Science communication often gets trapped between two unhelpful extremes. One version sounds more certain than the evidence allows. The other is so cautious that the audience cannot tell whether anything useful is known at all.

There is a better middle ground: communicate uncertainty as information.

Start with what is known, then define the boundary

“We do not know” can sound like a dead end. “We know A and B with reasonable confidence, but C is still uncertain because the available studies are small” gives the listener a map.

The CDC’s current risk-communication guidance explicitly recommends stating what is known and unknown and acknowledging uncertainty from the start. That guidance also emphasizes explaining numbers clearly and giving people meaningful options where possible.

Name the type of uncertainty

Uncertainty is not one thing. A conclusion may be uncertain because:

Naming the source matters because each kind of uncertainty implies a different next step. More data may solve one problem. Better measurement may solve another. Some questions remain uncertain because reasonable people weigh the same evidence differently.

Do not turn uncertainty into false balance

Being transparent does not mean presenting every position as equally supported. A 95-to-5 evidence split is still uncertainty, but communicating it as a 50-to-50 controversy would create a new distortion.

A useful phrase is: “There is still uncertainty around the exact size of the effect, but the evidence currently points more strongly in one direction.” That preserves both facts.

Use numbers when they genuinely clarify

Terms such as small risk, unlikely, or large effect can mean different things to different readers. When good estimates exist, absolute numbers often make uncertainty easier to understand.

For example, “the estimate is 8%, with a plausible range of roughly 5% to 12%” communicates something different from “the risk is around 8%.” The interval tells the reader how much confidence to place in the point estimate.

Explain why guidance can change

Changing a recommendation can look like inconsistency if the communication never explained that revision was possible. The National Academies has warned that hiding uncertainty can create a false sense of certainty and can undermine trust when information is later revised. Its work on science communication treats uncertainty as a normal part of scientific reasoning rather than something communicators should conceal.

It helps to say what would trigger an update: a larger trial, longer follow-up, better surveillance data, replication in another population, or a new safety signal.

A practical pattern: “Here is what the evidence currently supports. Here is the part we are less sure about. Here is why. Here is what we would look for next.”

Separate scientific uncertainty from decision uncertainty

Sometimes the evidence is fairly clear but the decision is still difficult because people have different goals, costs, values, or risk tolerance. That is not a scientific failure. It is a reminder that evidence informs decisions rather than making every decision automatically.

Conversely, a person may know exactly what they want to do while the evidence remains weak. Communication should keep those two dimensions separate.

Avoid using uncertainty as a rhetorical weapon

People sometimes invoke uncertainty selectively: demanding impossible certainty for evidence they dislike while accepting much weaker evidence for a preferred conclusion. Communicators can reduce this by applying the same standard in both directions and making the standard explicit.

The NIH’s communication guidance recommends conditional language when appropriate, context for emerging findings, and care not to overstate statistical or practical significance. Those habits make it easier to be appropriately confident without sounding evasive.

Respect helps uncertainty land

Uncertainty can be stressful when the topic affects health, family, money, or identity. A message that treats concern as foolish can make the audience less receptive, even if the evidence is sound. NIH explicitly recommends respectful language that does not stigmatize or assign blame.

You can be clear about the evidence and still acknowledge why a question is difficult. Those are compatible goals.

The goal is calibrated confidence

Good science communication is not a contest to sound cautious. It is an attempt to make confidence proportional to evidence. When the evidence is strong, say so. When it is preliminary, say that too. When the answer depends on context, explain which context matters.

For the broader communication framework, see what health communication is. For a research-specific example of why conclusions evolve, why one study rarely settles a health question goes deeper.