Limits of detection, identification and determination: a statistical approach for practitioners

Abstract

For the estimation of the limits of detection, identification and determination, considerations from analytical practice were applied to the statistics of the calibration line and its prediction interval. The detection limit was the concentration calculated from the maximum height of the prediction interval at zero spiking concentration. The identification limit was twice the detection limit and was the lowest concentration that could safely be detected. The determination limit was the lowest concentration fulfilling three criteria: 1. None of the signals resulting from determination limit concentration should interfere with any signal from detection limit concentration, thus providing an unambiguous separation between the two limits. 2. Recovery should be between 70% and 120%. 3. Lowest and highest predictable signal at determination limit concentration should not deviate more than B 30% from the average. Practical analytical guidance and the necessary mathematical formulae are presented.

Quantification of 1,8‐cineole and of its metabolites in humans using stable isotope dilution assays

“…To still use an authentic matrix and not a simple surrogate, we chose authentic urine and plasma, but without enzymatic hydrolysis. LOD and LOQ were determined according to the method of Vogelgesang and H. adrich [26]. LOD is the addition value referring to the 95% confidence limit of the calibration line at the zero addition level.…”

“…Detection and quantification limits were determined according to Vogelgesang and H. adrich [26]. 1,8-Cineole was added to saturated NaCl solution in four different amounts ranging from 133 to 1330 ng/L, triplicate each, and analyzed as described above.…”

“…To determine the LOD and LOQ, we applied the calibration procedure proposed by Vogelgesang and H. adrich [26]. The respective matrix devoid of the analytes was spiked with increasing amounts of analytes and ISs prior to analyzes.…”

Content of the Alternaria mycotoxin tenuazonic acid in food commodities determined by a stable isotope dilution assay

“…After addition of [ 13 C 6 , 15 N]-TA in the same amount as the respective amount of analyte, all samples underwent sample preparation and cleanup as described above and were finally analyzed by LC-MS/MS. LODs and LOQs were derived statistically from the data according to the literature (Vogelgesang and Hädrich 1998).…”

The Hubaux and Vos Method

“…From Table E.1, it is evident that the Hubaux and Vos method is inferior in every row. On the plus side, their method has a solitary advantage: its prediction interval hyperbolas, which result from the OLS slope and intercept being negatively correlated, typically result in slightly lower detection limits, for example, 2% or so [8]. This improvement is insignificant compared to the typically broad widths of x D distributions, for example, Fig.…”