Undesirable implications of disclosing individual genetic results to research participants.

Ravitsky and Wilfond (2006) correctly highlight the need for uniform guidelines to assist investigators and institutional review boards (IRBs) in determining whether to disclose individual genetic results to research participants. However, their results-evaluation framework, like other proposals that call for the disclosure of individual results to research participants (Partridge and Winer 2002; Rothstein 2006; Shalowitz and Miller 2005), is troublesome for two related reasons. First, their justification for disclosure rests on the mistaken view that principles of beneficence, respect, reciprocity, and/or justice ethically require researchers to offer participants individual genetic results. Whereas these principles and others obligate physicians to share individually relevant results with patients in the clinical care setting, they do not similarly obligate investigators to share such information with participants in the research setting. Second, because Ravitsky and Wilfond’s proposal conflates the aims of clinical research with those of clinical care, participants may suffer from a therapeutic or diagnostic misconception (Clayton and Ross 2006), researchers may be inclined to overstate the benefits of enrollment, and institutional review boards (IRBs) may face unforeseen difficulties in assessing trial risks and benefits. Beneficence, respect for persons, reciprocity, and justice are important components of ethical research, but none of these principles are violated or diminished when investigators do not share individual genetic results from research. Beneficence requires investigators to maximize possible benefits to participants and society at large while minimizing possible harms. Whether a clinical trial meets the requirement of beneficence is determined ab initio by an IRB before any participants are enrolled. At this stage of review, an IRB cannot weigh the possibility that some participants may gain individually useful results from participation. There is no guarantee that any given trial will produce such results. In contrast, if a study does generate individual information, it may apply only to participants with certain genomic characteristics. And even in this case it is not possible to determine whether a given participant will perceive the information as a benefit or harm once confronted with it. Because a clinical trial must meet the requirements of beneficence regardless of whether it returns individual research results, it cannot be the case that beneficence requires investigators to offer results. Ravitsky and Wilfond (2006) may respond that failure to disclose genetic information that is of direct clinical utility to participants violates nonmaleficence, the duty to minimize harms in research. However, except in the very rare circumstance in which failure to disclose individual genetic information to a participant poses a certain and near-immediate threat to morbidity (and thus should be disclosed), nonmaleficence does not trigger disclosure. Investigators have a duty not to expose participants to undue risk. This negative duty of nonmaleficence is quite different, however, from any positive duty to promote participants’ interests, the activity that seems to motivate Ravitsky and Wilfond’s proposal. They may reply that failure to offer beneficial results is itself a harm, but this understanding of harm as “failure to benefit” would, as previously noted, pose intractable difficulties for IRBs assessing probable harmbenefit ratios for research trials (Parker 2006). Ravitsky and Wilfond’s (2006) argument that respect for persons requires disclosure is similarly mislaid. Respecting participants means treating human beings capable of self-determination as autonomous agents. It requires researchers to present prospective participants with adequate information about a trial, to honor their voluntary decision about enrollment, and to accept participants’ decisions to withdraw from trials should they so choose. But respect for persons does not require researchers to account for participant interests or preferences about disclosure of individual genetic research results when designing a study. Investigators best respect participants as persons when they disclose the terms of the research and allow people to choose whether to enroll. This approach, rather than integrating participant preferences into research, is what The Belmont Report and the Common Rule mean by respecting participant selfdetermination. Just what Ravitsky and Wilfond (2006) envision when they claim that reciprocity in research requires disclosure is somewhat unclear. They contend that reciprocity requires researchers to consider the nature of their relationship with participants when deciding to disclose results, paying particular attention to “the duration and the intensity of this relationship” (2006, 8). This approach to reciprocity is somewhat more nuanced than the standard call for participants to receive something from researchers in exchange for participation. If researchers were to disclose individual results to participants, then arguably the degree