Protocol Design - Identification and Quantification of Research Risk

Definition of Research Risk

Research risks are potential events that might cause temporary distress, damage, or discomfort or that might result in more serious, long-term consequences for research subjects. The word “risk” in the context of human subjects research is shorthand for “risk of harm.” The degree of risk is defined by its magnitude (the degree of harm that can be anticipated) and probability (the likelihood that a particular harmful event might occur). It is further defined by its reversibility (e.g., with appropriate treatment). The types of risks, or potential harms, that may befall research subjects are varied. Harms can be categorized broadly as physical, psychological, social, economic, and legal. Research subjects may suffer bodily harm (physical); have their perception of self, their thoughts, or their behavior altered, or in some other way suffer emotionally (psychological); suffer employment discrimination or other forms of social stigmatization (social); directly or indirectly bear financial costs related to research participation (economic); or in the course of participation have illegal activities revealed (legal). Harm may be experienced by the subject, as well as by others not directly involved in the research, such as the subject’s family, social groups, and society in general.

Identification of Research Risk

Investigators must scrutinize their studies to identify opportunities for subjects to be exposed to risks (as a consequence of participation in the study) and must include a description of these opportunities in their consent forms. Investigators must also present a comprehensive review of these risks in their protocols. This compilation of risks is generally assembled following an adequate review of the literature. An adequate review is one that a peer would judge sufficient to properly identify the research risks given the nature of the research. “First-of-its-kind” research that could place subjects at considerable risk would require an extensive review, while survey-based research that is similar to previous studies would require a more modest review. The risks to be disclosed during the consent process should include at least those that a “reasonable volunteer” would find “material” to his or her decision as to whether to agree to participate in the research. All reasonably foreseeable risks should be disclosed. As investigators gain knowledge and experience in a research area, the risks considered reasonable and foreseeable will change, and disclosures in protocols and consent forms should be altered to reflect this new knowledge.

Quantification of Research Risk

Once identified, risk must be quantified. Risk quantification considers both the likelihood of occurrence and the potential severity of the harm. Severity depends on both the magnitude and the duration or permanency of the damage or harm. A frequent error in analysis of research risk is to focus on the severity of the outcome and overlook its probability of occurring. Just as the potential for two allergic reactions, a skin rash and anaphylactic shock, should be handled very differently, so should a one-in-a-million risk of serious injury be handled very differently from a risk of one in a thousand. Protocols should describe, to the degree possible, the odds of a subject experiencing a harm, based on information available in the literature as well as the investigator’s knowledge and experience. Information concerning the particular procedure or treatment to be utilized in the protocol or extrapolated from that concerning an analogous procedure or treatment can be used by investigators to determine the risks subjects may face. Accurate characterization of research risks by investigators must be performed to ensure that risks are minimized, that a favorable risk-benefit balance exists for the subjects, and that the risks of participation are adequately addressed in the informed-consent document. However, in some instances the quantification of risk is difficult, if not impossible, because there are gaps in knowledge of the risks associated with a procedure or treatment. In other cases, the potential harm can be identified but the probability of occurrence is unknown. In all cases the investigator and the IRB must assess the risk to the best of their ability, making best estimates when faced with uncertainty. When reviewing protocols, IRBs must assess whether an investigator has presented adequate data to allow for evaluation of his or her risk-benefit analysis.

Audio Extra