Skip to content
SearchProgress
Display

Display

Theme
Density
Text size

Sign in

Add your earlier progress to your account?

Study progress is waiting to be saved

After QC fails

14 min

  • Confirm corrective action before accepting the next QC run
  • Decide which patient results need review after rejected QC
  • Set a new control lot's mean from its crossover runs and its SD from the established CV

Read the full reference

Try first

Try first

A glucose run is rejected when both control levels fall below −2 SD. Without changing anything, you rerun both controls, and this time both are inside their limits. What do you do?

The next section explains it.

The next section explains it.

Right. The next section explains why.

The next section explains it.

Get the idea

A rejected QC run means the analyzer cannot be trusted for that analyte until someone shows otherwise. Reporting of affected patient results stops, and three pieces of work follow:

  1. Correct the fault.
  2. Prove the correction.
  3. Look back at what was already reported.

Correct and prove

The investigation points at a cause, such as:

  • a control vial made up wrongly
  • a reagent pack past its onboard stability
  • a drifted calibration
  • a clogged probe

The technologist fixes that cause and then runs QC again. Acceptable QC after a named correction is what returns the test to service.

Rerunning the same controls until they pass skips the first half. The rerun may land inside the limits by chance while the fault is still present, and the next run fails the same way.

CLIA requires the laboratory to document every corrective action it takes when control or calibration results fail its acceptance criteria, so the record says what went wrong and what was done.1

Look back

  • Patient results obtained in the unacceptable run, and all results reported since the last acceptable run, are evaluated to see whether they were affected.1
  • The window reaches back that far because a shift or trend can start before the run that finally breaks a rule.
  • Following the laboratory's procedure, specimens from the window are retested after the correction.
  • When a reported result proves to be in error, the laboratory notifies the ordering provider and issues a corrected report promptly.3

A new control lot

Recovery sometimes means opening a new lot, and every lot change needs its own limits.

Take the target mean from the new lot itself:

  • The new lot has its own concentration, so its target mean comes from its own crossover runs, tested in parallel with the current lot. About 10 measurements on separate days give a starting target.2
  • The old lot's mean would make every new-lot result look shifted.
  • The insert mean is assigned across many laboratories and can sit away from this analyzer's own center.

Set the SD:

  • A few days of crossover runs show only short-term scatter and give limits too tight to hold.
  • When the CV is steady across concentrations, the established long-term CV carries over:2

new SD = new mean × established CV ÷ 100

The mean and SD are then updated as stable data accumulate.

References
  1. Electronic Code of Federal Regulations. 42 CFR §493.1282: Standard: Corrective actions. Accessed September 23, 2026.
  2. Clinical and Laboratory Standards Institute. Statistical Quality Control for Quantitative Measurement Procedures: Principles and Definitions. 4th ed. CLSI guideline C24Ed4E. Clinical and Laboratory Standards Institute; 2016. Accessed September 23, 2026.
  3. Electronic Code of Federal Regulations. 42 CFR §493.1291: Standard: Test report. Accessed September 23, 2026.

Watch one

At 07:00 both alanine aminotransferase (ALT) control levels read below −2 SD, breaking the 2-2s rule.

  • The last acceptable QC was at 15:00 yesterday.
  • The reagent pack on board was opened 31 days ago, and its onboard stability is 28 days.

What has to happen before ALT results are reported again?

  1. Stop reporting ALT and keep the run's patient results pending release.

    Results from a rejected run cannot go out while their accuracy is in question.

  2. Take the expired reagent pack off and load a fresh pack.

    Both levels moving the same way points to a systematic cause, and a pack past its onboard stability is one.

  3. Run both control levels on the fresh pack and confirm they are inside their limits.

    Only acceptable control results show that the new pack fixed the fault.

  4. Document the rejection, the expired pack, the replacement and the acceptable QC.

    The record has to name the cause and the fix, so the recovery can be checked later.

  5. Evaluate the patient ALT results from the 07:00 run and every ALT reported since 15:00 yesterday.

    A drift in reagent can start before the run that finally fails.

Replace the expired pack, obtain acceptable QC, document both, then resume reporting and review every ALT result back to the last acceptable run at 15:00 yesterday.

Your turn

Problem 1 of 3

Potassium QC was acceptable at 10:00. Patient potassium results were reported through the late morning. At 14:00 a control breaks the 1-3s rule. After the fault is corrected, which patient results do you evaluate?

That covers the unacceptable run and everything since the last acceptable run, which is the whole stretch where the fault could have been present.

QC at 10:00 was acceptable, so it vouches for the results before it. The review starts at the last acceptable run.

The fault may have been present before the run that caught it, so results reported after 10:00 can be affected too.

Reviewed only the failed run's patient results

A shift or trend can begin before the run that fails. The review covers patient results from the unacceptable run and every result reported since the last acceptable run, so reviewing the failed run alone can leave affected results uncorrected in the record.

Hint
  1. The rule broke at 14:00, but the fault may have started earlier.
  2. Find the last point where QC showed the analyzer was working.

Review Resolving rejected QC

Problem 2 of 3

A new cholesterol control lot ran 10 times over 10 days beside the current lot. Its crossover mean is 212 mg/dL, and the SD of those 10 runs is 1.9 mg/dL. The insert lists a mean of 205 mg/dL with an SD of 8 mg/dL. The established CV for this level is 2.0%. Which starting limits do you set?

The insert values are assigned across many laboratories. Its mean sits away from this analyzer's own center, and its wide SD lets a real shift run longer before it is caught.

Set the provisional mean from the old lot or the insert

A new control lot has its own concentration, so every new-lot result judged against the old lot's mean reads as a shift. The insert mean is assigned across many laboratories and can sit away from this analyzer's own center. The mean comes from the new lot's own crossover runs.

The mean comes from the lot's own crossover runs, and the SD is 212 × 2.0 ÷ 100 = 4.24 mg/dL, about 4.2 mg/dL.

The mean fits this lot. An SD of 1.9 mg/dL reflects only 10 days of short-term scatter, so limits built on it are too tight and will reject good runs.

Set the provisional SD from the crossover runs or the insert

A short crossover study covers only short-term scatter and gives limits too tight to hold. The insert SD is built to cover many laboratories and lets a real shift run longer before it is caught. The established CV, applied to the new mean, carries the day-to-day, calibration, and reagent variation a long-term SD needs.

Hint
  1. The mean has to describe this lot on this analyzer.
  2. The SD should carry day-to-day, calibration and reagent variation, which a 10-day study barely sees.
  3. Apply the established CV to the new mean.

Review Control materials

Problem 3 of 3

A creatinine run is rejected under the R-4s rule. The technologist finds a small clot on the sample probe and clears it. What returns creatinine to patient testing?

The clot was a likely cause, and acceptable QC afterward shows whether the analyzer is working. The fix alone is a claim until controls confirm it.

Repeating controls until most pass lets chance decide. The correction is judged by one acceptable QC run after the fix, documented with it.

Accepted repeat QC without a documented correction

An in-range rerun after a rejection can pass by chance, and repeating controls until most pass lets chance decide. Either way, patient testing resumes on a system that may fail again, with no documented correction behind it. Recovery rests on a documented correction followed by acceptable QC.

A named correction followed by acceptable QC, both on record, returns the test to service. The lookback to the last acceptable run follows.

Review Resolving rejected QC

Use it

Try this at the QC workstationPendingQC run due07:00READY TO REVIEWCase 05Review QC

Keep

Sources checked