The conduct of clinical research is guided by three ethical principles set forth in the Belmont Report: respect for persons, beneficence, and justice.³ Articulated more than 20 years ago, the Belmont principles remain the paradigm for ethical problems in research. Accordingly, they are mirrored in the U.S. federal Common Rule,⁴ the Council for International Organization of Medical Sciences (CIOMS) international research guide-lines,⁵ and, in a somewhat expanded form, the Tri-Council Policy Statement. The principle of respect for persons requires that the wishes of autonomous persons be taken seriously, and those who are not autonomous are protected. The principle finds expression in requirements for informed consent and confidentiality. The principle of beneficence is typically expressed in terms of two complementary rules: (1) do no harm, and (2) maximize potential benefits while minimizing the risks to subjects. The principle is applied in the demand that a study present an acceptable balance of potential benefits and risks. The principle of justice entails that the benefits and burdens of research participation be distributed equitably. Accordingly, the REB must ensure that procedures for the selection of subjects are equitable.

Some Canadian researchers have commented recently that risk is an inherently biomedical concept that does not apply across the spectrum of research now captured in the Policy Statement. Undoubtedly, this is at least in part a reflection of the fact that, for the first time in Canada, a single ethics document governs biomedical, social sciences, and humanities research. But is there more to it than this? Let us consider three diverse examples of research studies.

Study A: placebo controlled trial of a drug for people with acutely-symptomatic schizophrenia. Chouinard reports a study on schizophrenic patients who are newly hospitalized with acute symptoms of their disease.⁶ Despite the existence of effective treatment for such symptoms, patients are randomized to a new antipsychotic drug (remoxipride), a standard drug (chlorpromazine), or placebo. Patients are treated in hospital for four weeks and a variety of psychometric scales are measured. A number of risks are presented by this study to research subjects: the new medication may have serious adverse effects, some may be irreversible; patients assigned to placebo (or the new drug if it proves to be ineffective) will be deprived of needed treatment for a month; patients may suffer from hallucinations or paranoia; they may be at increased risk of suicide; and, finally, they may pose a risk to others. The ethics of placebo controlled trials in schizophrenia is discussed in detail elsewhere.⁷

Study B: hypnotic induction of partial deafness to see whether paranoid symptoms result. Zimbardo and colleagues report a study in which a group of hypnotically-suggestible and otherwise healthy college students were randomized to three different hypnotic suggestions: partial deafness without awareness of the cause; partial deafness with awareness of the cause; and itchiness of an ear.⁸ The hypothesis was that persons in the first group would demonstrate symptoms of paranoia. Subjects were assessed by a variety of measures, including psycho-metric scales and scoring of observed behavior. Subjects were rehypnotized, debriefed at the end of the study, and reassessed at one month. The study posed a variety of risks to participants, including distress associated with paranoia and hearing loss, risk of suicide, the possibility of harm to others, and uncertain sequelae from hypnosis. Some of the ethical issues raised by this study are discussed elsewhere.⁹

As illustrated by these three examples, research participation may expose the study participant to a wide spectrum of risks. Risks are classified by Levine into four categories: physical, psychological, social, and economic.¹¹ Let us consider each briefly:

• Physical risks: the research subject may suffer bodily harm — minor or serious, temporary or permanent, immediate or delayed as a result of the study.
• Psychological risks: study participation may impact upon the research subject’s perception of self, cause emotional suffering, e.g., anxiety or shame, or may induce aberrations in thought or behavior.
• Social risks: research findings, or even study participation itself, may expose subjects to the possibility of insurance or employment discrimination, or other forms of social stigmatization.
• Economic risks: research subjects may directly or indirectly bear financial costs related to research participation.

So defined, risk is an inherently inclusive concept. As demonstrated by the above examples, a given study may present a variety of types of risk. For example, Study C (sex questionnaire) posed both psychological and social risks. Furthermore, no category of risk is exclusive to medical or non-medical research: Study B (deafness and paranoia) presented physical risks and Study A (schizophrenia trial) generated psychological risks. Finally, it is worth noting that despite the various disciplinary backgrounds involved, all three of the study examples posed non-trivial risk to research subjects.

When a study involves therapeutic procedures, the REB must ascertain that a state of clinical equipoise exists (TCPS page 7.1). Clinical equipoise exists when there is a state of uncertainty in the expert clinical community as to the comparative merits of the study treatments.¹³ In other words, there must be genuine uncertainty as to the preferred treatment. A trial is initiated to resolve this uncertainty. Clinical equipoise does not require that treatments present equal risk. Indeed, an experimental treatment may pose greater risk than standard treatment, so long as it also offers a greater prospect of benefit. In short, the ethical analysis of therapeutic procedures involves a risk-benefit calculus.

Non-therapeutic procedures, by definition, do not offer any prospect of benefit to the research subject and, thus a risk-benefit calculus is inappropriate. Two ethical requirements must obtain if non-therapeutic procedures are to be deemed acceptable (TCPS page 1.5). First, the risks associated with such procedures must be minimized. If the information can be obtained in a less risky way, through ‘piggy backing’ on other procedures or using available information, then this must be done. Second, the risks posed by such procedures must be proportionate to the knowledge that may reasonably be expected to be gained from the study. Thus, the ethical analysis of non-therapeutic procedures does not involve a risk-benefit calculus, rather a risk-knowledge calculus.

Many studies in the social sciences and humanities will not involve therapeutic procedures, and in these cases, the REB’s attention should focus on the analysis of non-therapeutic risks. When a research protocol involves both therapeutic and non-therapeutic procedures, the results of both the risk-benefit calculus and the risk-knowledge calculus must be favorable. A study may not counter-balance experimental therapy that is ab initio known to be inferior to standard treatment with claims that important knowledge may result.

Minimal risk "means that the risks of harm anticipated in the proposed research are not greater, considering probability and magnitude, than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests" (TCPS page 1.5).¹⁴ The definition has two parts: the first defines minimal risk as risks comparable to those "encountered in daily life;" the second gives an example of such risks, namely, risks encountered in "routine physical or psychological examinations or tests." Importantly, minimal risk only applies to the risks posed by non-therapeutic procedures.

The risks of daily life are familiar to us all. While it may be true that it is difficult to quantify the precise risk of disabling injury or mortality associated with, for example, driving to the grocery store or playing catch, we can nonetheless identify them as risks encountered in daily life. Properly understood, minimal risk is a categorical or qualitative determination made by the REB. As Freedman and colleagues observe: "We are, by definition, each acquainted with them; and, almost by definition, if we are unsure whether they belong within the set of common risks, then they don’t."¹⁵ The reasoning process to be used by the REB in the minimal risk determination is analogical. Research interventions may be determined to be minimally risky because either the procedure is in fact encountered in daily life or it is sufficiently similar to those routinely encountered.

It would be very useful to have lists of procedures that clearly present only minimal risk. A note of caution must be introduced here, however. Minimal risk refers to the sum total of risk posed by all the non-therapeutic procedures in a research study. Thus, a given procedure in isolation may be minimally risky, but in combination with other procedures, it may present more than minimal risk. Furthermore, the concept of risks of daily life is inherently flexible; its determination depends whose daily life we are talking about. The life experience of a diabetic is different from that of a cancer patient, and the life experience of a cancer patient is different from that of a healthy person. Certain procedures may be commensurate with the experience of one group of people but not others.

With these provisos in mind, at least one list of prima facie minimally risky procedures is available in the U.S. regulations. 4 Some of the procedures that may present minimal risk are as follows:

• Collection of hair and nail clippings;
• Collection of excreta and external secretions;
• Recording of data using non-invasive procedures routinely employed in clinical practice, e.g., ECG, EEG, and echocardiography;
• Venipuncture (not more than 450 ml in eight weeks and not more than two times per week);
• Collection of supra-and sub-gingival dental plaque and calculus;
• Voice recordings for research purposes such as investigation of speech defect;
• Moderate exercise by healthy volunteers; and,
• Research on individual or group behavior or characteristics of individuals, such as studies of perception, cognition, game theory, or test development, where the investigator does not manipulate subjects’ behavior and the research will not involve stress to subjects.

It is worth noting that none of the three studies we discussed obviously present only minimal risk to research subjects. Thus, the REB should avoid the erroneous presupposition that all social science and humanities research involves only minimal risk.

As suggested, the implications of minimal risk in the Tri-Council Policy Statement are numerous. Minimal risk serves at least two roles in the Policy Statement. It is used as a mechanism to focus the REB’s attention on studies that present more non-therapeutic risk:

• social science and humanities research that presents minimal risk need not be peer-reviewed (article 1.5);
• minimal risk is identified as the foundation of proportionate review and eligible protocols may be reviewed in an expedited fashion by the REB (article 1.6);
• research that poses minimal risk requires only a "minimal [continuing] review process" (article 1.13);
• naturalistic observation should usually be regarded as minimally risky (article 2.3);
• REBs should consider having a person independent of the research team obtain consent when more than minimal risk is present (article 2.8); and
• secondary use of data presenting more than minimal risk may require informed consent or community consultation (article 3.4) .

Minimal risk is also used to set a threshold for risk to which vulnerable populations may be exposed:

• some or all of the elements of consent may be waived for an eligible study (article 2.1c); and
• persons incapable of giving consent, including children and incompetent adults, may not be enrolled in studies that involve more than minimal risk (article 2.5).

Clearly, the challenges presented to the REB by risk analysis are substantial. In addition to NCEHR’s regional REB education workshops, individual institutions will need to undertake considerable educational efforts to ensure that REB members are fluent with the concept of risk and associated analytic techniques. As REBs face this challenge, it is imperative that they communicate their successes and failures with one another through forums such as NCEHR’s e-mail discussion group. Challenges are far better faced together than in isolation.