Why Reference Ranges Differ From Optimal Ranges

Laboratory reports mark results as within or outside a reference range, and much interpretation hinges on that boundary. How those ranges are constructed is less widely understood than it should be.

Ranges describe a reference population

A reference range is derived by measuring a group of people considered suitable and taking the central portion of their results, conventionally the middle ninety-five percent.

This means the boundaries are statistical positions rather than points where anything physiological changes. They mark where a result becomes unusual within that group.

A consequence follows directly: a proportion of entirely healthy people fall outside the range on any given test, purely because of how the boundaries were set.

The reference population shapes the result

Which people are included matters greatly. Some ranges derive from people who happened to have the test ordered, which is not the same as a healthy sample.

Where the underlying population has a high prevalence of a condition, the range widens to accommodate it, and the boundary drifts accordingly.

Ranges also vary by laboratory, because they depend on the specific assay and equipment. Comparing a result against another laboratory's range introduces error.

Optimal ranges make a different claim

Narrower optimal ranges are proposed on the basis that risk within the conventional range is not uniform, and that better outcomes cluster in part of it.

This is a coherent argument. For some measures, population data does show risk rising within what the reference range treats as normal.

The difficulty is that optimal ranges are frequently asserted without the outcome data required to support them, and different sources publish different bands for the same test.

Multiple testing multiplies flagged results

If each test flags a small percentage of healthy people, a panel of many tests will flag most people on at least one measure.

Narrowing the boundaries increases this sharply. A panel judged against optimal ranges will produce abnormal findings in almost everyone tested.

Each flagged result then invites investigation, and the yield from investigating results generated this way is low relative to the concern they cause.

Context outweighs the boundary

Clinical interpretation weighs a result against symptoms, history and other findings rather than against the boundary alone. A borderline value means different things in different contexts.

Trends across repeated measurements often carry more information than a single value, since a person's own stable level is more informative than population position.

Interpreting laboratory results, particularly borderline ones, requires a clinician with the full picture. The range on the report is a starting point rather than a conclusion.