- SPECIAL CO_V'IMUl\'IG'\TIO_‘l
`
`What Makes Clinical Research Ethical?
`
`Ezelciel J. Emanuel, Ml), Phi)
`
`David Weridler, l’lID
`
`Christine Grady, Pill)
`HAT MAKES RESEARCH IN-
`
`volving human subjects
`ethical? informed con-
`sent is the answer most
`
`US researchers, bioethicists, and insti-
`tutional review board (IRB) members
`would probably offer. This response re-
`llects the preponderance of existing
`guidance on the ethical conduct of
`research and the near obsession with
`
`autonomy in US bioethics.“ While
`infortned consent is necessary in most
`but not all cases, in no case is it suffi-
`cient for ethical clinical research.” in-
`KICCLI, SUIHC l}[l1lC lll\))l LUlllCIllil)l.l3 Ltlll‘
`
`temporary ethical controversies in
`clinical research, such as clinical
`
`research in developing countrics,"'”
`the use of placebos,”"“ phase 1 re-
`search,”‘‘’ protection for communi-
`ties,l"'“ and involvement of chil-
`
`dren,”'3" raise questions not of informed
`consent, but of the ethics of subject se-
`lection, appropriate risk-benefit ratios,
`and the value of research to society. Since
`obtaining infonned consent does not en-
`sure ethical research, it is imperative to
`have a systematic and coherent frame-
`work for evaluating clinical studies that
`incorporates all relevant ethical consid-
`erations.
`
`in this article, we delineate 7 require-
`ments that provide such a framework by
`synthesizing traditional codes, declara-
`tions, and relevant literature on the eth-
`
`ics of research with human subjects. This
`framework should help guide the ethi-
`cal (.lI’.‘VClU|)ll1Clll uutl cvaluatiuu ufLliui-
`cal studies by investigators, IRB mem-
`bers, funders, and others.
`
`Many believe that informed consent makes clinical research ethical. How-
`ever, informed consent is neither necessary nor sufficient for ethical clinical
`research. Drawing on the basic philosophies underlying major codes, dec-
`larations, and other documents relevant to research with human subjects,
`we propose 7 requirements that systematically elucidate a coherent frame-
`work for evaluating the ethics of clinical research studies: (1) value-
`enhancements of health or knowledge must be derived from the research;
`(2) scientific validity—the research must be methodological ly rigorous; (3)
`fair subject selection—scientific objectives, not vulnerability or privilege, and
`the potential for and distribution of risks and benefits, should determine com-
`munities selected as study sites and the inclusion criteria for individual sub-
`jects; (4) favorable risk-benefit ratio—within the context of standard clini-
`cal practice and the research protocol, risks must be minimized, potential
`benefits enhanced, and the potential benefits to individuals and knowledge
`gained for society must outweigh the risks; (5) independent review-
`unaffiliated individuals must review the research and approve, amend, or
`terminate it; (6) informed consent—individuals should be infonned about
`the research and provide their voluntary consent; and (7) respect for en-
`rollcd subjccts—subjccts should have their privacy protected, the opportu-
`nity to withdraw, and their well-being monitored. Fulfilling all 7 require-
`ments is necessary and sufficient to make clinical research ethical. These
`requirements are universal, although they must be adapted to the health,
`economic, cultural, and technological conditions in which clinical research
`is conducted.
`JAMA. 2000;283:2701-2711
`
`www.jama.com
`
`THE 7 ETHICAL
`REQUIREMENTS
`
`The overarching objective of clinical re-
`search is to develop generalizable
`knowledge to improve health and/or in-
`crease understanding of human biol-
`ogym‘ ; subjects who participate are the
`means to securing such knowledge.”
`By placing some people at risk of harm
`for the good of others, clinical re-
`search has the potential for exploita-
`tion of human subjects.“-“ Ethical re-
`quirements for clinical research aim to
`minimize the possibility of exploita-
`tion by ensuring that research sub-
`jects are not merely used but are treated
`with respect while they contribute to
`the social good.”
`
`For the past 50 years, the main sources
`of guidance on the ethical conduct of
`clinical research have been the Nurem-
`
`berg Code,“ Declaration of Helsinki,”
`Belmont Report,” International Ethical
`Guidelines for Biomedical Research In-
`
`volving Human Subjects,” and similar
`documents (TABLE 1). However, many
`of these documents were written in re-
`
`sponse to specific events and to avoid fu-
`ture scandals.5°5' By focusing on the in-
`stigating issues, these guidelines tend to
`
`Author Affiliations: Department of Clinical Bioeth-
`ics, Warren G. Magnuson Clinical Center, National In-
`stitutes of Health, Bethesda, Md.
`Corresponding Author and Reprints; Christine Grady,
`PhD, Warren G. Magnuson Clinical Center, Bldg 10,
`Room 1C118, National Institutes of Health, Bethesda.
`MD 20892-1156 (e-mail: cgrady@nih.gov).
`
`©2000 American Medical Association. All rights reserved.
`
`(Reprinted) JAMA, lsluy 2+/31, 2000—\"ol 283, No. 20 2701
`
`Page 1 of 11
`
`Biogen Exhibit 2060
`
`Coalition v. Biogen
`IPR2015-01993
`
`Page 1 of 11
`
`Biogen Exhibit 2060
`Coalition v. Biogen
`IPR2015-01993
`
`

`

`ETHICAL REQUIREMENTS FOR CLINICAL RESEARCH
`
`emphasize certain ethical requirements
`while eliding others. For instance, the
`Nuremberg Code" was part of thejudi-
`cial decision condemning the atrocities
`of the Nazi physicians and so focused on
`the need for consent and a favorable risk-
`bcucfit ratio but tualxcs uu uicutiuuuffaii
`
`subject selection or independent re-
`view. The Declaration of Helsinki” was
`
`developed to remedy perceived lacunae
`in the Nuremberg Code, especially as re-
`lated to physicians conducting research
`with patients, and so focuses on favor-
`able risk-benelit ratio and independent
`review; the Declaration of Helsinki also
`emphasizes a distinction between thera-
`
`peutic and nontherapeutic research that
`is rejected or not noted by other docu-
`ments.’‘’‘” The Belmont Report” was
`meant to provide broad principles that
`could be used to generate specific rules
`and regulations in response to US re-
`scar cli scaiidals such as Tuskegee“ and
`Willowbrook.5"-55 It focuses on in-
`
`formed consent, favorable risk-benefit ra-
`tio, and the need to ensure that vulner-
`able populations are not targeted for risky
`research. The Council for International
`
`Organizations of Medical Sciences
`(CIOMS) guidelines” were intended to
`apply the Declaration of Helsinki “in de-
`veloping countries .
`.
`. [particularly for]
`
`Table 1. Selected Guidelines on the Ethics of Biomedical Research With Human Subjects*
`Guideline
`Source
`Year and Revisions
`Fundamental
`
`Nuremberg Code”
`
`Declaration of Helsinki”
`
`Belmont Report”
`
`lntemational Ethical Guidelines for
`Biomedical Research Involving
`Human Subjects”
`
`45 CFR 46, Common Rule”
`
`Guidelines for Good Clinical
`Practice for Trials on
`Pharmaceutical Products“
`Good Clinical Practice:
`Consolidated Guidance“
`
`Nuremberg Military Tribunal
`decision in United States
`v Brandt
`World Medical Association
`
`National Commission for the
`Protection of Human Subjects
`of Biomedical and Behavioral
`Research
`
`Council for lntemational
`Organizations of Medical
`Sciences in collaboration with
`World Health Organization
`Other
`
`US Department of Health and
`Human Services (DHHS) and
`other US federal agencies
`Wortd Health Organization
`
`lntemational Conference on
`Harmonisation of Technical
`Requirements for Registration of
`Pharmaceuticals for Human Use
`
`Convention on Human Rights and
`Biomedicine“
`
`Council of Europe
`
`1947
`
`1964, 1975, 1983,
`1989, 1996
`1979
`
`Proposed in 1982;
`revised, 1093
`
`DHHS guidelines in
`1981 : Common
`Rule. 1991
`1995
`
`1996
`
`1997
`
`European Forum for Good
`Clinical Practice
`
`Medical Research Council.
`United Kingdom
`
`Uganda National Council for
`Science and Technology
`
`1997
`
`1998
`
`1998
`
`1998
`
`1999
`
`Guidelines and Recommendations
`for European Ethics
`Committees“
`Medical Research Council
`Guidelines for Good Clinical
`Practice in Clinical Trials“
`Guidelines for the Conduct of
`Health Research Involving
`Human Subjects in Uganda"
`Ethical Conduct for Research
`Involving Humans“
`National Statement on Ethical
`Conduct in Research Involving
`Humans‘°
`*CFR indicates Code 01 Federal Reguetions. More extensve lists of international guidelines on human subjects research
`can be found in Brody” and Flues.” An extensive summary of US guidelines can be lound in Sugarman et al."
`
`Tri—Council Working Group, Canada
`
`National Health and Medical
`Reswrch Council, Australia
`
`large-scale trials of vaccines and drugs.”
`The CIOMS guidelines lack a separate
`section devoted to risk-benelit ratios, al-
`
`though the council considers this issue
`in commentary on other guidelines. It
`also includes a section on compensa-
`liuu fut tcscaiclt lujutics uul fuuud in
`other documents. Because the Advisory
`Committee on Human Radiation Experi-
`ments was responding to covert radia-
`tion experiments, avoiding deception was
`among its 6 ethical standards and rules;
`most other major documents do not
`highlight this.” This advisory commit-
`tee claims that its ethical standards are
`
`general, but acknowledges that its
`choices were related to the specific cir-
`cumstances that occasioned the re-
`
`port.” Finally some tensions, if not
`outright contradictions, exist among
`the provisions of the various guide-
`1ineS_5.l9.30.5l.51.5758
`a universally aP_
`plicable ethical framework, investiga-
`tors, IRB members, funders, and others
`lack coherent guidance on determining
`whether specific clinical research pro-
`tocols are ethical.
`
`There are 7 requirements that pro-
`vide a systematic and coherent frame-
`wutk fut tlclciiititiiitg wlicllict Lliuiu:tl1t:-
`search is ethical (TABLE 2). These
`requirements are listed in chronologi-
`cal order from the conception of the re-
`search to its formulation and implemen-
`tation. They are meant to guide the
`ethical development, implementation,
`and review of individual clinical proto-
`cols. These 7 requirements are in-
`tended to elucidate the ethical stan-
`
`dards specific for clinical research and
`assume general ethical obligations, such
`as intellectual honesty and responsibil-
`ity. While none of the traditional ethi-
`cal guidelines on clinical research ex-
`plicitly includes all 7 requirements, these
`requirements systematically elucidate the
`fundamental protections embedded in
`the basic philosophy of all these docu-
`ments.” These requirements are not lim-
`ited to a specific tragedy or scandal or to
`the practices of researchers in l coun-
`try; they are meant to be universal, al-
`though Llicii application will tcquiit: ad-
`aptation to particular cultures, health
`conditions, and economic settings. These
`
`2702 JAMA, May 24/31, 2000—Vol 283, .\lo. 20 (Reprinted)
`
`©2000 American Medical Association. All rights reserved.
`
`Page 2 of 11
`
`Page 2 of 11
`
`

`

`ETHICAL REQUIREMENTS FOR CLINICAL RESEARCH
`
`7 requirements can be implemented well
`or ineffectively. However, their system-
`atic delineation is important and con-
`ceptually prior to the operation of an en-
`forcement mechanism. We need to know
`what to enforce.
`
`Value
`
`To be ethical, clinical research must be
`
`valuable,”5 meaning that it evaluates
`a diagnostic or therapeutic interven-
`tion that could lead to improvements
`in health or well-being; is a prelimi-
`nary etiological, pathophysiological, or
`epidemiological study to develop such
`an intervention; or tests a hypothesis
`that can generate important knowl-
`edge about structure or function of hu-
`man biological systems, even if that
`knowledge does not have immediate
`practical ramifications.*‘3° Examples of
`research that would not be socially or
`
`scientifically valuable include clinical
`research with nongeneralizable re-
`sults, a trifling hypothesis, or substan-
`tial or total overlap with proven re-
`sults.‘ In addition, research with results
`unlikely to be disseminated or in which
`the iutctvcutiuu cuultl ucvci be prac-
`tically implemented even if effective is
`not valuable.”*""“-5° Only if society will
`gain knowledge, which requires shar-
`ing results, whether positive or nega-
`tive, can exposing human subjects to
`risk in clinical research be justified.
`Thus, evaluation of clinical research
`should ensure that the results will be
`
`disseminated, although publication in
`peer-reviewed journals need not be the
`primary or only mechanism.
`There are 2 fundamental reasons why
`social, scientific, or clinical value should
`be an ethical requirement: responsible
`use of finite resources and avoidance of
`
`exploitation.‘ Research resources are lim-
`ited. Even if major funding agencies
`could fund all applications for clinical
`research, doing so would divert resources
`from other worthy social pursuits.
`Beyond not wasting resources, research-
`ct: sliuultl nut cxpusc liutuau lyciugs tu
`potential harms without some possible
`social or scientific benetit.*"°'”"“
`
`It is possible to compare the relative
`value of different clinical research stud-
`
`ies; clinical research that is likely to gen-
`erate greater improvements in health or
`well-being given the condition being
`investigated, the state of scientific
`understanding, and the feasibility of
`implementing the intervention is of
`higher value. Comparing relative value
`is integral to determinations of fund-
`ing priorities when allocating limited
`funds among alternative research pro-
`posals.“° Similarly, a comparative evalu-
`
`Table 2. Seven Requirements for Determining Whether a Research Trial Is Ethical‘
`Requirement
`Bcplanation
`Justifying Ethical Values
`Social or scientific value
`Scarce resources and
`Evaluation of a treatment. intervention,
`or theory that will improve health and
`nonexploitation
`well-being or increase knowledge
`Use of accepted scientific principles
`and methods, inrzluding statistical
`techniques, to produce reliable
`and valid data
`
`Scarce resources and
`nonexploitation
`
`Scientific validity
`
`Justice
`
`Fair subject selection
`
`Favorable risk-benefit
`ratio
`
`Independent review
`
`Selection of subjects so that stigmatized
`and vulnerable individuals are not
`targeted for risky research and the
`rich and socially powerful not favored
`for potentially beneficial research
`Minimization of risks; enhancement of
`potential benefits; risks to the subject
`are proportionate to the benefits to
`the subject and society
`Review of the design of the research
`trial, its proposed subject population,
`and risk-benefit ratio by individuals
`unaffiliated with the research
`
`Informed consent
`
`Provision of intonnation to subjects
`about purpose of the research, its
`procedures, potential risks, benefits,
`and altematives. so that the
`individual understands this
`information and can make a
`voluntary decision whether to
`enrol and continue to participate
`Respect for subjects by
`(1) permitting withdrawal from the
`research:
`(2) protecting privacy through
`confidentiality;
`informing subjects of newly
`discovered risks or benefits;
`informing st rhjects nf rest tits nf
`clinical research;
`(5) maintaining welfare of subjects
`*Ethica| requirements are listed in chronological order from conception of research to its lormtlation and irplementation.
`
`Respect for potential and
`enrolled subjects
`
`Respect for subject autonomy
`and welfare
`
`(3)
`
`(A)
`
`Nonmaleficence, benefioence,
`and nonexploitation
`
`Scientific knowledge; citizen's
`understanding of social values
`
`Public accountability; minimizing
`influence of potential conflicts
`of interest
`
`Respect for subject autonomy
`
`Expertise for Evaluation
`Scientific knowledge: citizen's
`understanding of social
`priorities
`Scientific and statistical
`knowledge: knowledge of
`condition and population to
`assess feasibility
`Scientific knowledge; ethical and
`legal knowledge
`
`Intellectual. financial, and
`otherwise independent
`researchers; scientific and
`ethical knowledge
`Scientific knowledge; ethical and
`legal knowledge
`
`Scientific knowledge; ethical and
`legal knowledge: knowledge of
`particular subject population
`
`©2000 American Medical Association. All rights reserved.
`
`(Reprinted) JAMA. May 24/31, 2000-)/ol 283, No. 20 2703
`
`Page 3 of 11
`
`Page 3 of 11
`
`

`

`ETHICAL REQUIREMENTS FOR CLINICAL RESEARCH
`
`ation of value may be necessary in
`considering studies involving finite sci-
`entific resources such as limited bio-
`
`logical material or the small pool of
`long-terrn human immunodeficiency
`virus nonprogressors.
`
`Scientific Validity
`To be ethical, valuable research must
`
`be conducted in a methodologically rig-
`orous manner.‘ Even research asking
`socially valuable questions can be de-
`signed or conducted poorly and pro-
`duce scientifically unreliable or in-
`valid results.“ As the CIOMS guidelines
`succinctly state: “Scientifically un-
`sound research on human subjects is
`ipso facto unethical in that it may ex-
`pose subjects to risks or inconve-
`nience to no purpose.”’“
`For a clinical research protocol to be
`ethical, the methods must be valid and
`
`practically feasible: the research must
`have a clear scientific objective; be de-
`signed using accepted principles, meth-
`ods, and reliable practices; have suffi-
`cient power to definitively test the
`objective; and offer a plausible data
`analysis plan.‘ In addition, it must be
`pusaible to execute the proposed Study.
`Research that uses biased samples, ques-
`tions. or statistical evaluations, that is un-
`
`derpowered, that neglects critical end
`points, or that could not possibly en-
`roll sufficient subjects cannot generate
`valid scientific knowledge and is thus
`unethical.""°"’l For example, research
`with too few subjects is not valid be-
`cause it might be combined in a mean-
`ingful meta-analysis vtn'th other, as yet
`unplanned and unperformed clinical re-
`search; the ethics of a clinical research
`study cannot depend on the research
`that others might but have not yet done.
`Of course the development and ap-
`proval of a valid method is of little use
`if the research is conducted in a sloppy
`or inaccurate manner; careless re-
`search that produces uninterpretable
`data is not just a waste of time and re-
`sources, it is unethical.
`Clinical research that compares thera-
`pies rrruat lrave “arr Irurrest rrull lryputlr-
`esis” or what Freedman called clinical
`
`equipoise.’°"" That is, there must be con-
`
`troversy within the scientific commu-
`nity about whether the new interven-
`tion is better than standard therapy,
`including placebo, either because most
`clinicians and researchers are uncertain
`about whether the new treatment is bet-
`ter, ur because >UlllC believe the starr-
`
`dard therapy is better while others be-
`lieve the investigational intervention
`superior.“ lf there exists a consensus
`about what is the better treatment, there
`is no null hypothesis, and the research
`is invalid. In addition, without clinical
`equipoise, research that compares thera-
`pies is unlikely to be of value because the
`research will not contribute to increas-
`
`ing knowledge about the best therapy,
`and the risk-benefit ratio is unlikely to
`be favorable because sorrre of the sub-
`
`jects will receive inferior treatment.
`Importantly, a “good question" can
`be approached by good or bad re-
`search techniques; bad research meth-
`ods do not render the question value-
`less. Thus, the significance of a
`hypothesis can and should be as-
`sessed prior to and independent of the
`specific research methods. Reviewers
`should not dismiss a proposal that uses
`inadequate rrretlrutls without first curr-
`sidering whether adjustments could
`make the proposal scientifically valid.
`The justification of validity as an ethi-
`cal requirement relies on the same 2
`principles that apply to value-
`limited resources and the avoidance of
`
`exploitation.’“” “Invalid research is un-
`ethical because it is a waste of re-
`
`sources as well: of the investigator, the
`funding agency, and anyone who at-
`tends to the research.” Without valid-
`
`ity the research cannot generate the in-
`tended knowledge, cannot produce any
`benefit, and cannot justify exposing
`subjects to burdens or risks.”
`
`Fair Subject Selection
`
`The selection of subjects must be
`fair.’‘’'"'’‘‘ Subject selection encom-
`passes decisions about who will be in-
`cluded both through the development
`of specific inclusion and exclusion
`criteria and the strategy adapted for
`recruiting subjects, such as which
`communities will be study sites and
`
`which potential groups will be ap-
`proached. There are several facets to this
`requirement.
`First, fair subject selection requires
`that the scientific goals of the study, not
`vulnerability, privilege, or other fac-
`tors urrrelartetl tu the purposes uf the re-
`search, be the primary basis for deter-
`mining the groups and individuals that
`will be recruited and enrolled."’°"7 In
`
`the past, groups sometimes were en-
`rolled, especially for research that en-
`tailed risks or offered no potential ben-
`efits, because they were “convenient”
`or compromised in their ability to pro-
`tect themselves, even though people
`from less vulnerable groups could have
`met the scientific requirements of the
`Study‘}0.37.53.S-l
`Similarly, groups or individuals should
`not be excluded from the opportunity to
`participate in research without a good sci-
`entific reason or susceptibility to risk that
`justifies their exclusion.“ It is impor-
`tant that the results of research be gen-
`eralizable to the populations that will use
`the intervention. Efficiency cannot over-
`ride fairness in recruitingsubjects.” Fair-
`ness requires that women be included in
`the research, urrIe>> there is guutl rea-
`son, such as excessive risks, to exclude
`them.""“’ This does not mean that ev-
`
`ery woman must be offered the oppor-
`ttrnity to participate in research, but it
`does mean that women as a class can-
`
`not be peremptorily excluded.
`Second, it is important to recognize
`that subject selection can affect the risks
`and benefits of the study.” Consistent
`with the scientific goals, subjects should
`be selected to minimize risks and en-
`
`hance benefits to individual subjects
`and society. Subjects who are eligible
`based on the scientific objectives of a
`study, but are at substantially higher
`risk of being harmed or experiencing
`more severe harm, should be ex-
`cluded from participation." Selecting
`subjects to enhance benefits entails con-
`sideration of which subjects will maxi-
`mize the benefit or value of the infor-
`
`mation obtained. If a potential drug or
`procedure i> likely to be prestr ibetl for
`women or children if proven safe and
`effective, then these groups should be
`
`2704 JAMA, May 24/31, 2000—V'ol 283, No. 20 (Reprinted)
`
`©2000 American Medical Association. All rights reserved.
`
`Page 4 of 11
`
`Page 4 of 11
`
`

`

`ETHICAL REQUIREMENTS FOR CLINICAL RESEARCH
`
`included in the study to learn how the
`drug affects them."""""’7 Indeed, part of
`the rationale lor recent initiatives to in-
`
`clude more women, minorities, and
`children in clinical research is to maxi-
`
`mize the benefits and value of the study
`by ensuring llliil l1lC3C EIUUPD alt C11‘
`rolIed."5"7-7"” It is not necessary to in-
`clude children in all phases of re-
`search. lnstead, it may be appropriate
`to include them only after the safety of
`the drug has been assessed in adults.
`Additionally, fair subject selection re-
`quires that, as far as possible, groups
`and individuals who bear the risks and
`
`burdens of research should be in a po-
`sition to enjoy its benefits,"-'3-3"-5°-7* and
`those who may benefit should share
`some of the risks and burdens.” Groups
`recruited to participate in clinical re-
`search that involves a condition to
`
`which they are susceptible or from
`which they suffer are usually in a po-
`sition to benefit if the research pro-
`vides a positive result, such as a new
`treatment. For instance, selection of
`
`subjects for a study to test the efficacy
`of an antimalarial vaccine should con-
`
`sider not only who will best answer the
`scientific question, but alsu whethet the
`selected groups will receive the ben-
`efits of the vaccine, if proven effec-
`tive.”-"'37'5°~“-7° Groups of subjects who
`will predictably be exclttded as benefi-
`ciaries of research results that are rel-
`
`evant to them typically should not as-
`sume the burdens so that others can
`
`benefit. However, this does not pre-
`clude the inclusion of subjects who are
`scientifically important for a study but
`for whom the potential products of the
`research may not be relevant, such as
`healthy control subjects.
`Fair subject selection should be
`guided by the scientific aims of the re-
`search and is justified by the prin-
`ciples that equals should be treated
`similarly and that both the benefits and
`burdens generated by social coopera-
`tion and activities such as clinical
`research should be distributed
`
`fairly. "”*”"““"“"’ This does not mean that
`itnlividual subjects and tnentbets uf
`groups from which they are selected
`must directly benefit from each clini-
`
`cal research project or that people who
`are marginalized, stigmatized, power-
`less, or poor should never be in-
`cluded. Instead, the essence of fair-
`
`ness in human subjects research is that
`scientific goals, considered in dy-
`natnic interaction with the putential fut
`and distribution of risks and benefits,
`should guide the selection of subjects.
`
`Favorable Risk-Benefit Ratio
`
`Clinical research involves drugs, de-
`vices, and procedures about which there
`is limited knowledge. As a result, re-
`search inherently entails uncertainty
`about the degree of risk and benefits,
`with earlier phase research having
`greater uncertainty. Clinical research
`can be justified only if, consistent with
`the scientific aims of the study and the
`relevant standards of clinical practice,
`3 conditions are fulfilled: the poten-
`tial risks to individual subjects are mini-
`mized, the potential benefits to indi-
`vidual subjects are enhanced, and the
`potential benefits to individual sub-
`jects and society are proportionate to
`or outweigh the risks.3°v3°-37
`Assessment of the potential risks and
`benefits ofclinical research by research-
`ers and review bodies typically in-
`volves multiple steps. First, risks are
`identified and, within the context of
`good clinical practice, minimized “by
`using procedures which are consis-
`tent with sound research design and
`which do not unnecessarily expose sub-
`jects to risk, and whenever appropri-
`ate, by using procedures already being
`performed on the subjects for diagnos-
`tic or treatment purposes.”
`Second, potential benefits to indi-
`vidual subjects from the research are de-
`lineated and enhanced. Potential ben-
`efits focus on the benefits to individual
`
`subjects, such as health improvements,
`because the benefits to society through
`the generation of knowledge are as-
`sumed if the research is deemed to be of
`
`value and valid. The specification and en-
`hancement of potential benefits to indi-
`vidual subjects should consider only
`ltealth-telatetl potential benefits de-
`rived from the research." Assessment of
`
`the research plan should determine if
`
`changes could enhance the potential ben-
`efits for individual subjects. For ex-
`ample, consistent with the scientilic ob-
`jectives, tests and interventions should
`be arranged to increase benefit to sub-
`jects. However, extraneous benefits, such
`ats paytnent, Ul adjunetive lllCl.liLul set-
`vices, such as the possibility of receiv-
`ing a hepatitis vaccine not related to the
`research, cannot be considered in delin-
`eating the benefits compared with the
`risks, otherwise simply increasing pay-
`ment or adding more unrelated ser-
`vices could make the benefits outweigh
`even the riskiest research. Further-
`
`more, while participants in clinical re-
`search may receive some health ser-
`vices and benefits, the purpose of clinical
`research is not the provision of health ser-
`vices. Services directly related to clini-
`cal research are necessary to ensure sci-
`entific validity and to protect the well-
`being of the individual subjects.
`In the final step, risks and potential
`benefits of the clinical research inter-
`
`ventions to individual subjects are com-
`pared. In general, the more likely and/or
`severe the potential risks the greater in
`likelihood and/or magnitude the pro-
`spective benefits must be, conversely,
`research entailing potential risks that
`are less likely and/or of lower severity
`can have more uncertain and/or cir-
`
`cumscribed potential benefits. If the po-
`tential benefits to subjects are propor-
`tional to the risks they face, as generally
`found when evaluating phase 2 and 3
`research, then the additional social ben-
`efits of the research, assured by the ful-
`fillment of the value and validity re-
`quirements, imply that the cumulative
`benefits of the research outweigh its
`risks.”
`
`Obviously, the notions of “propor-
`tionality“ and potential benefits “out-
`weighing" risks are nonquantifiable.”
`However, the absence of a formula to
`determine when the balance of risks and
`
`potential benefits is proportionate does
`not connote that such judgments arcin-
`herently haphazard or subjective. In-
`stead, assessments of risks and poten-
`tial benefits to the saline inclivitluals can
`
`appeal to explicit standards, informed
`by existing data on the potential types
`
`©2000 American Medical Association. All rights reserved.
`
`(Reprinted) JAMA, May 24/31, 2000—Vol 283, No. 20 2705
`
`Page 5 of 11
`
`Page 5 of 11
`
`

`

`ETHICAL REQUIREMENTS FOR CLINICAL RESEARCH
`
`of harms and benefits, their likelihood
`
`of occurring, and their long—term con-
`sequences.” People routinely make dis-
`cursively justifiable intrapersonal com-
`parisons of risks and benefits for
`themselves and even for others, such
`as children, friends, and employees,
`without the aid of mathematical for-
`mulae.”
`An additional evaluation is neces-
`
`sary for any clinical research that pre-
`sents no potential benefits to indi-
`vidual subjects, such as phase 1 safety,
`pharmacokinetic, and even some epi-
`demiology research, or when the risks
`outweigh the potential benefits to indi-
`vidual subjects." This determination,
`which Weijer7" calls a “risk-knowledge
`calculus,” assesses whether the societal
`
`benefits in temis of knowledge justify the
`excess risks to individual subjects. De-
`termination of when potential social ben-
`efits outweigh risks to individual sub-
`jects requires interpersonal comparisons
`that are conceptually and practically
`more dillicult.” However, policymak-
`ers often are required to make these kind
`of comparisons, for example when con-
`sidering whether pollution and its at-
`tertclant lrarrns tu surrre peuple are wur tlr
`the potential benefits of higher employ-
`ment and tax revenues to others. There
`
`is no settled framework for how poten-
`tial social benelits should be balanced
`
`against individual risks. Indeed, the ap-
`peal to a utilitarian approach of maxi-
`mization, as in cost-benefit analysis, is
`quite controversial both morally and be-
`cause many risks and benefits of re-
`search are not readily quantifiable on
`commensurable scales."“‘” Neverthe-
`
`less, these comparisons are made,” and
`regulations mandate that investigators
`and 1RBs make them with respect to
`clinical research. When research risks
`
`exceed potential medical benefits to in-
`dividuals and the benefit of useful
`
`knowledge to society, the clinical re-
`search is not justifiable.
`The requirement for a favorable risk-
`benefit ratio embodies the principles of
`nonmaleficence and beneficence, long
`r ecugnizecl as furrtlarrrerrtal values of
`clinical research."’°"""7 The principle of
`nonmaleficence states that one ought not
`
`to inflict hann on a person.’ This justi-
`fies the need to reasonably reduce the
`risks associated with research. The prin-
`ciple ofbenelicence “refers to a moral ob-
`ligation to act for the benefit of oth-
`ers.’‘’ in clinical research, this translates
`irrtu the need tu enhance the potential
`benefits of the research for both indi-
`
`vidual subjects and society."’°'" Ensur-
`ing that the benefits outweigh the risks
`is required by the need to avoid the ex-
`ploitation of subjects. "W
`
`Independent Review
`
`investigators inherently have mul-
`tiple, legitimate interests—interests to
`conduct high-quality research, com-
`plete the research expeditiously, pro-
`tect research subjects, obtain funding,
`and advance their careers. These di-
`
`verse interests can generate conflicts
`that may unwittingly distort the judg-
`ment of even well-intentioned investi-
`
`gators regarding the design, conduct,
`and analysis of research.‘’”‘7 Wanting
`to complete a study quickly may lead
`to the use of questionable scientific
`methods or readily available rather than
`the most appropriate subjects. inde-
`perrclent review by irrclivitluals urraffili-
`ated with the clinical research helps
`minimize the potential impact of such
`conflicts of interest.8°~“8 For some re-
`
`search with few or no risks, indepen-
`dent review may be expedited, but for
`much of clinical research, review should
`be done by a full committee of indi-
`viduals with a range of expertise who
`have the authority to approve, amend,
`or terminate a study.
`independent review of clinical re-
`search is also important for social ac-
`count