and these are not the official
opinions of anyone else.
I can tell you, having circulated the manuscript you saw
inside the NIH, many people certainly disagree with, but they
also thought it was important that these issues be aired and
discussed. I greatly appreciate their commitment to free inquiry.
For those of you who aren't sort of up, as it were, on human
subjects research and the regulations, I thought a brief historical
digression would be helpful. From basically the mid-sixties
to 1972, there were a series of research scandals in this
country that really began in 1966, where Henry Beecher, a
Professor of Anesthesia Research at Harvard Medical School
and the Mass. General Hospital, published a very important
paper in The New England Journal of Medicine that delineated
22 violations of research ethics at major research institutions.
He said he culled these 22 from a further 50 cases which
he couldn't produce because of reasons of space, and that
was from a further 100 cases. He made very clear that these
events were occurring at the premier research institutions
in the United States: Memorial Sloan-Kettering, NYU, Harvard,
For the next six years a series of scandals happened in the
country, culminating in the Tuskegee syphilis study, which
was stopped in 1972 by the then-Department of Health, Education
and Welfare after it was revealed in the press, although it
is important to emphasize Tuskegee was not hidden; it was
not done surreptitiously. Papers were coming out annually
from that. As recently as 1969, just three years before it
was stopped, it had been reauthorized by the Public Health
Tuskegee led directly to the appointment of a National Commission
for the Protection of Human Subjects of Biomedical and Behavioral
Research in 1974. That was chaired by Dr. Ken Ryan, an obstetrician
and gynecologist at the Boston Lying-In Hospital.
Over five years, the National Commission issued some very
major reports, including reports on research with prisoners
that has already been mentioned and many other topics. Its
probably most enduring report is so-called The Belmont Report,
which established the sort of framework, ethical framework,
for research ethics.
Soon thereafter, the federal regulations on research — that
should say, "on research" — 45 CFR 46, were enacted in 1981.
In 1991, they became what's called the Common Rule, when 17
federal departments and agencies adopted them, although it
is important to mention that there are some federal departments
that still don't adhere to these rules, notably the Department
What's cut off (referring to the viewgraph) should say that
the FDA maintains similar but slightly different rules regarding
research that is being applied for approval as a drug or a
All right, the Common Rule focuses predominantly on local
institution-based IRB review of research protocols and continuing
review, at least annual review, of those protocols. It is
also relies on informed consent of individual research participants,
and there are special rules that have been in place for research
with children, prisoners, and pregnant women.
With that brief historical background, I want to try to systematically
and coherently look at the problems of the system, because
I think a large part of the debate, certainly inside the Beltway,
focused on one or another of the problems, but not on a sort
of overall structure to the problems.
I think the problems can be divided into sort of three domains:
structural, the whole system is laid out in the regulations;
the IRB review process itself, and assessing the performance
of both clinical research and that process.
Before I proceed, I would need to insert two caveats. The
first is that there's a serious paucity of data documenting
these problems. We just don't have a lot of studies of the
whole system and of clinical research and how it is performed.
So my claims here that there are problems is born of my experience
as a clinical researcher and as an IRB member, discussion
with lots of clinical researchers and IRB members, collection
of the limited available data that's been published, and we'll
look at the actions by the federal oversight institutions.
I also personally think — and I'll get back to this at the
end — that the absence of performance data on the system
is also an indication of serious defects. If Ford didn't know
how many defects there were in its cars, you would think there
was something wrong with their production system.
Second, while the review of the problems here is systematic,
I don't by any means suggest it is comprehensive. There are
lots of other problems I'm not going to talk about, such as
the application of the regulations which were originally created
for biomedical research to things like social and behavioral
research and whether they fit or don't fit very well.
So with all of those caveats and hesitations, I want to talk
about some structural problems. The first structural problem
is that it may strike you and the public as ironic, but not
all research, biomedical research, in this country is actually
covered by the federal regulations. Research that's not funded
by the NIH, research that's not seeking FDA approval doesn't
have to get, isn't covered by oversight. This covers some
very important research.
A lot of reproductive research falls into this category,
and, in addition, research such as on dietary supplements
isn't covered by federal regulation. Research on surgical
procedures that aren't looking for approval as drugs or using
devices is not covered.
Second, currently in the system there's no mechanism in a
systematic way for addressing major ethical issues in research
or for assessing the social value of research. There was a
national commission; there was a President's commission; there
was NBAC; there is this Council; there is another group called
NHRPAC, which has just been disbanded — for occasionally
looking at major ethical issues in research, but these are
stop-and-start events. They're not systematically organized.
Importantly, assessing the social value of research — Is
it really going to contribute that much? What are the down
sides of research? Is it designed so that it will affect the
most citizens? — is not systematically evaluated in the system,
and to some degree is discouraged from being evaluated.
The third problem: We rely on local institutional-based IRBs.
When this was developed 25 years ago, that might have been
reasonable when a lot of the research happened at our major
academic settings like Johns Hopkins, the NIH, Harvard, UCSF,
but in fact a lot of research, or maybe even today a majority
of research, has now filtered out into private practices,
clinics, and a lot of other organizations. There is a mismatch,
many people feel, between the fact that we have institution-based
IRBs, but a lot of the research isn't even occurring anymore
Fourth, even within institutions there's no dedicated funds
for IRBs. When the NIH gives out grants for research, there
is no extra money for the IRB function. It comes out of what's
called indirect money, and there's not even a line item in
Many IRBs, certainly before these recent scandals, were resource-starved,
I think is an accurate way to put it. Now the scandals have
gotten the attention of many institutions, especially if you
can close down Johns Hopkins, and Johns Hopkins has announced
or indicated that they have doubled the amount of money they
devote to their IRB. But I think in general there's still
a lot of lack of resources there, and, importantly, IRBs have
to compete with all the other demands for medical institutions
that are under increasing price pressure.
Fifth, there's a lack of attention to institutional conflicts
of interest. One of the problems of having your IRB be institution-bound
is that the IRB is actually regulating the research within
the institution that is paying the IRB. That is an inherent
conflict of interest. I, myself, have twice experienced some
of the pressures that can be brought to bear by reviewing
research done at the institution and people, powerful people,
in the institution having some interest in getting that research
Sixth, there's a very poorly defined mechanism for investigator
conflict-of-interest rules and the IRB's role in enforcing
those rules. An investigator in this country can be subject
to as many as five different conflict-of-interest rules: one
from the institution where they work; one from the NIH; one
from the FDA; one maybe from their professional society, and
if they work at a state institution, there's frequently state
They all differ. It is very unclear how to follow them, and
it is very unclear how that is matched with submitting a protocol.
The IRBs also don't know what they are supposed to do.
Seven, there has been inadequate education of clinical investigators
and IRB members. The NIH has recently mandated that for its
grantees there be education and training, although there's
no curricula requirement; no one knows what the content is
supposed to do, and, importantly, therefore, no one knows
what the sort of core material they're supposed to understand
and any way of assessing whether it works or not. This, by
the way, is in marked contrast to animal research.
Eight, the system is replete with repetitive reviews. So,
for example, if I do a research study, which I once did at
Harvard, involving five other cities and hospitals in all
those other cities, and ultimately there were 40 different
hospitals, I actually had to get that protocol reviewed at
42 different places to do the research, all of them looking
at the exact same thing over and over again. Sometimes this
process can take a year and involve a lot of people's attention
and energy. I know that another project I am currently involved
on, we have one person who is full-time, worked for a year,
just in getting all the paperwork together.
Those are structural problems. Now I want to shift to the
IRB process problems. I think it is fair to say that a lot
of the attention that we have today is really focused on these
IRB process problems. A lot of the legislation, the accreditation
process, really take these as key. I, myself, think they're
important, but I don't think they're the whole story, and
that is one of the reasons — and the previous eight problems
are a large part of that reason.
Getting your research approved is a time-consuming process.
It has become longer and longer. I will say that most investigators
are increasingly frustrated. Six months from submission to
approval is not unusual. You often have to go through scientific
review, which can take several months. Then the IRB frequently
meets only monthly, and if you don't get on the schedule,
the IRB frequently has corrections or modifications they want
made. You then have to go back in front of the IRB. It is
a very lengthy process.
Ten, there's poor quality control of the IRB reviews. One
of the things we learned from the Johns Hopkins episode is
that the investigator submitted information about the drug
being used, but it wasn't complete or comprehensive information.
The IRB relied on that information, did not proactively investigate
further. This happens frequently.
Sometimes the IRB will get a protocol in an area. There's
no one on the IRB who's an expert. Well, IRBs always could
get ad hoc members, could actively investigate. That requires
time, resources, and it is frequently not done.
Eleven, there's is excessive focus on the informed consent
process at IRBs and excessive focus within that just on the
document and what is, I think, derisively but not inaccurately
called "wordsmithing" of the document. Now some of that wordsmithing
is important to make sure that the language and the level
of comprehensibility is correct, although we don't always
get it right, but a large part of it is just altering words
for preferences without any real focus on its importance.
There is good reason why this is true. Frequently, IRBs have
been criticized by federal regulators for leaving out something
when a scandal happens. This happened at Johns Hopkins. But
a large part of the deliberations are about the document and
really ignore many, many other important aspects of the research.
Twelve, there's substantial deficiency on how IRBs monitor
research and conduct these continuing or annual reviews. It's
frequently said, and very accurately said, it seems to me,
that IRBs see the research protocol at the start, and then
the ball rolls down the hill. They have no idea whether the
research is actually being implemented as stated, and continuing
review, the annual review, is a very perfunctory thing, done
frequently in just a handful of minutes for very complicated
studies. This is a very, very serious problem.
I think one of the most serious problems personally is we
have a very inefficient and ineffective adverse event reporting
system. Adverse events are those problems that arise in the
course of a trial, people having reactions to drugs or complications
that need to be reported.
It is now required that every IRB get informed. The FDA has
a very strict timeline, a strict rating criteria. The NIH
is a much more loose, very undefined system.
More importantly, if you have a large trial with lots of
centers, you have to send the same information to all the
IRBs and they have to make a decision, each individually,
what to do, frequently without knowing how many people have
been enrolled, are these three cases a large number of cases,
a small number of cases, how serious. It is a very time-consuming
process, but doesn't do a lot to protect the people on research
The last two problems I want to talk about — cut off at
the top, unfortunately (talking about viewgraph) — are what
I call performance assessment problems. I think these are
extremely important and not very well-appreciated.
We have almost no data about IRB function and how well they
do their function, and whether, in fact, they actually protect
patients. We don't know a lot about what goes on, where they
spend their time. Is there a relationship between bad continuing
review and harm to patients?
Fifteen, probably more serious is no one in this country,
not the Director of the NIH, no one at Pharma or at Bio, and
not the commissioner of the FDA, can tell you how many research
trials are going on today in the United States, how many people
are enrolled in research trials, how many people experience
serious toxicities, and, indeed, how many people die on research
trials. We don't record that information anywhere.
Again, it seems to me it is hard to know whether the system
is better, worse, going in the right direction, going in the
wrong direction, if you don't systematically, regularly, routinely
even know how many people are on research trials. We don't
even know how much we spend on it. I might say we don't even
know exactly how many IRBs there are. I am sure General Motors
knows exactly to the number how many dealerships it has. We
know that there's somewhere between 4,000 and 6,000 IRBs in
Let's talk about the reform proposals now. There are a number
of reform proposals: accreditation of IRBs, which I'll talk
about in-depth; credentialing of IRB professionals, which
I won't talk about because I think it addresses a very small
number of these problems; some of these legislative proposals,
which I'll talk about. OHRP, again, that federal oversight
body has some quality improvement initiatives and conflict-of-interest
initiatives. There's been an effort at the National Cancer
Institute to create a central IRB, which is in a pilot phase,
and I won't say more about that. In a few months we think
the Institute of Medicine is to report on its view of how
to change the oversight system.
So I want to say a few words about accreditation of IRBs.
There are two accreditation processes, again the alphabet
soup here. NCQA, the National Committee for Quality Assurance,
was hired by the Veterans' Administration to reform and look
into and to accredit veterans' IRB programs, and AAHRPP —
and I can never even remember what all the initials stand
for — is an independent group that is doing voluntary accreditation
of institutions and research hospitals.
There's some overlap in what they are evaluating. They are
looking at the educational programs of these institutions.
They are looking at the specification of IRB membership qualifications,
how they're trained, how they're kept current. They are looking
at policies on conflict of interest in other areas that the
IRB has to deal with, and they are developing guidelines and
assessing how many resources are provided to IRBs.
At the request of the Department of Health and Human Services,
the Institute of Medicine, as the first part of its oversight,
examination of the oversight system, reviewed these projects
on accreditation. Basically, they said that the standards
being developed do not apply to all types of research in all
types of settings, that they really focus on biomedical research
in the academic setting.
They argue that neither set of standards is tested or ready,
as it were, for prime time. They said that accreditation is
really best viewed as a pilot project, to be evaluated over
a substantial period of time.
I would say — and this is my summary; they didn't put it
in this language — at best, the IOM gave what I would call
tepid support for the accreditation initiatives. It was certainly
Importantly, which we can confess to you we just learned
yesterday, as of April 17th, the Veterans' Administration
suspended the NCQA accreditation process because of concerns
about the standards and the process itself. It is anticipated
they will resume in January, from what we have heard.
Part of what is happening is that the number of standards
is being decreased from 130 to 100, and the revised process
is supposed to be more VA-friendly. It was felt that the accreditors
were not, in fact, very sympathetic to the institutions and
that any deficiency on any one of those standards flunked
Now accreditation might affect a number of the concerns and
problems I raise, such as resources for IRBs, some of the
conflict-of-interest issues by forcing policies, some education
issues, quality control, but they, by and large, concentrate
on the reforms about the IRB process. They cannot affect the
structural issues and do absolutely nothing regarding collecting
I want to talk a little bit about legislative proposals,
and those of you who know about these things are much wiser
than I about these. But these are proposals. They're clearly
going to get reformed as they work their way through Congress.
They may never get passed. We know how these things happen.
But both Senator Kennedy and Representatives DeGette and
Greenwood have introduced bills into Congress, and they overlap
in many ways. They have some individual characteristics. All
the proposals want to apply the federal regulations to all
of human research regardless of the funding or the location
of the research. All mandate investigator education. Both
of the bills mandate disclosure of investigator conflicts
of interest. They permit central IRB review for these multi-center
trials, so that you don't have the repetitive review process,
and they mandate collection of some basic performance data.
Senator Kennedy's bill would require accreditation of IRB
six years after passage, and the DeGette-Greenwood bill talks
a little bit about requiring institutions to provide some
money for the IRBs and funding, although it doesn't actually
say where the money is supposed to come from.
Now, again, these legislative proposals I'm sure are going
to change as they evolve, but they do some important things,
but they obviously have some deficiencies, again, when we
look at all of the problems we have in the system. While they
will cover all research regulations and they will shift slightly
away from reliance on local IRB review, they do at least nod
towards providing some resources for IRBs, address issues
of conflict of interest and repetitive review process. There
are still many issues they don't address, such as the larger
ethical problems that IRBs are going to confront and need
Finally, OHRP, the federal oversight body, has undertaken
several initiatives, including a quality improvement program
and to get conflict-of-interest rules. As I understand that,
the conflict-of-interest rules are awaiting the Department
of Health and Human Services' final approval.
The quality improvement programs are an important self-assessment
program for institutions where they provide a survey that
institutions can do for themselves and see where they are
deficient. This at the moment is sitting at the Office of
Management and Budget and, as I understand it, hopefully,
by the end of the month, will be approved.
They also offer some site visits to particular institutions
to address particular problems, and they are beginning a program
on continuous quality improvement, which I think is quite
innovative. But this is really their form of accreditation.
The conflict-of-interest rules, at least as gone up to the
Department, apply to institutions, investigators, and IRB
members. Again, at best, many of these initiatives will concentrate
on the process problems, not really on the structural, and
certainly not on the performance assessment.
So those of you who know me as always being bold, a little
reckless, I thought I might offer a more comprehensive reform
proposal for this Council to consider and to enliven the debate.
I, again, hasten to add that many of my colleagues at the
NIH don't agree with me, but they think it is important to
elevate the debate and to begin to consider all possibilities.
So I want to suggest an alternative reform of the current
system that would have these qualities: It would extend federal
oversight regulation to all research. It would abandon local
IRB review. It would create Regional Ethics Board to conduct
all independent review of research studies, monitor research,
train investigators, and formulate ethical policies and assess
the value of research studies. It would entrust to a federal
agency coordination of these Regional Ethics Boards.
We imagine that the Regional Ethics Boards would be geography-based
in this country, that there would be 10 to 20 to cover the
whole country. They would have something like this structure:
Each Regional Ethics Board would have a number of Regional
Review Committees. It would have a liaison between the Board
and particular institutions. It would have ombudspersons to
interact with research subjects, and it would have an Ethics
Policies Committee that would develop comprehensive policies.
There would be a federal oversight body, as I said, that
would coordinate and monitor the function of these Regional
Ethics Boards. The federal oversight body would collect data,
publish performance data, and I give you some suggestions
as to what I think that data should include, so that sometime
someone in this country can actually say something authoritative
about the clinical research landscape.
Importantly, it would also coordinate the Regional Ethics
Boards to share best practices, to share operating procedures
and, I think importantly, share ethical policy reports, which
we really don't have today.
At the heart of these Regional Ethics Boards I think would
be Regional Review Committees. There would be multiple Regional
Review Committees per Ethics Board, I think possibly specialized
and focused, so you would have one on cancer, one on heart
disease, one on eye diseases, one on infectious diseases.
At least a quarter of their membership should be lay people.
They should meet very frequently, weekly, not monthly, and
they should conduct all prospective reviews of research studies
in the geographic region.
Importantly, I think each research study should be submitted
to one, and only one, Research Ethics Board. The Research
Ethics Board in the geographic region of the Chief Investigator
should review the study, and approval by that should be sufficient
for doing research anywhere in this country.
Importantly, I think one of the things we don't have, and
I didn't mention this as a problem but I think it is an important
problem, is we need an appeals process for any adverse rulings
by a Regional Review Committee. If you are going to concentrate
power, you need an examination of that decision.
So I would suggest that if a Regional Research Committee
rejects a study, the investigators should have the option
of appealing, and that the federal oversight body should assign
that research study to another group.
Importantly, I think the Regional Ethics Boards would collect
data from investigators and report the number of people approached
to enroll in research studies stratified by key variables
like race, sex, age, diseases, number of people who actually
enroll, number of participants removed from research and the
reasons why they were removed, number of people who stopped
participating in research, number of people who experienced
what we in the industry call grade IV and V toxicities —
that's serious, serious toxicities, usually accounting for
removal, but not always — number of participants who died
I also think that these Boards would be responsible for developing
educational material and educating all the investigators in
I think all of the Regional Ethics Boards should have to
establish a Regional Ethics Policy Committee that would create
a forum to examine major ethical issues arising in research.
Again, right now we do that haphazardly. We do it on a national
level, very distant from people.
The kind of problems or issues I have in mind are: Should
we pay children to be on research studies? A lot of studies
do, probably unbeknownst to you. What are the ethics of it?
What about xenotransplantation, transplanting tissues from
animals to people? What about having children to provide therapies
for a sibling, conceiving and having a child? These are big
issues that we really don't have a sort of systematic forum
These Regional Ethics Policy Committees should engage the
public in thoughtful discussion of the issues and develop
coherent, consistent policies. Again, part of the issue is
to educate the public and elicit their views. They could convene
public deliberations, develop a draft report that could be
the focus of public discussion, revise the report, establish
a policy, and begin sharing the policies. This is a way of
bringing the public into these discussions, educating the
public, but also developing policies in areas where we have
no policies today.
We also think that institution liaison is important. If you
remove this review process from institutions, you need to
connect with them. So we think each Regional Ethics Board
would appoint individuals to serve as liaisons with major
research institutions and probably liaison with smaller research
institutions and physicians who individually in their practices
Their responsibilities would include coordinating submission
of protocols, clarification of policies, feedback, alerting
the Regional Ethics Boards about problems and deficiencies.
Finally, I think an ombudsperson is very important. Right
now if you get a research trial, if you have a problem, the
number on the research form given to you to contact is the
principal investigator. That's the person who you contact.
We think there should be an ombudsperson who provides an independent
view, so people who participate in research can contact them
about any concerns they have about a research study.
Now any reform is going to have a lot of problems. I, by
no means, catalog all of them. But I think it is very important
to keep some in mind.
Again, I have mentioned the distance from and resentment
by investigators is possible. At least if it is in the institution,
the people you resent are your colleagues. If it is 100 or
200 miles away, it is some bureaucrat. I think that can cause
a lot of resentment.
The concentration of power, I don't have to tell this Committee,
can lead easily to the abuse of power. That is definitely
a potential problem.
Uniformity of policies among 20 Regional Ethics Boards can
certainly lead to rigidity, and that I think is also a problem.
Finally, no doubt, some people are going to worry about the
high cost of having multiple Regional Ethics Boards. Cost
is always a problem. Clinical research in this country is
estimated to be a $40-billion-a-year industry. That is the
research part of it. It seems to me a few hundred million
dollars is not too much to pay for protection, and that is
probably what we're paying now, although it is very hidden.
I thank you for your time and attention and welcome discussion.
CHAIRMAN KASS: Thank you very much.
The floor is open for discussion. Michael Sandel.
PROF. SANDEL: I would first like to ask
one or two naive factual questions, and I may be the only
one who doesn't know the answer to these questions, except
maybe my friend Gil. I hope maybe he doesn't either.
The IRBs, as they currently exist, that's only for federally-funded
DR. EMANUEL: It is a little more complicated
than that. It is for federally-funded research and research
that is seeking to get FDA approval for device or drug.
Now the federal oversight body has used its authority to
mandate, to create agreements, for example, with Harvard that
any research done at Harvard, even if it is not NIH-funded
or if it is not seeking FDA approval, will still abide by
the rules. But those are voluntary agreements. There's no
law that requires them.
Furthermore, there is a lot of research, research that's
not seeking FDA approval, like on dietary supplements; research
that is not NIH-funded, such as many reproductive research
studies, that don't have to abide by those regulations.
PROF. SANDEL: If there's a private company
that doesn't take federal money, they don't have —
DR. EMANUEL: And not seeking FDA approval.
PROF. SANDEL: They don't have any IRB?
DR. EMANUEL: They might.
PROF. SANDEL: They might?
DR. EMANUEL: They might, but, again, the
issue is whether they are legally required to.
PROF. SANDEL: And the people, let's say
there's some biotech company someplace that has an IRB. Can
they just appoint employees or do they have to appoint certain
people to it?
DR. EMANUEL: Usually companies don't appoint
or have their own IRBs. What they do is they contract out
with proprietary or for-profit IRBs, some of which are probably
the best IRBs in this country actually. So there is a private
IRB market out there, as it were.
PROF. SANDEL: And it only applies to research
on human subjects? IRBs don't deal with ethical questions
of research apart from research on human subjects?
DR. EMANUEL: Right, and human subjects are
defined in the regulations. They, for example, don't include
dead people, are not human subjects.
PROF. SANDEL: And just one last question:
This is no longer now in the category of the naive factual
questions, but about your proposal. The thing you're proposing,
that would apply, that would look at ethical questions beyond
necessarily human subject research?
DR. EMANUEL: That would depend on what's
crafted, but it seems to me one of the — it would certainly
be more than enough ethical questions regarding human subjects
research. Importantly, how human subject research interfaces
with many other aspects of clinical care and society are important
issues that we haven't looked at.
The inclusion, in our view, of the importance of examining
what we call social value of research, that's widely considered
to be part of the ethical requirements of research. It would
be wrong to do research that has no social value. It's now
currently not systematically looked at at all.
PROF. SANDEL: And for your setup, are you
only interested in the ethics of research, human subject or
otherwise, or are you interested in the ethics of biomedical
practice, and is there a clear distinction between the two?
The reason I ask is that you gave the example of people conceiving
a child for a blood match or a bone marrow match, or conceivably,
as we'll discuss later, sex selection.
Suppose an IVF clinic wants to engage in various practices
like those, not research practices but clinical practices.
Would your proposal cover those, too?
DR. EMANUEL: You're getting to the issue
of what distinguishes research from practice —
PROF. SANDEL: Right.
DR. EMANUEL: — and there's not a — I mean
there's some stipulative definitions, but it's an open question
as to exactly what the criteria are.
Certainly the early cases of conceiving a child for therapy
were done on research studies. My current conception is that
this is related, ought to be related to research, but, obviously,
the considerations and the rules that you set down and the
research setting would have some moral force, it seems to
me, to extend beyond.
One isn't going to engage necessarily in some of these things
in the clinical setting before you do it in the research setting.
So it's very frequently, it seems to me, that you will address
or at least confront the issue in the research setting before
you get to the clinical practice setting. Certainly you ought
CHAIRMAN KASS: Let me follow on this, too,
Zeke. I mean, I'm mindful of the fact that the talk is about
the reform of existing institutions that are problem-ridden,
and much of what you say in the proposal and the summary makes
sense to me. But what's missing in the discussion, and what
is somehow tacitly assumed, is some notion of what the goal
of the IRBs or of your Regional Ethics Boards would be. If
they remain defined as the protection of individual human
subjects of research, then it seems to me you have — I don't
mean inappropriately, but you've limited only one piece, you've
limited off only one piece of the larger ethical issues that,
for example, concern this body. Point one.
There is for many people a sense that the crucial ethical
issues in biomedical technologies are, in fact, the research
ethical questions, and one will try to force almost all of
the other questions into that model, and we had our hand at
that actually earlier this year.
But if you then go on to say, as the statute does, that what
you mean, what you're looking at, the protection of individual
human subjects, that you will consider risk-benefit analysis,
that you will consider questions of subject selection, and
that you will see to it that there is informed consent, that
is a further narrowing even perhaps of what it is you want
to consider when you're taking account of individual research
So the large question is, the largest way to put the question
is, assuming you wanted to design a system that would in fact
protect the individual subjects of research in as rich a way
as you can, would such an institution devoted to those goals
be reasonably expected to address some of the larger social
and ethical questions or do we need something different to
DR. EMANUEL: I don't think I said anywhere
that these groups should limit themselves in the way that
current — that we simply should take the way current IRBs
limit themselves to individual research subjects and simply
protecting them as necessarily the paradigm. I don't think
I meant that, and I certainly don't take that to be the case.
I'm very much on record in getting at, among other things,
concerns about how research affects communities. It's very
important and underplayed.
Second of all, it seems to me that many of the questions
that ought to be taken up do go beyond a narrow conception
of either risk or subject selection. Let me just point out
the way I conceive it, and maybe we're talking here a little
If you take the question of, should we pay kids to participate
in research, now you can conceive it very narrowly, but you
can also conceive it much more broadly. How does it express
how we value children in our society? How do we understand
the relationship between kids and their parents, if you pay
the child independent of the adults? What kind of payments
communicate different things? What's their both material and
It seems to me you can't answer those questions in any narrow
way, and part of my alluding to the kinds of things that I
think are on, or would have to be on the agenda, this is a
burning question for lots of IRBs, I take it, I know, in this
I would tell you that there are probably only two things
written on paying kids in research. We've written one of them,
and the American Academy of Pediatrics has written another.
Actually, the European Union has a third. But that's it. I
think discussing the underlying values of what you are trying
to achieve in policies is absolutely critical. That's part
of the reason for having an ongoing body.
Similarly, other topics that would have to be taken up that
are burning issues today have broad social meaning. For example,
what are the obligations of researchers from this country
who are going overseas to developing countries to provide
medical care beyond the end of their research for the people
who have been involved in their research studies? And maybe
not just on the disease that they've looked at, but on other
diseases. How long does that occur? Who is supposed to shoulder
the financial obligations? You cannot answer that kind of
question in a very narrow frame of just protecting subjects.
Third, it seems to me that when you include the notion of
assessing the value of research, the social value of research,
you go to the heart of the question: Why are we doing this?
One of the things we have certainly learned over the last
decade or 15 years, and I think was alluded to by Professor
Dresser, was a lot of the research we had structured was very
narrow, you know, assessed only white males between the age
of 50 and 59, and had no greater social implication, or couldn't
be extrapolated without significant danger.
How we understand the research we're doing, make sure that
it is maximally valuable to society, I think is very important.
Again, it is widely recognized as an ethical requirement for
research and simultaneously widely ignored in any review of
research. It seems to me that's exceedingly important for
us to examine.
So I think it is hard to see how, if you take one of these
issues seriously, if these Ethics Policy Committees took it
seriously, how they would stick to the narrow question of
just protecting individuals. Almost all these questions open
up larger questions of social values.
CHAIRMAN KASS: Just a very brief followup:
It does, then, seem to me that you've got enough trouble,
I suppose, peddling this proposal as it stands, but I think
it would be enriched if it began really with an articulation
of the kinds of things you're now saying, where you somehow
talk about your perception of what it actually means to call
something a Research Ethics Board as opposed to saying Institutional
Review Board, and to make some of these tacit goals explicit
and call attention to what might be different in the overall
ethical and social intent of this, I mean if you mean it.
DR. EMANUEL: Yes.
CHAIRMAN KASS: So that it becomes — people
will see really what's at issue. They may not like it. The
individualistic focus is very prominent in the society; communitarian
perspectives have an uphill battle. But, nevertheless, it
does seem to me that the proposal itself could be enriched
upfront with some kind of discussion that doesn't begin just
that there are dissatisfactions with the way the system works,
but to talk in a more positive way about what one is trying
PROF. SANDEL: Okay.
CHAIRMAN KASS: I'm sorry. Charles and then
DR. KRAUTHAMMER: Michael Sandel said he
was asking a naive practical question. My job is to ask the
naive philosophical questions.
Why regional and not national? I know that obscenity is defined
locally and regionally in this country. Is ethics also defined
DR. EMANUEL: Well, I think moving to a national
body would be, for practical reasons, unwieldy, just sheer
volume and considerations of that. I think many of the standards
are universal and aren't particular to a region. So part of
the regional focus is a practical focus.
It is also an attempt to try to avoid the rigidity problem.
While I think having 4,000 to 6,000 IRBs is probably more
akin to chaos, having 10 or 20 think through some of these
problems and some of the implications is probably a good way
to keep the pot stirring and keep innovative ideas.
I think in this country I don't have to tell you, a better
historian of American culture than I, national things obtain
a lot more resistance than even local things.
One of my colleagues did ask a similar question, which is,
well, then, let's look at the Ethics Policy Committees, why
are they regional and why that could be national. I would
consent to that.
DR. KRAUTHAMMER: Am I overinterpreting you
to say that this is more of a political, practical, what-is-feasible
consideration rather than philosophical? If you were creating
a system from scratch and had no political constraints, would
you have any — is there any reason in principle, other than
the idea of the laboratory of democracy, different regions
learning things and teaching each other, other than that,
which I think is a consideration?
It would seem to me more logical, if you were going to have
a set of ethical regulations and people applying them, that
it would make sense to have one system or at most one or two
or three regionally. But I don't quite understand, other than
what you're saying, that it is politically impossible to go
the other way. In principle, would it not be desirable to
have as few as possible?
DR. EMANUEL: I guess I am somewhat of a
student of political history. Concentration in one place I
think does tend to ossify. I think that is not a trivial concern
here, especially as research progresses in some areas. I am
not a big believer that it raises, you know, the argument
that it raises new questions, but it certainly puts them in
different lights and makes different kinds of challenges.
I think in that light, having everyone having to troop to
Washington or Chicago or something, it is probably a much
better idea to have different places looking at things.
DR. KRAUTHAMMER: Could I just have one quick
followup? So, in practice, if you have an investigator who's
headquartered, say, in Boston, under your proposal he goes
to the regional one there, but he might involve a hospital
in San Francisco that is operating under a different set of
rules. Yet, that protocol, as practiced in San Francisco,
would have to adhere to the Boston rules rather than the local
rules. Wouldn't that lead to all kinds of jurisdictional problems?
DR. EMANUEL: It might lead to jurisdictional
problems, but I think I don't see the different rules per
se; there might be different interpretations of some rules.
But, again, I think that's possible, but it is a problem that
is better lived with than either the current system or one
PROF. SANDEL: I hadn't thought about this
before Zeke's presentation, but one argument in favor of having
Regional Ethics Boards might be to allow not just a laboratory
of democracy, diversity for that reason, but different arguments
to be developed and positions taken on ethical questions,
analogous to the way it works with federal district courts
and courts of appeals.
Now there is a Supreme Court. And one question you might
have put to Zeke was, why not have a Supreme Ethics Review
Board to handle the appeals rather than some other regional
one, which is what I thought you were going to ask?
But there is currently the law now, as a result of regionally-based
federal courts of appeals, that you are allowed to count race
in admissions to law schools in Michigan but not in Texas,
because the courts of appeals have come out with different
positions on affirmative action. The Supreme Court could take
those cases up to resolve it, but so far they haven't.
There is something to be said — I don't know if it is enough
to be said — for having debatable questions like these medical
ethics questions and like affirmative action be reasoned and
debated differently by different regional courts.
DR. KRAUTHAMMER: But, Michael, that structure
really only makes sense because you ultimately do have a Supreme
Court which will in the end adjudicate on the critical issues.
If you didn't have that, you, I think would agree, that having
a system, a legal system in America of circuit courts and
no Supreme Court would be incoherent and ultimately hard to
PROF. SANDEL: Well, incoherent I'm not sure.
There are those who defended the federalist system who would
say there's something to be said for that degree of decentralization.
But you could add to Zeke's thing a Supreme Ethics Review
DR. KRAUTHAMMER: But you would have to find
a better name.
CHAIRMAN KASS: Mike Gazzaniga.
DR. GAZZANIGA: There are efforts going
on, or about to go on, to address some of these things. I
know that Dartmouth just received a large grant to start a
database of all the IRBs that are funneled through the NIH.
So some of these coordination efforts are started.
But I have three questions for you. One, are you knowledgeable
about any cross-cultural models that might help us think about
Two, have you really thought about the time to study? I'm
the clinical investigator who uses the current system, and
I'm about to inherent your system. Have you thought it through
to what's involved there?
Three, a little more thought, apparently, you're in total
agreement on this, but the monolithic problem of an agency
with an ethic. If I have learned anything in the last six
months, there's no one ethic; there's all kinds.
The ossification you talk about I think is a huge problem.
Science, for example, if it only had one vendor, one source
for all money, the beauty of the process would come to a screeching
Lastly, I don't think that local control is all bad. I see
in our IRB process that people learn quickly who is responsible,
who are good. Whereas, if we mailed this off to somebody in
Manchester, New Hampshire, or to Boston, or whatever, it sounds
like a nightmare.
So, as a friendly skeptic, maybe you could comment on some
DR. EMANUEL: First, in the cross-cultural
models, all through Europe there's different models. Some
countries have a national system, like Denmark. Some countries
have a regional system like this. I think it is fair to say
that there is a European directive now; they want to facilitate
review, and review is in a few years going to be required
to be 60 days, no longer than 60 days.
Places like The Netherlands are fulfilling that review by
basically getting rid of all IRBs at the bottom end, that
don't deal with a lot of protocols, don't have a lot of experience,
don't have a lot of standard operating procedures. They've
acknowledged that essentially it is going to be a regional
Second, I think we actually have home-grown examples of how
this might work. The proprietary IRBs are, in fact, a model
of a regional IRB system. At least in this country, as I alluded
to, they are quite good.
Now those people who are at academic centers I know are quite
skeptical of this, but, in fact, we have some good data about
that, because every time the federal government closes down
a major research institution like Hopkins, they don't send
their protocols from Hopkins to Harvard; they send them to
Western IRB out in Olympia, which is a for-profit, proprietary
IRB, which they think is good not just to review them, but
also to give Hopkins and other institutions ideas about how
to reform their system and remodel it.
Now if you look at the Western IRB, unlike most institutions,
they actually keep track of how long between submission of
a full protocol and actual action occurs. Their time on average
is 10 days.
I worked at the Dana-Farber Cancer Institute in Boston. We
couldn't get the 10 days to even get in front of the IRB,
and I'm sure at your institution it's no different.
So if you want to know, could it be more efficient, could
a regional system actually work, and would it be better than
what I live with, the answer is we actually have models that
suggest it is true. Now one of the virtues of these big regional
models is the Western IRB has 210 employees doing all the
background, you know, the back-room work, as it were. I don't
know what yours at Dartmouth has, but most IRBs have two-three
people even that are considering a thousand protocols a year.
So that can be a substantial big difference. I think, yes,
they are going to be distant from you, but, on the other,
they are also going to be — they might be more efficient.
As far as the single ethic versus multiple ethics, I think
it partially does draw into the discussion back and forth
between Charles and Michael. The fact of the matter is I am
sympathetic to your view. I don't think there is going to
be disagreement on some of these very, very tough issues.
I think one of the things we don't have in the current system
is an established, systematic way for addressing these issues
and discussing them. So you raise the issue of paying kids
for participating in research. You get a lot of gut reactions,
and I can tell you what the gut reactions are. Absolutely
no, it's wrong.
But you don't have a lot of reasoned thought about it and
reasoned argument. Is it wrong because you're exploiting the
kid? Is it wrong because it symbolically suggests that kids
are to be used? Is it perverting their incentive structures,
et cetera, et cetera?
It seems to me that part of what we do need in the system
are much deeper ways of thinking and developing those arguments
and considerations, laying them out, and seeing how we think
One of my other hopes, which I actually didn't emphasize,
is, you know, some of this actually is not just theoretical
arguments. It actually might depend upon data; you know, what
does the practice lead to? Are people more callous?
I mean, I heard earlier when someone asked the question,
well, you know, in the current research, if people don't identify,
do we get a lot more poor and ignorant on research? Well,
it might be good to know the answer to that question.
We don't even know the answer to the question that the Council
was debating in the morning: Do people identify with the research
studies they're on or not? We assume somehow that they don't.
Well, there's at least some evidence that they do, but it's
not great evidence.
Anyway, I think a large part of — I mean, there are multiple
considerations. In a pluralistic society we are going to have
debates, but on some things I think we might actually come
to some good consensus, and on other things at least we could
come to some greater clarity about what is at stake. Again,
I guess Charles' fear is different places might do different
things. To some extent, that might be good, and we might see
the consequences of that.
DR. GAZZANIGA: You wouldn't be opposed to
this being run by the private sector then, from your lionizing
of the Western —
DR. EMANUEL: Well, let me step back. I do
think Western is a great model actually, and I ought to say
here that I started out being a skeptic, went out there, and
its president now is a colleague and a friend of mine. I think
she has a great thing going, and partially because she has
been quite dedicated single-mindedly to it.
I'm not sure turning it over to the private sector is necessarily
the way to go. I do think the private sector ought to contribute,
since pharmaceutical and biotech companies do a lot of research,
but so does the federal government. Like everything else,
it is likely to be a public-private partnership because it's
got virtues on both sides.
CHAIRMAN KASS: Janet, please.
DR. ROWLEY: I just wanted to say that at
the University of Chicago we've gone from having a single
IRB with 15 members to three IRBs. They do meet monthly, but
now we have 45 of our faculty plus lay members involved in
this. I just wanted to add that because you mentioned that
one or two or three, or some such, as a number of IRB individuals
involved at some institutions. Rebecca was saying, I guess,
at WashU they've got four IRBs going simultaneously to meet
the increased demands.
DR. EMANUEL: Let me just clarify what I
was commenting on. It wasn't just two or three people on the
IRB. Yes, most IRB panels are good size. Especially at major
research institutions, 15 is not an unusual number. It's the
back room that I'm talking about, the people who are actually
monitoring, say, the regulations and trying to understand
how they apply or monitoring the literature to try to understand
how new thinking might apply to research, or actually getting
the protocol together and overseeing and collecting data like
adverse events. Those tend to be starved areas, areas that
don't get a lot of resources and I think actually are quite
important for protection, for actually thinking through the
Because, let's face it, most IRB members have lots of other
jobs; this isn't their primary function. They're not bioethicists.
When an issue of, you know, this protocol says they're going
to pay children $1400, you know, they're going to make a gut
reaction without having thought it through for very good reasons.
No one at the IRB or in the institution, might I say, and
dare I might say even in most of the country, has really thought
DR. ROWLEY: And I just follow up on that
in terms of the back room, which I have no way of knowing
right now how many we have. But these are costs that are actually
borne either by the medical school or the hospital, which
you alluded to, and you're saying, well, you get a certain
amount of indirect costs and so some portion of your indirect
cost is allocated to trying to compensate for this. But as
these costs increase dramatically, it does make it extremely
difficult to do this with the funds available.
DR. EMANUEL: Right. I agree with you 100
percent. The current system isn't made to support it.
DR. KASS: I have Rebecca Dresser, then Gil,
Bill May, and Paul.
PROF. DRESSER: I think on this issue of
how much diversity among regions or localities we would want,
it would get stickier the higher up we go with sort of broad
policy issues and value judgments, such as if one region says,
"Well, xenografts the risk is worth the percent of benefit,"
but another one doesn't. I think that could be problematic.
Or say, getting to this afternoon, one region says, "Well,
any more new drugs on sort of cosmetic psychopharmacology,
those aren't justified," you know, that could create some
difficulties. But that was just a sidebar.
One thing I would add, which you talk about publishing policies,
and so forth, I think we also need kind of a common law system.
We need IRBs who wrestle with difficult cases to then publish
them, and I think that would be excellent. Because there really
are a lot of these conflicts that can't simply be resolved
at the legislative level. I think IRBs need to see how different
ones handle these things.
I guess the final comment I would have is, I really agree
that the way the human subjects' protection system has been
applied has left out a lot of the more crucial values conflicts,
which I think underlie a lot of things. Belmont actually gets
to a fair number of them, but then they sort of were left
For example, I think so many judgments, say on the value
of studies, the regulations say the risks have to be reasonable
in relation to the benefit in terms of the knowledge that
will be advanced and any direct benefit, if there is one.
That is such a value judgment.
When we talk about reasonableness, in law we are talking
about the jury system, you know. They also refer in there
about, when you're talking about how well-informed should
people be in making a decision to be in a study, well, we
could talk about the standard of the reasonable subject. These
reasonableness judgments are throughout.
You talk about 25 percent lay people on your Regional Boards.
I don't think that fully gets to this problem of, how do we
inject the democratic element and input from people who are
not part of the scientific enterprise or the academic enterprise?
Robert Veatch wrote a really good article back in the seventies
alluding to some of these things. He said, basically, everyone
in universities is convinced that advancing knowledge is valuable.
As somebody who works at a university, I certainly buy into
However, not everyone in this country takes that point of
view. If we are talking about a democratic process, we ought
to have that point of view.
So I think it is not just a matter of what percent of public
representatives, lay people, you have on; it is also a matter
of: How do those people get educated? How do we get people
on there who are not intimidated and who don't learn everything
from the people who are part of the establishment, which I
think is often the case now, and really can represent an independent
point of view?
I actually think there are good patient advocates out there
who are knowledgeable enough, who are skeptical in ways that
I think are healthy, who might be able to help out here. But
I think that is something that is not so much addressed in
DR. EMANUEL: Let me first go to the point
about case law, which I think is absolutely a great point:
that if you are making lots of ethical choices, and there
are hundreds to be made in this process, you need some way
of sort of recording them, being consistent, having some institutional
memory. That is impossible, it seems to me, with 4,000 to
6,000 IRBs running around the country, all basically on a
starvation financial diet.
It becomes much more reasonable when you have 20 or something
and you have a coordination system. Coordinating 20 is a lot
different than coordinating 4,000 or 6,000. Part of that coordination
is sharing things like policies and practices and decisions.
Second, I'm not sure — and I would love to have your ideas
on it, and everyone else on the Council — I agree with you
it's very important to have people who have some independent
thinking and can think who aren't necessarily wedded to the
research enterprise or don't get their daily bread from it,
but are thoughtful about it. Other than specifying numbers,
it is hard for me to figure out what you do to engender that
Now we have tried, one, to increase the numbers so they are
not isolated and they are not intimidated necessarily. Another
element that we have tried to include is to try to think through
how an Ethics Policy Committee might actually involve the
public and create focus for discussion.
I think my at least peripheral or cursory look at the way
Britain handled much about the cloning and stem cell issue
is, you know, there over the course of about 15 years they
had a report. They had a lot of public discussion about the
report. They had a second commission looking at things.
First of all, it was an expanded period of time, but also
there was a focus for discussion. I mean, if we just talk
about some of these ideas without a focus and without someone
presenting a paper, I mean, it seems to me one of the — I
don't want to speak for Dr. Kass, but one of the reason to
have a paper at the start of a discussion is to provide a
focus for thoughts on the matter at hand, but we don't have
that system currently.
We have commissions coming and going, reports sort of collecting
dust because they haven't really — there's no ongoing continuity.
There's no actual implementation to test out the ideas, see
what the problems are, revision process.
So while I am sympathetic to your concern about having a
more open process and having people who thoughtfully engage,
it seems to me very hard to figure out, other than the sort
of structural and procedural elements, how you actually guarantee
CHAIRMAN KASS: Gil?
PROF. MEILAENDER: Just a question for a
little more information about — and I suddenly find myself
blanking on the name. Did you call them proprietary IRBs?
DR. EMANUEL: Yes.
PROF. MEILAENDER: You said we have several
illustrated examples of these that are really very good, although
you only mentioned one. But what makes them good? The only
thing I've gotten so far is it sounds like a resources issue;
namely, that they hire a bunch of people who do the kind of
legwork that is required. But is there something more than
that? That is one thing I want to know.
Then kind of where do their ethical principles come from?
Do they just tell you how to apply the Common Rule? I mean,
are they doing independent ethical thinking? So tell me a
little bit more about what makes you think they are so good,
and especially ways that extend simply beyond resources, although
maybe that is the answer; I don't know.
DR. EMANUEL: Like almost everything else,
I think there's a lot of reasons that make Western and a few
of the others very good. One is certainly leadership and dedication
to doing the right thing, and believing that by doing the
right thing, you'll be successful, and in their case profitable.
I do happen to know the, I guess, presidents — I don't know
exactly what their titles are — of several of them. They
really are dedicated to doing the right thing, and they actually
created their proprietary ones because many of them started
out in institutions and found that there were limits.
Now some of it is resources. If I did have a criticism of
these places, I think one of the criticisms is that they don't
do enough sort of independent ethics thinking.
At Western, which is by far the big gorilla in this area
— I mean they are the dominant player — they have, I think,
four to six people dedicated to their regulatory office, which
isn't just looking at regulations, but does scan the literature,
read articles, try to understand the impact of articles for
the protocols they are reviewing.
For one thing, that I think is admirable. It is certainly
more than the several IRBs I have been on do. But I think
it is inadequate. I have sort of chided the president there
about they have a wealth of information and knowledge and
experience, and maybe they ought to be developing their own
ethical guidance. I mean, they do have many standard operating
procedures which embody some ethical values on these complicated
But I think that the final thing I would say is they actually
may have an advantage in being institution-independent. This
might sound a little heretical. They don't have to kowtow
to any one place. As long as most of their money doesn't come
from one pharmaceutical company or one organization or something,
they can think independently of how things go.
Actually, an institution-based, you know, part of what they
think about is, how is this going to affect the institution?
You know, if I keep turning down those studies from Merck,
they're not coming back to us. If we keep turning down the
research of one of our investigators, it is hard for him to
get NIH grants or her to get career advancement. There are
those considerations. It seems to me you don't have any of
that in an independent IRB.
So there may be some structural things. There are certainly
some leadership things. There are certainly resource things.
But are they perfect? No. Do they provide a model for some
of what I have suggested? I think so.
But I do agree with you; I think if you are going to investigate
big ethical questions that have large social ramifications,
many of these places are hesitant under the current system
because they don't think they have the imprimatur to do it.
Certainly our regulations don't encourage it. If anything,
they discourage it.
I quote in the paper some of the language that says, you
know, limit very carefully what you're allowed to consider,
which I don't think is quite proper and certainly it limits
how you can really ethically evaluate the protocols.
CHAIRMAN KASS: Bill and then Paul, and we're
coming up to a break.
DR. MAY: It sounds like a budgetary question.
You have mentioned $200 million to mount this Regional Board
system. But I am interested in what kind of people you would
be calling on here. You said they would meet every week, and
if it is regional, it doesn't mean you slide into another
office in the hospital; you may have to go out of town, come
back into town.
DR. EMANUEL: Uh-hum.
DR. MAY: Institutional Review Boards I have
been a part of were largely pro bono. Now you are talking
about an amount of time invested. Are you really thinking
about a shift from pro bono work over to paid staff and some
kind of mix of ethicists and people out of institutions? The
regional problem is it is more convenient to get people still
in the city, where your regional office is located.
But I think the basic problem here is not whether we go with,
one were to go with, for-profit institutions, but the whole
shift over from volunteer to paid staff seems to me to be
a fundamental issue. There are two ways of conceiving it.
One is a staff that has preset standards and then efficiently
applies these standards. We know about that even in academic
institutions, which have committees, which are largely janitorial
I mean they apply what is already preset, and even curriculum
committees in universities like this, they process that material.
The other way of looking at a Regional Ethics Board is to
think of a deliberative body. That is a very different kind
of activity. I am wondering how you build that in, especially
since it seems to me you are moving in the direction of a
paid staff of specialists as you move toward a regional solution
to his problem.
DR. EMANUEL: Let me take the first part.
DR. MAY: May I add just one point? The local
problem is conflict of interest. What you may end up with
is flagging of interest. That is another kind of problem that
leads to inadvertent inattention to issues until you get a
public crisis that comes along, and then people get very excited
about it. That flagging of interest seems to me to be another
dimension or problem that one has to think about.
DR. EMANUEL: You are describing Washington,
I take it.
DR. MAY: Well, I am thinking of some corporations
as well, their boards of directors.
DR. EMANUEL: Let's take the first problem.
It certainly seems to me that we are going to have to have
more professional staff, but I would not want the Review Committees
to be professionalized and bureaucratized. I think that would
be a disaster.
DR. MAY: So they would be a staff serving
DR. EMANUEL: Yes. I think, again, in most
academic institutions I have been a part of, they undervalue
that, minimize its importance in proper and efficient functioning.
Nonetheless, I think if these Boards do meet weekly, the
question is, could you maintain a voluntary or at least nominally-paid
group? Again, institutions now are having to go to pay their
faculty to serve on these things because they are a big-time
commitment. So it is not even voluntary there.
Second of all, if you do look at the Western model, which
I have suggested here might be very instructive, they actually
get people for $150 a day. They meet four hours for the day,
separate from reading the protocols, which is at least another
four to eight hours. They re-evaluate these people each year.
They are not making money, these people, by participating
They have a number of physicians who want to do this to stay
intellectually active. They have a number of scientists who
want to do this. They have a number of just interested people
in the community.
We haven't experimented or seen how big an interest there
might be in doing this, and we assume that the options are
voluntary or a full-paid staff. I think that this is a very
rich country, and volunteerism, even for nominal pay, is something
that people do, and for adults who want to stay intellectually
active this may be very interesting.
I think, again, I don't like the model — we are trying to
create a hybrid model. There's clearly some things these groups
are going to have to do efficiently.
I have estimated that there's something like 40,000 research
protocols per year in this country. It seems you can't grind
the system to a halt; we're going to think deeply about every
single thing. Some of it has to just be applying rules.
On the other hand, we do need a system where we do confront
serious issues or we confront issues repeatedly that raise
deep ethical questions. We need a system for figuring out
how to reason about those. That is why we suggested sort of
differentiating those kind of functions into different kinds
of committees but having them under one rubric, so that there
is a lot of cross-talk. That is our view of how it would go.
I am under no illusion that any single Regional Ethics Board
with a regional sort of Ethics Policy Committee could address
all the issues that come up. I mean, there are just an unbelievable
number of serious issues and new issues or issues again coming
to the fore or we are being alerted to all the time.
But it does seem to me it needs to be systematic and ongoing
rather than sort of what we have had for the last 25 years:
start-stop; cancel that commission; put its findings on hold;
get another one. They are going to take up a different set
of issues. No continuity, no way of implementing in policy/practice
any of the decisions.
CHAIRMAN KASS: Paul McHugh?
DR. MCHUGH: I was very interested in what
you said and the series of problems that you think your proposal
will address. But what I haven't heard is how you think some
form of institutional review is going to coordinate with the
responsibilities of NIH study sections, which are, after all,
peer-reviewed and protocol-focused.
You, quite rightly, criticized the Johns Hopkins issue of
having overlooked a piece of important results and research
prior to the submission to the IRB, but I also want to point
out, as you should, that the peer review study section also
had passed that and the like.
If the issue that these organizations need is a better back
room where they can study the literature and back it up, well,
isn't that what the NIH study sections are supposed to do
with their peer review, protocol focus groups?
DR. EMANUEL: Well, I can safely say that
I have never received an NIH grant, and that entire process
is a complete labyrinth, of which I don't understand anything.
DR. MCHUGH: It is a terrific group. Having
been a chairman of one of the study sections, I am very proud
of it. I think it is a great achievement for the United States.
DR. EMANUEL: I wasn't criticizing it. It
is just I don't understand it. It is a little different. And
I don't understand —
DR. MCHUGH: Well, you should. You should
DR. EMANUEL: — how exactly it should interface
with this. I think that is probably another whole area where
the current system is just as opaque as our suggestion and
probably needs a little more thinking. But there is re-review
of science there, and that is obviously a very important element
of good research, but it is far from the only element.
DR. FOSTER: Just one quick question: We
have talked about the changing culture. I also chaired a study
section, sat on the Council, and so forth, but that culture
has drastically changed. So that now study sections do not
look just at the science. They look at the priorities. It
is a very different system. I don't know of a better system,
but it is drastically different from your glory days than
what it is now.
So even the time for the number of protocols is greatly enlarged.
You are worrying about budgets. You are doing all those things.
So that would take back-room people there, too.
CHAIRMAN KASS: Well, I want to thank you,
Zeke, for a very stimulating talk. I would hope that we might
down the road call on your help as we continue to talk about
this regulatory question, and in particular the kinds of questions,
not just the mechanical things which were the subject of very
interesting discussion, but also the large issue, for example,
that Michael Sandel and I tried to raise with you at the beginning:
To what extent this can be, for this Council, a source of
practical hope for the kinds of questions that we have been
Perhaps as this paper evolves, if you might see fit to try
to address that, or maybe we can have you down to the office
for further conversation, it would be very welcome.
But thank you very much.
We are adjourned until — what is it? — until 1:45, when
we will meet to have some discussion of happiness and sadness
and pharmacology — if that's not the wrong way to characterize
(Whereupon, the foregoing matter went off the record
for lunch at 12:20 p.m. and went back on the record at 1:50