QUALITY

CONTROL
IN THE
CLINICAL
LABORATORY
(revisited)
 Fill in the blank?

______________ in the laboratory is a

system designed to increase the
probability that each result reported is
valid and can be used with confidence
by the physician making a diagnostic or
therapeutic decision.
 Quality Control

The statistical process requires:

• Regular testing of quality control products

along with patient samples

• Comparison of quality control results to

specific statistical limits (ranges)
Quality Control

• QC results are used to validate patient results

• Ex. Serum potassium = 1.9 mmol/L
• Is the result truly reliable?
• What can be done to increase the confidence
that the patient is truly HYPOKALEMIC?
• Resolved by regular use of quality control
materials and statistical process control
 Quality Control Products

• Liquid or freezed-dried materials of human,

animal, or chemical origin that are used to
monitor the quality and consistency of the
analytical process.

• A normal control product contains normal levels

for the analyte being tested; An abnormal control
product contains the analyte at a concentration
above or below the normal range for the analyte
 Regular Testing

• Regular testing of quality control products

creates QC database to validate patient results
• Validation occurs by comparing daily QC
results to the laboratory-defined range of QC
values.
• Lab-defined range is calculated from QC data
collected from testing of normal and abnormal
controls
 LEVEL I LEVEL II PATIENT RESULTS
NORMAL ABNORMAL
CONTROL CONTROL
RANGE 3.7 – 4.3 6.7 – 7.3
mmol/L mmol/L
DATE 3/1/07 4.0 7.0 4.2, 4.0, 3.8, 5.0, 5.8, 4.2

3/2/07 4.1 7.0 3.8, 4.4, 4.6, 3.9, 4.8, 4.4, 3.9

3/3/07 4.0 6.9 4.4, 3.9, 3.7, 4.7

3/4/07 4.2 7.1 4.7, 5.6, 4.2, 3.7, 4.3

3/5/07 4.1 7.0 4.2, 4.3, 4.1, 4.3

3/6/07 4.1 7.0 4.6, 4.4, 5.5, 3.8, 3.2

3/7/07 4.2 8.0 2.8, 4.6, 4.2, 3.2, 3.9, 4.1, 6.0, 4.3
• On March 1 through March 6, both controls
were “in control” and patient values could be
reliably reported
• However, the laboratory was “out of control”
for abnormal high potassium on March 7; the
value obtained for the QC material (8.0
mmol/L) was outside the acceptable range (6.7
– 7.3 mmol/L)
 LEVEY-JENNINGS
3 SD CHART
THE STANDARD DEVIATION
REPRESENT THE EXPECTED
2 SD IMPRECISION FOR THE
METHOD
1 SD
MOST LABS SET
THEIR DECISION
MEAN LIMITS AT THE +/-
2SD RANGE

1 SD

THE MEAN REPRESENTS THE

2 SD
TARGET VALUE FOR THE
RESULT
3 SD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
 Calculation and Use of QC Statistics

Mean
• The laboratory’s best estimate of the analyte’s
true value for a specific level of control
Mean:
 LEVEL I LEVEL II PATIENT RESULTS
NORMAL ABNORMAL
CONTROL CONTROL
RANGE 3.7 – 4.3 6.7 – 7.3
mmol/L mmol/L
DATE 3/1/07 4.0 7.0 4.2, 4.0, 3.8, 5.0, 5.8, 4.2

3/2/07 4.1 7.0 3.8, 4.4, 4.6, 3.9, 4.8, 4.4, 3.9

3/3/07 4.0 6.9 4.4, 3.9, 3.7, 4.7

3/4/07 4.2 7.1 4.7, 5.6, 4.2, 3.7, 4.3

3/5/07 4.1 7.0 4.2, 4.3, 4.1, 4.3

3/6/07 4.1 7.0 4.6, 4.4, 5.5, 3.8, 3.2

3/7/07 4.2 8.0 2.8, 4.6, 4.2, 3.2, 3.9, 4.1, 6.0, 4.3
 Calculation and Use of QC Statistics

• Mean
Sum = 28.7 mmol/L
n=7

4.1 mmol/L
 Calculation and Use of QC Statistics

• Standard Deviation
• Quantifies how close numerical values are in
relation to each other
• Provides the laboratory an estimate of test
consistency at specific concentrations
• The repeatability of a test may be consistent (low
SD, low imprecision) or inconsistent (high SD,
high imprecision)
Standard Deviation:
 LEVEL I LEVEL II PATIENT RESULTS
NORMAL ABNORMAL
CONTROL CONTROL
RANGE 3.7 – 4.3 6.7 – 7.3
mmol/L mmol/L
DATE 3/1/07 4.0 7.0 4.2, 4.0, 3.8, 5.0, 5.8, 4.2

3/2/07 4.1 7.0 3.8, 4.4, 4.6, 3.9, 4.8, 4.4, 3.9

3/3/07 4.0 6.9 4.4, 3.9, 3.7, 4.7

3/4/07 4.2 7.1 4.7, 5.6, 4.2, 3.7, 4.3

3/5/07 4.1 7.0 4.2, 4.3, 4.1, 4.3

3/6/07 4.1 7.0 4.6, 4.4, 5.5, 3.8, 3.2

3/7/07 4.2 8.0 2.8, 4.6, 4.2, 3.2, 3.9, 4.1, 6.0, 4.3
 Calculation and Use of QC Statistics

• Standard Deviation

S = 0.082 or 0.10 (Rounded)

 Creating a Levey-Jennings Chart

• Used to graph successive quality control

values
• 1st step: calculate the decision limits
• These limits are +/-1s, +/-2s, and +/-3s from the
mean
 Creating a Levey-Jennings Chart

• The +/-1s, +/-2s, and +/-3s quality control limits are

calculated as follows:
• The +/-1s range is 4.0 to 4.2 mmol/L:
4.1 – 0.10(1) = 4.0
4.1 – 0.10(1) = 4.2
• The +/-2s range is 3.9 to 4.3 mmol/L:
4.1 – (0.10)(2) = 3.9
4.1 – (0.10)(2) = 4.3
• The +/-3s range is 3.8 to 4.4 mmol/L:
4.1 – (0.10)(3) = 3.8
4.1 + (0.10)(3) = 4.4
 LEVEY-JENNINGS
3 SD CHART
THE STANDARD DEVIATION
REPRESENT THE EXPECTED
2 SD IMPRECISION FOR THE
METHOD
1 SD
MOST LABS SET
THEIR DECISION
MEAN LIMITS AT THE +/-
2SD RANGE

1 SD

THE MEAN REPRESENTS THE

2 SD
TARGET VALUE FOR THE
RESULT
3 SD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Calculation and Use of QC Statistics

• QC statistics for each test performed in the

laboratory are calculated from the QC
database collected by regular testing of control
products

• Mean
• Standard deviation
 Using the Levey-Jennings Chart to Evaluate
Run Quality
• The laboratory needs to document that quality
control materials are assayed and that the
documentation is accomplished by:
• Maintaining a QC Log
• Using Levey-Jennings chart on a regular basis
• The log should identify the name of the test, the
instrument, units, the date the test is performed,
the initials of the person performing the test, and
the results for each level of control assayed
 Systematic Error

• Systematic error is evidenced by a change in

the mean of the control values.

• The change in the mean may be gradual and

demonstrated as a trend or it may be abrupt
and demonstrated as a shift.
 Systematic Error

• Trend
• Indicates a gradual loss of reliability in the test
system
• Causes:
• Deterioration of the instrument light source
• Gradual accumulation of debris in sample/reagent
tubing
• Gradual accumulation of debris on electrode surfaces
 Systematic Error

• Trend
• Causes:
• Aging of reagents
• Gradual deterioration of control materials
• Gradual deterioration of incubation chamber
temperature (enzymes only)
• Gradual deterioration of light filter integrity
 Systematic Error

• Shift
• Abrupt changes in the control mean
• Represent a sudden and dramatic positive or
negative change in test system performance
 Systematic Error

• Shifts
• Causes:
• Sudden failure or change in the light source
• Change in reagent formulation
• Change of reagent lot
• Major instrument maintenance
• Sudden change in incubation temperature (enzymes
only)
• Change in room temperature or humidity
 Systematic Error

• Shift
• Causes:
• Failure in the sampling system
• Failure in reagent dispense system
• Inaccurate calibration/recalibration
 Random Errors

• Any deviation away from an expected result

• Any positive or negative deviation away from
the calculated mean
• Acceptable random error
• Unacceptable random error
 Westgard Rules

• Dr. James Westgard (1981)

• Six basic rules of the Westgard scheme
• 12s Rule
• 13s Rule
• 22s Rule
• R4s Rule
• 41s Rule
• 10x Rule
 Westgard Rules

• 12s Rule
• Warning rule which is violated when a single
control observation is outside the +/-2s limits
• Warns that random error or systemic error may be
present in the test system
• Relationship between this value and other control
results w/in the current and previous analytical runs
must be examined
• Assume that a single control value outside the +/-2s
limits is an acceptable random error
12s Rule:
 Westgard Rules

• 13s Rule
• Identifies unacceptable random error or possibly the
beginning of a large systematic error
• Any QC result outside +/- 3s violates this rule
13sRule:
 Westgard Rules

• 22s Rule
• Identifies systematic error only
• Criteria for violation:
1. Two consecutive QC results
2. Greater than 2s
3. On the same side of the mean
22sRule:
 Westgard Rules

• 22s Rule
• Applications:
1. Within run
• Violation indicates that the systematic error is
present and that it affects potentially the entire
analytical curve
2. Across run
• Violation indicates that only a single portion of the
analytical curve is affected by the error
 Westgard Rules

• R4s Rule
• Identifies random error only, and is applied only
w/in the current run
• If there is at least a 4s difference between control
values w/in a single run, the rule is violated for
random error
R4sRule:
 Westgard Rules

• Violation of the ff rules does not necessarily

require rejection of the analytical run:
• 31s Rule
• Criteria:
1. Three consecutive results
2. Greater than 1s
3. On the same side of the mean
• 41s Rule
• Criteria:
4. Four consecutive results
5. Greater than 1s
6. On the same side of the mean
41sRule:
 Westgard Rules

• 7x, 8x, 9x,10x, and 12x Rule

• Rules are violated when there are:
1. 7 or 8, or 9, or 10, or 12 control results
2. On the same side of the mean regardless of the
specific standard deviation in which they are located
• Applications:
3. Within control material
– Systematic bias in a single area of the method curve
4. Across control materials
– Systematic bias over a broader concentration
10xRule:
Exercise:
Exercise:
Exercise:
Exercise:
 THE WESTGARD RULES
RULE MEANING DETECTS
12S One observation exceeds 2 SD from the
target value (mean)
Warning for random or
systematic error

13S One observation exceeds 3 SD from the

mean
Random error

22S Two sequential observations exceed 2SD

from the mean in the same direction
Systematic error

R4S Range between two observations in the

same run exceed 4SD
Random error

Ten sequential observations are on the Systematic error, trend

10x same side of the mean

41S Four sequential control observations

exceed the mean by 1 SD in the same
Systematic error

direction
LOGIC DIAGRAM
 ACROSS RUNS
ACROSS MATERIALS?
WITHIN RUNS
WITHIN MATERIALS?
• THERE IS A 22S
VIOLATION FOR BOTH
THE HIGH AND
NORMAL CONTROL
LEVEL FOR THIS
ANALYTE.
• WE SAY THAT THERE IS
A 22S VIOLATION
ACROSS MATERIALS
AND WITHIN RUN (3).
• THERE IS AN R4S
VIOLATION FOR THE
CONTROL LEVEL FOR
THIS ANALYTE.
• WE SAY THAT THERE IS
AN R4S VIOLATION
WITHIN MATERIAL
AND ACROSS RUNS
(FROM 5-6).
• THERE IS A 22S
VIOLATION FOR ONLY
THE HIGH CONTROL
LEVEL FOR THIS
ANALYTE.
• WE SAY THAT THERE IS
A 22S VIOLATION
ACROSS RUNS (FROM
9-10) AND WITHIN
MATERIAL.
• THERE IS A 41S
VIOLATION FOR BOTH
THE HIGH AND
NORMAL CONTROL
LEVELS FOR THIS
ANALYTE.
• WE SAY THAT THERE IS
A 41S VIOLATION
ACROSS RUNS (FROM
11-12) AND ACROSS
MATERIALS.
• THERE IS AN R4S
VIOLATION FOR BOTH
THE HIGH AND
NORMAL CONTROL
LEVELS FOR THIS
ANALYTE.
• WE SAY THAT THERE IS
AN R4S VIOLATION
ACROSS MATERIALS
AND WITHIN RUN (14).
• THERE IS A 10X
VIOLATION FOR BOTH
THE HIGH AND
NORMAL CONTROL
LEVELS FOR THIS
ANALYTE.
• WE SAY THAT THERE IS
A 10X VIOLATION
ACROSS MATERIALS
AND ACROSS RUNS
(FROM 16-20).
 DAY 5
• 13s violation.

DECISION:
• REJECT RUN
• LIKELY
RANDOM
ERROR
 DAY 6
• 12S WARNING

DECISION:
• ACCEPT RUN.
 DAY 8
• 22S VIOLATION
ACROSS
MATERIALS

DECISION:
• REJECT RUN
• LIKELY
SYSTEMATIC
ERROR
 DAY 11
• R4S VIOLATION
ACROSS
MATERIALS

DECISION:
• REJECT RUN.
• LIKELY
RANDOM
ERROR.
 DAY 13
• 12S WARNING.

DECISION:
• ACCEPT RUN.
 DAY 14
• 22S VIOLATION
ACROSS RUNS &
WITHIN
MATERIAL.

DECISION:
• REJECT RUN.
• LIKELY
SYSTEMATIC
ERROR.
 DAY 27
• 10X VIOLATION
ACROSS RUNS &
WITHIN
MATERIAL.

DECISION:
• REJECT RUN.
• LIKELY
SYSTEMATIC
ERROR.
 DAY 29
• 13S VIOLATION
ON THE HIGH
CONTROL
LEVEL.

• 22S VIOLATION
ACROSS
MATERIALS.

DECISION:
• REJECT RUN.
• LIKELY
SYSTEMATIC
ERROR.
➢ When the control system gives a rejection
signal, a problem-solving procedure should
be initiated.
➢ The particular control rule violated
may give an indication of the TYPE of
ERROR that is occuring.
➢ 22s, 41s, 10x – systematic error.
➢ 13s, R4s – random error.
 SOURCES OF RANDOM
1.
ERROR
Power Supply
2. Double pipetting of control sample
3. Misplacement of Control sample within
the run
4. Random air bubbles in reagent or
sample pipette system
5. Incorrect reconstitution of the control
product
6. Inappropriate Storage of control
7. Inadequately mixed reagents
8. Individual operator variation in
pipetting, timing , etc.
 SOURCES OF SYSTEMATIC
1.
ERROR
Improper alignment of sample or reagent
pipettes resulting to change in reagent and
calibrator volume.
2. Drift or shift in incubator chamber
temperature, deterioration of photometric light
source
3. Change of Calibration /reagent lot
4. Deterioration of reagent /calibrator/control
product while in use, storage or shipment
5. Incorrect handling of control product
6. Inadequate storage of reagent or
calibrators/control products
7. Change in test operator
➢ Once PROBLEM is identified:
➢ Hold patients’ results.
➢ Correct the problem.
➢ Re-analyze the patients’ samples.
➢ Consult a supervisor for any decision
to report data.
 OVERVIEW OF THE QC
PROCESS
TESTING THE
MEASUREMENT
PROCESS/METHOD

QUALITY CONTROL
PROCEDURES

NO YES
CORRECTIVE
ACTIONS
?
STABLE
REPORT PATIENT
RESULTS
YES
REPEAT TESTING OF
PATIENTS
 What rule is violated?
3 SD

2SD

1SD

Mean
10 x rule is violated.

-1SD

-2SD

-3SD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
 What rule is violated?
3 SD 13s rule is violated.

2SD

1SD

Mean

-1SD

-2SD

-3SD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
 What rule is violated?
3 SD

2SD

1SD

Mean

-1SD
No rule is violated.
-2SD

-3SD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
 What rule is violated?
3 SD

2SD

1SD

R4s rule is violated.

Mean

-1SD

-2SD

-3SD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
 What rule is violated?
3 SD

2SD

1SD

Mean
12s rule is violated.

-1SD

-2SD

-3SD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
 What rule is violated?
3 SD

2SD

22s rule is violated.

1SD
A SHIFT is noted.
Mean

-1SD

-2SD

-3SD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
 What rule is violated?
3 SD

2SD
22s rule is violated.
1SD
A TREND is suggested.
Mean

-1SD

-2SD

-3SD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20