Key Topics

When is method validation required?

Method validation usually begins during the method development process when some method performance parameters are evaluated approximately to determine whether the chosen test method is likely to be fit for purpose. There are a number of situations where a validation study is required. The extent of the work required will differ in each case. It is important to remember that all test methods require some form of validation before they are used by a laboratory for the analysis of samples for customers.

Validation of a new method

Where a new method is needed to meet a particular measurement requirement, the development and subsequent validation of the method may be coordinated by a national or international standards body (e.g. BSI or ISO), or by a sector organisation. In such cases the method may be developed by a single laboratory but the validation is usually carried out by means of a collaborative study (round robin) involving a number of laboratories. This approach generally leads to ‘standard’ published methods which are available to all laboratories to use (see ‘How are methods validated?’). However, a laboratory wishing to use such a method will need to demonstrate that it can operate the method satisfactorily and achieve the specified performance criteria. The process of demonstrating that an existing performance standard can be met is sometimes referred to as verification.

If an individual laboratory encounters a new measurement requirement, and a suitable test method is not available (either as a standard method or a method published in the literature), then it falls to the laboratory to develop and validate an appropriate method (see ‘How are methods validated?’).

Validation of minor changes to an existing method

On occasions, a laboratory may need to use a previously validated method outside of its stated scope. For example, the concentration of the analyte in the sample may by higher or lower than that which was studied in the original validation of the method, or the nature of the sample (its matrix) may have changed. In such cases, the laboratory will need to demonstrate that the method performance is acceptable for the new sample types. If changes to equipment, reagents or the working environment are made then the laboratory will again have to demonstrate fitness for purpose of the modified method.

Replacing one method by another

Sometimes - for technical, environmental or operational reasons - it is necessary to replace a validated test method with one that is significantly different. The validation of the new method needs to demonstrate that the test results obtained are comparable with those produced by the existing method. Where the measurement results from the two methods are not directly equivalent, it may be possible to establish an empirical relationship between them by analysing a range of samples using both methods.