Total bilirubin controls are run at 07:00, 15:00 and 23:00. For all three runs, the technologist uses aliquots she poured at 07:00 into clear sample cups, which stay in an open rack on the bench under the room lights. Level 1 has a mean of 1.0 mg/dL (17 µmol/L) and level 2 a mean of 15.0 mg/dL (257 µmol/L). Results in SD units (level 1, level 2): 07:00, +0.4 and +0.2; 15:00, -1.6 and -2.3; 23:00, -2.4 and -3.4. At 23:15, fresh aliquots poured from the refrigerated, foil-wrapped vials of the same lots read +0.2 and -0.3. What best explains the 23:00 rule violation?

Light has broken down bilirubin in the poured aliquots

The method's calibration has drifted since the morning

The bilirubin reagent on the analyzer has deteriorated

Evaporation has concentrated the controls in the open cups

A new potassium analyzer goes into service. To set control limits, a technologist runs each control 20 times in one afternoon: level 1 mean 4.00 mmol/L, SD 0.02 mmol/L; level 2 mean 6.50 mmol/L, SD 0.03 mmol/L. The manufacturer's between-day CV for potassium is about 1.2%. Over the next 2 weeks, 11 of 30 runs are rejected under 1-3s or R-4s, with control results scattered on both sides of the mean and no shift or trend. Electrode slopes are normal, daily patient medians match the old analyzer, and the first proficiency testing event was fully acceptable. The most likely problem and its correction are:

SD set too small; recalculate it from 20 separate runs

Calibration error; recalibrate before every run

Failing K electrode; replace it and recalibrate

Unstable controls; change to a new control lot

The two-level Levey-Jennings chart shows calcium controls. The pattern in run 15 indicates:

Random error: reject and investigate

The chart shows two ALT control levels over 20 days. The instrument log records: day 6, photometer lamp replaced; day 11, new reagent lot loaded and a new vial of level 1 control opened; day 16, sample probe cleaned. The most likely cause of the control pattern is:

The new reagent lot loaded on day 11

The photometer lamp replaced on day 6

The new level 1 control vial on day 11

Random error from probe cleaning on day 16

Quality control (Westgard rules, Levey-Jennings)

23 ASCP MLS practice questions on Quality control (Westgard rules, Levey-Jennings), part of Laboratory Operations. The first 23 are below; the first 5 with answers and explanations (tap Show answer after you have chosen), the rest in Learn with a free account.

Question 1 · Hard
Total bilirubin controls are run at 07:00, 15:00 and 23:00. For all three runs, the technologist uses aliquots she poured at 07:00 into clear sample cups, which stay in an open rack on the bench under the room lights. Level 1 has a mean of 1.0 mg/dL (17 µmol/L) and level 2 a mean of 15.0 mg/dL (257 µmol/L). Results in SD units (level 1, level 2): 07:00, +0.4 and +0.2; 15:00, -1.6 and -2.3; 23:00, -2.4 and -3.4. At 23:15, fresh aliquots poured from the refrigerated, foil-wrapped vials of the same lots read +0.2 and -0.3. What best explains the 23:00 rule violation?
1. AThe method's calibration has drifted since the morning
2. BThe bilirubin reagent on the analyzer has deteriorated
3. CLight has broken down bilirubin in the poured aliquots
4. DEvaporation has concentrated the controls in the open cups
Show answerHide answer
Answer: C · Light has broken down bilirubin in the poured aliquots
At 23:00 level 2 breaks 1-3s and both levels break 2-2s within the run, after a steady fall in both levels since 07:00, which points to a systematic error. Fresh aliquots from the same lots, kept cold and dark, read near their means, so the analyzer, reagent and calibration are working correctly and the fault lies in the material that sat out all day. Bilirubin is broken down by light, so bilirubin controls, calibrators and neonatal samples must be protected from it. The technologist should document the cause, discard the old aliquots, accept the run on the fresh controls, and pour protected aliquots for each run from now on.
- A: Calibration drift would affect the fresh aliquots too, and they are in range.
- B: A failing reagent would make the fresh aliquots read low too, but they read near the mean.
- D: Evaporation raises concentrations, but these controls fell steadily.
Question 2 · Hard
A new potassium analyzer goes into service. To set control limits, a technologist runs each control 20 times in one afternoon: level 1 mean 4.00 mmol/L, SD 0.02 mmol/L; level 2 mean 6.50 mmol/L, SD 0.03 mmol/L. The manufacturer's between-day CV for potassium is about 1.2%. Over the next 2 weeks, 11 of 30 runs are rejected under 1-3s or R-4s, with control results scattered on both sides of the mean and no shift or trend. Electrode slopes are normal, daily patient medians match the old analyzer, and the first proficiency testing event was fully acceptable. The most likely problem and its correction are:
1. ASD set too small; recalculate it from 20 separate runs
2. BCalibration error; recalibrate before every run
3. CFailing K electrode; replace it and recalibrate
4. DUnstable controls; change to a new control lot
Show answerHide answer
Answer: A · SD set too small; recalculate it from 20 separate runs
An SD from 20 replicates in a single run measures only within-run imprecision, here a CV of 0.5%, which is much smaller than the method's real day-to-day CV of about 1.2%. Limits built on it are too narrow: the 3 SD limit for level 1 (±0.06 mmol/L) is barely more than one true SD (about 0.05 mmol/L), so many good runs are rejected falsely. Normal electrode slopes, stable patient medians and acceptable proficiency testing show that the method itself performs well. CLSI C24 advises setting a new control's mean and SD from at least 20 results obtained in separate runs, ideally on different days, and refining them as more data accumulate.
- B: A calibration error causes a shift in one direction, not scatter on both sides of the mean.
- C: Electrode slopes are normal and patient medians and proficiency testing agree, so the method is not failing.
- D: Unstable control material drifts in one direction; these results scatter randomly about a stable mean.
Question 3 · Easy
The Levey-Jennings chart shows 20 runs of a glucose control. Which Westgard rule does run 20 break?
Figure: MedExams
1. AR-4s
2. B2-2s
3. C1-3s
4. D10x
Show answerHide answer
Answer: C · 1-3s
Run 20 reads 91.5 mg/dL [5.1 mmol/L], 3.4 SD below the mean of 100 mg/dL [5.6 mmol/L], so it falls outside the -3 SD limit and breaks the 1-3s rule. A 1-3s violation rejects the run and usually reflects random error, though it can also be a large systematic error. The run is held, the cause is found and corrected, controls are rerun, and patient samples since the last good run are reviewed.
- A: R-4s needs two results 4 SD apart within one run, usually across two levels.
- B: 2-2s needs two consecutive results beyond the same 2 SD limit; run 19 is within 2 SD.
- D: 10x needs ten consecutive results on one side of the mean.
Question 4 · Medium
The two-level Levey-Jennings chart shows calcium controls. The pattern in run 15 indicates:
Figure: MedExams
1. ALevel 1 vial problem only
2. BRandom error: reject and investigate
3. CSystematic error: recalibrate
4. DAcceptable: both within 3 SD
Show answerHide answer
Answer: B · Random error: reject and investigate
In run 15, level 1 reads +2.3 SD and level 2 reads -2.2 SD, a spread of 4.5 SD within one run, which breaks the R-4s rule. Controls moving apart in opposite directions point to random error, such as a bubble, a short sample or a pipetting fault. The run is rejected even though neither result is beyond 3 SD. A systematic error would move both levels in the same direction.
- A: A bad level 1 vial would not explain the low level 2 result.
- C: Systematic error moves both levels the same way; here they moved apart.
- D: The R-4s rule rejects the run even though neither control is beyond 3 SD.
Question 5 · Hard
The chart shows two ALT control levels over 20 days. The instrument log records: day 6, photometer lamp replaced; day 11, new reagent lot loaded and a new vial of level 1 control opened; day 16, sample probe cleaned. The most likely cause of the control pattern is:
Figure: MedExams
1. AThe photometer lamp replaced on day 6
2. BThe new level 1 control vial on day 11
3. CThe new reagent lot loaded on day 11
4. DRandom error from probe cleaning on day 16
Show answerHide answer
Answer: C · The new reagent lot loaded on day 11
Up to day 10 both levels scatter on both sides of their means; from day 11 every result of both levels lies above the mean, a shift that marks systematic error starting that day. By day 15, ten consecutive results across the two levels lie above the mean, breaking 10x. Two changes were made on day 11, but a new level 1 vial could move only level 1, while level 2 shifted too, so the common cause is the new reagent lot. Its lot-to-lot difference should be checked and the method recalibrated, and patient results since day 11 reviewed. The lamp change was followed by five days in control, and the probe cleaning came after the shift had begun.
- A: Both levels stayed in control for five days after the lamp change.
- B: A new level 1 vial cannot explain the shift in level 2.
- D: The shift began before day 16 and is systematic, not random.
Question 6 · Medium
The Levey-Jennings chart shows a creatinine control. Which rule is broken at run 20, and what type of error does it suggest?
Figure: MedExams
1. A1-3s: random error
2. BR-4s: random error
3. C10x: systematic error
4. D4-1s: systematic error
Answer and explanation in LearnFree account
Question 7 · Hard
After recalibration with a new calibrator lot, triglyceride controls with means of 50 mg/dL (0.56 mmol/L) and 200 mg/dL (2.26 mmol/L) read 55 mg/dL (0.62 mmol/L) and 220 mg/dL (2.48 mmol/L), stable from run to run. The most likely cause is:
1. AA reagent blank reading too high
2. BAir bubbles in the sample probe
3. CAn error in the calibrator's set value
4. DDeterioration of the control material
Answer and explanation in LearnFree account
Question 8 · Medium
A laboratory is putting a new lot of control material into use. How should its mean and SD be set?
1. AAssay it with the current lot over about 20 runs
2. BUse the manufacturer's range without testing
3. CCopy the mean and SD of the current lot
4. DAssay it 20 times within a single run
Answer and explanation in LearnFree account
Question 9 · Easy
In the classic Westgard multirule procedure, a control result between 2 SD and 3 SD from the mean is treated as:
1. AA reason to open a new control lot
2. BA reason to reject the run
3. CA warning to check the other rules
4. DA sign that recalibration is needed
Answer and explanation in LearnFree account
Question 10 · Hard
Level 1 glucose control, last 10 runs, in SD from the mean (oldest first): +0.5, +1.2, +0.9, +1.5, +0.3, +0.8, +0.6, +1.8, +1.1, +2.2. Level 2 stays within ±1 SD. Which rule is broken?
1. A4-1s, showing systematic error
2. B10x, showing systematic error
3. CNone; the run can be accepted
4. D2-2s, showing systematic error
Answer and explanation in LearnFree account
Question 11 · Medium
Which problem typically causes random rather than systematic error?
1. AA photometer lamp losing its intensity
2. BA wrong value entered for the calibrator
3. CA new reagent lot with lower reactivity
4. DAir bubbles in the sample syringe
Answer and explanation in LearnFree account
Question 12 · Hard
Controls were acceptable at 06:00. At 14:00 a 1-3s violation is found; the cause is fixed and repeat controls are in range. Which patient samples must be evaluated for possible retesting?
1. AOnly samples tested in the hour before 14:00
2. BNone, since controls are now back in range
3. CAll samples tested since the 06:00 run
4. DOnly samples with results outside the reference interval
Answer and explanation in LearnFree account
Question 13 · Hard
A cholesterol method has an allowable total error of 10%, a bias of 2.0% and a CV of 1.6%. What is its sigma metric?
1. A2.5
2. B6.3
3. C5.0
4. D7.5
Answer and explanation in LearnFree account
Question 14 · Medium
Since a sample probe was replaced, both control levels show much wider scatter around their usual means, with no shift. This pattern indicates:
1. AIncreased random error
2. BA systematic shift
3. CImproved precision
4. DA gradual systematic trend
Answer and explanation in LearnFree account
Question 15 · Medium
Which Westgard rule is applied only within a single run, not across runs?
1. A4-1s
2. B10x
3. CR-4s
4. D2-2s
Answer and explanation in LearnFree account
Question 16 · Easy
A Levey-Jennings chart shows:
1. ANew-method results plotted against the old method's
2. BPatient results plotted against the reference interval
3. CControl results plotted over time against SD limits
4. DAbsorbance plotted against standard concentration
Answer and explanation in LearnFree account
Question 17 · Hard
Potassium QC: level 1 mean 4.00 mmol/L, SD 0.10; level 2 mean 6.00 mmol/L, SD 0.15. In today's run, level 1 reads 4.23 mmol/L and level 2 reads 5.69 mmol/L. What is the correct interpretation?
1. AR-4s violation: reject the run
2. B2-2s violation: reject the run
3. C1-3s violation: reject the run
4. D1-2s warning only: accept the run
Answer and explanation in LearnFree account
Question 18 · Hard
Since a new vial of level 2 control was reconstituted this morning, it has read +2.6 and +2.4 SD in two runs. Level 1 on the same method stays near its mean, and no reagent or calibrator has changed. The first step should be to:
1. AAccept the runs, as level 1 shows the method is in control
2. BHold results and test a freshly reconstituted level 2 vial
3. CChange the reagent lot and repeat both control levels
4. DRecalibrate the method and retest all of today's patients
Answer and explanation in LearnFree account
Question 19 · Easy
In Westgard rules, a single control result beyond 3 SD from the mean breaks which rule?
1. AR-4s
2. B2-2s
3. C1-3s
4. D1-2s
Answer and explanation in LearnFree account
Question 20 · Hard
Run 1: level 1 +1.4 SD, level 2 +1.2 SD. Run 2: level 1 +1.6 SD, level 2 +2.1 SD. No earlier results are above +1 SD. The correct interpretation is:
1. A10x violation: systematic error
2. BR-4s violation: random error
3. C4-1s violation: systematic error
4. D1-2s warning only: accept the run
Answer and explanation in LearnFree account
Question 21 · Hard
Right after a new reagent lot is loaded, both control levels read above +2 SD (+2.3 and +2.5 SD) in the same run. Neither exceeds 3 SD. What is the correct action?
1. AAccept the run, since neither control exceeds 3 SD
2. BReject the run, check and calibrate the new lot, then rerun QC
3. CReport patients and move the control means to the new values
4. DRepeat the controls and report if both are within 2 SD
Answer and explanation in LearnFree account
Question 22 · Medium
If 1-2s is used as a rejection rule with two control levels in each run, about what share of good runs will be rejected falsely?
1. AAbout 9%
2. BAbout 13%
3. CAbout 5%
4. DAbout 0.3%
Answer and explanation in LearnFree account
Question 23 · Medium
Two consecutive control results fall above +2 SD on the same side of the mean. Which Westgard rule is broken, and what type of error does it suggest?
1. AR-4s, suggesting random error
2. B2-2s, suggesting systematic error
3. C1-3s, suggesting random error
4. D10x, suggesting systematic error
Answer and explanation in LearnFree account

Quality control (Westgard rules, Levey-Jennings)

The Levey-Jennings chart shows 20 runs of a glucose control. Which Westgard rule does run 20 break?

The two-level Levey-Jennings chart shows calcium controls. The pattern in run 15 indicates:

The chart shows two ALT control levels over 20 days. The instrument log records: day 6, photometer lamp replaced; day 11, new reagent lot loaded and a new vial of level 1 control opened; day 16, sample probe cleaned. The most likely cause of the control pattern is:

The Levey-Jennings chart shows a creatinine control. Which rule is broken at run 20, and what type of error does it suggest?

After recalibration with a new calibrator lot, triglyceride controls with means of 50 mg/dL (0.56 mmol/L) and 200 mg/dL (2.26 mmol/L) read 55 mg/dL (0.62 mmol/L) and 220 mg/dL (2.48 mmol/L), stable from run to run. The most likely cause is:

A laboratory is putting a new lot of control material into use. How should its mean and SD be set?

In the classic Westgard multirule procedure, a control result between 2 SD and 3 SD from the mean is treated as:

Level 1 glucose control, last 10 runs, in SD from the mean (oldest first): +0.5, +1.2, +0.9, +1.5, +0.3, +0.8, +0.6, +1.8, +1.1, +2.2. Level 2 stays within ±1 SD. Which rule is broken?

Which problem typically causes random rather than systematic error?

Controls were acceptable at 06:00. At 14:00 a 1-3s violation is found; the cause is fixed and repeat controls are in range. Which patient samples must be evaluated for possible retesting?

A cholesterol method has an allowable total error of 10%, a bias of 2.0% and a CV of 1.6%. What is its sigma metric?

Since a sample probe was replaced, both control levels show much wider scatter around their usual means, with no shift. This pattern indicates:

Which Westgard rule is applied only within a single run, not across runs?

A Levey-Jennings chart shows:

Potassium QC: level 1 mean 4.00 mmol/L, SD 0.10; level 2 mean 6.00 mmol/L, SD 0.15. In today's run, level 1 reads 4.23 mmol/L and level 2 reads 5.69 mmol/L. What is the correct interpretation?

Since a new vial of level 2 control was reconstituted this morning, it has read +2.6 and +2.4 SD in two runs. Level 1 on the same method stays near its mean, and no reagent or calibrator has changed. The first step should be to:

In Westgard rules, a single control result beyond 3 SD from the mean breaks which rule?

Run 1: level 1 +1.4 SD, level 2 +1.2 SD. Run 2: level 1 +1.6 SD, level 2 +2.1 SD. No earlier results are above +1 SD. The correct interpretation is:

Right after a new reagent lot is loaded, both control levels read above +2 SD (+2.3 and +2.5 SD) in the same run. Neither exceeds 3 SD. What is the correct action?

If 1-2s is used as a rejection rule with two control levels in each run, about what share of good runs will be rejected falsely?

Two consecutive control results fall above +2 SD on the same side of the mean. Which Westgard rule is broken, and what type of error does it suggest?