Return to Index Page

Institutional Review Board


A. IntroductionD. Medical Devices
Biomedical ResearchE. Use of Radioactive Materials and X-Rays
Behavioral Research F. AIDS/HIV-Related Research
FieldworkG. Transplants
Social Policy ExperimentationH. Human Genetic Research
B. Drug TrialsI. Alcohol and Drug Research
C. Vaccine TrialsSuggestions for Further Reading


Most of the research reviewed by IRBs falls within the broad categories of biomedical or behavioral research. IRBs should be sensitive to specific aspects of biomedical and behavioral research in their review of protocols.

Biomedical research includes both studies designed primarily to increase the scientific base of information about normal or abnormal physiology and development and studies primarily intended to evaluate the safety, effectiveness or usefulness of a medical product, procedure, or intervention. The terms "behavioral research" or "the behavioral sciences" may be used to refer either to studies of the behavior of individuals, or to studies of the behavior of aggregates such as groups, organizations, or societies. The broad objective of the behavioral and social sciences is similar to that of the biomedical sciences: to establish a body of demonstrable, replicable facts and theory that contributes to knowledge and to the amelioration of human problems.

It is neither possible nor necessary to draw a clean line between biomedical and behavioral research. Some biomedical research pertains to behavior (e.g., in psychiatry, neurology, or epidemiology), and many of the methods used in behavioral research, such as observation and the questioning of subjects, are also used in biomedical research. Research may be designed to evaluate the behavioral changes that result from a biomedical intervention (e.g., lessening of depression after taking a particular medication or changes in psychiatric disorders following hemodialysis) or to examine physiological responses to behavioral interventions (e.g., lowering of blood pressure through biofeedback or weight loss through hypnosis). Some studies involve functions that are not easily defined as either behavioral or physiological (e.g., sleep, exercise, or diet). Thus, although it is sometimes useful to refer to biomedical or behavioral and social research as if they involve distinct activities, there is considerable overlap among the three areas. (While the use of such terms as "behavioral and social research" may imply that social research is distinct from behavioral research, this distinction generally has little utility for the work of IRBs and is not applied here.) The questions that are of concern to IRBs stem not from the label attached to the research but from the nature of the interventions and the characteristics of the subjects in any given study. It is for this reason that institutions and federal agencies are concerned that IRB members be knowledgeable about the various types of research reviewed by that IRB.

Return to Top of Page


Biomedical research employs many methods and research designs. Studies designed to evaluate the safety, effectiveness, or usefulness of an intervention include research on therapies (e.g., drugs, diet, exercise, surgical interventions, or medical devices), diagnostic procedures (e.g., CAT scans or prenatal diagnosis through amniocentesis, chorionic villi testing, and fetoscopy), and preventive measures (e.g., vaccines, diet, or fluoridated toothpaste). Research on normal human functioning and development can include studies of the human body while exercising, fasting, feeding, sleeping, or learning, or responding to such things as stress or sensory stimulation. Some studies compare the functioning of a particular physiological system at different stages of development (e.g., infancy, childhood, adolescence, adulthood, or old age). Others are directed at defining normal childhood development so that deviations from normal can be identified. Sometimes research, particularly records research, is used to develop and refine hypotheses. Research on specific disease processes is often needed before improved methods of prevention, diagnoses, and treatment can be developed (e.g., research on the biochemical changes associated with AIDS or schizophrenia, or the neurological changes associated with senile dementia of the Alzheimer type). Research on the human genome and genetic markers is expected to create new avenues for understanding disease processes and their eventual control.

Subjects of some biomedical studies engage in ordinary tasks (e.g., exercise, learn a series of words, or respond to various sensory stimuli) while measurements of physiological and bodily functions are made. Although many procedures used in biomedical research are similar to those used in routine physical examinations, at times more invasive procedures (e.g., "spinal taps," skin or muscle biopsies, or X-rays used in conjunction with contrast dyes) must be used if a desired measurement is to be made. Although research designed to generate information about normal physiology or a disease process is not concerned with evaluating a medical intervention, it may still require the use of invasive procedures. When the research deals with subjects whose condition is not normal, the research can have either therapeutic or nontherapeutic purposes.

Other biomedical studies do not involve human subjects or are exempt from the human subjects regulations, and, therefore, do not require IRB review. This category includes research with animals and research on preexisting samples of materials (tissue, blood, or urine) collected for other purposes, where the information is recorded by the investigator in such a manner that subjects cannot be identified, directly or through identifiers linked to the subjects [Federal Policy ___.101(b)(4)]. It also includes research based on records, when the data are recorded in such a manner that the individuals to whom the records pertain cannot be identified, either directly or through identifiers linked to them [Federal Policy ___.101(b)(4)]. [See Guidebook Chapter 4, "Considerations of Research Design."]

Some biomedical studies, particularly those conducted to evaluate new therapies or treatments, use such rigorous experimental methods as random assignment to treatment and control groups. Other studies, such as those directed at establishing the normal range of some element in the blood, may involve no experimental intervention and no assignment of subjects to groups. [See Guidebook Chapter 4, "Considerations of Research Design."]

The fact that much biomedical research is conducted for the purpose of evaluating new therapies or treatments leads to two problems for IRBs. The first is to some degree a problem of IRB jurisdiction; the second is a problem of risk/benefit assessment. The distinction between research and treatment can become blurred in patient care settings, as well as in some educational and training settings. This distinction raises questions of IRB jurisdiction over the research: Is the proposed activity one that requires IRB review (pursuant either to federal regulations or institutional policy)? A discussion of this issue appears in the Guidebook in Chapter 1, Section A, "Jurisdiction of the Institutional Review Board."

The second distinction between research and therapies that may pose a problem for IRBs concerns risk/benefit assessments in research on therapies. Often, the risks of a study may seem justified by a therapy provided as part of the study. IRBs should determine, however, whether the anticipated therapeutic benefits would be available to persons who are not participating in a study that presents additional risks. As is discussed in the Guidebook Section on risk/benefit analysis [Chapter 3, Section A], such benefits should not be used to justify risks presented by the research.

The IRB's general responsibilities in reviewing biomedical research are discussed in other chapters of the Guidebook. [See Chapter 3, "Basic IRB Review," and Chapter 4, "Considerations of Research Design."] Special concerns arising in the conduct of certain types of biomedical research are discussed in the following Sections of this chapter on "Drug Trials," "Vaccine Trials," "Medical Devices," "Use of Radioactive Materials and X-Rays," "HIV-Related Research," "Transplants," "Human Genetic Research," and "Alcohol and Drug Research." The additional IRB responsibilities that arise when the subjects of biomedical research are other than healthy, normal adults are set forth in Chapter 6, "Special Classes of Subjects."

Return to Top of Page


The scope and diversity of research areas in the behavioral and social sciences is quite broad. Some research is readily applicable to human affairs; other studies may broaden understanding without any apparent or immediate application. Some research is designed to test hypotheses derived from theory; other research is primarily descriptive. Still other research may be directed at evaluating an intervention or social program.

Theories and methods vary both across and within disciplines; the same problems may be approached by researchers trained in different disciplines. For example, some research psychologists work in laboratories studying the neurology, anatomy, and physiology that underlies perception, learning, instinctual behavior, and emotional responses. Other psychologists may perform survey research, observational studies, or small group experiments that differ little from work done by some sociologists. Within anthropology, physical anthropology overlaps with paleontology, anatomy, and genetics, while the social or cultural anthropologist studies the organization, institutions, and belief and value systems of societies or groups of people.

Behavioral research involving human subjects generates data by means of questionnaires, observation, studies of existing records, and experimental designs involving exposure to some type of stimulus or intervention. Many variations of these four basic methods are used. Questions may be asked in person, over the telephone, or by means of a questionnaire. Observation may or may not be covert, and the observer may or may not be a participant in the activity being studied. Records studied in research may be public (e.g., vital statistics, motor vehicle registrations, or court records) or non-public and sensitive (e.g., medical or educational records in which the subjects are identified). Experimental studies may be conducted in public places, in private settings (e.g., a clinic or therapist's office), or in laboratories. Interventions in such studies range from the innocuous, such as varying the package design of commercial products, to the potentially significant, such as varying behavior modification techniques in studying the treatment of alcoholism. Not all behavioral research involves human subjects. Studies of human migration are often undertaken using anonymous U.S. Census data, and much research in behavioral psychology is done with animals. In addition, many categories of behavioral research that do involve human subjects are exempt from the federal regulations for protection of human subjects. [See Federal Policy ___.101.] This exemption does not imply that investigators have no ethical responsibilities to subjects in such research; it means only that IRB review and approval of the research is not required by federal regulations.

Most behavioral research involves no physical intervention and no physical risk. However, some studies do present a risk of social harm (e.g., harm to a subject's reputation, which is sometimes a danger if confidentiality is not maintained) or psychological harm, which may occur if the research involves deception or provides subjects with unwelcome and disturbing information about themselves. When deception is involved, the IRB needs to be satisfied that the deception is necessary and that, when appropriate, the subjects will be debriefed. (Debriefing may be inappropriate, for example, when the debriefing itself may present an unreasonable risk of harm without a counterveiling benefit.) The IRB should also make sure that the proposed subject population is suitable. [See Guidebook Chapter 3, Section A, "Risk/Benefit Analysis."]

Some studies involve the possibility of a moral wrong, which is what some commentators have labeled the ethical problems posed by deception of subjects or invasions of their privacy. Although some psychologists have overemphasized the value and necessity of using deception, deception or incomplete disclosure may be the only scientifically valid approach for certain research. An example of such research would be a study designed to determine the effect of group pressure (i.e., responses of others) on a subject's estimate of the length of a series of lines. In some groups, pseudo-subjects would be told in advance to give incorrect answers to questions about the length of the lines to determine the effect of such misinformation on the real subjects' responses. Obviously, if the subjects were told all about the research design and its purpose in advance, it would not be possible to do the research. IRBs need to determine whether any deception or invasion of privacy involved in a research protocol is justified.

Some social and behavioral researchers are concerned that IRB judgments at times seem to be influenced more by the subject matter of the study than by concerns about informed consent or risks to subjects. Researchers cite examples of studies that involve minimal risk and pose no consent questions, but that encounter difficult with some IRBs, particularly IRBs in medical settings. Some researchers believe that IRBs are more likely to object to research on the behavior or values of the powerful (e.g., physicians, professors, or managers) than to research using similar methods but on subjects of lower status (e.g., patients, students, or workers). Other researchers believe that IRBs sometimes perceive research on controversial topics, such as deviant sexual behavior or fraud in science, as presenting ethical problems because of the nature of the activity being studied, rather than because of research methods, risks, or the rights of subjects. Still others complain of a less specific prejudice against social and behavioral research on the grounds that it is "soft" or concerned with trivial questions.

Some behavioral research involves human subjects in studies of heredity and human behavior, genetics, race and IQ, psychobiology, or sociobiology. Vigorous ethical debates about these studies arise out of the fear that scientific data may be used to justify social stratification and prejudice, or that certain groups will appear to be genetically inferior. The possible use or misuse of research findings, however, should not be a matter for IRB review, despite the importance of this question.

The incidence of such problems may well have decreased because the regulations exempt much social research and provide additional flexibility regarding informed consent. IRBs should resist placing restrictions on research because of its subject matter; IRBs should instead be concerned about research methods and the rights and welfare of research subjects. IRBs must differentiate disapproving a research proposal because of qualms about the subject being explored or its possible findings, such as genetic differences in intelligence, from disapproving research involving the performance of illegal or unethical acts. The former raises serious issues of academic freedom; the latter is quite different and appropriate. Whatever the propriety of institutional administrators prohibiting research to protect the institutions from being associated with controversial or sensitive subjects, it is generally agreed that this is not an appropriate concern for an IRB, whose function is to protect human subjects.

Return to Top of Page


Fieldwork, or ethnographic research, involves observation of and interaction with the persons or group being studied in the group's own environment, often for long periods of time. Since fieldwork is a research process that gains shape and substance as the study progresses, it is difficult, if not impossible, to specify detailed contents and objectives in a protocol.

After gaining access to the fieldwork setting, the ongoing demands of scientifically and morally sound research involve gaining the approval and trust of the persons being studied. These processes, as well as the research itself, involve complex, continuing interactions between researcher and hosts that cannot be reduced to an informed consent form. Thus, while the idea of consent is not inapplicable in fieldwork, IRBs and researchers need to adapt prevailing notions of acceptable protocols and consent procedures to the realities of fieldwork. IRBs should keep in mind the possibility of granting a waiver of informed consent.

Return to Top of Page


Social policy experimentation involves interventions in social or economic systems for use in planning public policy. Such experimentation often involves studying the costs and benefits of alternative ways of providing health, educational, or welfare services at national, state, or local levels. Some of this research may be exempt from IRB review under ___.101(b)(5) of the Federal Policy. That section exempts research and demonstration projects that are conducted by or subject to the approval of department or agency heads, and that are designed to study, evaluate, or otherwise examine: (1) public benefit or service programs; (2) procedures for obtaining benefits or services under those programs; (3) possible changes in or alternatives to those programs or procedures; or (4) possible changes in methods or levels of payment for benefits or services under those programs.

Return to Top of Page



Drug trials provide the transition from promising basic or laboratory research to helpful therapeutic or diagnostic procedures for patients. New drugs that offer the hope of some beneficial response in afflicted patients are first tested in animal models. But animal trials do not necessarily demonstrate what the physiological, pharmacological, or toxicological effects of a new drug will be in human beings. Only by careful testing in human subjects can the safety and effectiveness of a new drug be evaluated. The Food and Drug Administration (FDA) is responsible for monitoring the testing of new drugs in humans, for determining whether a new drug can be marketed, and for observing drugs after marketing to be sure that they are safe, effective, and properly labeled [21 CFR 312 and 21 CFR 314].

See also Guidebook Chapter 4, Section H, "Clinical Trials," and Section J, "Assignment of Subjects to Experimental and Control Groups."


Clinical Trial: A controlled study involving human subjects, designed to evaluate prospectively the safety and effectiveness of new drugs or devices or of behavioral interventions.

Drug: Any chemical compound that may be used on or administered to humans as an aid in the diagnosis, treatment, cure, mitigation, or prevention of disease or other abnormal conditions.

Investigational New Drug or Device: A drug or device permitted by FDA to be tested in humans, but not yet determined to be safe and effective for a particular use in the general population, and not yet licensed for marketing.

Investigator: In clinical trials, an individual who actually conducts an investigation [21 CFR 312.3]. Any interventions (e.g., drugs) involved in the study are administered to subjects under the immediate direction of the investigator. (See also: Principal Investigator.)

Phase 1, 2, 3, 4 Drug Trials: Different stages of testing drugs in human, from first application in humans (Phase 1) through limited and broad clinical tests (Phase 3), to postmarketing studies (Phase 4).

Phase 1 Drug Trial: Phase 1 trials include the initial introduction of an investigational new drug into humans. These studies are typically conducted with healthy volunteers; sometimes, where the drug is intended for use in patients with a particular disease, however, such patients may participate as subjects. Phase 1 trials are designed to determine the metabolic and pharmacological actions of the drug in humans, the side effects associated with increasing doses (to establish a safe dose range), and, if possible, to gain early evidence of effectiveness; they are typically closely monitored. The ultimate goal of Phase 1 trials is to obtain sufficient information about the drug's pharmacokinetics and pharmacological effects to permit the design of well-controlled, sufficiently valid Phase 2 studies. Other examples of Phase 1 studies include studies of drug metabolism, structure-activity relationships, and mechanisms of actions in humans, as well as studies in which investigational drugs are used as research tools to explore biological phenomena or disease processes. The total number of subjects involved in Phase 1 investigations is generally in the range of 20-80.

Phase 2 Drug Trial: Phase 2 trials include controlled clinical studies conducted to evaluate the drug's effectiveness for a particular indication in patients with the disease or condition under study, and to determine the common short-term side effects and risks associated with the drug. These studies are typically well-controlled, closely monitored, and conducted with a relatively small number of patients, usually involving no more than several hundred subjects.

Phase 3 Drug Trial: Phase 3 trials involve the administration of a new drug to a larger number of patients in different clinical settings to determine its safety, effectiveness, and appropriate dosage. They are performed after preliminary evidence of effectiveness has been obtained, and are intended to gather necessary additional information about effectiveness and safety for evaluating the overall benefit-risk relationship of the drug, and to provide an adequate basis for physician labeling. In Phase 3 studies, the drug is used the way it would be administered when marketed. When these studies are completed and the sponsor believes that the drug is safe and effective under specific conditions, the sponsor applies to FDA for approval to market the drug. Phase 3 trials usually involve several hundred to several thousand patient-subjects.

Phase 4 Drug Trial: Concurrent with marketing approval, FDA may seek agreement from the sponsor to conduct certain postmarketing (Phase 4) studies to delineate additional information about the drug's risks, benefits, and optimal use. These studies could include, but would not be limited to, studying different doses or schedules of administration than were used in Phase 2 studies, use of the drug in other patient populations or other stages of the disease, or use of the drug over a longer period of time [21 CFR 312.85].

Principal Investigator: The scientist or scholar with primary responsibility for the design and conduct of a research project. (See also: Investigator.)

Sponsor: A person or entity that initiates a clinical investigation of a drug usually the drug manufacturer or research institution that developed the drug. The sponsor does not actually conduct the investigation but rather distributes the new drug to investigators and physicians for clinical trials. The drug is administered to subjects under the immediate direction of an investigator who is not also a sponsor. A clinical investigator may, however, serve as a sponsor-investigator. The sponsor assumes responsibility for investigating the new drug, including responsibility for compliance with applicable laws and regulations. The sponsor, for example, is responsible for obtaining FDA approval to conduct a trial and for reporting the results of the trial to the FDA.

Sponsor-Investigator: An individual who both initiates and actually conducts, alone or with others, a clinical investigation. Corporations, agencies or other institutions do not qualify as sponsor-investigators.


Once a chemical (drug) is identified as having a potential effect on a disease state, it is subjected to testing in animals. Initial animal tests are designed to see whether the chemical has any desired drug effects, what dosage levels are poisonous, what the safe dosage range might be in humans, and whether there is a reason to test the chemical in humans. Additional animal tests may be required as human tests progress. If initial animal tests indicate that the drug can be safely tested in humans and that the chemical may be therapeutically useful, the drug sponsor will submit an Investigational New Drug Application (IND) to the FDA. In the IND, the sponsor must describe the complete composition of the drug, its source, and how it is made. In addition, the sponsor must submit the results of all animal studies that support the drug's potential usefulness in humans and that define its toxicity in animals. The data should indicate that no human subject will be exposed to an unreasonable risk. The IND must also include a protocol describing the plan for testing in humans. To permit the FDA to review the materials and make sure subjects will not be exposed to unreasonable risks, the sponsor may not begin clinical tests for 30 days after submitting the IND. At the end of that period, the sponsor may begin the proposed clinical trial unless the FDA has asked for a delay because of a potential safety problem involving use of the drug.

Clinical trials are conducted by clinical investigators (usually physicians) who have entered into an agreement with a sponsor to conduct the study. All physicians administering an investigational drug agree to conditions regarding the conduct of the study outlined by FDA regulations. Clinical investigators agree to these conditions by signing an FDA form that certifies that the investigator has obtained IRB review and approval prior to conducting the study.

Investigational new drugs may be available outside of a clinical trial, through a treatment protocol, to patients with life-threatening or other serious diseases for which no satisfactory alternative drug or other therapy exists. Established by the FDA in 1987, the Treatment Investigational New Drug exemption (Treatment IND) is a treatment protocol that is added to an existing IND. The Treatment IND allows physicians to treat qualifying patients according to the protocol. Treatment INDs are discussed in greater detail in Guidebook Chapter 2, Section B, "Food and Drug Administration Regulations and Policies."

For further information concerning human subjects research to which FDA regulations apply, contact:

Mr. Richard M. Klein
Office of Health Affairs (HFY-20)
Food and Drug Administration
5600 Fishers Lane
Rockville, MD 20857
Tel: (301) 443-1382


In reviewing proposed drug research, IRBs must first consider whether the protocol is scientifically sound. Since this decision is not the IRB's primary concern, however, an IRB may rely on the FDA, institutions, scientific review committees, funding agencies (e.g., NIH), or others for this determination. [See the Introduction to Guidebook Chapter 4, "Considerations of Research Design" for a discussion of this question.] Evaluating the risks and benefits of drug trials requires IRBs to consider many aspects of the study design, paying special attention to the study population, the trial phase, and mechanisms for data analysis and surveillance. Risk/benefit analysis and review of the procedure for obtaining informed consent must be performed in all IRB reviews. [See Guidebook Chapter 3, Section A, "Risk/Benefit Analysis," and Chapter 3, Section B, "Informed Consent."] In addition, subjects participating in studies involving investigational drugs must be told that the FDA may have access to their medical records as they pertain to the study.

The obligation of IRBs and investigators to assure that subjects understand the purposes, methods, and possible hazards of the research is more difficult to fulfill when prospective subjects are seriously ill and in need of therapy. The consent process may require additional efforts and attention for research involving particularly vulnerable subjects such as the seriously ill. [See Chapter 6, Section G, "Terminally Ill Patients."]

Phase 1 trials are historically safest because they usually involve administering a single dose to healthy volunteers. However, Phase 1 trials may pose the highest level of unknown risk because they involve the drug's first administration to humans. (With highly toxic drugs such as cancer chemotherapies, Phase 1 trials are usually conducted with cancer patients as subjects.) Insofar as possible, risks should be identified from previous laboratory experiments and animal trials. The FDA, which reviews Phase 1 trials submitted in the initial IND application, may have valuable information and recommendations on particular protocols.

Subjects in Phase 2 trials are usually patients with the condition that the new drug is intended to detect or treat. IRBs should recognize that although Phase 2 testing is preceded by earlier clinical trials, the physiological responses of healthy volunteers to a therapeutic drug may not be reliable indicators of how safe the drug is for persons who are ill, taking other medication, or have immunodeficiencies. Since the primary purpose of a Phase 2 trial is to test the drug's effectiveness in achieving its purpose, the responses of subjects receiving the drug are usually compared with those of subjects who are not receiving the drug (control subjects). Whether control subjects receive some existing therapy or a placebo is a research design issue with serious ethical implications. Where an alternate safe and effective drug is available for a serious condition being studied, it should generally be given to the control subjects; however, existing therapies may be inadequate because they are of limited effectiveness against the disease, they have relatively high levels of toxicity, or because they are inconvenient to administer. When determining the acceptability of a proposed research design, IRBs must examine the risks and effectiveness of existing therapies, as well as the risks associated with providing no therapy (or a placebo). [See Chapter 4, "Considerations of Research Design."]

While most drug trials involve agents that the FDA has not yet approved for marketing, some drugs may be the subject of further testing concurrent with or following FDA approval. Post-marketing investigations, also called Phase 4 trials, are conducted to develop further information about the article's safety or effectiveness. Such studies might, for example, seek to establish the safety or effectiveness of using the drug for a new indication, with a new dosage level or a new route of administration [21 CFR 312.85].

Phase 4 studies should be distinguished from use of a marketed product by a physician for an indication not in the approved labeling as part of the "practice of medicine." Investigational use of a marketed product differs from such uses by physicians in that the principal intent of the investigational use of a test article is to develop information about its safety or efficacy; the submission of an IND or IDE may therefore be required. The criteria for submission of an IND or IDE for investigational use of a marketed product is described in the FDA's IRB Information Sheet entitled, "Investigational Use of Marketed Products," (1989, pp. 70-71).

Throughout drug trials, the distinction between therapy and research must be maintained. A physician who participates in research by administering a new drug to consenting patients must ensure that the patients understand and remember that the drug is experimental, and that its benefits for the condition under study are unproven. Furthermore, whereas the principal investigator's primary allegiance is to the protocol, the physician's allegiance is to the patient. Where an individual is both an investigator and the subject's treating physician, these two allegiances may conflict. The subject must recognize that the person with whom he or she is dealing may have such conflicting interests. The IRB should be aware of the need to inform the patient of the potential conflict.

If the trial is to collect accurate and timely data concerning the drug's safety and effectiveness, procedures for identifying positive and negative responses to the drug should be in place, and all participating physicians should be well integrated into a reporting system. The principal investigator is responsible for keeping all subjects informed of material changes in the design and conduct of the research, and must communicate new information that might affect their willingness to continue as subjects [Federal Policy ___.116]. The IRB may assist the investigator in deciding when information from accumulating data should be disclosed to participating or prospective subjects. The disclosure of information gained during the conduct of the trial is especially important with patients entering a study when it is nearing completion.

As part of their determination of the appropriate methods for conducting continuing reviews of ongoing studies, IRBs should be aware of the arrangements made for monitoring the study results. In FDA-regulated clinical investigations, arrangements for data monitoring are the sponsor's responsibility. The sponsor may designate an independent person or group (often called a data and safety monitoring board) to assume this responsibility. An IRB may function in such a capacity; however, most IRBs do not have the necessary expertise. Independent monitoring is most appropriate when the study is double-masked (i.e., neither the subjects nor the investigators know which drug a subject is receiving) or if the trial is multicentered. Ongoing monitoring of drug trials includes review of data on therapeutic effects, side effects and the effects of any changes in the study design. [See also Guidebook Chapter 3, Section E, "Monitoring and Observation."] Sponsors must notify the FDA and all participating investigators of any adverse experiences associated with the use of an investigational new drug that is both serious and unexpected [21 CFR 312.32].

Occasionally, hazards are discovered after a trial is concluded. If the drug has since been marketed, the FDA and the drug manufacturer are usually responsible for notifying users and physicians.


1. Is the proposed research scientifically sound?

2. Has sufficient information been obtained from the literature, experimental and animal studies, and the FDA to define, as far as possible, the potential risks of and the precise need for studies involving human subjects?

3. Does the principal investigator have the appropriate qualifications, experience, and facilities to ensure that all aspects of the trial and follow-up will be conducted rigorously and with due regard for the safety and well-being of the subjects?

4. Have appropriate measures been adopted to ensure that subjects understand the objectives and consequences, particularly the risks, of their participation?

5. Are sufficient safeguards provided to ensure the confidentiality of data generated during research?

6. Are adequate procedures provided for the ongoing surveillance of the drug's effectiveness and safety, and for notifying subjects and physicians of significant risks?

7. Has appropriate FDA review and clearance been obtained?


Federal Policy for the protection of human subjects

21 CFR 50 [FDA: Informed consent]
21 CFR 56 [FDA: IRB review and approval]
21 CFR 312 [FDA: Investigational new drugs]
21 CFR 52 [FDA: Sponsor and monitor (proposed)]
21 CFR 54 [FDA: Clinical investigators (proposed)]

Return to Top of Page



Vaccines are used to prevent infectious diseases. Successful vaccine trials have resulted in the development of safe and effective vaccines for polio, measles, rubella, hepatitis B, pneumococcal pneumonia, and other serious diseases. Currently, vaccines are being evaluated to prevent infectious diseases such as AIDS (or transmission of HIV), malaria, tuberculosis, trachoma, cytomegalovirus, herpes simplex, and influenza. Vaccines must undergo clinical testing prior to approval and licensure by the FDA. The regulations governing the conduct of clinical trials on investigational vaccines are the same as those governing the conduct of investigational new drug research [see Guidebook Chapter 5, Section B, "Drug Trials"]; however, the risks and benefits associated with vaccine trials may differ from those of drug trials.

A vaccine is a biologic; its use in trials involving human subjects is similar to the use of any drug. Vaccines do, however, differ from therapeutic drugs in two important ways. As used here, they are not designed to diagnose or cure disease in afflicted individuals; their purpose is to prevent a particular disease in healthy human beings. Vaccines are also used to protect people with a high statistical risk for contracting a particular disease or for suffering especially serious consequences from a disease. Vaccines trigger the body's normal immune response, producing antibodies that protect against future infection. Some vaccines (e.g., those containing active microorganisms or live-attenuated vaccines) have a small but real disease-producing capacity. Thus, one rare risk of a new vaccine is the possibility of infecting a healthy subject with the very disease researchers are seeking to prevent. More often, however, subjects involved in vaccine trials temporarily suffer from some of the symptoms and effects of the disease (e.g., polio, German measles) as they acquire immunity.



The development of vaccines is of considerable benefit to society, especially in the case of devastating or highly infectious diseases. The direct benefit to the individual subject receiving a new vaccine is the possibility of immunity (i.e., protection against future disease). The benefits of such immunity will vary depending on: (1) the severity of the disease to be avoided; (2) the likelihood that the subject will be exposed to the infectious disease; and (3) in the case of certain diseases, the likelihood that the subject would suffer adverse consequences should he or she contract the disease. Some populations will be at greater risk of contracting an infectious disease than others, either because they are more likely to be exposed to the disease or because they have an increased susceptibility to it. Among those who contract an infectious disease, there may be some sub-groups that are particularly vulnerable to adverse consequences (e.g., children, persons of advanced age, or persons suffering from other illnesses).

For most diseases, participation in vaccine trials carries the generally small risk of contracting the disease. [In some vaccine trials (e.g., HIV) there is no such risk. In the case of HIV vaccine research, the lack of risk is due to the manner in which the vaccine is derived.] The risks of participating in a vaccine trial also include adverse effects unrelated to the disease in question (e.g., slight fever, headache, muscle soreness, or muscle aches). Such side effects are usually short-lived, tolerable, and not life-threatening. Again, the degree of risk associated with participating in a vaccine trial varies depending on the subjects' vulnerability to the adverse side effects of the vaccine. Some subjects may have an allergic or anaphylactic (i.e., a decrease rather than an increase in immunity) reaction to the vaccine. Anaphylactic reactions to vaccines cause the recipient to be hypersusceptible to the disease. Such reactions are generally unpredictable, and may be serious or potentially life-threatening.

The IRB should be aware of other risks associated with vaccine trials, including the possibility that vaccines produced synthetically or using recombinant DNA techniques may present risks as yet unknown, that groups often most likely to benefit from receiving a vaccine are often the most vulnerable to coercion (e.g., institutionalized persons or children), and that subjects in control groups may erroneously assume that they have been immunized.

When determining whether the risks are reasonable in relation to the benefits, IRBs should consider the severity of the disease, the risk of contracting the disease, and any special vulnerability of the subject population to the potential adverse effects of the vaccine. The most difficult cases are those in which the subjects most likely to benefit from participating in the vaccine trial are also the subjects at the greatest risk of suffering from the vaccine's potential adverse effects.

Some of the risks inherent in vaccine trials can be minimized. Before a vaccine is approved for testing with human subjects, IRBs should receive satisfactory evidence that animal trials and laboratory tests have, to the extent possible, demonstrated its safety. Since the sponsor must submit such information to the FDA as part of its investigational new drug application (IND), IRBs can readily obtain evidence of safety as well.

Mechanisms for protecting human subjects from some risks can be built into the vaccine study design. For example, with careful screening, investigators can avoid enrolling persons who may be susceptible to certain adverse reactions. Furthermore, trials can be designed to involve subjects who are most likely to be exposed to the infectious agent and who stand to benefit most from the protection afforded by the vaccine. Selecting subjects in this way avoids exposing those who may not be in need of its protective benefits to the risks of the vaccine. In many situations, however, Phase 1 trials should be designed to evaluate low risk subjects. For example, an effective hepatitis B vaccine already exists. It would therefore be appropriate to determine that an investigational vaccine for hepatitis B is immunogenic in humans prior to use in high risk subjects.

Vaccine trials require careful monitoring of human subjects for both immune status and adverse reactions. The monitoring reflects the dual goals of any trial to determine both the effectiveness and the safety of the investigational substance or device. Although subjects in vaccine trials should be advised beforehand of known or anticipated side effects, rare or unknown reactions may occur. FDA regulations require that subjects be provided with written instructions about whom to contact in the event of serious adverse reactions or research-related injury.

IRBs should also be aware that large-scale field trials of a vaccine may involve many thousands of subjects, making monitoring difficult. The IRB should make sure that the sponsor has made provisions for monitoring the progress of the research, the immune status of participants, and side effects reported. Maintaining careful records is important both for monitoring the safety and effectiveness of the vaccine and for locating subjects for follow-up. If a vaccine either does not immunize the subject or does so for too limited a time, subjects may erroneously assume they are protected and fail to seek necessary medical attention. In addition, members of a control group may (incorrectly) assume they are immune from the disease because they believe they have received an effective vaccine (which they have not). IRBs sometimes require that control group subjects be given the first opportunity to receive the vaccine once its safety and effectiveness have been established. If such arrangements are not part of the research design, at the end of the trial control subjects should be informed of both their status vis a vis the vaccine, and the outcome of the trial: e.g., that the vaccine was shown to be safe and effective, but that they either did not receive the vaccine or did not receive an effective dose of the vaccine.

For a discussion of ethical issues related to the clinical testing of AIDS vaccines, see Guidebook Chapter 5, Section F, "AIDS/HIV-Related Research."


1. Has appropriate FDA clearance and an approved IND been obtained?

2. Is there evidence that the vaccine has been adequately tested in animal trials and in the laboratory?

3. Where appropriate, are subjects clearly told in the consent process that they might receive a placebo or ineffective dose of the vaccine, and thus may not be protected against the disease?

4. Does the protocol provide adequate plans to monitor all subjects for immune status and adverse reactions, respond to problems, and disseminate results?

5. Will subjects be informed about what to do and whom to contact in case of a serious adverse reaction or research-related injury?


Federal Policy for the protection of human subjects

21 CFR 50 [FDA: Informed consent]
21 CFR 56 [FDA: IRB review and approval]
21 CFR 312 [FDA: Investigational new drug research]
21 CFR 600-800 [FDA: Standards for biological products]
21 CFR 630 [FDA: Standards for viral vaccines]

Return to Top of Page



Comprehensive federal regulations governing investigations involving medical devices are comparatively new. In addition to their other duties, IRBs reviewing certain device investigations must also determine whether a device study presents a significant or nonsignificant risk to the human subjects participating in the study. When making determinations of significant versus nonsignificant risk, IRBs must consider not only the risks associated with use of the device itself, but also the risks associated with the investigational device study as a whole.


510(k) Device: A medical device that is considered substantially equivalent to a device that was or is being legally marketed. A sponsor planning to market such a device must submit notification to the FDA 90 days in advance of placing the device on the market. If the FDA concurs with the sponsor, the device may then be marketed. 510(k) is the section of the Food, Drug and Cosmetic Act that describes premarket notification; hence the designation "510(k) device."

General Controls: Certain FDA statutory provisions designed to control the safety of /marketed drugs and devices. The general controls include provisions on adulteration, misbranding, banned devices, good manufacturing practices, notification and record keeping, and other sections of the Medical Device Amendments to the Food, Drug and Cosmetic Act [21 U.S. Code 360(c) (Food, Drug and Cosmetic Act 513)].

Investigational Device Exemptions (IDE): Exemptions from certain regulations found in the Medical Device Amendments that allow shipment of unapproved devices for use in clinical investigations.

Medical Device: A diagnostic or therapeutic article that does not achieve any of its principal intended purposes through chemical action within or on the body. Such devices include diagnostic test kits, crutches, electrodes, pacemakers, arterial grafts, intraocular lenses, and orthopedic pins or other orthopedic equipment.

Nonsignificant Risk Device: An investigational medical device that does not present significant risk to the patient. (See also: Significant Risk Device.)

Postamendments Devices: Medical devices marketed after enactment of the 1976 Medical Device Amendments.

Preamendments Devices: Medical devices marketed before the enactment of the 1976 Medical Device Amendments.

Predicate Devices: Currently legally marketed devices to which new devices may be found substantially equivalent under the 510(k) process.

Premarket Approval: Process of scientific and regulatory review by the FDA to ensure the safety and effectiveness of Class III devices.

Significant Risk Device: An investigational medical device that presents a potential for serious risk to the health, safety, or welfare of the subject. Such a device is:

intended for use as an implant and presents a potential for serious risk to the health, safety, or welfare of the subject; or
purported or represented to be of use in supporting or sustaining human life and presents a potential for serious risk to the health, safety, or welfare of the subject; or
intended for a use that is of substantial importance in diagnosing, curing, mitigating, or treating disease, or otherwise preventing impairment of human health, and presents a potential for serious risk to the health, safety, or welfare of the subject; or
otherwise presents a potential for serious risk to the health, safety, or welfare of a subject.


The 1976 Medical Device Amendments (the Amendments) to the Federal Food, Drug and Cosmetic Act (the Act) were passed to give the FDA additional authority to assure safety and effectiveness in devices intended for human use. New medical devices must be cleared by the FDA prior to being placed on the market. As part of the clearance process, all medical devices are classified into one of three categories by the FDA based on the extent of control necessary to ensure the safety and effectiveness of each device [21 U.S. Code 360(c) (Food, Drug and Cosmetic Act 513)].

Medical devices are classified as Class I, Class II, or Class III devices depending on several criteria. Devices are classified as Class I medical devices if their safety and effectiveness can be assured by the general controls of the Amendments. The general controls include the provisions of the Act pertaining to adulteration, misbranding, banned devices, notification, repair, replacement or refund, records and reports, and restricted devices. In addition, general controls require device manufacturers or other designated persons, unless specifically exempted, to register their establishment, list their device, submit a premarket notification application, and be in compliance with the good manufacturing practices (GMPs). If a device cannot be classified as a Class I device because the general controls are insufficient to provide reasonable assurance of the safety and effectiveness of the device, the device may qualify for Class II classification. A Class II device must comply with general controls, and, in addition, the sponsor must provide sufficient information about the device to establish special controls that are sufficient to provide such assurance. Examples of special controls include the promulgation of performance standards, postmarket surveillance, the establishment of patient registries, and the development and dissemination of guidelines.

Devices are classified as Class III devices when: (1) their safety and effectiveness cannot be reasonably assured through either general or special controls; and (2) they are life-sustaining, life-supporting, implanted in the body, or of substantial importance in preventing impairment to health.

A new device that a manufacturer claims is substantially equivalent to a currently legally marketed device may be marketed after the FDA is notified of the intent to market, and the agency concurs with the manufacturer's claim of equivalence to other marketed devices. If the FDA determines that the new device is not substantially equivalent to a predicate device, the new device is automatically placed in Class III, and the manufacturer must obtain premarket approval from the FDA. Alternatively, the sponsor (or others) may petition the FDA to reclassify the device into Class I or II.

Investigational devices are medical devices that are the object of clinical research to determine their safety or effectiveness. Clinical investigations are necessary to support a request for premarket approval. Studies involving human subjects that are undertaken to develop safety and effectiveness data for medical devices must be conducted according to the requirements of the Investigational Device Exemption regulations [21 CFR 812] or Investigational Exemptions for Intraocular Lenses [21 CFR 813]. An approved IDE exempts a device from certain sections of the Act (e.g., misbranding under 502; registration, listing, and premarket notification under 510; special controls under 513; premarket approval under 515; banned devices under 516; records and reports under 519; restricted devices under 520(e); good manufacturing practices under 520(f); and color additive requirements under 706).

The IDE regulation describes two types of device investigations: significant risk device studies and nonsignificant risk device studies. Clinical trials involving significant risk devices require both FDA and IRB approval; sponsors must meet the full IDE requirements, including obtaining an FDA-approved IDE. Approval of studies involving nonsignificant risk devices require only IRB approval; no IDE is required to be formally submitted to the FDA. However, the sponsor must comply with the abbreviated regulatory requirements for such devices [21 CFR 812.2(b)]. The FDA may overturn IRB determinations that a device presents no significant risk.


In reviewing studies involving medical devices, IRBs should recognize that they must make two determinations: (1) whether a device study presents significant or nonsignificant risk; and (2) whether the study should be approved. These questions should be considered separately because the issues involved in making these decisions are quite different. Determining whether a device study poses a significant risk is based solely on considerations of risk to subjects, while IRB approval of the study is based on many factors. The discussion in this Section first considers IRB determinations of significant risk.

The FDA reviews and approves IDEs for significant risk device studies; it exercises less regulatory control over nonsignificant risk device studies. The initial responsibility for making the nonsignificant risk assessment for studies lies with the sponsor. If the sponsor believes that a particular device study presents a nonsignificant risk, the sponsor should provide the IRB with the study proposal, an explanation of why the device study presents a nonsignificant risk, and any other supporting information, such as reports of prior investigations. The sponsor should also tell the IRB whether the FDA or any other IRB has made a risk assessment and what the results of those assessments were. The IRB reviews the information, and may or may not agree with the sponsor's determination. If the IRB finds that the device study presents a nonsignificant risk, the investigation may begin without submission of an IDE application to the FDA. If the IRB disagrees with the sponsor's determination that a device study presents nonsignificant risk to human subjects, the sponsor must so notify the FDA, whether or not the sponsor ultimately conducts the study at that institution.

If the study comes to the attention of the FDA, the agency's Office of Device Evaluation may reach a different conclusion on the risk presented by a device study than that reached by the IRB. If the FDA overrules an IRB's decision that a device study presents nonsignificant risk, the sponsor must then submit an IDE application to the FDA. The IRB must then review the investigation as a significant risk device study, and the investigator will be subject to more stringent recordkeeping and reporting requirements.

In determining whether a device study presents a significant or a nonsignificant risk, both the risks of the device and the risks associated with the procedure for using the device (e.g., surgery for installing an implant) must be considered. The comparison of risks is the basis for the other decision the IRB must make: whether to approve the research.

The clinical investigator should provide the IRB with adequate information about a device's regulatory status and the results of any risk assessment the FDA may have made. The IRB may also ask the sponsor whether other IRBs have reviewed the study and what determinations were made. IRBs may also request the sponsor or clinical investigator to provide documentation of appropriate FDA clearances, and may consult the FDA for its opinion on risk.

In the past, clinical investigations of intraocular lenses (IOLs) differed from other medical device studies in that there were few restrictions on the total number of subjects in an IOL investigation. Unlimited "adjunct" studies were phased out when enough approved IOLs became commercially available. IOL studies are now limited in enrollment size, as are other medical device studies.

Clinical investigations involving IOLs that commenced before July 27, 1981, are exempt from investigational device requirements [21 CFR 812], since they are subject to specific regulations on intraocular lenses [21 CFR 813], which specify procedures for IRB review and informed consent.

The IRB's second responsibility is to decide whether to approve the proposed research. In general, full IRB review is required for both significant and nonsignificant risk studies. However, some studies involving nonsignificant risk devices may also be considered minimal risk studies, and thus may be reviewed through the expedited review procedure established by the IRB.

IRBs need to keep in mind the difference between the risk/benefit evaluation made in the context of approving the research and the IRB's assessment of whether use of the device poses significant or nonsignificant risk. The latter decision categorizes the degree of risk of harm based upon the seriousness of the harm that may result from the use of the device; the former is a balancing of those risks (plus the risks of the research process) against the potential benefits to be gained from conducting the research.

The criteria for deciding whether a medical device study should be approved are the same as those used to evaluate research involving any FDA-regulated product. The IRB should determine that risks to subjects are minimized and are reasonable in relation to anticipated benefits and knowledge to be gained, that subject selection is equitable, informed consent procedures and documentation are adequate, and that provisions for monitoring the study and protecting subjects' privacy and confidentiality of data are acceptable. As in other clinical investigations, an IRB's decision to approve the research must take into account the risks and benefits of the investigational device as compared with the other available therapies. However, the IRB should not simply consider the increase in risk over standard treatment, but rather the risk of the procedure as a whole.

For further information and guidance on studies involving medical devices, contact:

Dr. Michael J. Blackwell
Chief, IDE Section (HFZ-403)
Office of Device Evaluation
Center for Devices and Radiological Health
Food and Drug Administration
1390 Piccard Drive
Rockville, MD 20850
Tel: (301) 427-1190

Mr. Richard M. Klein
Health Assessment Policy Staff
Office of Health Affairs (HFY-20)
Food and Drug Administration
Room 11-44, 5600 Fishers Lane
Rockville, MD 20857
Tel: (301) 443-1382


1. What risks are presented by the device? Are they significant or nonsignificant?

2. Have other IRBs reviewed and made decisions regarding this device? (Such information should be available from the sponsor or clinical investigator.)

3. What is the status of the device with the FDA? Has the device been approved for marketing? Is the device approved for other indications? Is it now being studied for a different indication? Is an IDE needed for this device? If so, has it been approved?


Federal Policy for the protection of human subjects

The Food, Drug and Cosmetic Act, as amended [codified at U.S. Code, Title 21]

The Medical Device Amendments of 1976 [P.L. 94-295, 90 Stat. 539 (May 28, 1976)]

The Safe Medical Devices Act of 1990 [P.L. 101-629]

21 CFR 50 [FDA: Informed consent]
21 CFR 56 [FDA: IRB review and approval]
21 CFR 812 [FDA: Investigational device exemptions]
21 CFR 813 [FDA: Investigational exemptions for intraocular lenses]

Return to Top of Page



Radiopharmaceuticals and X-rays are widely used in medicine today for both diagnostic and therapeutic purposes. Certain aspects of human physiology can only be studied through exposure to radiation, or can be studied more safely by radiation than by alternative methods.

The types of radiation used most frequently in medical investigations and treatments are X-rays, gamma rays, and beta radiation. In addition to passing X-rays through the body to produce an image, some procedures use contrast agents to outline or define the shape of internal structures, or to image metabolic processes. Nuclear medicine uses procedures in which radioactive materials (i.e., radiopharmaceuticals) are injected, ingested, or inhaled into the body. Most medical institutions have a radiation safety committee responsible for evaluating the risks of medical projects involving radiation and limiting the radiation exposure of employees and patients. Nevertheless, IRBs should have an understanding of radiation and its biological effects so they can evaluate the relative risks and benefits of research proposals utilizing radioactive materials or X-rays.


Radioactive Drug: Any substance defined as a drug in 201(b)(1) of the Federal Food, Drug and Cosmetic Act that exhibits spontaneous disintegration of unstable nuclei with the emission of nuclear particles or photons [21 CFR 310.3(n)]. Included are any nonradioactive reagent kit or nuclide generator that is intended to be used in the preparation of a radioactive drug and "radioactive biological products," as defined in 21 CFR 600.3(ee). Drugs such as carbon-containing compounds or potassium-containing salts containing trace quantities of naturally occurring radionuclides are not considered radioactive drugs.

Radioactive Drug Research Committee (RDRC): An FDA-approved institutional committee responsible for the use of radioactive drugs in human subjects for certain research purposes [21 CFR 361.1]. Research involving human subjects that proposes to use radioactive drugs must be approved by the RDRC and must meet various FDA requirements, including limitations on the pharmacological dose and the radiation dose. The research must be basic research, not intended for diagnosis or treatment of a disease. Furthermore, the exposure to radiation must be justified by the quality of the study and the importance of the information it seeks to obtain. The committee is also responsible for continuing review of the drug use to ensure that the research continues to comply with FDA requirements, including reporting obligations. The committee must include experts in nuclear medicine as well as other medical and scientific members.

Radiopaque Contrast Agents: Materials that stop or attenuate radiation that is passed through the body, creating an outline on film of the organ(s) being examined. Contrast agents, sometimes called "dyes," do not contain radioisotopes. When such agents are used, exposure to radiation results only from the X-ray equipment used in the examination. The chemical structure of radiopaque contrast agents can produce a variety of adverse reactions, some of which may be severe and possibly life-threatening in certain individuals.

Radiopharmaceuticals: Radioactive drugs that are labeled or tagged with a radioisotope. These materials are largely physiological or subpharmacological in action, and, in many cases, function much like materials found in the body. The principal risk associated with these materials is the consequent radiation exposure to the body or to specific organ systems when they are introduced into the body.

REM: Acronym for Roentgen Equivalent in Man; the unit of measurement for a dose of an ionizing radiation that produces the same biological effect as a unit of absorbed dose (1 rad) of ordinary X-rays. One millirem is equal to 1/1000 of a rem.


The quantity of natural background radiation to which we are exposed varies considerably (e.g., radiation exposures are much lower at sea level than they are at higher altitudes). The average annual natural background radiation from all sources in the United States is approximately 100 to 125 millirems (mrem) per year, while some individual exposures may be more than 400 mrem per year. Diagnostic medical procedures are the most likely source of additional radiation exposure. Estimates suggest that medical procedures increase the total exposure by 50 to 70 mrem per person per year.

Experts disagree, however, over the fundamental concepts that affect how radiation risks from medical procedures and other sources are estimated. The disagreements include debate about the existence of a theoretical threshold level below which no harmful effects occur. The National Council for Radiation Protection and Measurement (NCRPM) takes the position that there is no absolutely safe radiation dose. Generally, only approximations of risk from exposure are available; they are based on extrapolations from known exposures to high levels of radiation. The NCRPM has recommended dose standards; the Nuclear Regulatory Commission (NRC) has established occupational dose limits. The occupational dose limits vary according to the part of the body exposed to radiation.

The NRC is responsible for those radioactive materials considered to be "source material," "byproduct material," or "special nuclear material" [10 CFR Parts 30, 40, and 70]. The NRC directly regulates these materials in 21 states; the other 29 states, known as "Agreement States," have entered into an agreement with the NRC to regulate uses within their states of byproduct material, source material, or special nuclear material involving less than certain quantities. Agreement States may have unique policies or standards concerning the use of radioactive materials in research that could, in some cases, be more restrictive than those of the NRC. Naturally-occurring or accelerator-produced radioactive materials (NARM), such as Thallium-201, are not covered by the Atomic Energy Act; therefore they are not regulated by the NRC. Those radioactive materials (NARM) may be dealt with under specific state regulations (in both Agreement States as well as non-Agreement States) governing the use of radioactive materials.

The FDA requires investigators to submit an Investigational New Drug Application (IND) for radioactive drugs, kits, or generators that are to be used for investigational diagnostic or therapeutic purposes (including testing to establish their safety and effectiveness). An exception is made for radioactive drugs to be used in certain research designed to study the metabolism of the drug or to gather information about human physiology, pathophysiology, or biochemistry, but not intended for immediate therapeutic, diagnostic, or similar purposes [21 CFR 361.1]. If the radiation dose will not exceed the limits set forth in these regulations, the study design meets other research criteria, and the protocol is approved by a Radioactive Drug Research Committee (RDRC), the investigator does not need to submit an IND. Current radiation limits for the use of such drugs in research (including radiation doses from X-ray procedures that would not have occurred but for the study) are as follows [21 CFR 361.1]:

For an adult research subject, radiation to the whole body, active blood-forming organs, the lens of the eye, or the gonads may not exceed a single dose of 3 rems or an annual cumulative dose of 5 rems.
The amount of radiation to other organs may not exceed a single dose of 5 rems or an annual cumulative dose of 15 rems.
Permissible doses for children (persons under age 18) are 10 percent of those for adults. The FDA must approve studies involving children before the study begins.

[See also 21 CFR 312.2(b), providing certain exemptions from IND application requirements.]

In addition to the RDRC, most medical institutions also have an Institutional Radiation Safety Committee, which assesses the risks that may be associated with exposure to radiation, both for research subjects and employees. In some states or institutions, review by the Radiation Safety Committee is mandated by law or policy; in others, the committee's review is offered as an opinion to the IRB to help it assess the risks and benefits of a given study involving radiation exposure.


An IRB should distinguish between radiation exposure resulting from routine medical management of a patient and radiation exposure that is part of research, including a clinical investigation. Although the occupational dose limits may not necessarily be appropriate when applied in a research setting, they do provide some guidance when exposure to radiation for research purposes is contemplated.

The likelihood of adverse effects associated with radiation exposure is generally considered to be low, but adverse effects can be serious when they do occur. Some effects rarely present themselves until many years after the subject has been exposed to radiation. The two adverse effects most commonly associated with radiation exposure are certain types of cancer and genetic damage.

The increased risk of genetic damage is of particular concern because exposure to radiation may involve substantial risk to the subject's unborn offspring. When the proposed research poses risk of genetic damage, an IRB should pay particular attention to the subject selection criteria. The human embryo is known to be particularly susceptible to damage from exposure to radiation; research involving pregnant or possibly pregnant women has therefore been of particular concern. Pregnancy tests could be required where doubt exists as to the presence of pregnancy, or the subject might be asked to use an effective contraceptive method during the course of the research. [See Guidebook Chapter 3, Section C, "Selection of Subjects," and Chapter 6, Section B, "Women."] Recent studies have suggested that male sperm cells are also adversely affected by radiation. Thus, no radiation dose should be considered risk-free if it is directed toward, or absorbed by, the reproductive organs.

Research involving radiation may also pose risks to lab personnel, nursing staff, and family members. This increased risk usually results from exposure to nuclear sources of radiation used in a medical device or nuclear medicine or radiotherapy. For example, when nuclear-powered artificial heart implants were under consideration, a federal panel expressed concern over the possible exposure and resultant risk to the patient's spouse.

Additional risk may be associated with the intravascular administration of contrast agents used in X-ray procedures (e.g., intravenous pylograms (IVP), venograms, and cardiac catheterizations). The risks vary depending on the dose of the contrast agents, the chemical nature of the contrast agent used, and the age and disease state of the subject. Conditions such as advanced age, renal disease, diabetes, cardiac, or cerebrovascular disease, asthma, or chronic obstructive pulmonary disease may greatly increase the risk associated with the proposed study. Unsuspected anaphylactic reactions may also, although rarely, occur.

Radiopharmaceuticals present relatively low risks of adverse reactions unrelated to their radioactivity. The principal risks associated with radiopharmaceuticals are posed by the radioisotope's energy, its half-life, the radiosensitivity of the organ system being studied, and the radiation dose to the target organ, adjacent organs, and the whole body. Other factors are, however, also relevant. For example, the dose of a labeled brain receptor agent or the status of a subject's brain receptors must be considered.

In addition to determining the level of risk associated with exposure to radiation, IRBs must be concerned with informed consent. Specifically, IRBs must determine what subjects should be told: how properly to communicate the uncertainty about the risk of harm posed by exposure to the level of radiation involved in the study. Since subjects must be given sufficient information on which to decide whether to participate, consent should be based on information that the subjects may reasonably be expected to want to know. The question for the IRB is how much risk must there be before a "reasonable volunteer" would want to know about it. Given the sensitivity of our society to the uncertainty surrounding the risks associated with radiation exposure, IRBs should require that subjects be told that participation in the research involves exposure to radiation.

Several ways of explaining the risks associated with exposure to radioactive materials to potential subjects have been suggested, but none are totally satisfactory. One method used is comparing the risk of death from radiation exposure to that of more familiar activities such as air travel or cigarette smoking. A second method compares the incidence of death per year from radiation exposure with the mortality rates of various occupations. Comparisons may also be made between the proposed research exposure and the dose received from cosmic and background radiation to which a subject is naturally exposed. The proposed research exposure may also be compared with the annual maximum permissible exposures suggested by the NCRPM for occupational workers. Finally, the research exposure can be compared with exposures from more familiar medical procedures, such as chest X-rays.

The major problem with expressing risks in comparative terms is that the actual risk from low levels of exposure is not known. This uncertainty should be communicated to research subjects. Even in cases where the risks from exposure are considered to be minimal and not reasonably foreseeable, the IRB may determine that the information concerning exposure and its possible effects is something that research subjects might reasonably want to know.

The IRB should ensure that the risks of radiation exposure are minimized. In an attempt to minimize radiation exposure, experts have developed a principle known as ALARA: As Low As Reasonably Achievable. IRBs should ensure that the ALARA principle is observed. [See also 21 CFR 361.1(b)(3) (limit on radiation dose).]


1. Can the information to be gained from the research project be gathered using methods that do not expose subjects to more radiation than that to which they would naturally be exposed?

2. Could the research be performed on patients undergoing the procedures for diagnostic or therapeutic purposes?

3. Will the smallest exposure (dose) possible be used in the study?

4. Have investigators taken steps to avoid re-exposure? Are procedures in place to ensure that investigators will use a minimum number of re-exposures in the event that the study needs to be repeated?

5. Are adequate radiation safety measures being taken to protect research subjects and others who may be exposed to radiation?

6. Have the investigators taken adequate precautions to screen subjects and exclude those not essential to the research project and those at increased risk from exposure to radiation or contrast agents?

7. Will both men and women be informed of the risks to future offspring due to possible genetic damage?

8. Will women of childbearing potential be adequately informed of the risks to an embryo associated with radiation exposure in early pregnancy, and of the importance of disclosing a possible pregnancy to the investigator? Does the protocol make adequate provisions for detecting pregnancies?


Federal Policy for the protection of human subjects

10 CFR 19 [NRC: Notices, instructions, and reports to workers; inspections]
10 CFR 20 [NRC: Standards for protection against radiation]
10 CFR 35 [NRC: Medical use of byproduct material]
21 CFR 50 [FDA: Informed consent]
21 CFR 56 [FDA: IRB review and approval]
21 CFR 361.1 [FDA: Radioactive drugs for certain research uses]
21 CFR 312 [FDA: Investigational new drug application]

State laws regarding radioactive materials licensure

Return to Top of Page



The human immunodeficiency virus (HIV) is a pathogenic retrovirus that causes acquired immunodeficiency syndrome (AIDS) and its related diseases in humans. Because of its high rate of mortality, AIDS has become the center of worldwide attention; research into the development of safe and effective therapies, as well as methods of prevention of this fatal disease, is currently a national public health priority.

HIV-related research centers on both biomedical and behavioral questions. Biomedical research has been characterized as falling into five major scientific categories: "(1) the study of the distribution of HIV infection and AIDS in the population (epidemiology) and the pattern of disease progression (natural history); (2) the identification and characterization of the virus that causes AIDS (etiologic agent); (3) delineation of the mechanisms by which the virus destroys the immune system and produces disease (pathogenesis); (4) the development and testing of potential therapies for HIV infection and its complications; and (5) the development and evaluation of potential AIDS vaccines" [Hamburg and Fauci (1989), p. 22].

Behavioral research on HIV focuses on: (1) identifying the social, psychological, and behavioral conditions of disease transmission and prevention; (2) the effects of psychological state on immunosuppression; and (3) the role of psychology in alleviating the distress experienced by persons affected by HIV infection (including families, friends, and persons at risk).


Research designed to answer the many biomedical and behavioral questions presented by HIV poses numerous ethical concerns. Primary among them are considerations of privacy, confidentiality, and justice (fairness in the distribution of the benefits and risks of research). The subjects involved in HIV-related research, HIV-infected individuals, and persons at risk of HIV infection, are particularly vulnerable, both because of their disease status, and because the disease disproportionately affects certain populations: male homosexuals and bisexuals, intravenous drug users, minorities, and, increasingly, women and children. [See Guidebook Chapter 6, "Special Classes of Subjects."]

An overriding concern in HIV research is confidentiality. Subjects included in HIV-related studies are understandably concerned about the confidentiality of the data, since breaches in confidentiality could have severe adverse consequences such as loss of employment or insurance coverage, or criminal charges. OPRR guidance on HIV studies states that:

where identifiers are not required by the design of the study, they are not to be recorded. If identifiers are recorded, they should be separated, if possible, from data and stored securely, with linkage restored only when necessary to conduct the research. No lists should be retained identifying those who elected not to participate. Participants must be given a fair, clear explanation of how information about them will be handled.

As a general principle, information is not to be disclosed without the subject's consent. The protocol must clearly state who is entitled to see records with identifiers, both within and outside the project. This statement must take account of the possibility of review of records by the funding agency.... [OPRR Reports, Dear Colleague Letter (December 26, 1984), p.3.]

IRBs should also consider whether and how information from HIV-related studies will be recorded in subjects' medical records, and may decide to impose limits on the recording of such data. Before agreeing to participate in an HIV study, subjects should be informed of exactly what information will be recorded, and whether any state laws require the reporting of HIV infection or other disclosures of information. The research protocol should also deal with the possibility of attempts under compulsory legal process to force disclosure of records, how such attempts will be responded to, and whether individuals will be notified of such attempts. [See also the Guidebook Chapter 3, Section D, "Privacy and Confidentiality," which deals with certificates of confidentiality and subpoenas.] The protocol should specifically set forth how to respond to requests by third parties who have authorizations for disclosure of information signed by subjects. An extensive set of guidelines for confidentiality in research on HIV has been developed by a group of prominent scholars, practitioners, and community members, and may be helpful to IRBs considering HIV-related protocols. [See Bayer, Levine, and Murray (1984).]

The PHS has an established policy on the issuance of certificates of confidentiality to projects that are subject to the reporting of communicable diseases to state and local health departments. The policy applies to projects that intend routinely to determine whether its subjects have communicable diseases, and that are required to report them under state law. Certificates will be issued: (1) where the referring treating physicians assure the project that they have complied with reporting requirements; (2) the investigator has reached an agreement with the health department about how he or she will cooperate with the department to help serve the purposes of the reporting requirements (unless the investigator can show why such cooperation is precluded); and (3) only where disclosures of identifiable information about subjects comply with regulations on subject protection, and are explained clearly to subjects prior to their participation [Mason (August 9, 1991)]. [See also Guidebook Chapter 3, Section D, "Privacy and Confidentiality."]

The giving of voluntary consent, axiomatic to all research involving human subjects, applies equally in HIV-related research. Complicating the consent issue, however, is that HIV-related illness, particularly in its later stages, can cause dementia, thus affecting the ability of subjects to give consent or continue to consent to ongoing research. Research protocols should deal with this possibility; IRBs should ensure that subjects in this particularly vulnerable condition are adequately protected. [See also Guidebook Chapter 6, Section D, "Cognitively Impaired."]

Research on vaccines and treatments poses some of the most difficult questions, including the level of acceptable risk to subjects when the disease is fatal and no effective therapy is available; whether HIV-infected patients can be used as a placebo group that is not given experimental treatments; how subjects should be selected to receive experimental therapies; whether and under what circumstances healthy and at-risk but not-yet-HIV-infected persons can ethically be asked to participate in vaccine trials.

Clinical Trials of HIV-Related Therapies. Randomized clinical trials (RCTs) and the ethical problems surrounding their use is discussed in Guidebook Chapter 4, Section H and related Guidebook Sections. This Section will focus on questions of particular concern for research involving HIV-infected individuals.

Randomized, controlled clinical trials are considered the research design most likely to yield valid scientific results for the evaluation of the safety and effectiveness of experimental therapies. Ethical use of RCTs depends on the existence of both the ability to state a null hypothesis (also called "theoretical equipoise") and that there be no other therapy known to be more effective than the one being studied in the RCT. A report produced by a working group on clinical HIV research convened by the American Foundation for AIDS Research argues, however, that when no known effective alternative therapy exists, as is presently the case with HIV, it may be justified to consider the use of other forms of controls such as historical controls (that is, to compare the effects of the therapy in the trial population with the treatment experiences of patients with the same disease before use of the experimental therapy) [Levine, Dubler, and Levine (1991), pp. 3, 6]. The justification for this position is that the conditions of "clinical equipoise" (a situation in which there is a "current or likely dispute among expert members of the clinical community as to which of two or more therapies is superior in all relevant respects," and which is also necessary for an RCT to be ethical) are not satisfied [id.]. The working group issued a document that included 57 recommendations on the conduct of clinical research on HIV, which IRBs may wish to consult [id.].

The use of placebo controls is particularly problematic. As a general matter, where the disease is lethal or seriously debilitating, as in the case of HIV, the use of placebo controls in place of an active control is difficult to justify ethically, despite the possibility that the experimental therapy is harmful (e.g., toxic) rather than therapeutic. In the language of the Belmont Report, the question of the use of control groups in this situation is one of beneficence: Are potential benefits maximized in all arms of the trial? The fatal nature of the disease leaves patients in a desperate position in which many seek any promising treatment. It has been suggested that the question may be resolved in favor of placebo controls only under two conditions: (1) when there is either no known effective therapy that can be used as an active control, or subjects are persons who cannot tolerate a known effective therapy; and (2) the trial therapy is "so scarce that only a limited number of patients can receive it" [Levine, Dubler, and Levine (1991), p. 8]. A fair way to then assign subjects to the active and control arm(s) is through a lottery [id.] [See also Macklin and Friedland (1986), pp. 277-79, and Guidebook Chapter 4, Section H, "Clinical Trials," and related Guidebook Sections.]

Once there is sufficient evidence of either a beneficial therapeutic effect, unacceptable side effects, or indication that there is a very low probability of establishing statistically significant research results, the trial should be stopped or the protocol should be modified [Macklin and Friedland (1986), pp. 177-78]. Where an experimental therapy is shown to have a beneficial therapeutic effect, the control group should be offered access to the experimental therapy. Prospective subjects should be informed of the probability of being assigned to the control group, the risks associated with being assigned to either the treatment or control group, the criteria that will be used for determining a beneficial effect sufficient to discontinue the control arm of the trial, and the consequences of discontinuing the control arm (e.g., will control subjects be added to the experimental group, will they be given the experimental therapy on a treatment basis, will they be offered the experimental therapy only if they pay for its cost, or will they be dropped from the study without access to the experimental therapy). It should be made clear to prospective subjects that the likelihood of the experimental therapy having harmful effects may well be as great as the likelihood of its having beneficial effects.

The selection and recruitment of subjects is also of concern. Subjects for clinical trials are often recruited on the recommendation of treating physicians. Unable or unwilling to obtain medical care, many individuals have been excluded from participation in trials. Others, not aware of the existence of trials, are also left out. Care should be taken to ensure the appropriate inclusion of women, children and adolescents, and minority groups in HIV-related clinical trials. Note also that IRBs must follow the additional protections provided in the DHHS regulations wherever applicable. [See Subpart B (fetuses, pregnant women, and human in vitro fertilization), Subpart C (prisoners), and Subpart D (children).]

When reviewing protocols involving HIV-infected or at-risk individuals or persons, IRBs should consider including (as consultants, if they are not already members) persons knowledgeable about and experienced in working with such subjects [Federal Policy ___.107]. Some investigatory groups have used "community advisory committees" as a means both of better understanding the concerns of the subject population and of educating the HIV-infected community about clinical research.

Vaccines. The testing of AIDS/HIV vaccines in human subjects raises substantial ethical issues. First and foremost is the question of risks and benefits. Limited availability of animal data means that many of the risks that might be associated with an AIDS/HIV vaccine (e.g., vaccine-induced immunotoxicity) are unknown. Nonetheless, the importance of developing an AIDS/HIV vaccine is felt to outweigh these uncertainties. From the standpoint of protecting the welfare of human subjects, however, the lack of knowledge about risk and the potential for the existence of serious risk must be clearly communicated and consented to by prospective subjects.

While all viral vaccines pose risks, HIV vaccines may, in addition, increase the risk of acquiring the disease when subsequently exposed to HIV. Also, because of potential immune tolerance, subjects may not be able to be vaccinated with a different AIDS/HIV vaccine if the experimental one proves ineffective. Persons with whom the subject is in close contact may also be at risk of transmission of recombinant viruses (through the injection site). IRBs should consider the degree to which investigators have minimized these risks, and ensure that subjects are adequately informed of and consent to these and other potential physical risks.

Another issue about which subjects must be informed is the effect of participation in the trial on their HIV serostatus and the potential social ramifications of changes in HIV serostatus. Just as persons infected with HIV through more usual means of transmission (e.g., sexual activity, the use of intravenous drugs, or blood transfusions) will test positive on antibody screening tests, so too will persons immunized with experimental AIDS/HIV vaccines. There may be limited access to diagnostic methods for distinguishing between persons who are HIV-infected and persons who have received HIV vaccinations. One way to help alleviate this problem is for trial sponsors to follow the lead of the National Institute of Allergy and Infectious Diseases (NIAID), and provide subjects with documentation certifying participation in the vaccine trial. Nonetheless, participation in AIDS/HIV vaccine trials in itself may carry a social stigma.

Informing Subjects of Their HIV Serostatus. Some research protocols involve screening blood samples for HIV seroprevalence or other procedures through which subjects' HIV serostatus will be discovered. In addition to ensuring that the confidentiality of this information and all research data is scrupulously provided for, and that subjects will be informed that they will be tested and of the risks and benefits involved, IRBs will need to consider the circumstances under which subjects should or must be told of their HIV serostatus. PHS policy requires that where HIV testing is conducted or supported by the PHS, individuals whose test results are associated with personal identifiers must be informed of their own test results and provided the opportunity to receive appropriate counseling unless the situation calls for an exception under the special circumstances set forth in the policy. Under the PHS policy, individuals may not be given the option "not to know" their test results, either at the time of consenting to be tested or thereafter. The acceptable "special circumstances" include such compelling and immediate reasons as an indication that a given individual would attempt suicide if informed that he or she was HIV seropositive; that extremely valuable knowledge might be gained from research involving subjects who would be expected to refuse to learn their HIV antibody results; or research activities conducted at foreign sites where cultural norms, the health resource capabilities, and official health policies of the host country preclude informing subjects of their HIV serostatus. Subjects should also be informed early in the consent process of any plans to notify subjects' sexual or needle-sharing partners. [See OPRR Reports ("Dear Colleague" letters dated December 26, 1984 and June 10, 1988).] Several commentators have taken issue with the position that subjects should be told of their serostatus regardless of their wishes. [See, e.g., Novick (1986) and Dubler (1986); compare Landesman (1986).] While this issue may be controversial, opportunities for early intervention weigh in favor of policies that require informing subjects of their HIV serostatus.

Counseling. Whenever subjects will be informed of their HIV serostatus, appropriate pretest and post test counseling must be provided. Counselors should be qualified to provide HIV test counseling and partner notification services. IRBs should ensure that such provisions are made. [See OPRR Reports ("Dear Colleague" letters dated December 26, 1984 and June 10, 1988)]

See also Guidebook Chapter 2, Section B, "Food and Drug Administration Regulations and Policies" (discussing expanded availability of investigational agents), and Chapter 4, "Considerations of Research Design."

Behavioral Research. Research on behavioral questions related to HIV often centers on what behavioral factors contribute to disease transmission and dissemination, as well as other psychosocial factors related to HIV (e.g., the relationship of stress to immunosuppression). The American Psychological Association has expressed concerns for subjects' privacy, protections against the intrusive nature of behavioral research (because research on risk factors and modes of disease transmission often probes intimate details of subjects' lives such as sexual practices and past history of illicit drug use), confidentiality, and the need to carefully debrief subjects.

Vulnerability of Subjects. In addition to the ethical issues raised by the conduct of HIV-related research itself, the involvement of HIV-infected subjects presents special concerns to which IRBs should be sensitive. As noted above, homosexual and bisexual men, intravenous drug users, minorities, and, increasingly, women and children constitute the bulk of the HIV-infected population. Their vulnerability as subjects arises primarily because their HIV status presents special concerns of confidentiality and privacy. Knowledge of a person's HIV status can lead to discriminatory practices on the part of employers, landlords, insurance companies, and others. That HIV disproportionately affects certain populations heightens the threat of inappropriate disclosure of HIV-related data. In addition, characteristics of the progression of AIDS, which can include both physical incapacity and loss of mental capacity, can impinge on subjects' ability to exercise their right to autonomy in the course of the research. IRBs can ensure that AIDS patients and other HIV-infected subjects are adequately protected by viewing each subject first and foremost as an individual. Researchers working with HIV-infected persons must be capable of dealing with social, emotional, and psychological, as well as physical factors. Taking such a multifaceted approach to working with this subject population is a means of incorporating the various necessary cultural and filial influences into the research relationship. Researchers should seek the advice and consultation of experts in these and other relevant fields as necessary.

Another factor that heightens the vulnerability of HIV-infected individuals is the lack of available treatment alternatives. At present, HIV infection is believed uniformly to progress to AIDS; no available treatment cures AIDS, although some therapies postpone the onset and severity of opportunistic infection. Prospective subjects in HIV-related studies may, therefore, agree to participate in research out of a hope for a cure, which may or may not be realistic. But while IRBs should protect subjects against exposure to excessive risk, they must also guard against paternalism. Despite the fatal nature of the disease, there may be risks to which individuals should not be asked to subject themselves; despite their vulnerability, however, prospective subjects should be given the opportunity to participate and obtain whatever benefits may be available. IRBs should consider protocols and make their evaluation of the requisite factors (i.e., the level of risk involved, a positive risk/benefit ratio, equitable selection of subjects, informed consent, and protection of privacy and confidentiality) with this concern in mind. The additional protection that IRBs can provide is to ensure that the protocol, its goals, and the research benefits and risks are clearly and simply delineated and communicated to the subject. It is important that participation in the research not engender either false hopes or a sense of hopelessness. Furthermore, IRBs should try to ensure that access to health care does not serve as a lure for participation.

IRBs need to review participant eligibility requirements closely and extensively monitor the data collection and analysis process. The consent process should also be carefully considered, with special attention to provisions for determining mental capacity to consent and alternative means for obtaining consent, where necessary. [See Guidebook Chapter 6, Section D, "Cognitively Impaired."] The duration of any health care to be rendered through participation, including counseling, should be thoroughly reviewed with subjects. As noted above, subjects must be clearly and explicitly informed of any applicable law or policy that requires either partner notification or notification to health authorities of subjects' HIV serostatus or disease status.

Finally, many HIV-infected persons are economically and/or educationally disadvantaged, and may need adjunct services or other help to be able to participate in research. To ensure that all affected groups have an adequate opportunity to participate, IRBs should give some thought to how investigators might meet these needs, thereby encouraging a broader distribution of the risks and benefits of HIV-related research.

Availability of Drugs and Other Therapeutic Agents for AIDS and HIV-Related Conditions. The availability of experimental drugs and other therapeutic agents for the treatment of AIDS and other HIV-related conditions has been highly controversial. Two mechanisms, Treatment INDs, and a subset of Treatment INDs, Parallel Track programs, have been developed by the FDA to meet this concern. They are discussed in the Guidebook in Chapter 2, Section B, "Food and Drug Administration Regulations and Policies."


1. Pre-screening clinical study participants for HIV antibody status: See the list of questions provided in OPRR Reports, "Points to Consider for Institutional Review Boards (IRBs) Regarding the Screening of Volunteers for HIV Antibody Status," (circa August, 1989).

2. Is the composition of the IRB membership appropriate for an adequate review of the protocol? Should the IRB seek consultation with laypersons, persons with AIDS or who are HIV-infected, or members of the HIV-affected community?

3. Are subjects' privacy and confidentiality adequately protected? Are certificates of confidentiality appropriate?

4. Does the consent process provide adequately for the special needs of subjects participating in HIV-related research, including subjects with impaired mental capacities and the difficulties of communicating the risks presented by drug and vaccine trials?

5. Will the informed consent process clearly inform the subject of all pertinent information (e.g., the circumstances under which the investigator may terminate the subject's participation without the subject's consent; the circumstances under which the subject may withdraw from participation and the costs associated with withdrawal; the financial costs of participation; how medical care will be handled in the event of injury or onset of opportunistic illness; whether partner notification and/or disease status reporting to health authorities will occur)?

6. Is there a mechanism for dealing with changes in mental capacity and continuing consent? Who will give consent in the event of diminished mental capacity or lack of majority (in the case of children)? Is it necessary to obtain subjects' assent?

7. Are protections against coercion in place?

8. If the protocol involves a clinical trial, have appropriate FDA clearances and an approved IND been obtained?

9. Does the protocol provide for adequate monitoring of all subjects for adverse reactions? Are provisions made for early termination?

10. Will subjects be informed about what to do and whom to contact in case of a serious adverse reaction or research-related injury?

11. Will subjects involved in behavioral research be adequately debriefed? Are intrusions into subjects' privacy minimized?


Federal Policy for the protection of human subjects

21 CFR 50 [FDA: Informed consent]
21 CFR 56 [FDA: IRB review and approval]
21 CFR 312 [FDA: New drugs for investigational use]
45 CFR 46, Subparts B-D [DHHS: Protection of human subjects]

Federal Register 57 (April 15, 1992): 13250-13259 [FDA: Parallel track policy]

State and local laws concerning the reporting of HIV-related information

Public Health Service policies related to AIDS research:

U.S. Public Health Service. National Institutes of Health. "Guidance for Institutional Review Boards for AIDS Studies" [Dear Colleague Letter]. OPRR Reports (December 26, 1984).

U.S. Public Health Service. National Institutes of Health. "Policy on Informing Those Tested About HIV Serostatus" [Dear Colleague Letter]. OPRR Reports (June 10, 1988).

U.S. Public Health Service. National Institutes of Health. "Points to Consider for Institutional Review Boards (IRBs) Regarding the Screening of Volunteers for HIV Antibody Status" [Dear Colleague Letter] OPRR Reports [circa August, 1989].

James O. Mason [Assistant Secretary for Health]. "Certificates of Confidentiality Disease Reporting" [Memorandum]. (August 9, 1991.)

Return to Top of Page



Numerous ethical issues confront IRBs considering research that involves the transplantation of organs or tissues into human subjects. Transplanted organs may be either natural or artificial; natural organs or tissue may be of either human or animal origin. Ethical issues include the physical and psychological risks to the donor and recipient, informed consent, coercion, and the selection of recipient-subjects (i.e., the distribution of organs or tissue to needy recipients).

The ethical considerations surrounding the transplantation of organs concern two basic problems: the scientific basis of the procedure (i.e., risk to the recipient-subject) and the procurement of organs for transplantation. The first problem raises issues with which IRBs are familiar: determining whether the proposed research poses an acceptable risk; balancing that risk with the potential benefits; ensuring that the patient-subject and the donor give their informed consent; and ensuring that the decision to participate is free from coercion and undue influence. The second problem has several facets, including the appropriate selection of recipient-subjects and the obtaining of organs. Equitable subject selection for research on transplantation raises unique questions because of the involvement of the donor in the process and because of the scarcity of appropriate materials (e.g., organs, tissue, or bone marrow) for the transplant procedure. The use of fetal tissue in transplantation is dealt with in Guidebook Chapter 6, Section A, "Fetuses and Human In Vitro Fertilization."


Experimental transplants are performed using a number of techniques: An organ or tissue can be obtained from a living relative, a living nonrelative, or a deceased person (usually a nonrelative). Transplants can also be performed using organs or tissue from animals (called xenografts); portions of organs have also been transplanted from living relatives into patient-subjects. The use of artificial implants is another method of replacing diseased organs that has been pursued.

The transplant procedure requires the matching of various factors between donor and recipient (e.g., blood and tissue types). To increase the likelihood of a match (i.e., to decrease the likelihood that the organ or tissue will be rejected by the recipient's system), living relatives are a preferred source of organs or tissue. For some organs, such as a heart, such an arrangement is obviously impossible. Furthermore, the subject may not have a living relative who provides an appropriate match or who is willing to donate the organ or tissue.


Candidates for experimental transplant procedures are usually under threat of imminent death; experimental transplant procedures are a last hope for survival. The highly vulnerable status of potential subjects makes stringent review of proposed transplant research essential. Transplant investigations involving children as subjects are governed by Subpart D of the DHHS regulations [45 CFR 46.401-409]. [See Guidebook Chapter 6, Section C, "Children and Minors."]

The first issue with which IRBs must concern themselves is whether the risk of the transplant procedure is outweighed by the potential benefits of the research. [See Guidebook Chapter 3, Section A, "Risk/Benefit Analysis."] The benefits take two forms: intended therapeutic benefit for the individual subject and the benefit to society from the knowledge gained from the research. An important factor when considering the benefit to individual subjects is the availability and quality of therapeutic alternatives for potential subjects. The subjects' prospects for survival and quality of life, with or without the transplant, will be particularly relevant to the IRB's decision.

Transplants involving living donors present a second level of risk that must be evaluated: the risk of obtaining the organ from the donor. That risk entails the risk of the removal procedure itself, plus the long-term risks of living without the donated organ or tissue. When balancing those risks against the potential benefits, one can see that the relationship of the donor to the recipient may be relevant. The donor will not therapeutically benefit from the donation; quite the contrary. The benefit comes, rather from the direct good the donor gives the recipient. In this regard, the living related donor will benefit more directly than will the living nonrelated donor: He or she is increasing the likelihood that the relative (about whom he or she presumably cares more than would a nonrelated donor) will live longer.

As with any research involving human subjects, IRBs need to ensure that subjects give informed consent that is free from coercion or undue influence. Potential subjects for studies involving experimental transplants must be clearly informed of the highly experimental nature of the procedure, including the state of knowledge about the prospects for long-term viability of the organ or tissue.

Complicating the question of consent when the research involves transplants is the involvement of a donor. Where the donor is living, his or her consent must be obtained; the regulations concerning research subjects apply fully to the donor as well as to the recipient. Where the donor is deceased, his or her next of kin must be consulted: State and federal Required Request Laws mandate that the treating physician ask if the family wishes to donate organs from the patient upon his or her death; the deceased may also have indicated a desire to donate his or her organs in the event of death by, for instance, signing an organ donation card.

Technological innovations that allow for the preservation of cadavers and organs has led to concerns about treating brain dead persons as research objects. Some question exists whether deceased donors come within the jurisdiction of IRBs because the federal regulations define subjects as "living individual[s]" [Federal Policy ___.102(f)]. Nevertheless, the President's Commission [(1983), p. 41] suggested that IRBs consider requiring review of research on brain dead persons "to determine is consistent with 'commonly held convictions about respect for the dead.'" [See also Levine (1986), p. 78.] Considerable controversy surrounds the use of anencephalic infants as a source of organs for donation, with most commentators arguing against their use.

The involvement of living related donors also raises concerns of coercion and undue influence. The pressure on relatives to donate needed organs or tissues is unquestionably great; IRBs must carefully scrutinize the proposed consent process. Some investigators have provided for both medical and psychiatric evaluations and counseling as part of the donor consent process, as well as a waiting period (if feasible) before the transplantation, during which the donor may withdraw consent. Some investigators have also provided for a consent advocate for the donor who is not directly involved in the donor's operation. [See, e.g., Singer, et. al. (1989).]

A further complication to the consent process for organ donors is the minor who is a potential donor for a relative a sibling, for instance. Where the donor is a minor, the regulations concerning children and minors as research subjects apply [45 CFR 46.401-46.409]. Organ donations from minors raise concerns about the ability of the minor to comprehend the risks of donation, as well as the possibility of coercion or undue influence. [See Guidebook Chapter 6, Section C, "Children and Minors."] IRBs may want to consider requesting the guidance of a court of law before allowing a given donation to be made.

Experimental xenografts have been particularly controversial. The celebrated Baby Fae case, in which an infant received the transplanted heart of a baboon, raised serious questions about IRB review of research involving human subjects. Any research involving transplants should be carefully reviewed by an IRB regardless of the source of funding. The extremely risky nature of the procedure and the special vulnerability of the subjects demand that their welfare be scrupulously protected. Subjects must be clearly informed of the state of knowledge about the long-term viability of the transplant, of alternatives to the procedure, and of all possible physical and psychological effects that may result from the transplant and any other procedures that will be undertaken as a part of the transplant. Consent to the transplant must be carefully documented. [See Caplan (1985), p. 3343].


1. Does the consent process adequately protect both the donor and the recipient? Is sufficient information provided regarding the risks of all procedures involved? Is adequate provision made for incompetent subjects by providing for trustworthy proxy decision makers?

2. Have both donors and recipients been adequately protected against coercion and undue influence?

3. Are special regulatory provisions applicable, e.g., Subpart D governing children as subjects?


Federal Policy ___.111(a)(3) [Criteria for IRB approval of research: equitable selection of subjects]

Omnibus Budget Reconciliation Act of 1986 (Pub. L. 99-509) enacted sec. 1138, Social Security Act (Required Request Law)

The Uniform Anatomical Gift Act

The Uniform Definition of Death Act

State and local laws pertaining to organ donation

Return to Top of Page