A Survey of Sigma Metrics across Three Academic Medical Centers

Sigma metric calculations provide laboratories an objective means to assess analytical method performance. Methods with higher sigma values are desirable because they are more reliable and may use less frequent quality control in order to maintain optimal performance. Sigma metrics can also serve as a tool when comparing method performance across assay and manufacturer platforms.

Sigma values were calculated for 28 common chemistry and 24 immunoassay assays across 3 academic medical centers. Method imprecision and percent bias relative to peer group means was tabulated from Bio-Rad quality control (QC) data. Sigma values were calculated for each method using allowable total error (TEa) from either the CLIA evaluation limits or desirable biological variation. Average sigma values were generated for each site and graded as optimal: >6 sigma; good: 5-6 sigma; marginal: 3-5 sigma; or poor: <3 sigma. Analysis of NIST SRM1950 standards for a subset of analytes allowed an estimation of absolute bias.

Clinical chemistry assays displayed similar method performance across all 3 study sites. Immunoassays showed significant differences between manufacturers, and a majority of assays failed to meet an optimal level of performance. Different TEa values produced different sigma metrics with more stringent TEa limits based on biological variation, resulting in poorer performance estimates than the wider CLIA limits. Analysis of NIST standards revealed similar performance.

Sigma metrics are comparable for chemistry but not immunoassay platforms. The selection of total allowable error goals led to differences in sigma metrics.