Maternal serum alpha-fetoprotein screening: further consideration of low-volume testing

Unrecognized assay drift that may occur during low-volume (fewer than 500 specimens per week) maternal serum alpha-fetoprotein testing could result in either underestimation or overestimation of the number of pregnant women who are at increased risk of fetal malformations and genetic anomalies. Quality control software programs that incorporate the use of a multirule Shewhart chart are designed to detect assay drift. Careful selection of quality control sera for inclusion in analytic assays and appropriate application of a multirule quality control procedure to values that are obtained on these control materials should detect assay drift, regardless of the volume of patients' specimens in the run.