(IQC) - Internal Quality Control.pptx

Internal Quality Control
Dr. RAJEEV RANJAN
Resident, Dept. of Lab. Medicine
AIIMS, New Delhi
10th Dec, 2018
1

Overview
• What is Quality Control
• Internal Quality Control
• Reference Material & Control Material
• Process for IQC
• Interpretation
• Systematic Error
• Random Error
• Root Cause Analysis
• Concept of Lab Mean
• IQC vs EQA
2

INTRODUCTION
• Accurate analytical results are important in confirming diagnosis and to
monitor therapy.
• Evidence Based Medicine
• Laboratory Results must be Trustworthy
• Experience has shown that all Analytical results are subjected to errors
arising from a variety of causes.
• It is essential that these errors be kept to a minimum otherwise Lab becomes
an additional cost of Healthcare without adding value to it.
• So, Lab should maintain the Quality Results
3

Problems May Occur Throughout the Testing Process
1) Pre-analytical Pre-Examination  Examination Ordering
 Sample Collection
 Sample Transport
 Sample receipt and Processing
2) Analytical Examination  Examination
 Results Review and Follow Up
 Interpretation
 Reagents, Test methods, Competency of staff
QC (Quality Control)
 Internal quality control (IQC)
 External quality assessment (EQA)
3) Post-analytical Post-Examination  Results reporting and Archiving
 Sample Management
In a medical lab, we have three main stages that need control:
All of them should be under tight control 4

What is Quality Control ?
• QC in the laboratory involves
• Systematic monitoring of analytic processes
• In order to detect analytic errors
• That occur during analysis &
• To ultimately prevent the reporting of incorrect patient test results.
• QC is part of the performance monitoring that occurs after a test has
been established
5

Operation of Quality Control System
• The QC system in the clinical laboratory is used to monitor the analytic
variations that can occur.
• The QC program can be thought of as a 3-stage process:
• Establishing allowable statistical limits of variation for each analytic method
• Using these limits as criteria for evaluating the QC data generated for each test
• Taking action to remedy errors when indicated
a) Finding the cause(s) of error
b) Taking corrective action
c) Reanalyzing control and patient data
6

Different Types of IQC Materials
• Usually we procure commercial control materials
• Materials should be available for atleast 1 year
• Same matrix as patient samples
• These materials are often lyophilized (dehydrated to powder for stability)
• Can be reconstituted in specific diluents or matrices representing urine, blood, or CSF.
• Diluent should be carefully added and mixed.
• Incomplete mixing yields a partition of supernatant liquid and underlying sediment
and will result in incorrect control values.
• Control material concentrations should span the clinically important range of the
analyte at appropriate decision levels. (L1,L2,L3….PPU,PNU)
• Labs can use indigenous materials for IQC
• Pooled Serum : Lyophilized for later use
• Retained samples can also be used
7

Control Material Criteria to be considered :
1.Matrix • BASE from which the control material is prepared.
• Ideally the same matrix as specimen so that they behave like a specimen.
• Controls available are HUMAN based or BOVINE based
2.Reconstitution • Liquid / Lyophilized (Liquid controls better)
• Reconstitution material
• Quantity of solvent (Volume?)
• Mixing
• Waiting (when to use the reconstituted control)
• Vial to Vial variation
• (Standardization of these factors necessary)
3.Stability • Expiry date and stability after reconstitution should be considered
• Desirable is 1 year supply of the same lot, so that the lab can have its own QC range for a longer period of time.
4.Assayed or Un-assayed • Assayed QC material are the usually used type in our labs
• They have manufacturer values for each instrument / method
• Those values should be considered as guidelines only till each lab establishes its own QC range
• Minimum period of time for a lab to establish its own range is 20 days
• Ideally, a new QC lot should be run, hand in hand with the old lot, its values are established before it is run
routinely.
5.Appropriate Analyte level • The used QC material should be covering the measured range of the analyte, both in its normal and pathological
range.
• It is preferable to have a control value near the medical decision level.
• Ideally, there are 2 levels of QC material used in Chemistry analysis and 3 levels for Hormones and Tumor
markers.
8

Reconstitution of Control
• Secondary Reference Material
• QC – Range Value
• PNU (Preci Norm Universal) [Preci Control Clinchem – MULTI 1] – GREEN cap
• Normal
• Denoted as (C1)
• Lyophilized form – (stable at 2-80C until expiry date)
• Reconstituted in 5ml distilled water (1000µL x 5)
• Left for 30 min. to get mixed, in between mix them properly by shaking them
• PPU (Preci Path Universal) [Preci Control Clinchem – MULTI 2] – RED cap
• Pathological
• Denoted as (C2)
• Lyophilized form
• Reconstituted in 5ml distilled water
• After reconstitution, 500µL is alliquoted in cups
• These alliquotes to be used as Control
 These 500µL alliquoted cups are freezed in freezer for later use.
(Before using it, we keep if for 15min. outside the refrigerator to get it automatically thawed.)
9

When QC Material is Deficient
 Retained Patients Samples
Original samples must be assayed in the lab under accepted control run conditions
Stored aliquoted immediately after finishing 1st analysis
Storage conditions must meet the requirements for analytes to be measured
Avoid repeated warming/ thawing
Stability of the measurand must be carefully respected
Better to be alternating with QC material
Result of the 1st assay is used as the base for comparison and calculation (as if gold value)
Allowable analytical imprecision is used for judgment of acceptance or rejection
Data calculated from biological variations/ CLIA limits are used for the issue
11

Quality Control using Patient data
 Individual Patient Results:
• Clinical Correlations
• Correlation with other laboratory tests
• Intra laboratory duplicates
• Delta check with previous test results
• Limit Check
 Multiple Patients:
• Test distribution statistics
• Monitoring patients means
12

Process for IQC
 Frequency – NABL guidelines :
Minimum of One Level QC at least once a day
If Patient’s sample >25/day : 2 Level of QC at least once a day
If Patient’s sample >75/day : 2 Level of QC at least twice a day at appropriate intervals
13

Tools for Analysis of IQC
• Mean –
• Measurement of Central Tendency.
• Statistical mean is an arithmetic average.
• Standard Deviation (SD) –
• Standard deviation is a statistic that quantifies how close
numerical values are in relation to each other.
• Coefficient of Variation (CV) –
• (SD/Mean) x 100
• Ratio : can be used for comparison because it lacks unit.
14

Levey- Jennings Control Chart
• Graphical representation of observed values of a control material over
time in the context of the upper and lower control limits.
15

Interpretation of IQC : Westgard’s - Multirules
Dr. James Westgard
16

Westgard Multi-rules 12s Rule
• 1 reading exceeding 2 SD
• Warning Rule – does not cause Rejection of a run
• Alerts for possible method of Instrument malfunction
17

• 1 reading exceeding 3 SD
• Random error, Rejection
• A run is rejected when a single control measurement exceeds the
mean plus 3s or the mean minus 3s control limit
18

• 2 readings exceeding 2 SD
• Systematic error , Rejection
• 22s – reject when 2 consecutive control measurements exceed
the same mean plus 2s or the same mean minus 2s control limit.
19

Westgard Multi-rules R4s Rule
• 2 readings, their sum exceeds 4 SD
• Random error, Rejection
• R4s – reject when 1 control measurement in a group exceeds
the mean plus 2s and another exceeds the mean minus 2s
20

• 4 readings exceeding 1 SD, on one side of mean
• Systematic error, Rejection
• 41s – reject when 4 consecutive control measurements exceed the
same mean plus 1s or the same mean minus 1s control limit
21

Westgard Multi-rules 10x Rule
• 10 readings on the same side of the mean
• Systematic error, Rejection
• 10x – reject when 10 consecutive control measurements fall on
one side of the mean
22

Westgard Multi-rules 7T Rule
23

ERRORS
Systematic Error (SE) Random Error (RE)
o Systematic error is evidenced by a change in the mean of the
control values.
o Change may be gradual and demonstrated as a trend in control
values or it may be abrupt and demonstrated as a shift in control
values.
o Technically, random error is any deviation away from an
expected result.
o For QC results, any (+) or (-) deviation away from the calculated
mean is defined as random error.
o There is acceptable(or expected) random error as defined and
quantified by standard deviation.
o There is unacceptable (unexpected) random error that is any
data point outside the expected population of data
(e.g., a data point outside the ±3s limits)
 Systematic error, SE, or inaccuracy is an error that is always in
one direction, displacing the mean of the distribution from its
original value and cause all the test results to be either high or
low.
 Either constant or proportionate
 Random error, RE, or imprecision is described as an error that
can be either positive or negative, whose direction and exact
magnitude cannot be predicted, where the distribution of results
when replicate measurements are made on a single specimen.
Causes :
1. Deterioration of Calibrator/Calibration
2. Deterioration of Control Material
3. Deterioration of Reagents
Causes :
1. Pipetting Error
2. Temperature Error
3. Mixing Defect
4. Machine need Troubleshooting
24

Systematic Error…
Trend Shift
 A trend indicates a gradual loss of reliability in the test
system.
 Trends are usually subtle.
 Causes of trending may include:
 Abrupt changes in the control mean are defined as shifts.
 Shifts in QC data represent a sudden and dramatic positive
or negative change in test system performance.
 Shifts may be caused by:
• Deterioration of the instrument light source
• Gradual accumulation of debris in sample/reagent tubing
• Gradual accumulation of debris on electrode surfaces
• Aging of reagents
• Gradual deterioration of control materials
• Gradual deterioration of incubation chamber
temperature (enzymes only)
• Gradual deterioration of light filter integrity
• Gradual deterioration of calibration
• 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
• Failure in the sampling system
• Failure in reagent dispense system
• Inaccurate calibration/recalibration
An example of trending on a Levey-Jennings chart is
provided in Figure 5.
An example of a shift in test system performance is provided in
Figure 5.
25

Random or Systematic
• 13S and R4S usually associated with Random Error
• 22S, 41S, and 10X most often associated with Systematic Error
 3 are mandatory rules and the other 3 are warning rules.
Error Condition Westgard Rule
False Rejection 12s
Random Error 13s, R4s
Systematic Error 22s, 41s, 10x
27

When a Rule is violated: Root Cause Analysis
Change one variable at a time to troubleshoot an out of control test
 The variables are usually common to all instruments or procedure systems
and include the following:
1. Try a fresh vial of QC
2. Try a new reagent (same lot)
3. Try a new consumable (water, part, solution, tubing) if applicable
4. Try a new reagent (new lot)
5. Recalibrate (current calibrator)
6. Recalibrate (new lot calibrator)
 If a system cannot be brought into control, notify senior staff
 Do not process patient samples for the test affected
29

Concept of Laboratory Mean
 When even after all corrective action IQC is not coming at target mean.
 And EQC is within acceptable limits.
 Lab Mean and SD can be calculated using at least 20 controls points.
 New Mean and SD are used instead of manufacturer provided Mean and SD.
 New QC data is analyzed using Lab Mean and SD.
 Lab Mean is continuous process and needs periodic evolution.
 Setting a Lab Mean is not a substitute for root cause analysis.
30

Quality Control = Internal Quality Control + External Quality Assessment
31
Frequency Daily Usually once a month or periodically
Value Target Mean and SD is known Result is not provided to lab before hand
Timing IQC is done before patient samples EQC sample tested between patient sample
Treatment After reconstitution proper storage As patient sample, no special treatment
Interpretation Using L-J chart & Westgard’s rules Using Z score, SDI with other participating labs
Corrective action Root Cause Analysis, Day-to-Day action Retrospective, Future corrective action
Primary goal To monitor Precision To monitor Accuracy
Requirement Must Depends on Local Law

Consequences of Poor Quality
 Inappropriate action
 Over-investigation
 Over-treatment
 Mistreatment
 Inappropriate Inaction
 Lack of Investigation
 No treatment
 Delayed action
 Loss of credibility of Laboratory
 Legal action
32

Youden Plot
 In 1964, Dr. W.J. Youden proposed a method of evaluating Precision as well as Accuracy using
paired survey samples.
 2 samples sent to the participating laboratories.
 1 sample would have a low concentration, other a high concentration.
 Results are grouped by analyte & methodology & Mean & SD is calculated for each sample.
 A control chart is prepared so that 1 sample is on X-axis & the other on Y-axis.*
 **An easier to construct, Modified Youden Plot can be made by using uniform scales on both axes
& drawing rectangle that encompasses the area within the 3s limits of both samples.(Fig: 8.4 & 8.5) William John Youden
33

Implementation of New Lot of QC Materials
• Procedure:
• Over a 20 days period, assay 1 set of new QC material on each day along with the existing QC materials.
• After the runs are accepted based on the exciting QC materials, calculate the mean, SD and coefficient of
variance (%CV) for the new QC.
• Compare the SD and %CV with the List of Allowable Limits of Error (ALE) charts
• Make sure that the SD or %CV is less than half of the ALE (1/2 ALE) value that set in the ALE chart. For assayed
controls, the tentative mean should fall within the manufacturer’s quoted mean ± ALE. Should the
requirement not be met, compare values throughout the analytic range. If the difference is consistent, there
may be a standardization problem, which should be investigated. If the difference is inconsistent, the method
may not be usable or usable only over a narrower analytical range than the manufacturers’ claim. Follow up
investigation by contacting the manufacturer of the reagent system and the QC agency to verify the quoted
mean and compare the group mean of the other users using the same company kit.
• Prepare tentative QC chart for the analyte. Set mean ±2SD, the 95% confidence limit, as the temporary
target ranges, which should be less than the mean ± ALE.
• Ideally the new lot of QC material should overlap with the existing lot of QC for at least 20 batch of assays.
• Sometime this is not feasible e.g.:
• a) For manual and /or infrequent tests:
Step 1 may be replaced with a single evaluation in which 20 replicates of the new QC are run in a single batch.
• b) For a new method or when there is a lack of practical time:
Step 1 may be reduced to a minimum of 10 days.
34

Procedure for Setting up QC Limits for Multiple Analyzers
• Procedure:
1. There will always be a different mean and SD from one analyzer to the next. This is the result of
random variability. The differences should not be significant. Evaluating the significance of the
difference in mean values, however, is very important. Any medically significant difference
observed should be reported to the instrument manufacturer for action.
2. Separate means should be determined for each analyzer. This is important if statistical rules
(e.g. 13s, 22s) are used to monitor the performance. However, the same baseline SD could be used
for each analyzer.
3. The SD may be determined by averaging the observed SD from each instrument or by merely using
the largest SD observed across the instruments.
4. By using the same SD to monitor daily performance, one will be able to control multiple analyzers
as “one system”. This is important since patient samples can, practically, be analyzed on any
instrument in the laboratory.
5. Always use the same lot reagents and the same calibrators to calibrate both analyzers to minimize
calibration differences.
6. It is important to review maintenance log periodically. Perform any maintenance before calibration
rather than afterwards. For example, if the source lamp is about to change in one analyzer, one
should consider changing it on both analyzers.
35

(IQC) - Internal Quality Control.pptx

In this document

More Related Content

What's hot

Similar to (IQC) - Internal Quality Control.pptx

Recently uploaded

(IQC) - Internal Quality Control.pptx