TNI Logo The NELAC Institute

Standard Interpretation



Standard: 2003 NELAC
Section: Appendix D.1.1.2.1
Link to relevant standard

REQUEST:

Can control limits include 0? Marginal exceedances must be random. If control limits don't include 0, poor performing analytes will consistently be marginal exceedances.

Same question for demonstration of capability.

TNI FINAL RESPONSE:

(Quality Systems Expert Committee/NELAP Board, 10-25-08)

The laboratory would have to demonstrate how its data can meet all other aspects of the method and still generate control limits that include a 0% recovery. However, there is no restriction on control limits using 0 - look at PT criteria for soil to see examples of this occurring.

As for marginal exceedances, there's no such thing as a "consistent marginal exceedance", since they are required to be random. Consistent means that there is something wrong in the process or the method just doesn’t work for that analyte.