You are Here: Home Page > Safeguarding Healthy Research Subjects > Safeguarding Healthy Research Subjects: Protecting Volunteers from Harm  

Safeguarding Healthy Research Subjects: Protecting Volunteers from Harm

RISK-BENEFIT ASSESSMENT

Under the federal regulations governing human subject research (the “Common Rule”),¹² an IRB may approve a research protocol only if it determines that any risks associated with the research are minimized and that the risks are reasonable in relation to the anticipated benefits.¹³ To fulfill this requirement, IRBs must engage in a three-step process. First, they must identify the risks associated with the research. Next, they must determine the benefits likely to be achieved by the research. Finally, they must evaluate the risks and benefits, to determine that the risks are minimized and that they are reasonable in relation to the benefits likely to be achieved.

STEP 1: IDENTIFYING RISKS

In determining the risks and benefits of biomedical research involving normal healthy subjects, IRBs must recognize the specific meaning of the concepts of “risk” and “benefit” in the context of human subject research.

• Magnitude of harm — The magnitude of harm refers to both the severity and the duration of any possible harm. For example, the danger of an indwelling catheter in a central blood vessel is greater than the danger of a venipuncture to sample blood from a peripheral blood vessel. Similarly, the scar from taking a biopsy may persist for a lifetime; a bruise from a venipuncture may last a few days. In some cases, an antidote or intervention may remedy or alleviate the harm. For example, transient lightheadedness after several tubes of blood are drawn may be momentary and easily remedied, while the headache resulting from a lumbar puncture may be more difficult to relieve.
• Probability of harm — The probability of harm refers to the measurable likelihood that harm will occur. Determining the probability of harm often is complicated by (1) the unpredictable risks of some drugs, biological agents, or procedures; (2) the idiosyncratic responses of some individuals (for example, a subject may experience an anaphylactic reaction to a drug); or (3) the reality that some “healthy” subjects may have unknown or undisclosed health conditions that put them at increased risk (for example, a subject may have an eating disorder and chronic electrolyte imbalance, which could cause a dangerous cardiac arrhythmia). (See page 29 regarding researchers’ obligations to elicit such information from subjects.)

IRBs also should consider the effect of additive and cumulative risks associated with participation in biomedical research.

• Additive risks — The IRB should consider each element of risk in a protocol separately. For example, a procedure such as bronchoscopy entails certain risks from the physical intervention and separate risks from the anesthesia that is used to make the procedure more tolerable. In addition, if the interaction between the two risks is likely to be greater than the sum of their parts, the IRB should include this fact in the risk-benefit calculation. For example, in a Phase One trial of two new drugs expected to potentiate each other in preventing replication of a dangerous virus, there is a possibility that both drugs together could aggravate their known side effect of reducing the subject’s white blood cells, thereby reducing immunity.
• Cumulative risks — IRBs also should be aware that there may be cumulative risks involved in participating in sequential protocols or more than one protocol at the same time (for example, cumulative exposure to X-rays). IRBs should alert researchers to the possibility of cumulative risk and, when appropriate, require researchers to provide procedures for avoiding such risks. IRBs should especially ensure that researchers consider the cumulative risks involved in utilizing subjects whose occupation is, in part or whole, being a research subject. (See page 30)

All elements of the protocol should be subjected to risk analysis. The protocol should include a description of all interventions that a subject will experience. This includes interventions that are experimental and part of the research design as well as those ancillary interventions that are not experimental but are essential for accomplishing the study.

Discomfort and/or inconvenience often are reasonably foreseeable consequences of research for some or all subjects. In this context, the term “discomfort” refers to unpleasant sensations or emotions of short duration and minimum to moderate severity. Examples include shortness of breath in a study of maximal exercise tolerance or the anxiety associated with being stuck with a needle. The term “inconvenience” refers to any interference with the subject’s ability to carry out usual activities. Examples include time spent travelling to the research facility or the need to get up in the middle of the night to take a pill.

When researchers anticipate that subjects may experience discomfort or inconvenience, that fact should be made explicit in the protocol provided to the IRB for review. In addition, the possibility of discomfort or inconvenience should be disclosed to potential subjects as part of the process of obtaining informed consent. (See page 16)

In most cases, the discomfort and/or inconvenience associated with research does not constitute a significant “risk” to subjects, as that term is defined above, and is therefore not weighed heavily in the IRB’s consideration of whether a protocol should be approved.¹⁶ In some types of research, however, tolerance for discomfort and inconvenience may vary considerably, causing what may be perceived as ordinary discomfort or inconvenience by some subjects to escalate to significant harm for others.¹⁷ Examples include a bronchoscopy or bone marrow biopsy, which may be experienced as unpleasant by some subjects and as severe discomfort by others, or a sleep study that reverses day and night, which upon completion may require no readjustment by some subjects and a psychologically difficult readjustment by others.

IRBs should always be cognizant of these possibilities. In addition, when researchers anticipate that an intervention may cause greater-than-normal distress for some subjects, they should inform IRBs of this fact. IRBs should consider this information in their assessment of the protocol’s risks and ensure that such risks are explained in the informed consent process.

The IRB should distinguish between situations where risks are known and situations where the probability and/ or nature of the possible harm is uncertain, and it should ensure that the informed consent process makes this distinction clear. An example of a known risk is the complication rate of a well-established diagnostic procedure, such as lumbar puncture. An example of an uncertainty in research is the potential for adverse effects from a new drug.

STEP 2: DETERMINING BENEFITS

IRBs should recognize that there are certain aspects of research that subjects are likely to perceive as benefits but that do not fall within the definition of “benefit” set forth above. For example, subjects may derive altruistic satisfaction from the expectation that others, including family and friends, may benefit from the knowledge that might be gained from the study. In addition, subjects may benefit financially from participating in the research.¹⁸ While some subjects may attach considerable importance to these factors, they do not constitute the type of “benefit” that IRBs should consider in evaluating the risk-benefit ratio of a protocol.¹⁹

STEP 3: EVALUATING RISKS AND BENEFITS

Unnecessary risks to subjects, no matter how small, can never be justified. Accordingly, IRBs should ensure that researchers minimize risks. To fulfill this obligation, IRBs should inquire about the following:

• Necessity of proposed procedures — Are the procedures proposed as part of the research design necessary to the question under investigation?
• Safety of proposed procedures — Is there a safer way to carry out the proposed procedure? Does the equipment that will be utilized meet safety standards? Will procedures be performed by competent and properly qualified individuals? Investigators are responsible for identifying the safest method of conducting their proposed research. If the investigators know or learn of a potentially safer method than the one proposed, even if it requires substantially more costly equipment or procedures, they should provide that information to the IRB along with a discussion of the advantages and disadvantages of the alternative methodology.
• Number of subjects exposed — Have the researchers ensured that the fewest people necessary will be exposed to any given risk? To answer this question, statistical tools such as power analysis should be considered by the IRB in the approval process.

Because knowledge is the sole benefit against which risks are balanced, the IRB should determine not only that the knowledge sought by the research is important and otherwise unavailable (i. e., that the desired knowledge is worth the risk even if that risk is very small) but also that the proposed research will contribute to obtaining such knowledge. Among other factors, the IRB should consider the strength of the research design and the likelihood of recruiting sufficient numbers of appropriate subjects to reach a valid conclusion in a reasonable period of time.

The process of assessing risks and benefits for normal healthy subjects is analogous to comparing apples to oranges. Because risks and benefits cannot be balanced on the same scale, the IRB needs to exercise caution when reviewing any risk-laden protocol. Where there is significant risk and where the IRB is uncomfortable with that risk, the researchers should be encouraged to return to the IRB with enhanced protections or alternative research designs.

When performing interventions in the clinical setting, practitioners sometimes depart from standards suggested or established in the professional literature when doing so is anticipated to promote the patient’s health or well-being. For example, when performing a bronchoscopy to establish a tissue diagnosis in a patient with suspected cancer, a physician may administer a dose of anesthesia that is greater than that recommended in the professional literature, if the patient cannot tolerate the intervention with anesthesia at the recommended dose and a tissue diagnosis is needed to determine an appropriate course of treatment. Provided the dose administered is based on a sound clinical rationale (for example, the anecdotal experience of well-regarded practitioners in the field), the benefit to the patient of completing the bronchoscopy may justify the physician’s decision to exceed the standard dose.

When research interventions are performed on subjects who do not stand to benefit directly, however, departing from standards established in peer-reviewed published literature is more problematic. If researchers believe that an intervention cannot be performed successfully without departing from the standards recommended in the scientific literature — whether increasing the dose of a drug, extending the duration of an invasive procedure, or renewing attempts to carry out an intervention after repeated failures — they should describe their justification for departing from established standards in the protocol submitted to the IRB and seek specific IRB approval for doing so. The IRB should carefully review the nature and degree of risk that can reasonably be expected. In some cases, the IRB may conclude that the intervention should be performed only in the manner supported by published data.

In all cases, the IRB should rigorously scrutinize the risks and benefits of all protocols involving normal healthy subjects and ensure that the risks are not out of keeping with the potential scientific gain. Thus, badly designed research that will produce no useful results can support no risk — not even minimal risk. Research likely to produce significant knowledge that could not otherwise be obtained might justify substantial, although not unlimited, risks. As the risk increases, the importance of the knowledge to science and to society must increase accordingly to justify the IRB’s approval.

If the level of risk is high, the IRB should scrutinize closely the incentives offered to potential subjects, regardless of the benefits of the study. Incentives should not be so great as to blind the subject to the degree of risk. (See page 27)

Some groups, such as the British Royal College of Physicians, have concluded that it is unacceptable to conduct research with normal healthy subjects if the research involves greater-than-minimal risk.²¹ For two reasons, the present guidelines do not set a maximum level of acceptable risk. First, personal autonomy should enable an individual to choose to participate in a scientifically valid more-than-minimal-risk study as long as the risks are disclosed to the individual, the individual understands the extent of those risks, any incentives offered are appropriate, and the individual provides informed consent. ²² Second, an absolute prohibition of greater-than-minimal risk research with normal healthy subjects might simply encourage the characterization of all risks as minimal. A process that assesses, minimizes, and explains the nature and probability of the risk is far more useful to subjects and the enterprise of research than an across-the-board prohibition of research involving greater-than-minimal risk.

One limit to risk that applies to all protocols, regardless of the degree of benefit likely to be achieved, is that researchers may not deliberately inflict a lethal or disabling injury upon subjects. This principle, grounded in the Nuremberg Code, requires IRBs to reject any protocol involving a certainty or high probability of serious physical harm, regardless of the potential for acquiring scientifically valuable data.²³

In general, IRBs should seek membership to achieve the broadest possible range of experience. Many IRBs find it helpful to have a primary reviewer who has directly relevant knowledge about the research and whose responsibilities include talking with the investigator to have questions answered and contacting other experts as needed. Institutions should ensure that IRBs have sufficient resources to obtain expert consultations as needed. (See page 32)

In addition to the review and approval of each protocol and informed consent process prior to a study’s initiation, the IRB must periodically review all ongoing research. Federal regulations require review at intervals appropriate to the degree of risk but not less than once a year.²⁵ It is critical that protocols with no prospect of direct benefit to subjects are reviewed frequently to ensure that there are no increases in risk or reported adverse events, as well as to ensure that the research still holds the prospect of significantly enhancing knowledge. At a minimum, researchers should prepare an annual progress report for the IRB. If the study is longstanding and has undergone a number of protocol changes, it is essential that the IRB review the protocol as a whole to assess risks and benefits.

The IRB should

• establish a schedule for the frequency of continuing review that takes into account the risk-benefit assessment of a given protocol
• require that researchers report any and all changes in protocol to the IRB, regardless of whether researchers view the changes as involving a change in the level of risk to subjects
• provide guidance to researchers on specific information to be included in annual reports, e. g., a summary of research procedures, changes in the protocol, adverse events, and findings over the entire duration of the study

¹² NIH Office for Protection from Research Risks, Institutional Review Board Guidebook (Bethesda, MD: National Institutes of Health, 1993), A4-12. ¹³ 45 C. F. R. § 46.111( a)( 1)-( a)( 2) (1998); 21 C. F. R. § 56 (1998).
¹⁴ Institutional Review Board Guidebook, 3-1.
¹⁵ Ibid., 3-3, 3-5.
¹⁶ R. J. Levine, “Clarifying the Concepts of Research Ethics,” Hastings Center Report 10, no. 3 (1979): 21, 24.
¹⁷ J. E. Sieber, “Ethical Considerations in Planning and Conducting Research on Human Subjects,” Academic Medicine 68 (1993): S9, S12-S13.
¹⁸ For additional discussion of the motives that prompt healthy individuals to participate in research, see A. Ferriman, “Why I Am a Guinea-Pig,” Daily Telegraph, January 28, 1994, 13; L. Petrecca, “For Students, These Lab Tests Pay,” USA Today, March 13, 1995, 6D; L. T. Koritz, “I Was a Human Guinea Pig,” The Saturday Evening Post, July 25, 1953, 27; A. Shuster, “Why Human Guinea Pigs Volunteer,” New York Times Magazine, April 13, 1958, 62; H. Simons, “They Volunteer to Suffer,” The Saturday Evening Post, March 26, 1960, 33.
¹⁹ 45 C. F. R. § 46.111( a)( 2) (1998); 21 C. F. R. § 56.111( a)( 2) (1998).
²⁰ Institutional Review Board Guidebook, 3-1.
²¹ Royal College of Physicians, “Research on Healthy Volunteers,” 245; see also Advisory Committee on Human Radiation Experiments, “Research Ethics and the Medical Profession: Report of the Advisory Committee on Human Radiation Experiments,” Journal of the American Medical Association 276 (1996): 403, 407-408 (“ nontherapeutic research that puts subjects at significant risk is rarely justifiable”).
²² A. Jonsen, “The Ethics of Using Human Volunteers for High-Risk Research,” Journal of Infectious Diseases 160 (1989): 160, 205-208 (“ In our culture we place high value on the autonomy of the individual. Personal autonomy, interpreted in many different ways, has become the linchpin of our ethics: The actions of a freely consenting and uncoerced person are right, provided those actions do not infringe upon the free actions of another.”).
²³ The Nuremberg Code, reprinted in B. R. Furrow et al., Health Law: Cases, Materials and Problems, 3d ed. (St. Paul, MN: West, 1997), 1205-1206; G. Annas and M. Grodin, “Legacies of Nuremberg: Medical Ethics and Human Rights,” Journal of the American Medical Association 276 (1996): 1682.
²⁴ 45 C. F. R. § 46.107( f) (1998); 21 C. F. R. § 56.107( f) (1998).
²⁵ 45 C. F. R. § 46.109( e) (1998); 21 C. F. R. § 56.109( f) (1998).