Statistical analysis of the stability of the standard curve for some Syva EMIT assays.

Derivation of standard curves for the EMIT therapeutic drug monitoring system involves several mathematical algorithms, all of which can be rewritten in the form of a linear equation y = mx + b. We examined the stability of the standard curve in terms of slope and y-intercept for three drug assays (procainamide, gentamicin, and carbamazepine) by generating calibration curves intermittently for periods as long as 90 days. Controls at three concentrations were assayed after each calibration to validate the standard curves. On the basis of 98% confidence intervals, the slopes of standard curves for procainamide, gentamicin, and carbamazepine were stable for 89, 80, and 57 days, respectively. Control values generated from standard-curve manipulations (adjustments to the y-intercept) indicated consistent accuracy and precision throughout the entire study, as compared with control values determined after each calibration. The increased utility of the standard curve and reagents suggests that full recalibration on a regular basis is not always necessary.