Quality Assurance

in Medical Laboratory Sciences

4: Analytical & Post-analytical
Phase of QA

By:- Derbie A.(BSc, MSc )

Nov. 2024 G.C.
 Learning objectives
• Upon completion of this chapter the student will
be able to:
– Describe a quality control sample in terms of
assayed versus un-assayed, internal and external
and their use.
– Indicate the main components of a Levey-Jennings
chart including values and markings on the x and y
axis.
– Describe the following QC rules in terms of
appearance with control values and their ability to
detect systemic shifts and trends) versus random
precision errors:
– 1 3s, 2 2s, R4s, 6x, 7t & 4 1s 2
 Learning objectives …
• Upon completion of this chapter the student
will be able to:
– Differentiate quality assurance measures such
as absurd value check, duplicate analysis and
delta check.
– Discuss the use of mixed samples
– Describe the false sense of security in laboratory
analysis
– Describe correct documentation, interpretation
and reporting of tests results
3
 Outline
• Internal and external controls
• Source of External controls
• Internal quality control
• Types of internal quality control materials
• Characteristics of good control materials
• Control versus calibrator
• Preparation of quality control samples
• Practical evaluation & interpretation of quality control data
– Qualitative tests
– Quantitative tests systems
– Levey Jennings or the Shewhart control charts
– Shifts and trends
– Basic quality control rules
4
 Outline …
• Similarities in performance characteristics for quality
control and diagnostic tests
• Approaches used to interpret patient samples in
quality control
– Absurd value check
– Duplicate analysis
– Delta check
– Mixed samples
– False sense of security
• Post Analytical Quality Assurance
– Documentation of tests results
– Interpretation and Reporting of tests results 5
 4: Analytical & Post-analytical Phase
of QA
4.1. Internal and external controls
• QC: essential to ensure that laboratory tests produce
accurate and reliable results
• Quality control has two main components
– Internal Quality Control
– External Quality Assessment
• IQC: involves routine, daily testing of control samples
that are processed along with patient samples.
• EQA: an independent check on the quality & reliability
of the lab's results
– Allows the lab to benchmark its performance against
others
4: Analytical & Post-analytical QA …

The Need for Quality Control

• Patient results are unknowns.
• Analysis produces a number.
• Is the result valid?
• Patient may have previous result.
– Change from previous may be expected
– No change from previous result may be
expected
• Quality control samples are knowns.
4: Analytical & Post-analytical QA …

Quality Control Materials

• Similar to patient specimens
• Preserved for consistency
• Assayed: provided with mean and s for
analytes
• Un-assayed: you must determine mean
and sd
• QC materials may be made in-house
• QC materials may be made by
manufacturer
4: Analytical & Post-analytical QA …

Internal QC materials: Patient replicates

• Previously analysed patient sample
• Easily obtained
• Cost effective
• Results and samples readily available
4: Analytical & Post-analytical QA …
Calibrators
• Contain known concentration of the analyte
dissolved in a simple matrix
– Use to adjust the output of an instrument to
analyte concentration (calibration)
– Calibration occurs when first setting up an
instrument.
– Calibration occurs periodically
• according to manufacturer recommendations
• according to guidelines written in operating procedure
• Similar to standard
4: Analytical & Post-analytical QA …
Calibrator & Control?
• Both have known concentrations of the analyte(s)
specified
Feature Calibrators Controls
Establishes Monitors ongoing
Purpose
measurement accuracy performance
Sets reference points Validates calibration
Use
for quantification and system function
Known concentrations Known samples to
Composition
across a range check system reliability
Adjusts instrument Indicates if results
Outcome output based on known remain within expected
values limits
4: Analytical & Post-analytical QA …

Quality Control Use

• Qualitative testing of QC materials:
– Provide same type of result as patient
specimens
• Positive or Negative
• Recorded daily on monthly log but not plotted on a
QC chart
• Quantitative testing of QC materials:
– Analyzed solely for quality control purposes
– Not for calibration
– Plotted on monthly charts to track over time
4: Analytical & Post-analytical QA …
Preparing Quality Control Materials
• External QC material: Commercially purchased
– Foll ow manufacturer’s instructions
– Dilute or reconstitute with diluent
– Mix well and store according to instructions

• Internal QC material:
– Prepared within a lab
– Pool salvaged (At least 100 mL of serum)
• Exclude: obvious diseased, haemolysed, lipemic and
jaundiced)
– Determine mean and s
– Preserve for use in next run (-20 degrees C)
4: Analytical & Post-analytical QA …
Qualitative QC Results
• Month: Year:
• Lot # Positive Control
• Lot # Negative Control
• Lot # Kit Exp. Date:

Day Positive Negative Technologist

control control
4: Analytical & Post-analytical QA …
Quantitative QC Results and Expectations
• Gaussian distribution
– About 68% of the points fall within 1 SD of the
mean, 95% within 2 SD of the mean & 99.7%
within 3 SD of the mean
4: Analytical & Post-analytical QA …

QC Rules are Used as an Alarm System

• QC result should
– Indicate a true alarm: to "alert me
when the method has a problem

– Not indicate a false alarm: "don't

alert me when the method is
working okay
4: Analytical & Post-analytical QA …
QC Outside of Expected Range
• QC value > +3SD or < -3SD from the mean
– Commonly indicates a problem
• QC value > + 2SD or < -2 SD from the
mean
– 5% could be due to acceptable result
– False alarm if reject without further checks
• QC value > +1 SD or < -1 SD from the
mean
– Expected most of the time
4: Analytical & Post-analytical QA …
Historical Use of QC
• Walter A Shewart: developed statistical tools
to monitor quality
– The father of statistical quality control
• Levey & Jennings: develop Levey-Jennings
chart
– Quality standards for the laboratory
– Paved the way for implementing various QC rules
• Westgard: establishing rules for monitoring
quality control
4: Analytical & Post-analytical QA …

Shewart Chart
• Shewart set up separate
mean & standard deviation or
range charts to evaluate
quality.
• Mean of daily QC results are
plotted.
• S of daily QC results are
plotted on a separate chart.
• Forerunner to Levey-
Jennings charts
4: Analytical & Post-analytical QA …
Levey-Jennings Chart
4: Analytical & Post-analytical QA …
Levey-Jennings Chart …
• A graphical method for displaying control results &
evaluating whether a measurement procedure is in-
control or out-of-control.
– Control results are plotted versus time or sequential
run number
– lines are generally drawn from point to point to accent
any trends, systematic shifts, and random excursions
• Control limits are lines drawn on a control chart to
provide graphical criteria for assessing whether a
measurement procedure is in-control or out-of-
control
4: Analytical & Post-analytical QA …
Levey-Jennings Chart …
• Steps
1. Collect initial data (run at least 20
control samples)
2. Calculate the mean & Sd, &
establish control limits
3. Prepare the chart (draw)
4. Collect QC data (measure)
5. Plot the data point (locate time on
the x-axis & measured value on
the y-axis)
6. Connect the data points
7. Interpret the results
4: Analytical & Post-analytical QA …
Levey-Jennings Chart …
Instrument: _________________ Test: ____________________ Units: ________
Method: ___________________ Month/Year: ______________ Tech: ________
Control: ____________________ Lot #: ____________________ Exp: _________
Control Mean: __________ +1 Std Dev: _________ +2 Std Dev: _______ +3 SD : _______

+ 3s(90)

+ 2 s (85)

+ 1 s (80)

Mean (75 mg/dL)

- 1 s (70)

- 2 s (65)

- 3 s (60)

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

Days
4: Analytical & Post-analytical QA …
Quality Control Limits
• Perform multiple runs (at least 20) of the assay
using a control material to collect initial control
values.
• This initial data set will help establish the mean and
standard deviation.
• Determine Control Limits
• Mean is the central line.
– One standard deviation above = + 1 s
– One standard dev below = - 1s
– 2 standard dev above = + 2s
– 2 standard dev below = -2s
– 3 standard dev above = + 3s
– 3 standard dev below = -3s
4: Analytical & Post-analytical QA …
Calculating Confidence Limits on the Chart
• For example, Control 1 has a mean of 200 and a
standard deviation of 4 mg/dL.
• What is the 95% confidence limit or mean + 2 s
limit?
– The upper control limit would be:
200 + 2*4 = 208 mg/dL
– The lower control limit would be:
200 - 2*4 = 192 mg/dL
4: Analytical & Post-analytical QA …
QC with each analytical run
• Run, analytical run, or run length: the interval/period of time
/ group of samples for which a decision on control status is
to be made.
– Each batch is a run for a manual procedure ( 2 controls should
be run with each batch)
– 24 hours is usually the longest run time for automated chemistry
methods (2 controls should be run at least once every 24 hr)
– 8 hours is the longest run time for automated haematology
methods (3 controls should be run at least once every 8 hr)
4: Analytical & Post-analytical QA …

QC Procedure
• Written guidelines (SOPs) should be
developed to define the QC procedure in
detail including;
– How many controls
– How often run
– How document results and problems
– How problems are solved
4: Analytical & Post-analytical QA …

Quality Control Rules

• Control rule means decision criteria for judging
whether an analytical run is in-control or out-of-
control
• Should detect random errors
– Imprecision
• Should detect systematic errors
– Shift (6 results that settle around new mean)
– Trend (7 results upward or downward )
4: Analytical & Post-analytical QA …
QC Rules …
12s rule: (One Point Outside ±2 SD)
• One control measurement exceeds ±2 standard
deviations (SD) from the mean.
– This rule alone is not sufficient to reject a run
– Indicates a warning; further investigation needed if
repeated
4: Analytical & Post-analytical QA …
QC Rules …
13s rule: (One Point Outside ±3 SD)
• Patient results are held/ rejected when a single
control measurement exceeds + 3 S
– Indicates a random error
– Look for the cause of the random error, solving the
problem before reporting the patient results
4: Analytical & Post-analytical QA …
QC Rules …
R4s Rule:(One Point Outside +2 SD & One Point Outside -2 SD)
• Patient results are held/ rejected when 1 control
measurement in a group exceeds the mean plus 2s and
another exceeds the mean minus 2s.
• The range (difference) between two consecutive
control measurements exceeds 4 SD
• This could mean the same control
sample 2 runs in a row or it could
be both controls in one run.
• Extreme random error indicated
4: Analytical & Post-analytical QA …
QC Rules …
22s Rule:(Two Consecutive Points Outside ±2 SD)
• Patient results are held/ rejected when 2 consecutive
control measurements exceed the same mean plus
2s or the same mean minus 2s control limit.

• Indicates systematic error

• Stop & look for the cause,
solving the problem before
reporting the patient results
4: Analytical & Post-analytical QA …
QC Rules …
6 x Rule:
• The patient results are held/ rejected if six consecutive
control measurements fall on one side of the mean
(either all above or all below)
– Shifts
– Indicates a systematic error
4: Analytical & Post-analytical QA …
QC Rules …
• 7t Rule:
• Patient results are held/ rejected, if 7 consecutive
control measurements show an upward or
downward pattern.
– Trend
– Systematic error detected
4: Analytical & Post-analytical QA …

Westgard Multi-rules System

• QC rules work best in a multi-rule system
• Warning rules trigger the need to look for other
violations
– 1 2s
4: Analytical & Post-analytical QA …
Multi-rules

• This decision tree help to decide whether to reject or accept a

run.
• If you answer yes, it means that rule has been violated.
– 4 1s may be replaced by 7t and 10x may be replaced by 6x.
4: Analytical & Post-analytical QA …

Summary
• Control materials can be described in
terms of
– Assayed (mean and s provided)
– Un-assayed (mean and s not provided)
– Use (validation of run, not as a calibrator)
4: Analytical & Post-analytical QA …
Summary
• Levey-Jennings chart including values and markings on the x
(days or time) and y axis (concentration mean and s).
• Errors such as shifts and trends can be detected.
4: Analytical & Post-analytical QA …
Summary …
• Plotting QC over time and applying QC rules can
help to detect random errors in patient results.
Thank you !!!
40
4: Analytical & Post-analytical QA …
Approaches used to interpret patient
samples in quality control
• Includes;
– Absurd value check
– Duplicate analysis
– Delta check
– Use of mixed samples
– False sense of security
• These play key roles in identifying & preventing
potential errors
4: Analytical & Post-analytical QA …
Absurd value check
• Biological Limit check
– Examples of results not compatible with life
• Sodium result of 100 mmol/L
• Hemoglobin result of 2 g/dL
• Glucose result of 2 mg/dL
– Internal quality assurance measure
– These results indicate the need to check the
instrument, reagent or specimen for errors and
retest specimen.
4: Analytical & Post-analytical QA …

Duplicate analysis
• Test 10 consecutive patient specimens in
duplicate
– Standard deviation (s)
– 2 s is the limit for difference between
duplicates
• Useful as internal quality assurance
measure
4: Analytical & Post-analytical QA …

Delta Check
• Compare patient parameters with previous
run parameters from the same patient
• If a difference considered an analytical error
– Precision and accuracy
• Example
– Hemoglobin and MCV
– Na
• Internal quality assurance measure
4: Analytical & Post-analytical …
Use of Mixed samples:
• Created by pooling patient specimens with known
analyte levels
• To assess laboratory test consistency and to ensure
all processes are working as expected
• Serve as a routine QC check or help troubleshoot
when results seem unusual.
• Ideal for routine performance checks
– Eg: If a pooled sample result falls outside of expected
ranges, it may suggest issues with reagents or
instrumentation
4: Analytical & Post-analytical …
False sense of security
• A robust QC system involves more than just passing
QC checks
• It requires a comprehensive approach that
incorporates
– Critical evaluation, external verification, and
continuous monitoring
• Minimizing the risk of a false sense of security by
– Understanding the limitations of QC procedures and
– Implementing strategies to detect subtle issues
4: Analytical & Post-analytical …
False sense of security …
• Key factors that can contribute to a false sense of
security:
– Overreliance on QC passing without cross-verification
– Limited QC samples that don’t reflect real patient variability
– Infrequent/inadequate QC Testing
– Misinterpretation of statistical QC Tools
– Ignoring QC trends and subtle shifts
– Assumption of manufacturer quality control sufficiency
– Failure to account for human errors in QC processes
– Relying solely on internal QC without EQA
– Underestimating sample matrix effects
– Inadequate corrective action plans
4: Analytical & Post-analytical …
Interpretation and Reporting Patient Results
• For interpretation of microscopic or gross
examination results
– Expected normal results should be established
– Morphological changes should be determined based on
standardized criteria
• Reference ranges should be established for all
quantitative test results
• For Interpretation of patient quantitative results
– Compare the Patient Result with the Reference Range
4: Analytical & Post-analytical …

Reporting the patient result

• 3 Rs
– Reliability
– Rapidity
– Relevance
4: Analytical & Post-analytical …
Reporting of Patient Results
• Use approved form
• Fill in relevant information:
– Patient name and accession number
– Ordering physician
– Specimen collection information
– Patient Result
– Reference Range
– Name of reporting technologist
– Date and time of report
– Comments by reporting technologist
– Reviewer name, date and time
4: Analytical & Post-analytical …
Standardization in reporting test results
• Results and units
– Interpretation of results
• Benefit of standardized reporting
– Comparison of results
– Efficiency of laboratory service
– Referral of patient from another health facility
4: Analytical & Post-analytical …
Patient
Report Form
4: Analytical & Post-analytical …

Reporting of Patient Results

• Results should be carefully review before
reporting to clinicians.
• Avoid common Post-analytic Errors:
– Wrong name
– Incorrect units or reference range
– Transcribing wrong number
– Report too late
4: Analytical & Post-analytical …

Document Examples
Record book
Specimen receipt log
4: Analytical & Post-analytical …
Documentation and Record Management
• Manage patient test records in whatever method
works best
– Carbon copies of reports
– Worksheets
– Register or logbooks
• Manage test records in worksheets or registers
– Work reports
– Laboratory workload
– Instrument maintenance
– Quality control logs
– Occurrence logs
4: Analytical & Post-analytical …
Documentation
• Document storage and retrieval
– Use format that works best
 Summary
– Absurd value check, duplicate analysis and delta
checks are used for post-analytical quality
assurance before patient results are reported.
– Mixed samples
– A false sense of security in laboratory analysis
– Correct documentation in specimen and result
registers, interpretation of patient results
compared to reference ranges and reporting of
patient tests results on the approved test report
form are important for post-analytical quality
assurance.
Thank you !!!
58