Harmonisation of performance assessment in qualitative PT/EQA

Evaluation of participant performance is a prerequisite in EQA scheme design. At its most basic the evaluation is used to determine whether the result reported is correct or not. This of course requires a means of determining the correct result. For qualitative testing where the result is presence or absence, assignment of the correct result is simple. Where the result requires accurate identification/interpretation, there may be a sliding scale for grading how close the reported result is to the assigned result. For quantitative PT/EQA schemes variability of the reported result from the correct result (assigned value) can be assessed in terms of precision, accuracy and bias using statistical tools to determine acceptability. The statistical tools commonly used are described in ISO 13528. Performance assessment is often based on the difference between the reported result and the assigned value and a performance score assigned. For qualitative PT/EQA schemes there is no commonly accepted statistical evaluation. Some providers apply scoring schemes to transform qualitative results into quantifiable data that can then be used to apply a performance score. However, scoring schemes vary and are not universally applied, challenges are not necessarily equivalent and many providers only assess acceptability of performance on an annual basis following completion of several PT/EQA rounds. Use of a harmonised approach across all PT/EQA schemes is seen by some as the Holy Grail with the performance understood by participants, their customers and other stakeholders. The issues associated with harmonisation of performance assessment will be discussed.