Designing quality control for neonatal screening assays

The purpose of neonatal screening is to find those that have a high risk for a disorder and therefore need further action for diagnosis and treatment. The separation of the high-risk and low-risk groups is typically achieved by establishing cut-off values for interpretation of the test result. The guidelines should preferably include both a low cut-off and a high cut-off value, with a grey zone between them. The width of the grey zone can be used for defining the quality required by the assay. The present work is based on the use of the EZ Rules software (from Westgard QC, Madison, WI), which automatically selects control rules. The user enters the needed parameters, such as the precision (as % CV) and bias of the assay. By using the grey zone as the medical decision interval the program will calculate possible control rules for the assay. The program was used to calculate the control rule for the AutoDELFIA neoTSH assay (from PerkinElmer Wallac, Turku, Finland). The grey zone was taken as 10 - 20 mU/L TSH, which is the recommendation of the American Academy of Pediatrics (Pediat. 91, 1203 - 1209). The entered parameters were: a total imprecision of 9%, which is typically seen with the AutoDELFIA neoTSH kit, a bias of 0% and a preanalytical variation of 20%. With the number of controls chosen as two, as often is used, a 1 3.0 s rule can be applied. The program also gives alternative control rules. Many laboratories lack a documented definition of the required quality, and tend to use a 2 SD control rule, which however leads to many unnecessary rejections. The EZ Rules program provides a tool for selection of QC rules. With the quality of the AutoDELFIA neoTSH kit two controls and a 1 3.0 s rule is sufficient. Runs are rejected only if one control out of two exceeds the 3.0 SD limit.