B-266 A multisite review across the USA of major competitive biochemistry platforms reveals significant differences in analytical performance

Abstract

Broad surveys of diagnostic biochemistry instruments are extremely rare. Proficiency testing surveys provide aggregated performance data, but group standard deviations are not fully reflective of individual instrument performance. The most helpful information for a clinical laboratory is to know how other individual laboratories experience diagnostic performance. Westgard QC collected routine long-term clinical chemistry instrument performance data across 4 major diagnostic manufacturers and assessed it using analytical Sigma metrics. Starting in mid-2023 and through 2025, we collected performance data from 89 individual instruments from 57 different laboratories across the USA. Multiple types of controls were used, from Bio-Rad to Technopath to Thermo-Fisher MAS, and this data was used to determine cumulative imprecision. Bias was determined through comparison of cumulative means to the cumulative peer group means from peer group comparison reports. Updated CLIA 2025 proficiency testing criteria were used for to establish performance requirements. Where CLIA does not directly regulate a particular clinical chemistry assay, a variety of extant performance requirements were used, including the European Federation of Laboratory Medicine (EFLM) biological variation database minimum analytical performance goals. The standard Westgard analytical Sigma-metric equation was used to assess the performance of the method on the Six Sigma scale. Instrument performance was displayed on Sigma metric Heat Maps. Medians were calculated as well as averaged. Westgard Sigma Rules were then used to determine the appropriate number of control rules and control measurements for the method. The largest difference between diagnostic competitors occurs at the low end of performance, where assays below 3 Sigma vary significantly by company. The poorest performing instrument vendor has more than twice the percentage of their biochemistry menu operating below 3 Sigma than the leading vendor. The next significant difference occurs at the 6 Sigma level, where the leading vendor has 150% more 6 Sigma performance than the bottom vendor. The percentage of the biochemistry menu operating at 4 and 5 Sigma appears to be roughly similar across all diagnostic manufacturers. The lower Sigma metric performance indicates that more Westgard Sigma Rules are necessary for QC on the bottom performers. A majority of clinical chemistry performance across these 4 major competitors are at 4, 5, and 6 Sigma, but clinical chemistry platforms are not identical or interchangeable. Analytical quality is not a commodity. Some platforms have more low analytical Sigma metric methods. Other platforms have more world-class methods. Regardless of the instrument chosen, laboratories can reduce their current QC practices and implement Westgard Sigma Rules, since the greatest number of methods for all platforms perform at 6 Sigma. Given that the typical laboratory employs the 1:2s control rule with 2 controls and 9% false rejection, Westgard Sigma Rules will recommend those methods switch to 1:3s with 2 controls, which generates less than 1% false rejection. For 6 Sigma methods, that means a reduction in outliers by more than 90%. This implies significant practical savings are possible for every laboratory, regardless of platform choice.