Reference ranges are clinic-specific. Treat them that way.

A reference range looks like a fixed fact: a lower limit, an upper limit and a unit. In practice, ranges depend on the analyser, the reagents, the method, the population the laboratory serves and sometimes the patient’s age and sex. A range that is correct for one laboratory can be wrong for another.

That is why importing a generic test catalogue, with ranges copied from a textbook or another system, is one of the most common sources of quiet errors after a laboratory goes live.

How generic ranges go wrong

Flags fire on normal results. If the range is narrower than your method’s, healthy patients are flagged as abnormal, and staff learn to ignore flags.

Flags fail on abnormal results. If the range is wider, genuinely abnormal values pass without a flag.

Units do not match. A range stored in one unit and a result reported in another produces nonsense comparisons.

Age and sex are ignored. Many analytes need different ranges for children, adults and older patients, or for men and women.

Each of these failures is silent. Nothing crashes; the report simply says the wrong thing.

What to define before go-live

Before the first result is entered, the laboratory should confirm each of the following for every test it runs:

The test name and code as the laboratory uses them.

The unit, matching what the analyser reports.

The reference range for each relevant age band and sex, taken from the laboratory’s own method or the manufacturer’s documentation.

Critical limits: the values that must be reported to a clinician immediately.

The sample type and any preparation, such as fasting.

Tests the clinic sends to an external laboratory should carry that laboratory’s ranges, not the clinic’s own.

Critical limits deserve their own attention

Reference ranges describe what is usual. Critical limits describe what is dangerous. They should be set deliberately, agreed with the clinicians who will receive the calls, and kept short enough that a critical flag always means action. A result beyond a critical limit should stop the workflow until someone verifies it and confirms the clinician was told.

Keep ranges under change control

Ranges change when an analyser, reagent or method changes. Treat every change as an event: record who changed it, when and why, and from which date it applies. Old results should keep the range that applied when they were reported, so a report printed today still says what it said at the time.

Check before you trust

Before go-live, run a set of known samples or past results through the new catalogue and compare the flags with what the laboratory expects. Pay special attention to borderline values just inside and just outside each limit. This takes an afternoon and prevents months of flags that nobody believes.

The short version

Build the catalogue from your laboratory, not from someone else’s. Define units, age and sex bands and critical limits for every test. Record every change. Test with real values before the first patient. A flag is only useful if staff can trust it.