Six Sigma revisited: We need evidence to include a 1.5 SD shift in the extraanalytical phase of the total testing process

Abstract

The Six Sigma methodology has been widely implemented in industry, healthcare, and laboratory medicine since the mid-1980s. The performance of a process is evaluated by the sigma metric (SM), and 6 sigma represents world class performance, which implies that only 3.4 or less defects (or errors) per million opportunities (DPMO) are expected to occur. However, statistically, 6 sigma corresponds to 0.002 DPMO rather than 3.4 DPMO. The reason for this difference is the introduction of a 1.5 standard deviation (SD) shift to account for the random variation of the process around its target. In contrast, a 1.5 SD shift should be taken into account for normally distributed data, such as the analytical phase of the total testing processin practice, this shift has been included in all type of calculations related to SM including non-normally distributed data. This causes great deviation of the SM from the actual level. To ensure that the SM value accurately reflects process performance, we concluded that a 1.5 SD shift should be used where it is necessary and formally appropriate. Additionally, 1.5 SD shift should not be considered as a constant parameter automatically included in all calculations related to SM.