Quality Standards

A study of the new Cobas Pure detailed imprecision and bias. Can these instruments meet the new measurement uncertainty goals?

Over 45% of 17 assays in a multisite study of cobas pure assays cannot meet desirable uncertainty goals

Sten Westgard, MS
March 2025

An extensive study across Europe and Asia looked at one of the latest instruments to hit the market: cobas pure.

Performance evaluation of cobas pure integrated solutions at multiple sites in Europe and Asia. Baum H, Chun S, Findeisen P, Fleurkens H, Gu H, Jinyoung H, Prevot F, Radziszewski D, Rossier MF, Vogt N, Furrer J, Klopprogge K, Scheider-Thauren C. J Lab Med 2023; 47(3): 129-140. The study was conducted in Germany(2 sites), Poland, Switzerland, and South Korea.

"At all five sites, the Roche-manufacturerd and BioRad QC materials were measured using selected applications [tests] at two analyte concentration levels over 14-16 weeks. Pooled control materials (Roche-manufactured and BioRad) were used to minimize the potential influence of sample handling between sites and operators; two control samples were measured per application (low and high concentration)....Coefficients of variation (CVs) for repeatability (within-run precision), intermediate imprecision (within-laboratory precision), and reproducibility (across-laboratory precision) were calculated and compared...."

A summary of what can meet the uncertainty goals

The bright side is that a majority of these assays can hit desirable and minimum measurement uncertainty goals. The dark lining of that cloud is that the VISP laboratory, which calculated the longer term estimate of imprecision, reproducibility, did not. The more realistic operating performance, in other words, might not achieve these new goals.

Still another problem is that the u:ref and u:cal are not available to be included in a my calculation. The control CVs are taken as a surrogate, but the addition of the other uncertainties might push more assays into unacceptability. A final challenge is that all of these calculations don't take into account any bias. If bias is not zero, then even fewer of these assays will be able to hit the new mu goals.

The study did in fact determine bias, and we'll tackle that in a future article.