Panels and reference ranges
Panels
/tests/panels — what gets ordered together, and in what order it reports.
A panel is a named group of catalogue tests with a declared reporting order, plus one sentence saying why the panel exists, written for whoever reviews the panel list in three years.
Reordering is two buttons, deliberately — there is no drag handle, and there will not be one. A reporting order that changes because a finger slipped on a touchscreen is a report that reads differently for no recorded reason.
Reference ranges
/reference-ranges — a reference range is a function, not a column.
This is the difference between a system that prints an interval beside a number and one that has already resolved it:
| Printed | Haemoglobin 9.1 g/dL · Normal: 12.0–15.5. The reader does the arithmetic. Nothing states which population that interval belongs to, whether it matches this patient, or which method produced it |
| Resolved | Haemoglobin 9.1 g/dL Low · ref 12.0 to 15.5 · adult female · this method. The interval was selected by age, sex and method, and the flag is part of the result rather than a rendering of it |
Ranges are held as bands. A band carries its own low and high, and optionally its own critical low and high — leave both critical fields empty and that band can never raise a critical flag.
:::info The intervals are yours LabFlow resolves against the ranges your laboratory has verified for its own methods and population. It does not supply clinical reference intervals. :::
A released result keeps the range it was read against
Ranges change. Editing a band changes what the next result means; it does not reach back. A released result stores the interval that applied when it was produced, exactly as an audit row stores the role that applied when it was written.
Personalised ranges
/results/personalized-ranges — the patients for whom the population range is the wrong answer.
Some patients need their own interval: the population band would flag every result they ever produce, or flag none of the ones that matter. A personalised range overrides the band for that patient and that test.
Changing an override never rewrites a result that already exists. The then-and-now comparison arrives with released results, so a reader can see both what was used and what would be used now.