Ethically Speaking

One of the most resilient debates in science, medicine and ethics surrounds best practices for handling incidental human research findings that may be clinically significant. Whether in imaging, genetic testing, behavioral studies or a combination of modalities of inquiry, at the heart of the debate about these findings are questions about the blurring of boundaries between research and clinical practice and the risk of causing human subjects anxiety when a researcher finds or suspects an abnormality that may be benign or ultimately prove to be a false positive.

Further fueling these discussions are issues about the duty to warn participants about the possibility of an incidental finding in advance of a research study, the extent to which investigators are obligated to look for abnormalities, and how these unanticipated findings should be communicated to participants.

Here we highlight two major categories of challenges with respect to incidental findings that psychologists ought to be thinking about: planning for incidental findings in research and responding to them when they occur. We use neuroimaging and genetics research to situate the discussion.

Planning for incidental findings

Both empirical and consensus-based attempts in the past to wrangle with the issue of incidental findings in neuroimaging and genetics research have yielded practical recommendations for investigators and clinicians who engage in this type of research. Guidance has created awareness about incidental findings and compelled explicit and transparent consideration of their management during study design, requests for ethics and human subjects approval, the informed consent process and study implementation. This approach is highlighted most recently and significantly in the December report of the Presidential Commission on the Study of Bioethics Issues, "Anticipate and Communicate: Ethical Management of Incidental and Secondary Findings in the Clinical, Research, and Direct-to-Consumer Contexts." The practical details about how to handle incidental findings are largely left up to investigators and local institutional review boards, which have benefited from both writing in the peer-reviewed literature and online reports and materials.

In the domain of brain imaging, and fMRI in particular, one major issue in planning for incidental findings is whether a physician who is qualified to interpret these scans, such as a neuroradiologist or a neurologist, should be a required collaborator on research protocols, and whether he or she should read all MRI scans or only scans on which the investigator detects a suspected anomaly. This has been a hotly debated issue and is an especially complicated question for investigators in psychology departments that have MRI scanning capabilities but are unaffiliated with a medical center where qualified physicians can interpret their scans.

There are several approaches these psychology-based imaging programs can adopt. If the institutional approach is "don't look, don't tell," and this is communicated to participants through an informed consent process, then there is full transparency and the ethical duty to warn is met. However, there remains the nagging possibility that something life-threatening and treatable — such as an arteriovenous malformation or certain tumors — could be missed by this approach. The incidence of medically actionable brain findings is about 2 percent, potentially generating anxiety and guilt among investigators and a false sense of security for participants.

Many have argued, however, and we are generally among this group, that investigators in neuroimaging studies should always have access to physicians qualified to interpret scans. If these physicians are only reviewing scans identified by the investigator as having a suspicious finding, then a new question arises: Should an investigator who is not qualified to clinically read brain MRI scans — such as the PhD investigator, postdoctoral fellow or trainee — review the images to search for possible structural abnormalities in the brain? If the answer is yes, then this approach may put an undue burden on an untrained individual and could create either anxiety over potentially missed findings or a sense of guilt if a missed finding caused preventable harm. Health economics analyses suggest that the answer to this question is unequivocally no, based on the estimated costs that false positives would bring to society. The other approach, then, is to report to a collaborating physician only an anomaly that is indisputably apparent even to the most naive eye, or have qualified physicians formally review all scans.

Assuming that a physician can review either all scans or scans with suspicious findings, should participants be given the opportunity to explicitly opt out of being informed of an incidental finding, irrespective of the type of lesion and its clinical significance? We highlight this possibility even against a backdrop of data from our own work, which has shown that an overwhelming number of participants would want to know about an abnormal brain finding, whether or not it has clinical significance. For moral and ethical guidance on these issues of detection and disclosure of incidental findings, however, there are perhaps as many answers and opinions as people asked.

Beyond testing individuals who are ostensibly healthy, additional challenges around anticipating incidental findings are raised by studies that involve children and people with mental health disorders or dementia. Risks related to incidental findings in neuroimaging and genetics research with vulnerable populations include, for example, the complexities of informed consent; the heightened susceptibility to anxiety, depression and fear concerning the possible or actual discovery of an abnormality; and justice-related issues, such as re-contacting participants after a study or insurance coverage following the unexpected discovery of clinically significant incidental findings.

Additional concerns pertain to people whose professional duties or activities — such as members of the military, operators of public transportation or competitive athletes — may be affected or delayed by an incidental finding, regardless of its ultimate clinical significance. And, finally, as views on consent, ownership and reporting of findings may differ widely among participants from diverse cultural backgrounds, thinking and planning ahead will ensure the integrity of the research, the protection of participants and the respect for the diverse pools from which they may be drawn.

Responding to incidental findings

Within the domain of functional neuroimaging, most would argue that a suspicious finding should never be reported directly to a participant prior to review of this finding by a physician qualified to clinically read and interpret brain MRI scans, even if only to encourage medical follow-up. Thus, physicians who are qualified to read MRI scans should always confirm a suspicious finding as abnormal before relaying this information to participants. When being informed of an incidental finding, irrespective of its identity and clinical significance, participants will undoubtedly have questions about the finding's clinical implications, possible interventions and long-term consequences. A qualified physician, such as a neurologist or neurosurgeon, is the preferred person to communicate an abnormal finding to a participant in conjunction with the research investigator, even if the physician does not have a prior therapeutic relationship with the participant. It may be beneficial to have psychologists be in the front line of counseling research participants in whom an incidental finding has been detected, both for life-threatening findings and, maybe more important, for abnormalities of unclear significance, where follow-up imaging and testing are required.

With increasing numbers and applications of neuroimaging and genetics studies, there is a growing mandate to bank data. While the significance of banked data that contain functional brain activity maps for predicting future neurocognitive performance or neurological disease is unclear, genetic profiles may provide more concrete predictions of disease or the risk of transmitting a condition to future generations. The disclosure of genetic information is also a controversial topic, and new recommendations are emerging. One set from the American College of Medical Genetics (ACMG) suggests that laboratories and physicians must report the results of at least 56 genes from whole exome sequencing, even the results that are unrelated to a primary inquiry.

Although genetic counselors are the obvious professionals to provide support when unexpected genetic findings are detected and disclosed, they will likely need to partner with psychologists to be able to address anxiety-provoking questions, such as whether people should participate in genetics research, especially when the choice to be told about their genetic profile has been stripped, as the new ACMG guidelines direct. By working with psychologists, genetic counselors, physicians and others will be more equipped to counsel study participants when unanticipated discoveries are made, and they can begin to address such issues as how to time the communication of these findings, the psychological impact on participants and their children, and the personal and psychological implications for third parties.

Advances and sound policy at the interface of biomedicine and psychology hinge not only on good answers to these questions but also on the way that scientists and health providers approach them in thoughtful partnership, with openness and foresight. As we see it, an excellent metaphor to invoke here is: All hands on deck.

Dr. Judy Illes is a professor of neurology at the University of British Columbia. Dr. Matthew P. Kirschen is a pediatric neurologist and fellow in pediatric critical care medicine at The Children's Hospital of Philadelphia.

The authors wish to acknowledge the NIH/NIMH and NIH Task Force on Bioethics Research, Training, and Translation; the Canadian Institutes of Health Research; NeuroDevNet Inc.; and the Canada Research Chairs Program.

This installment of "Ethically Speaking" was invited by members of APA's Committee on Human Research. Email your comments and questions.