Why Does Terminology Matter?
In the world of laboratory analyses, linguistic precision is as important as measurement precision. Improper understanding of basic concepts can lead to incorrect interpretation of results, problems with accreditation, or misunderstandings with laboratory clients.
When a laboratory receives a test result with the value "detected below the limit of quantification," every team member must understand what this exactly means and how it differs from "not detected." When an auditor asks about the difference between repeatability and reproducibility, the answer must be unambiguous.
Consequences of Misunderstandings
Many times it has turned out that a client testing the same batch of product in different laboratories received seemingly different results. Usually, the cause is not technical errors, but fundamental misunderstandings about what and how we actually measure, and how we interpret results or method limits.
Verification vs Validation - Similar, but Not the Same
These two terms are often used interchangeably, which leads to confusion. Meanwhile, they represent two fundamentally different processes in the laboratory.
Method Verification
Verification answers the question: "Does this known method work correctly in our laboratory?"
When a laboratory implements a standard method, for example an ISO standard, it is not creating something new. The method has already been developed and thoroughly tested. Verification consists of confirming that in the specific conditions of a given laboratory - with its equipment, reagents, personnel, and procedures - this method works as expected.
Verification Example
A laboratory implements a standard spectrophotometric method for determining iron in water, described in a standard. It performs a series of tests with certified reference materials, checks whether the results fall within acceptable limits. This is verification - confirmation that a known method works for us as it should.
Method Validation
Validation is a much more complex process, answering the question: "Is this new or modified method suitable for its intended purpose?"
Validation is performed when a laboratory develops its own method, modifies an existing one, applies a known method to a new type of sample, or uses a method that does not have full primary validation. The validation process requires thorough examination of all parameters characterizing the method and proving that it meets the requirements. It is sufficient to perform it once, when introducing the method; subsequent checks of the method's performance should be based on method verification.
| Aspect | Verification | Validation |
|---|---|---|
| Type of method | Standard method with full primary validation | New or modified method |
| Objective | Confirmation of performance | Full characterization |
| Scope | Shorter process | Comprehensive process |
| Parameters | Fewer parameters | All parameters |
Precision and Accuracy - The Heart of Measurement Quality
Understanding the difference between precision and accuracy is crucial for interpreting laboratory results. Both concepts describe measurement quality, but from completely different perspectives.
Precision - Clustering of Results
Precision tells us how much the results of repeated measurements of the same sample agree with each other. A precise method produces results clustered close to each other, even if they are all shifted from the true value.
Imagine shooting at a target - precision is the clustering of all shots in a small area. They may hit beside the center, but if they are close to each other, precision is high.
Repeatability and Reproducibility
Repeatability is the agreement of results obtained under identical conditions: the same analyst, the same equipment, the same reagents, analysis time as close as possible.
Reproducibility is the agreement of results obtained under changed conditions: different analysts, different equipment, different time. This is a more realistic picture of what we can expect in practice.
Accuracy - Closeness to Truth
Accuracy determines how close the measurement result is to the true value. An accurate method gives results close to the actual content of the substance in the sample.
Returning to the target analogy - accuracy is hitting the center. We can have high accuracy (the average of many shots is in the center), even if individual shots are quite dispersed.
Four Scenarios of Measurement Quality
- High precision and high accuracy: Ideal situation - results clustered and close to the true value
- High precision, low accuracy: Results close to each other but systematically shifted (systematic error)
- Low precision, high accuracy: Results dispersed but close to the true value
- Low precision and low accuracy: Chaotic and incorrect results - method requires intervention
Limit of Detection and Quantification - Where Our Vision Ends
Every analytical method has its limitations. Below a certain concentration level, we are unable to detect or measure anything with appropriate certainty. This is where two key concepts come in.
Limit of Detection (LOD)
The limit of detection is the lowest concentration of a substance that we can still detect with a specified probability (e.g., 95%) - distinguish from background noise - but we cannot accurately measure.
It's like looking at stars at night - you can see a very faint point of light and say "yes, something is there," but you are unable to determine how brightly it shines or how large that star is.
Limit of Quantification (LOQ)
The limit of quantification is the lowest concentration of a substance that we can not only detect but also measure quantitatively with appropriate precision and accuracy.
Continuing the analogy - this is the brightness level at which you can not only confirm the presence of a star but also estimate its brightness and size with reasonable certainty.
Practical Scenario
A method for determining pesticides in drinking water has an LOD of 0.01 µg/l and an LOQ of 0.03 µg/l. The legal limit is 0.1 µg/l.
A result of 0.02 µg/l means detection of the pesticide, but it cannot be considered a reliable quantitative value. The report should contain the information "detected below the limit of quantification," not a specific number.
Linearity and Range - The Method's Comfort Zone
Method Linearity
Linearity is the method's ability to provide results directly proportional to the concentration of the analyte in the sample. In simple terms - when you double the concentration, the signal should also double.
Perfect linearity is the ideal we strive for. In practice, there are always some deviations from the ideal straight line. The key question is: are these deviations small enough that we can accept them?
Working Range of the Method
The working range of the method defines the interval of substance concentrations in which the method works with appropriate linearity, precision, and accuracy. This is the method's "comfort zone" - the area in which we can trust it.
Danger of Extrapolation
The most common error in laboratory practice is extrapolation beyond the established range of the method. The result may seem sensible, but its reliability is questionable. It is always better to dilute or concentrate the sample to fit within the working range of the method than to risk uncertain calculations.
Selectivity and Specificity - Are We Measuring What We Think?
These two terms are often used interchangeably, although technically they describe slightly different aspects of the same feature of the method.
Specificity
Specificity is the ideal situation in which only and exclusively our analyte generates a signal. No other substances present in the sample affect the result or give false results. This is the highest level of selectivity.
Selectivity
Selectivity is a more pragmatic approach - the method can determine the analyte or group of analytes of interest to us in the presence of other substances that may be present in the sample.
How to Test Selectivity?
The laboratory tests selectivity by analyzing samples containing the analyte (positive test) and samples without the analyte but with potential interfering substances (negative test). If the method is selective, the negative test should not give a false positive result.
Recovery - Are We Getting Everything We Added?
Recovery is one of the simplest and most intuitive indicators of method quality. It shows what percentage of a known amount of added analyte we are able to measure.
What Affects Recovery?
- Losses during sample preparation: Extraction, evaporation, filtration - each step can lead to losses
- Matrix effect: Sample components can affect the determination efficiency
- Analyte degradation: Some substances decompose during preparation or analysis
- Systematic errors: Incorrect calibration or interferences
Acceptable Recovery Values
For major components, we usually expect recovery in the range of 95-105%. For trace analytes, we accept a wider range - 80-120%. These limits take into account greater technical difficulties at very low concentrations.
Method Stability - Can I Trust It Tomorrow?
Stability (sometimes called flexibility or robustness) of a method determines how sensitive it is to small, deliberate changes in measurement conditions.
Why Test Stability?
Testing stability allows answering key questions:
- Which method parameters require strict control?
- How precisely must I control temperature, pH, reaction time?
- Can I introduce minor modifications without losing result quality?
- What will happen when something doesn't go as planned?
Example of Stability Testing
The method requires a pH of 7.0. The laboratory tests what happens at pH a few tenths different. If the results remain acceptable, the method is stable with respect to small pH changes. If results significantly change already at a pH a few tenths different, pH requires very strict control and should be precisely described in the procedure.
Measurement Uncertainty
Measurement uncertainty is an estimate of the range of values within which, with a specified probability, the true value of the measured quantity is located.
Sources of Uncertainty
Measurement uncertainty comes from many sources:
- Method precision (scatter of repeated measurement results)
- Activities performed on the sample (sampling, weighing, etc.)
- Accuracy of standards and reference materials
- Equipment parameters (pipettes, balances, volumetric flasks)
- Environmental conditions (temperature, humidity)
- Personnel competencies
- Sample matrix effects
Why Is Uncertainty Important?
When a laboratory reports a result as "10 mg/l ± 2 mg/l," the client knows that the true value is most likely between 8 and 12. This allows them to make informed decisions, especially when the result is close to a legal limit or quality standard.
Confirmation of Result Validity - Ensuring Test Quality
It should be noted that introduced methods, validated and verified, must be covered by a result validity monitoring program. All component elements should be considered and fulfilled in accordance with section 7.7 of the PN-EN ISO/IEC 17025:2018-02 standard.
The laboratory should regularly assess the effectiveness of applied result validity monitoring methods and document the results of these assessments. If it is found that the applied procedures do not provide sufficient quality control, they should be appropriately modified or supplemented.
Summary - Proper Nomenclature
Terminology regarding method characteristics is not a collection of dry definitions to memorize for an exam. It is a common language of communication between laboratories, auditors, clients, and regulatory bodies. Precise use of these concepts is a sign of professionalism and a guarantee of laboratory work quality.
Understanding these terms allows:
- Properly design and implement analytical methods
- Communicate method limitations and capabilities
- Interpret results in the context of their uncertainty
- Meet standard and accreditation requirements
- Build client trust through transparency
Key Conclusion
In an analytical laboratory, it is not enough to measure well - you must also understand and communicate well what is being measured, how well it is done, and what the method's limitations are. This is what distinguishes a professional laboratory from an amateur approach to analysis.
The article was prepared based on the requirements of the PN-EN ISO/IEC 17025:2018-02 standard "General requirements for the competence of testing and calibration laboratories" and current recommendations presented during the XXVIII POLLAB 2023 Symposium regarding validation, verification, and ensuring result validity. The content has been verified for substantive and linguistic compliance to support laboratories in the practical application of metrology and measurement quality principles.