The federal regulations of Title 45 CFR, Part 46 provide researchers and
institutional review boards with a definition of research on human subjects:
"[A] systematic investigation, including research development, testing
and evaluation, designed to develop or contribute to generalizable
knowledge" [45 CFR 46.102(d)]. Although this definition provides clear
guidance for the majority of human research that is conducted, there is a
contingent of activities that seems to adhere to this definition but
ultimately does not require institutional review board oversight. This subset
of research comprises certain quality improvement and quality assurance
initiatives as well as certain case reports and case reviews. Quality
improvement assessments that involve structured data collection and analysis
but have the fundamental goal of improving specific processes and systems
within specific organizations may not be considered research and thus may not
require institutional review board approval prior to
inception24.
There are two general criteria that indicate that a quality improvement and
quality assurance initiative constitutes human research according to Food and
Drug Administration (FDA) definition: (1) the investigators will seek
publication of the data in a national journal or presentation at a national
meeting and/or (2) the results of the initiative will have application beyond
the confines of the investigators' local institution or practice. If either of
these criteria is met, then the investigators are considered to be collecting
"generalizable" knowledge and the study therefore constitutes
human subjects research and requires institutional review board approval. When
an investigator is in doubt, the local institutional review board should be
contacted to discuss the situation.
It is a bit more challenging for institutional review board administrators
to determine whether case reports and case reviews constitute research on
human subjects and thus require institutional review board review.
Unfortunately, there is no regulatory guidance regarding this issue, and there
continues to be national debate among institutional review board professionals
about whether case reports and reviews constitute "generalizable"
knowledge or simply unique information about rare conditions that deserves
greater exposure through presentation or publication. Because this is largely
a philosophical issue, the policies of institutional review boards vary from
institution to institution. It is advisable to consult one's local
institutional review board to determine the definitions of a case report
requiring approval. Review boards may review individually or refer the
protocol to a privacy officer or privacy board at each institution that is
charged with overseeing compliance with the HIPAA (Health Insurance
Portability and Accountability Act) regulations.
There is another level of screening to which investigators can look for
guidance. As the national attention on human subject research has intensified,
journal editors and conference planners are more likely to screen for
institutional review board approval before accepting articles for publication
and posters for presentation. When an investigator does not require
institutional review board approval to conduct a project that will generate
data that he or she intends to publish or present, the investigator should
provide a statement with the article or presentation explaining that the
institutional review board and the investigating site or institution did not
require that the particular project be approved.
One of the main reasons for assessing the need for institutional review
board approval is that the consequences of violating research regulations can
be severe. In one example, the development of a medical curriculum was being
evaluated with a survey of students. During the initial phase of the survey,
no identifier was used and hence the survey was exempt from institutional
review board
approval25. During
the second and third years of the study, and in order to collect demographic
information from the participants, the students' badge numbers were recorded.
The faculty and staff conducting that survey faced possible suspension and
other disciplinary action because they did not obtain institutional review
board approval.
Submission Process
The submission process can be an arduous task and may contribute to the
sometimes strained relationship between institutional review boards and
investigators. It is not uncommon for studies to be approved (or disapproved)
after requests for a number of modifications of the protocol and consent form.
Common reasons for modifications to be requested include improperly designed
consent forms, poor study design, unacceptable risk to human subjects, and
questions regarding ethical or scientific
merit9. Although
there is a lack of reported data on the rate of revisions of research
submissions, Kent concluded that only 24% of research submissions were
approved without
modifications26.
There are no metrics with which to assess the success and effectiveness of
an institutional review board, as its ultimate success depends on its ability
to protect human subjects. However, an efficient institutional review board
infrastructure that allows easy submissions and faster turnaround is to be
commended.
Levels of Review
Federal regulation identifies three levels of review that may be used for
human subjects research. These are expedited review, full or convened review,
and exemption from review (Table
I). A study suitable for expedited review is one that poses only
"minimal risk" to a research subject. As defined by federal
regulation, "minimal risk means that the probability and magnitude of
harm or discomfort anticipated in the research are not greater in and of
themselves than those ordinarily encountered in daily life or during the
performance of routine physical or psychological examinations or tests"
[45 CFR 46.102(i)]. Many ethics committees provide a simplified application
for, and an expedited review of, studies that involve only a review of patient
data, records, or
specimens27.
However, considerable variability in institutional review board processes for
minimal risk studies has been
reported28.
Expedited studies include chart and radiograph reviews, observational studies,
and studies involving routine physical exercise, blood drawing, and other
minimal risk procedures as identified in the Office for Protection from
Research Risks (OPRR) guidance sheet "Categories of Research That May Be
Reviewed by the Institutional Review Board (IRB) through an Expedited Review
Procedure"
(1998)7. Expedited
studies are not reviewed by the full institutional review board committee;
instead, they are reviewed by a subcommittee or administratively within the
office.
Studies requiring full review are those that pose "greater than
minimal" risk to research subjects; these include most clinical studies
involving investigational drugs, devices, or procedures and many studies
involving vulnerable populations. Full reviews require evaluation by the
convened institutional review board committee
(Table I).
An exempt study is one that meets the narrow federal criteria allowing a
study to be exempted from the purview of the human subjects regulations and
therefore from the review of an institutional review board. Examples include
retrospective reviews of de-identified data or collection of de-identified
tissue samples. However, the exemption designation must be made by the
institutional review board. Once a study is exempted from institutional review
board review, no further notification of the institutional review board is
required except if there are revisions in the protocol. In certain instances,
revisions may alter the protocol to the degree that the study can no longer be
exempted and will require a new review designation. In any case, we recommend
consulting one's local institutional review board when in doubt about the
level of review that an individual study requires and what paperwork must be
completed prior to submission.
Training Programs
At present, most, if not all, human subject protection programs in the
United States have training requirements that must be satisfied before
submission of the protocol to the institutional review board. Investigators
and research staff (including students, fellows, and research assistants)
usually are required to successfully complete proper training in the basic
ethical principles underlying protection of human subjects, the history of
protection of human subjects, and the practical applications in research (such
as informed consent, integrity of data, subject recruitment, and equitable
selection of subjects). Certification to allow human subject research
following such training is federally required. Additional requirements include
various forms of annual training modules, the main purpose of which is to
provide interim training regarding regulations and recent developments.
Finally, with the advent of
HIPAA29 in 1996,
which dictated full compliance by April 14, 2003, a one-time HIPAA training
module must be completed prior to human subjects research. Most institutional
review boards have online training for investigators and key personnel.
Impact of HIPAA Regulations on Modern Research
The implementation of HIPAA regulations and the need for full compliance at
the present time have introduced restrictions that influence the ease with
which research is conducted. For example, in a previous Current Concepts
Review article written several years before these regulations were
implemented, Smith and Watts described mechanisms available to locate patients
for the purpose of long-term follow-up for clinical
studies30. Some of
the described legal and acceptable methods, such as the use of local telephone
companies, the Internet, the Social Security Administration, the United States
Postal Service, and credit bureaus to locate patients, would now be subject to
the regulations introduced by the Patient Privacy
Act31 and require
institutional review board approval. Research studies that commenced prior to
the introduction of HIPAA regulations now require an amendment to the protocol
requesting a waiver of subject authorization for collection of protected
health information. The approval letter may also serve as one proof of the
legitimacy of the request.
Conflicts of Interest
Disclosure of conflicts of interest is important as such conflicts must be
managed prior to one's involvement in a study. In medicine and research, the
patient's welfare and the investigators' integrity should be the primary
concerns of all involved. Private industry funding for clinical trials,
particularly in orthopaedic research, increased considerably between 1985 and
200232. Disclosures
of conflict of interest provide a system of checks and balance, the aim of
which is to maintain the public's trust in medical
research33.
Conflict of interest has made headlines in recent years with allegations in
the press of possible inadequacies of clinical trials. The involvement of
investigators and industry has called some trial results into question. Close
attention to conflict of interest has evolved since the Bayh-Dole
Act34 in 1980,
which encouraged cooperation between industry and academia by allowing
inventions made with federal funding to be licensed to industry and brought to
the marketplace. While this fostered innovative research, it also inevitably
resulted in conflicts of interest. Researchers who developed new techniques,
devices, or treatments licensed the innovation to industry and then became
investigators for industry-sponsored trials employing these same techniques,
devices, and treatments. By definition, conflicts of interest may exist when
financial or other personal considerations have the potential to compromise or
bias professional judgment or objectivity. While physicians and surgeons
intend to "do no harm" to their research subjects, a conflict of
interest may potentially influence decisions, albeit unintentionally, and lead
to coercion or, at worst, undue harm to subjects. Conflict of interest may be
tangible or intangible. Tangible conflicts include personal financial gain and
funding for research. Intangible conflicts include pressure to obtain
publications, promotion, and prestige.
Regulations regarding conflict of interest represent attempts to ensure
that research is not biased by financial or other interest and to maintain
public trust in the integrity of the researcher, the research, and the
institution. Management of conflict of interest generally requires the
individual and/or institution to do any or all of the following: disclose
significant financial interests, allow independent reviews, modify protocols
to remove or distance the individual with the conflict from research, divest
oneself of significant financial interests, sever relationships that create
actual or potential conflicts, and/or disqualify investigators from
participation in relevant sponsored research. In collaborative research
involving investigation of a drug or a device produced by a company with which
an individual has a relationship that creates a conflict of interest, the
conflict also needs to be disclosed to the other research co-investigators.
Annual conflict-of-interest disclosures are typically required by
institutions, and any conflicts should be revealed in the informed consent
form. The usual thresholds for reporting conflicts, based on the
recommendation of a task force assembled by the Association of American
Medical Colleges
(),
are >5% equity ownership interest in a company and >$10,000 of
investment or other financial interest. Acceptance of gifts, gratuities, or
entertainment that reach the $10,000 threshold should also be reported, as
should being a manager, a scientific advisor, or a board member of the entity
for which one is performing research. Conflict of commitment varies, but
generally more than three to four days per month away from a primary job
should be reported to institutional officials. The trend now is for zero
tolerance for investigators who have a significant interest in a publicly
traded company or any interest whatsoever in a non-publicly traded one. Zero
tolerance in this context means that the investigator will not be allowed any
involvement with the study in question.
The Protocol
A protocol synopsis must be succinctly written in clear, nonscientific
language with all of the acronyms defined, since lay people, non-physicians,
and scientists with different affiliations are members of institutional review
boards. Institutional review boards may furnish so-called lay-language
glossaries on their websites or in their procedure manuals that investigators
can use when they prepare submissions. The protocol should include a clear
description of the purpose of the research and the definition of the end
point(s) to be measured as well as clear rules for either temporarily or
permanently removing a subject from the trial. It should also indicate the
number of subjects who will be involved, both at the national or international
level and at the institutional level. Members of the institutional review
board need to assess whether a sufficient number of subjects is involved to
achieve the goals of the research and whether too many subjects are being put
at risk by the use of a particular drug, device, or procedure. The protocol
should contain a clear description of the study procedures along with
information regarding risks and other key aspects of the protocol. Protocol
information should exactly match the information provided to the potential
subject in the informed consent form.
Protocols also need to address the nature of the drug or device and whether
it has FDA approval for the intended use in the research study or is being
used under an investigational new drug (IND) or an investigational device
exemption (IDE) number. The risks associated with the test article must be
clearly presented, ideally in a format indicating the expected incidence of
these events (e.g., >10%, 2% to 9%, or <2%).
If an FDA-approved device is being used in a research study for an
off-label indication, a separate IDE application to the FDA for that research
use may be required. The FDA-approved uses explicitly stated in the device
labeling should be consulted to determine the need for additional
applications. Inquiries can be made by contacting the Office of Device
Evaluation (ODE) at the FDA (telephone number: 301-594-1190).
Monitoring of Research
Once a protocol is approved, the research can be initiated. During the
conduct of the research or after its completion, the investigator is obliged
to report any adverse events that have occurred in the subjects in their
institution or in other institutions conducting the same research. Each
institution usually has a safety and monitoring board or committee in place
that is charged with overseeing the conduct of research and reporting adverse
events. The same board or a separate committee may also be involved with
internal audits to confirm that the research was conducted in accordance with
the approved protocol. Hence, both the safety and the protection of the
subject as well as the high quality of research are ensured.
Involvement of Vulnerable Populations
Populations of research subjects that for various reasons are not
considered to be fully autonomous are designated by the FDA as
"vulnerable populations." This lack of autonomy may be a result of
a subordinated relationship, as with prisoners; may apply to persons with
diminished or not yet fully developed mental capacity, such as cognitively
impaired individuals or children, respectively; or may apply to pregnant women
(because the fetus lacks autonomy). Because population-specific data yields
are required to fulfill the ethical principle of justice, as outlined in the
Belmont Report, certain vulnerable populations (pregnant women, children, and
subjects with different cultural and linguistic backgrounds) must be included
in research unless there is a scientific rationale for their exclusion. Other
vulnerable populations such as prisoners, cognitively impaired persons, and
drug users may be included in studies only after careful consideration of the
level of autonomy and/or impairment they possess and whether adequate
protections can be implemented for them.
The National Institutes of Health (NIH)
()
provides specific explicit criteria for exclusion of children from a research
study, and these can be used as general guidelines when one is considering
whether a non-NIH-funded study should include children. These exclusion
criteria are as follows:
The research topic is irrelevant for children.Children are barred by law from participation because of the risk.The study is redundant; knowledge is being obtained in another study or is
already available.A separate age-specific study of children is preferable to draw
conclusions.The rarity of a disorder makes inclusion of children extremely
difficult.The limited numbers of available children are already enrolled in a
nationwide pediatric disease network.The study design precludes direct applicability to children.Insufficient adult data exist to evaluate potential risk for children.The study design is a follow-up of an adult study.
The research topic is irrelevant for children.
Children are barred by law from participation because of the risk.
The study is redundant; knowledge is being obtained in another study or is
already available.
A separate age-specific study of children is preferable to draw
conclusions.
The rarity of a disorder makes inclusion of children extremely
difficult.
The limited numbers of available children are already enrolled in a
nationwide pediatric disease network.
The study design precludes direct applicability to children.
Insufficient adult data exist to evaluate potential risk for children.
The study design is a follow-up of an adult study.
While children should be included in research, institutional review boards
must ensure that risks are acceptable in relation to the anticipated benefits.
As children are a vulnerable population because they are (1) not autonomous
and (2) do not yet have adult cognitive ability, the parameters of allowable
risk are much narrower than those for an autonomous adult. Subpart D of Title
45 CFR 46 stipulates that the institutional review board can approve research
involving children if it poses minimal risk, greater than minimal risk with
the prospect of direct benefit to the child subjects, and a "minor
increase over minimal risk" when there is no prospect of direct benefit
to the subjects but the research is likely to yield generalized knowledge
about the subjects' disorder or condition.
The same stipulation for inclusion, however, does not apply to cognitively
impaired populations. Also, no specific exclusion criteria for this population
are provided by federal guidelines or regulations. There is a general
consensus among institutional review board professionals that cognitively
impaired persons should not be involved in research that can be adequately
conducted with non-impaired subjects, unless the research poses the
possibility of direct benefit to the impaired subjects. Furthermore, no
impaired persons should be involved in research involving greater than minimal
risk when there is no potential for benefit to the
subjects35. As
there are no current federal regulations addressing inclusion of cognitively
impaired subjects in research, special care should be taken when considering
these issues and deciding whether it is ethical to enroll them in a particular
research study. The institutional review board can provide guidance on this
issue and will make the final decision regarding whether this population can
be included.
Ethical practices are of paramount concern for vulnerable populations in
developing countries, especially when industrialized countries sponsor
clinical trials in these
regions36,37.
International clinical trials often span various jurisdictions; however,
research must be founded on ethical principles and should not exploit human
subjects38. Some
have suggested that the ethical standards espoused by the United States and
other countries, such as prior review and review board approval (including
informed consent), should be adopted in developing nations in conjunction with
local customs and culture to ensure protection of human
subjects36.
An important point worth mentioning relates to the potential abuse of
patients by subjecting them to multiple simultaneous research projects that
may place an undue burden on their time, energy, and attention. In addition,
the inability to comprehend the consent form, due to linguistic barriers and
so on, could also result in potential abuse of subjects. It is the
responsibility of the investigator to ensure that the potential for subject
abuse is avoided at all times.
The Consent Form
One facet of the submission that generally requires modification following
the initial review is the consent form. These modifications typically require
further explanation of the risks and benefits for the participants. In a
five-year review of the actions of one institutional review board, Sansone et
al. found the majority of modifications required for study approval were
related to the consent
form39. Moreover,
at least one modification was requested for 84% of the submissions.
Interestingly, Sansone et al. also concluded that submissions for approval of
pharmacological studies were substantially more likely to require at least one
modification of the consent form than were submissions for approval of
non-pharmacological studies; however, Kent found that submissions for approval
of pharmacological studies were less likely to require a modification of the
study protocol than were submissions for approval of non-pharmacological
studies26.
Consideration of the aforementioned during the drafting of the consent form
may provide guidance and decrease the potential for modification requests.
Prior to submission to the institutional review board, proofreading and
spell-checking documents will result in fewer required modifications.
Providing a lay-language title for the research and avoiding lengthy
descriptions of the background supporting the research are useful suggestions,
particularly for the informed consent forms. Additionally, avoiding irrelevant
details about drug or device development is recommended for this form. Federal
guidelines suggest that consent forms be written between a sixth and
eighth-grade reading level. While challenging, practical solutions may include
a review of the form by a non-medical colleague, especially to indicate
medical terminology that may need to be changed. Software packages that
analyze and modify the reading level are also available to the
investigator.
In most cases, medical terms should be replaced with lay terminology. Also,
distinction between the standard of care and the research drug, device, or
treatment needs to be made in the appropriate sections of the consent form
(Introduction, Procedures, and Costs). The subject's
alternatives—namely, the ability not to participate in research and yet
receive the standard of care—should be clearly explained. These elements
are vital to the participant's understanding of the research. If he or she is
not able to distinguish the investigational procedure, drug, or device from
the standard application, then he or she cannot truly give informed consent to
participate. The investigator needs to adhere as closely as possible to local
institutional review board guidance, instructions, and recommendations for
formatting on the consent form. On signing a protocol and informed consent
form as the principal investigator, one assumes full responsibility.
Description of study procedures should include the number of visits, the
duration of visits, the estimated time required for filling out
questionnaires, and whether a pregnancy test, blood drawing (including how
much, in teaspoon equivalents), and additional procedures such as radiographic
and imaging studies will be required. Subjects should have costs clearly
defined, including research costs incurred by the sponsor and costs billed to
insurance providers or individuals. The consent form must include disclosure
of conflicts of interest as well as guidelines for indemnification.
If the study involves long-term follow-up of patients—e.g., to
determine the success or failure of a device or patient
survival—provisions for contacting subjects should be included in the
consent form. Such provisions could include the subject providing the name and
contact information of a relative or friend who would always know the current
address of the subject or whether the subject was alive or dead. In addition,
HIPAA regulations require that patients' consent for long-term follow-up be
sought.
Consent Process
Valid informed consent is a requirement for ethical research involving
human subjects. Studies of informed consent forms in wealthy countries
demonstrated that subjects' understanding of research often is incomplete or
at times
inaccurate40,41.
Informed consent is a process of decision-making designed to protect the
rights of human research subjects and is not limited to obtaining a signature
on the informed consent
form42. While
efforts to improve the participant's understanding are continuous, one study
has suggested that a neutral educator spending time with the subject one on
one may be the most effective way of bridging the gap between investigators'
and subjects' awareness of the study objectives and the risks and benefits
associated with
participation43.
The consent form is important in the consent process, although arguably
institutional culture overemphasizes the importance of this document.
Investigators who assume that once they have obtained the participant's
signature on the form their obligations regarding the consent process have
ended are misunderstanding the informed consent process and ultimately may be
committing a serious disservice to the participant by not observing the
ethical standards. A failure to communicate adverse effects arising from the
use of a research drug or procedure detected during the course of research is
one such disservice. Another example, and one unfortunately seen on a frequent
basis, is the unavailability of the principal investigator to address issues
and questions that the research subject may develop.
The informed consent document is important because it serves as the initial
contract between the researcher and the participant as well as a detailed
reference document for the participant. However, informed consent is a
principle that should underlie all interactions between the participant and
the research team. A participant's questions at any time during the study
should be addressed immediately and thoroughly by research personnel. The
participant should be able to contact the principal investigator (hence the
need to provide a telephone number) at any time that he or she has new
concerns about the study. The participant should also be able to contact the
institutional review board at any time if he or she has concerns about the
research and would like to consult with advocates independent of the research.
New risks should be communicated to the institutional review board in a timely
fashion, and the subjects should be notified accordingly if necessary. The
consent form may also be revised, reflecting the new risk, and submitted to
the institutional review board for reapproval. Revised informed consent should
be obtained from new and all active participants. If a subject decides not to
continue participating once he or she has been informed of new risk data, it
must be clear that he or she is free to do so without risk of retribution or
of not being able to receive additional care at the institution. Common sense
dictates that an ethical study is one in which the lines of communication
between the research staff and participants are always open.