Analytic and post-analytic quality control
Hello, welcome to this 5th class of the Hematology course, provided by the University of Guanajuato. In this section we are going to review the analytical and post analytical part of the quality control, both still Important to guarantee reliable results.
In the analytical part we are going to learn some techniques to check the reliability of the automatized results as well as some of the manual results. These techniques are the Levey-Jennings graphs, where the Westgard’s rules must be applied in order to check that all the control measurements can be accepted. Every complete blood count parameter must be checked and controlled, therefore must be graphed and observed, and the Westgard’s rules applied.
In the post analytical part, presentation of the results in a special format is required. Special attention must be paid to the reference values and the units of each result, in order to report them correctly.
Everything must be written.
- QUALITY MANUAL: Policy on WHAT is to be done, WHO is to do it.
- OPERATING PROCEDURES: HOW to perform the functions. Detailed information about the different activities carried out in the laboratory. The objective is that anyone can easily and precisely know how the laboratory works.
- TECHNICAL PROCEDURES: HOW to carry out operations, detailed, clear, and understandable.
- CONTROL SHEETS. Records of reports, maintenance, and equipment calibrations.
ISO 15189 STANDARD
Quality control analytical stage
- Selection of quality control
- Selection of the analytical method
- Sensitivity, specificity, linearity, precision, accuracy.
- Selection of equipment
- Physical space, test menu, performance and speed, maintenance, sample volume, possibility of facilities, cost, competitiveness, evaluation, suppliers.
- Sample processing
Variables affecting the analytical stage
- Deterioration of reagent composition
- Failure in the actual mixing of the sample with the reagents * Failure in temperature control
- Exposure of the reagents to light and air
- Error in instrument performing
- Use of an incorrectly prepared reagent or calibrator
- Calibration of the instrument with expired reagents or reagents with exceeded shelf life in the instrument
- Training errors.
Proximity to the «true value».
In biological material it is not possible to know this.
The average value of many laboratories using the same method is assumed to be the «true value».
Reproducibility of the values of a series of measurements.
In practice, the degree of inaccuracy is evaluated through the sd or cv describing the dispersion between measurements.
Types of errors
The average result of a large number of repeated measurements of the same sample minus the true value of that sample.
A systematic error increases or decreases all results (in the same proportion), causing all results to move in the same direction.
The result of one measurement minus the average result of a large number of repeated measurements of the same sample.
An increase in random error causes the results to have a large variation making the distribution of error or imprecision very wide.
Control material: Substance routinely used in the laboratory to monitor the performance of the analytical process.
Reference material: Material or substance used for the calibration of an instrument or the evaluation of a measurement method.
Analysis of data resulting from the implementation of a quality system.
Levey – Jennings charts
The SD is the parameter to create the chart in which the daily values of the controls are indicated.
This graph is created for each test and for each level of control.
The limits of the graph are ± 1 SD, ± 2 SD and ± 3 SD with respect to the arithmetic mean.
- Provides a good visual representation of accuracy and precision.
- Easy to interpret.
- Time required to plot the data.
- Different plots are required for each determination and level of control.
a) Control material (commercial hematology material) or patient samples are used to verify the performance of the equipment.
b) Statistics are used to decide which of the results are not acceptable.
c) Red blood cells indices hace a small biological range, therefore the indices obtained when testing patient samples can be used to monitor Q.C.
Internal quality control
To be correlated:
- Cumulative results reports
- Extended ratio/cellular discounts
- Correlation with clinical events
At daily intervals:
Testing with control sample
- Enter control samples in each run
- Prepare control charts
- Enter duplicate samples from a few patient samples.
Internal quality control. Duplicate samples.
10 samples are read in duplicate e.g:
Hb determination (g/L)
The SD is calculated and NONE of the duplicates should differ by more than 2 SD (2x 2.37= 4.75).
Internal quality control
Statistical analysis of patient data
- With automated counters: VCM, HCM, CHCM averages
- With manual methods: mean CHCM
- At daily or weekly intervals: Check automated counters, spectrophotometers and other instruments.
- At monthly or quarterly intervals: Participate in an external quality assessment scheme.
- When necessary calibrate hematology counters and check other equipment.
The sd and cv are measures of precision.
The acceptable level of precision should be such that the random errors caused do not significantly affect the clinical interpretation of the results.
How often are they performed?
- CCI: every day or at least twice a week.
- SCC: once a month or in some programs on a quarterly basis.
Quality control is no longer an optional tool but a defensive tool when it comes to validating our results.
It should not be an obligation but a necessity in our quest to provide accurate and reliable results.
Post analytical stage
- Handling and treatment of results
- Correct use of units
- Reporting of results: clear, precise and avoid ambiguities. Report with reference values for each analyte and annotations that can help the physician to understand the reported value.
- Result printout
- Delivery of results: on time.
- Confidentiality of results
How to report the results obtained from a complete blood cell count?
The Levey-Jennings chart are useful to visualize the performance of the analytical part of the CBC.
The Westgard’s rules are crucial to ensure the reliability of the results.
The reference values and the units are important to specify In the results report