Inviting researchers to attend institutional review
board sessions designed to approve these
same investigators' requests to conduct research involving
human subjects doesn't seem to affect the
efficiency of the process one way or the other, a new study
led by Johns Hopkins bioethicists
suggests.
The findings are the result of one of the few studies
to date that have sought to verify or
challenge a fairly wide perception that inviting
participation by so-called principal investigators, or
PIs, could introduce more inefficiencies in what already is
a lengthy and detailed process beset by
scheduling problems, poor investigator-IRB relationships
and administrative delays. Some researchers
have suggested an opposing view: that inviting PIs can
improve efficiency.
"The limited data on IRBs indicate they do not
routinely invite PIs to attend convened
meetings," said Holly Taylor, assistant professor in the
Department of Health
Policy and Management
at the Johns Hopkins Bloomberg School of Public Health and
assistant director of empirical research
at the Johns
Hopkins Berman Institute of Bioethics. She and her
co-authors on the review of IRB
practices at The Johns Hopkins University say one national
estimate found that fewer than 9 percent
of IRBs require PIs to attend the meetings.
Under federal law and regulations, and to assure the
safety and welfare of research volunteers,
all institutions that receive federal funds to conduct
human subject research require review and
approval by an IRB, a group generally composed of senior
scientists not involved in the research under
review, along with individuals who represent the lay
community. Bioethicists and others familiar with
human research protocols also may be involved.
Among other things, IRBs carefully consider questions
such as whether the study's science is
valid and generalizable, whether its benefits outweigh
risks that volunteers might encounter and
whether volunteers will be adequately informed about the
study to consent to participate.
While serving as members of four IRBs at the Johns
Hopkins School of Medicine, Holly A.
Taylor, Nancy E. Kass and other bioethicists at the Berman
Institute of Bioethics noticed that some
IRBs regularly invite PIs when their research plans are
discussed, while others do not.
Wondering whether there was any difference in
inefficiency between them, Taylor and Kass,
along with former Johns Hopkins master's degree student
Peter Currie, now a law student at
Georgetown University, looked back at 125 reviews conducted
by four Johns Hopkins School of
Medicine IRBs between March 2002 and June 2005. Two of the
IRBs did not regularly invite PIs to
their meetings, one did, and a fourth switched midway
through the examination period from not
inviting PIs to inviting them.
The team wondered, for example, whether PIs in
attendance could more efficiently answer any
questions that arise, quickly and directly, rather than by
replying to multiple calls and e-mails from
different board members after a meeting takes place. So
they checked the total time it took to
approve the research plans, how many pieces of
correspondence passed between the IRB and the PI
and how many meetings took place where a particular study
was discussed.
Their analysis, published in the January-February
issue of IRB: Ethics & Human Research,
showed few differences between IRBs that invited PIs to
attend meetings and those that didn't. All
took an average of 65 days to approve a study's plans, had
about five pieces of correspondence
between the IRB and the PI, and reviewed a study at an
average of 1.6 meetings.
Taylor noted that in the IRB that switched from not
inviting PIs to inviting them, time to
approval went down from an average of 114 days when PIs
weren't present at meetings to 70 days
when they attended. Additionally, the number of meetings
where each study was discussed changed
from an average of 2.4 to 1.7. The researchers aren't sure
whether the presence of the investigator
was a factor in this improved efficiency, but they suggest
that it could be one of many factors that
led to the change.
"PIs are really busy, and some IRB members might worry
that requiring PI attendance could
delay scheduling. We didn't find that was the case," she
said.
She and her colleagues plan to eventually test PI
attendance at multiple research institutions by
randomly assigning PIs to be present or absent at meetings.
Taylor notes that finding ways to improve
the efficiency of IRB approval can help researchers begin
their research faster.