ALSO IN THIS ISSUE
Experts Examine Vulnerability and Risk in Behavioral Research
On November 9-10, 2007, the APA Science Directorate and Columbia University co-sponsored a working meeting on Defining Vulnerability in Minimal Risk Behavioral and Psychological Research. The goal of the meeting, held at the Columbia University in New York City, was to engage relevant stakeholders in the research enterprise -- investigators, experts on Institutional Review Board (IRB) and human research protection issues, and regulatory officials -- in an effort to systematically explore the concept of vulnerability in behavioral and psychological research. The workgroup was jointly chaired by André Ivanoff (Columbia University School of Social Work) and Barbara Stanley (New York State Psychiatric Institute/Columbia University & John Jay College, CUNY), both experts on IRB issues and active researchers at Columbia University, and Sangeeta Panicker, Director of the APA Research Ethics Office.
Although much research in the behavioral and psychological sciences can be characterized as posing minimal (or even less than minimal) risk to the participants, IRBs often find the regulatory definition of “minimal risk” difficult to apply in a consistent manner to behavioral and psychological research protocols. Furthermore, because of the lack of guidance for making this determination, the definition is more open to interpretation by individual IRBs. Anecdotal evidence suggests that IRBs often overestimate risks in behavioral and psychological research. Such overestimation of risk subjects behavioral research protocols to unnecessary and time-consuming full panel IRB review. In addition, a conservative approach to research risk can lead to implementation of participant protections that may be inappropriate or even harmful, dilute the scientific validity of the research, or discourage behavioral research involving certain populations.
IRBs frequently confounded assessment of research risks with judgments regarding the vulnerability of the populations under study. Vulnerability has typically been analyzed within the framework of research participants’ inability to make decisions that are in their own best interests (e.g., cognitively disabled individuals, children) and/or a power differential that subjects participants to potential coercion and undue influence (e.g., students, employees, prisoners). Regulations currently call for additional protections for “vulnerable” research participants; these may include such protections as full IRB review, surrogate consent, and disallowing waivers of consent. In other words, in order to qualify for expedited IRB review or alteration or waiver of consent requirements, the research should entail only minimal risk and not involve a population deemed “vulnerable.”
Typical usage of the term “vulnerability” in research conflates one or more of the following participant characteristics: (1) susceptibility to coercion, (2) cognitive disability leading to lack of capacity to make autonomous decisions, and (3) susceptibility to potential harms. The current usage of the term also raises the issue of whether the concept should be applied to entire populations or whether research participants should be assessed on a case-by-case basis. While the former application makes regulatory control possible (as detailed in current Department of Health and Human Services regulations for Protection of Human Research Subjects, 45 CFR 46 Subparts B, C, and D), the latter would fall outside the regulatory structure and would rely on the professional judgment of investigators and IRBs.
At the meeting, participants discussed the importance of a clearer definition of vulnerability in research; types of vulnerabilities, such as individual characteristics, setting, temporal, and relationship that may be relevant in the research environment; whether IRBs should evaluate vulnerability at the population level or at an individual level; whether IRBs should consider research to involve vulnerable participants only when participants are targeted due to their membership in a vulnerable group (e.g., pregnant women) or also when a study has a significant likelihood of including vulnerable participants (e.g., the occasional pregnant woman); and finally whether IRB deliberations about vulnerability should be separate from assessments of risk. Guiding the meeting’s discussions were the 1979 Belmont Report, “Ethical Principles and Guidelines for the Protection of Human Subjects of Research,” the basis for current human research protection regulations; the current regulations themselves; and recent scholarship on research risk and vulnerability.
The discussions at this working meeting, which may serve to stimulate further examinations of the topic as well as inform development of new guidance on human research protections, will be published in a series of articles currently in preparation. “We hope that the ultimate outcome of this meeting is to aid IRBs to assess the risks of behavioral research more accurately, to provide guidance to investigators to help them communicate more effectively about risks of behavioral research, particularly minimal risk research, to ensure protections for all participants, and promote research that is both ethically and scientifically sound,” says Stanley, one of the meeting organizers.
Other participants of the workgroup included Tracy Arwood (Clemson University), George Bigelow (Johns Hopkins University), Geraldine Downey (Columbia University), James DuBois (Saint Louis University), Cynthia Gomez (San Francisco State University), Dale Hammerschmidt (University of Minnesota), Mary Ann McCabe (Society for Research on Child Development), Jamie Ostroff (Memorial Sloan-Kettering Cancer Center), Ivor Pritchard (Office for Human Research Protections), Wendy Rogers (Georgia Institute of Technology), David Strauss (Columbia University), Cathy Widom (John Jay College, CUNY) and Philip Yanos (John Jay College, CUNY).
The genesis of this meeting was a previous working meeting convened in April 2005, cosponsored by the APA Science Directorate and the Fordham University Center for Ethics Education, which focused on operationalizing the regulatory definition of “minimal risk” in research with normal populations. The goal of the 2005 meeting was to develop a tool that would enable IRBs to more accurately identify and classify minimal risk research protocols. The discussions at that meeting raised important questions about the inclusion of populations that might, in some cases, be considered “vulnerable” in research that would otherwise be considered minimal risk. In other words, the research is regarded as “minimal risk” if some participants or populations with a perceived “vulnerability” were excluded.