Every echocardiogram number hides a cloud of uncertainty, commentary warns

All clinical measurements carry noise, and a new commentary in Cardiovascular Ultrasound argues that medicine’s habit of reporting single point estimates hides that uncertainty with real consequences for patient care.

Their core argument is simple but uncomfortable. Research studies enrol many patients to average out noise and use randomisation and double-blinding to counter bias. Clinical medicine rarely enjoys these protections: an individual patient typically has a variable measured only once. Yet that single measurement is reported as a precise figure, an excess of decimal places that, in the authors’ words, gives a spurious impression of accuracy.

The authors draw on 95% confidence intervals published by Manna and colleagues in 2024 in the Journal of the American Society of Echocardiography. A typical report might describe severe aortic stenosis when the point estimate for peak velocity is 4 metres per second and valve area is 1.0 square centimetres. Applying those confidence intervals, the peak velocity could actually indicate moderate or severe disease. The valve area is worse still: consistent with anything from mild to critical stenosis.

The two variables differ dramatically in how much noise surrounds them, and that variation matters for utility. The commentary’s central claim is that adding aortic valve area to the quantification of aortic stenosis may result in poorer, noisier estimates rather than better ones. Conventional presentation of data as point estimates masks this wide variability between variables, so a clinician reading a report cannot see which numbers deserve confidence and which barely constrain the diagnosis.

To make the philosophical point vivid, the authors reach for physics. In 1913, Niels Bohr’s planetary model offered a simple, precise picture of the atom, with electrons located like moons orbiting a planetary nucleus. By 1925, Erwin Schrödinger had replaced it with the current model, in which an electron’s position is a fuzzy cloud of probabilities. Less satisfying to the human desire for certainty, but a more faithful description of reality. Every clinical variable, they argue, similarly comes with an invisible cloud of uncertainty of varying size across variables.

Treating the point estimate as the true value of a variable, the authors conclude, confers the certainty decision-makers crave but does not reflect reality and may not lead to optimal clinical decisions. They suggest that being mindful of how noise varies across variables should facilitate better choices at the bedside. The article is a hypothesis-generating commentary rather than a study with new data; the authors state there are no new data associated with it, and its quantitative illustration rests on previously published confidence intervals. Even so, its message travels well beyond one valve: any single measurement reported without its uncertainty invites clinicians to mistake imprecision for fact.

Subject of Research: Measurement noise and uncertainty in echocardiographic assessment of aortic stenosis severity

Article Title: Noise in clinical data: insights from assessment of aortic stenosis by echocardiography

Article References: Lloyd, J. N., & Armstrong, G. P. (2026). Noise in clinical data: insights from assessment of aortic stenosis by echocardiography. Cardiovascular Ultrasound, 24(1), Article 11. https://doi.org/10.1186/s12947-026-00372-8

Image Credits: AI Generated

DOI: 10.1186/s12947-026-00372-8

Keywords: aortic stenosis, echocardiography, measurement noise, clinical decision-making, point estimates, confidence intervals, valve area, peak velocity, diagnostic uncertainty, cardiovascular imaging, Noise, clinical