← All guides
42 CFR 493.15(e)(1) ↗

CLIA-waived testing: what the rule actually requires

For a CLIA-waived test — a rapid strep, a urinalysis strip, a glucometer — the entire federal quality-control requirement is one sentence long. Almost everything else you've heard about how often to run controls comes from somewhere else.

The rule, in full

42 CFR 493.15(e)(1): “Follow the manufacturer's instructions for performing the test.” That's the whole federal quality requirement for a waived test.

CLIA doesn't specify a control frequency, a QC schedule, or an acceptable range for a waived test — the manufacturer does, in the package insert. A rule that says “run a control once a week” is real, and it's binding — it's just not coming from CLIA. It's coming from the box the strips shipped in.

When to run a control, beyond the schedule

Whatever interval the insert specifies, plus a control run for each of these, because each one is a reason the last passing control no longer tells you anything about today's strip:

  • A new strip lot. Manufacturing varies lot to lot; a control passed on the old lot says nothing about the new one.
  • A new shipment, even of the same lot — shipping and storage conditions in transit are outside the manufacturer's control.
  • A new operator.
  • A result that doesn't fit the patient — the test disagreeing with the clinical picture is itself a reason to question the strip before trusting the result.

The half nobody quality-checks: the strips themselves

A reagent strip analyzer reads what the strip did and reports it — it has no way to know the strip itself was already compromised. Strip pads oxidize on contact with air and humidity, so a bottle left open, a missing desiccant packet, or a pad that's already started to discolor produces a result that's wrong in a way the machine cannot detect and will not flag.

Checking the bottle — desiccant present, cap closed promptly after each strip, no discoloration against an unused strip — is not housekeeping. For a waived test, it's the only check that exists on the input at all.

“Not run” is a real, recordable answer

Sometimes a control genuinely wasn't run. A form that can't record that forces an invented “in range” instead — which is worse than an honest gap, because it looks clean. A monthly QC record showing both controls skipped and filed as passing is exactly the hollow record an audit exists to catch; one line saying why the control wasn't run is a better answer every time.

Medicin Binder asks for the lot, the bottle, and the reason

Not just “pass” or “fail” — the strip lot, when the bottle was opened, whether the desiccant's still there, and which analyte was off when something isn't. That's the record that actually answers a surveyor's next question.

Start the 30-day trial