What if an Incidental Finding Is Potentially Important but Highly Uncertain?

What if the Finding Could Be Serious, but You Are Not Sure What It Means?

Some incidental findings are relatively clear. Others occupy a much more uncomfortable space: the observation might indicate something important, perhaps even something serious, but the evidence is not strong enough to know.

A suspicious area appears on a research image but could be an artifact. A genomic variant is real, but its relationship to disease is uncertain. A physiological measurement is markedly abnormal once, then normal when repeated. The researcher now faces risks in both directions.

Disclose too quickly, and an uncertain research observation may be mistaken for established health information. Ignore it too readily, and a participant could lose the opportunity to investigate something genuinely consequential.

Uncertainty Usually Calls for Evaluation Before Disclosure

“Uncertain” Can Mean Several Very Different Things

Start by asking what exactly is uncertain

Researchers sometimes speak of an “uncertain finding” as though uncertainty were one property. It is not. Several different questions can be uncertain, and they require different responses.

This distinction prevents a common mistake. A finding that may simply be a technical artifact is not ethically equivalent to a confirmed finding whose long-term clinical consequences are incompletely understood.

Analytic uncertainty comes first

If researchers do not know whether the underlying observation is technically reliable, they should be cautious about moving immediately to its possible medical meaning.

The return-of-results literature shows broad agreement that acceptable analytic validity is a basic condition for responsible informational return. A technically unreliable result can expose a participant to anxiety and unnecessary follow-up without a sound evidentiary basis.

Depending on the research method, uncertainty might be reduced through repetition, independent confirmation, specialist review, a higher-quality measurement, or an appropriate clinical test. The correct approach depends on the modality and governing requirements.

A real finding can still have an uncertain meaning

Sometimes the technical result is not in doubt. What remains uncertain is its interpretation.

Genomic variants provide a familiar example. Sequencing may accurately establish that a participant carries a particular variant while available evidence remains insufficient to determine whether that variant causes disease. Similar problems arise when imaging abnormalities have uncertain significance or biomarkers have poorly established relationships to future outcomes.

This distinction is essential because confirmation of the measurement does not necessarily confirm the feared interpretation.

Researchers should avoid sliding from the first statement into the second without sufficient evidence.

Uncertainty weakens the case for routine return

Influential frameworks for returning incidental findings generally give substantial weight to established meaning, analytic validity, significant risk, and actionability. Recommendations for genomic biobank research, for example, have proposed not returning findings when their likely health or reproductive importance or personal utility cannot be ascertained and return therefore offers unlikely net benefit.

This makes ethical sense. Returning information is not neutral. Once a researcher tells someone that a finding may indicate serious disease, the information can influence medical decisions, anxiety, family communication, finances, and future planning even when the researcher carefully says “we aren't sure.”

But uncertainty does not always justify doing nothing

The opposite rule would also be too simple. Research rarely offers perfect certainty, and waiting for absolute certainty could itself expose participants to preventable harm.

Suppose a research image contains a suspicious abnormality that a qualified reviewer cannot diagnose from the research scan but believes warrants prompt clinical evaluation. The correct message may not be “You have disease X.” It may instead be “An observation on the research scan should be evaluated clinically.”

The uncertainty remains, but its ethical significance changes. Researchers are no longer claiming to return a diagnosis. They are communicating that there is a sufficiently credible reason for further assessment.

Separate the threshold for further evaluation from the threshold for definitive return

This is one of the most useful distinctions in difficult cases. An observation may be too uncertain to return as an established individual research result while still being important enough to trigger expert review, confirmation, or clinical referral.

Those thresholds need not be identical. Treating them as identical creates two bad options: disclose premature information or ignore anything that has not yet reached clinical certainty.

Seriousness and urgency affect how much uncertainty can be tolerated

The potential consequences of waiting matter. A mildly concerning observation that can be reviewed routinely presents a different problem from a finding suggesting an immediate threat to life.

As the potential harm becomes more serious and time-sensitive, the threshold for escalating the observation for expert evaluation may reasonably become lower. This does not mean that low-quality evidence should suddenly be presented as fact. It means the research team may need to investigate the uncertainty more quickly.

If an observation might represent an immediate threat to a participant's health or safety, the study's urgent safety procedures should take priority over the ordinary pace of return-of-results deliberation.

Uncertainty should be communicated as uncertainty

If an uncertain finding is appropriately returned, the uncertainty is part of the finding. It should not disappear in translation.

Participants may need to understand what was actually observed, how reliable the observation is, what interpretation is possible, what remains unknown, whether confirmation is required, and what next step is reasonable.

The wording matters. “Your research scan showed something that warrants clinical evaluation” conveys a different epistemic claim from “We found a tumor.” Likewise, “This variant has uncertain significance” is not equivalent to “This variant means you will develop the disease.”

The more difficult the interpretation, the stronger the case for having an appropriately qualified professional explain the finding to the participant.

Variants of uncertain significance illustrate why restraint matters

Genomics makes the uncertainty problem particularly visible. A variant can be identified accurately while evidence about pathogenicity remains insufficient or conflicting.

Returning such variants indiscriminately can invite participants or clinicians to treat uncertainty as risk information. This is precisely why frameworks for return tend to emphasize established meaning and net benefit rather than the mere availability of sequence information.

The broader principle applies beyond genetics: technical detection is not the same as established individual significance.

Knowledge can change after the study

Some findings are uncertain today but may become interpretable later. NHGRI notes that genomic information can be interpreted differently as scientific knowledge develops, creating difficult questions about infrastructure and resources for returning results.

That does not automatically create an indefinite duty to reinterpret every participant's data forever. Reanalysis and recontact require their own policy decisions about duration, resources, consent, data identifiability, and the research relationship.

If a study plans to reinterpret findings or recontact participants when classifications change, that commitment should be stated prospectively rather than casually implied during disclosure.

Participant preference does not eliminate uncertainty

A participant may strongly prefer to receive every possible finding. That preference deserves respect, but it cannot make ambiguous information more scientifically interpretable.

Conversely, researchers should not assume that participants would rather remain uninformed whenever evidence is imperfect. The relevant question is whether the information meets a defensible return threshold and, if so, whether the participant has chosen to receive that category of information.

This is where the broader question of participant choice about receiving incidental findings becomes important.

A Research Scan Shows Something Suspicious but Not Diagnostic

The team has acted despite uncertainty without pretending that uncertainty has disappeared. That distinction is often the core of responsible management.

Common Mistakes When a Finding Is Uncertain

Manage the Uncertainty Before You Manage the Message

When an incidental finding is uncertain, do not begin by asking how to tell the participant. Begin by identifying what you actually know, what you do not know, and whether additional evaluation can reduce the uncertainty.

The objective is not certainty at any cost. Research rarely provides that luxury. The objective is to ensure that the strength of the communication matches the strength of the evidence.

Before Acting on an Uncertain Incidental Finding

Questions About Highly Uncertain Findings

Uncertainty Should Change the Response, Not End the Analysis

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.