There is usually no single expiry date
People expect a printed deadline and are surprised not to find one. In most systems a laboratory order carries no fixed statutory lifespan of its own. What governs it is a mixture of the ordering provider’s instructions, the laboratory’s own policy, and, where payment is involved, whether the test is still medically necessary at the moment it is performed.
That last condition is the one that bites. Coverage generally depends on a test being ordered by a provider and being necessary now, not on the date printed at the top of the page.
What actually limits an old order
The clinical question. An order written for a symptom you had three months ago was designed to answer a question about that moment. Running it now produces a valid measurement of an outdated question.
The follow-up interval. Orders written to track something have their own rhythm: too early and the value has not moved, too late and you have lost the point of the series.
And anything that changed in between. A new medicine, a new diagnosis, a pregnancy or a hospital stay can all make the original list the wrong list, even though every line on it is still a real test.
What to do with an order you did not use
If it is recent and nothing has changed, take it. Laboratories deal with this daily and will tell you at reception if something on it needs confirming.
If months have passed, or if anything has changed, ask the prescriber before you go rather than after the draw. A two-minute message is cheaper than a panel that answers the wrong question, and the prescriber may simply confirm it or adjust one line.
Repeat orders are a special case: some are written to be used several times over a defined period. If yours says so, the period is on the page.
The practical rule
Treat an order as valid while the reason it was written is still your reason. That test is more useful than any number of months, and it is the one a prescriber applies when you ask.