Comparison of the Results of Proficiency Testing in Seven Medical Laboratories - a Benchmark Project to Define a Uniform Key Performance Indicator

Background: In medical laboratories, it is mandatory to ensure the analytical quality of the measurement procedures by proficiency testing (PT). The aim of this study was to evaluate whether the PT results of seven medical laboratories in South Tyrol - as a measure of the analytical performance - were different. Methods: As a measure for the analytical performance of the individual laboratories, we used the PT results (RIQAS, Randox international quality assessment scheme) of one year for 34 analytes. We calculated annual 'total scores' of each analyte for all participating laboratories and compared them statistically. Results: In 2018, there was a highly significant difference between the seven laboratories in the 'total scores' for the 34 analytes (p < 0.001). The laboratories had a 'cumulative, annual total score' of 75 - 91% of the maximum achievable values. Essentially, two groups could be distinguished. Laboratories 1 - 3 achieved better results (90 - 91%) than laboratories 4 - 7 (77 - 82%). In particular, the non-participation of the laboratories 4 - 7 in several PT cycles in 2018 and the registration in the wrong homogeneous group for some analytes in the laboratories 4 and 5 seem to be responsible for the worse results. Conclusions: The analytical performance as assessed by the PT results was different across the seven participating laboratories of the South Tyrolean Medical Service. Based on our study results, we defined a uniform key performance indicator for the seven laboratories with a limit value for the 'cumulative, annual total score' of > 80%.

Automated summary

The study found significant differences in proficiency testing results among the seven medical laboratories in South Tyrol, with laboratories 1-3 outperforming laboratories 4-7. The lack of participation in several PT cycles by laboratories 4-7 in 2018 and registration in the wrong homogeneous group for some analytes by laboratories 4 and 5 seem to be responsible for the poorer results.