Quality Requirements and Standards

A ground-breaking new study of EQA performance demonstrates the vast distance between real laboratory performance and EFLM optimal, desirable, and minimum goals. A quick review of 25 common chemistry tests shows a gap unlikely to be bridged by today's instrument performance.

 Reality Check: EQA peer Goals reveal a great distance between real world performance and EFLM biological goals

Sten Westgard, MS
September 2026

EQA programs from France and Germany combined their data in a recent study to derive reality-based performance specifications:

Analytical Performance Specifications Derived from External Quality Assessment Data: Comparison with Biological Variation-Based Models for Intermediate Precision Assessment. Claudio Ilardo, Emmanuel Reynaud, Nathalie Benaily. Ann Lab Med https://doi.org/10.3343/alm.2026.0092

The authors state: "We propose an alternative ramework for deriving APSs based on pooled CVs, addressing several limitations of existing approaches. BV-derived specifications may not accurarly reflect routine laboratory performance, whereas EQA-based approaches....To our knowledge, this is the first study to apply pooled CVs to establish APSs across more than 280 measurements spanning three major medical biology disciplines."

One unique aspect of these goals is that they are  generated for low, middle, and high levels of the working range for each test. This is a level of granularity that even the split low-end-unit/upper-end-percentage goals from CLIA and RCPA don't provide.

A brief comparison of the EFLM biological TEa goals vs. these  new Peer Group EQA goals (which for convenience we will shorthand as PGEQA goals) shows the inability of today's methods to achieve most of EFLM goals:

No analytes can hit the optimal BV TEa goals. 3 of the 25 analytes can meet the BV desirable goals, and one analyte can partially meet the BV desirable goals. 9 of 25 methods can meet the BV minimum goals, and 4 methods can partially meet the BV minimum goals. Finally, 8 of 25 methods cannot hit any of the BV goals.

In other words, the reality of laboratory performance is that BV goals, even the minimum goals, are out of reach. State of the Art goals are necessary to set realistic, practical, achievable goals for performance. 

We thank Ilardo, Reynaud, and Benaily for completing this ground-breaking study, and also for allowing us to reproduce part of these study results on our website.