Test Method Validation
Test method validation is the documented process of ensuring a test method is suitable for its intended use. It involves establishing the performance characteristics and limitations of a method and the identification of influences which may change those characteristics. The experimental results are subjected to statistical analysis and a series of pre-defined acceptance criteria are applied to the results. Establishing that a test method consistently produces reliable results is a critical element of assuring product quality and safety.
TMV is an important element of quality control. Without validation, there can be no assurance that the test results will be reliable and fit for the purpose. In some fields, validation of methods is a regulatory requirement. Generally, any method used to produce data in support of regulatory (e.g., FDA) filings or the manufacture of devices for human use should be validated.
Test methods can be destructive or non-destructive, based on the disposition of the test samples. The methods can measure attribute data (qualitative) or variable data (quantitative). All of them are candidates for validation, though the process for each can vary. Most test methods exist as validated standards, methods developed by technical standard organizations (ANSI, ASTM, ISO…) to establish uniform methods and procedures for testing. But standard methods do not always fit the requirements of the tests to be performed.
These test methods may require validation:
- standard (verifying lab version of standard is acceptable) - the lab specific version of a method must meet or exceed results of the standard
- standard methods outside normal scope - when using an established standard for testing outside the original scope of the standard
- amplified/modified standard methods - validation of a version of a standard, with modifications • laboratory designed/developed - validation of a method developed specifically for the test lab, to cover a requirement not sufficiently met by a standard
TMV is a risk based activity. The extent of the activity is often dictated by the potential level of patient harm weighed against the business risk of not performing the activities. The device risk index or harm classification dictates the minimum level of statistical confidence required. Higher risk requires more rigorous testing. In most cases, TMV is not mandated in the medical device industry (ISO 11607 testing is one exception). But demonstrating the safety and effectiveness of a device is difficult to do if the methods for establishing these parameters are not shown to be appropriate and reliable.
Conditions when TMV may be required:
- New method is developed
- Revision of established methods
- when established methods are used in (or transferred to) different test facilities
- For comparison of methods
Validation characteristics that should be considered include but are not limited to:
- Accuracy: a generic concept of exactness related to the closeness of agreement between the average of one or more test results and an accepted reference value.
- Precision: the closeness of agreement among test results obtained under prescribed conditions.
- Repeatability: closeness of the agreement between the results of successive measurements of the same measurement and carried out under the same conditions of measurement.
- Reproducibility: closeness of the agreement between the results of measurements of the same measurement and carried out under changed conditions of measurement.
- Linearity: Indicates that gage response increases in equal increments to equal increments of stimulus
- Range: The values within which a measuring instrument is capable of measuring or which a generating instrument can generate.