May 13, 2020
Incidental findings – what are they?
Imagine you are conducting a usability study of a blood pressure monitor. You developed a well-thought-out screener to identify participants. Your study was approved by the Institutional Review Board and is now up and running. While a participant is measuring his own blood pressure as part of the study, the blood pressure monitor detects an unusually high systolic blood pressure (i.e., hypertension), which is associated with an increased risk of a heart attack or stroke. You have just encountered an incidental finding (IF) in your study. The participant correctly interprets the device’s warning regarding hypertension, but he is surprised to see the warning and expresses concern about his blood pressure. As the study moderator, how are you to respond?
What is an IF?
According to the United States Department of Health and Human Services, IFs “are discoveries of individual-level findings that are unrelated to the goals of the study.” Such findings are more common in medical or psychological evaluations than in human factors (HF) research. For example, during a computed tomography (CT) scan of a rib fracture, a tumor can be detected in an otherwise healthy subject. However, as the blood pressure monitor example illustrates, IFs can also occur during HF research.
How does HF research risk the discovery of IFs?
When an individual participates in an HF validation (i.e., summative) study of a medical device, the FDA requires participants to interact with production-equivalent devices. Using a production-equivalent medical device might result in someone using a medical device on themselves (e.g., the blood pressure monitor study described above) and getting real—rather than simulated—results. These real results are where an IF could occur.
The easiest solution to preventing IFs is to limit device functionality, to present simulated results unrelated to the participant, or to avoid having participants perform actual diagnostics when they do not directly relate to the study’s goal. However, occasions might arise in which simulated outcomes cannot be used, and actual diagnostic information and results are generated and visible to the participants. In other scenarios, an ancillary device or assessment might be used to supplement a usability study that could produce real results. These real results (e.g., visual or cognitive impairment test scores, blood pressure) could produce IFs and should be considered as part of the study. As such, HF researchers might be in a position where they feel responsible to respond to an IF, which could include notifying participants of IFs.
Why notify participants of IFs?
Because IFs potentially have very serious implications for a participant’s health and well-being, there is good reason to notify participants of an IF. However, it is important to realize the consequences of doing so. Firstly, because HF researchers are generally not medical professionals, they cannot and should not provide a diagnosis to a participant. Secondly, though the intent of notifying the participant of an IF is to help uncover and subsequently treat a medical condition, learning unsolicited health information might evoke negative emotions from a participant (e.g., stress). Furthermore, there is no certainty the IF is accurate, as the test setting and device fidelity might prevent accurate results. As such, researchers should instead simply notify the participant of the IF and recommend that participants see a doctor to discuss the IF further and receive medical information.
Study moderators must understand how to respond to IFs when they occur, and participants must understand the risk of encountering an IF during a study so that they can make fully informed decisions about whether to enroll. Therefore, HF researchers should have a well-defined protocol to guide responding to IFs.
Little official guidance exists on responding to IFs, let alone during HF studies. As an example, the Food and Drug Administration (FDA) refers to the Code of Federal Regulations (CFR) for regulations and law applicable to human subjects research (21 C.F.R. 2020, Part 50), yet the CFR does not mention IFs. Our own search for organizations in the United States, European Union, and United Kingdom that govern or advise on human subjects research does reveal some guidance, protocols, and recommendations on dealing with IFs (see Resources for a list of what we found). However, no consensus currently exists regarding how to best handle IFs. Additionally, because the existing guidance focuses primarily on medical and psychological research, the approach for addressing IFs in HF research is still undetermined; this is in part because how IFs appear during HF research is generally different from how they appear in medical or psychological research. Fortunately, because HF research falls under the umbrella of human subjects research, we can consider how other fields suggest responding to IFs and apply suggestions to HF research.
First, attempt to eliminate the possibility of IFs occurring. Avoiding IFs, if feasible, is ideal. If eliminating the possibility of IFs is not feasible, consider the following recommendations:
Consider any IF-related best practices provided by your institution/organization and provided by the legal entity overseeing your research (e.g., IRB). Other strategies for responding to IFs can be implemented in absence of, or in addition to, these best practices.
Plan for IFs.
- Determine the potential for IFs at the outset of test planning.
- Identify IFs that are not harmless (i.e., IFs that could potentially lead to harm).
- Ensure all project stakeholders (e.g., clients, third-party researchers) agree on the approach to responding to IFs.
Determine how you will respond to IFs.
- Develop a script for responding to IFs, and ensure the script receives approval from the IRB and from project stakeholders.
- Consider when and how to incorporate participant preferences, such as whether participants want to be informed of any IFs.
- Consider participant confidentiality requirements (e.g., General Data Protection Regulation [GDPR]) and any limitations that might impose on IF reporting.
Act appropriately during the study.
- Before a session begins (e.g., in the informed consent form), inform participants that you might discover something about their health of which they might be unaware, and describe the implications (e.g., the potential need to follow up with a doctor).
- Enable participants to opt in or out of learning information related to an IF before their session begins (e.g., in the informed consent form).
- If you encounter an IF during a study, remain calm to help keep participants at ease.
- Do not address IFs until after all study activities are completed to avoid compromising the data.
- Follow the protocol and script you developed for responding to IFs.
- Inform participants of the facts and how to proceed.
- Do NOT give any medical advice of any type; HF researchers are not qualified to do so.
- Encourage participants to direct any questions about the IF to a doctor.
- At the end of the study session, remind participants of the contact information for the study so that they can reach out with any questions.
Below is a list of the resources we found when searching for existing guidance:
- Recommendations on Reporting Incidental Findings from the United States Department of Health and Human Services
- Managing incidental findings and research results in genomic research involving biobanks and archived data sets from Genetics in Medicine
- Incidental findings in research with human participants: Ethical challenges for psychologists from the American Psychological Association Committee on Human Research
- Guidance Note for Researchers and Evaluators of Social Sciences and Humanities Research from the European Commission
- Framework on the feedback of health-related findings in research from the Medical Research Council
Ruben Post and Alexandra Benbadis are Senior Human Factors Specialists at Emergo by UL's Human Factors Research & Design division.