This Practice can be used in a single laboratory for trace analysis (that is, where: 1) there are concentrations near the lower limit of the method and 2) the measurements system‘s capability to discriminate analyte presence from analyte absence is of interest). In these testing situations, a reliable estimate of the minimum level at which there is confidence that detection of the analyte by the method represents true presence of the analyte in the sample is key. Where within-laboratory detection is important to data use, the WDE procedure should be used to establish the within-laboratory detection capability for each unique application of a method.

When properly applied, the WDE procedure ensures that the 99 %/95 % WDE has the following properties:

Routinely Achievable Detection The laboratory is able to attain detection performance routinely, using studied measurement systems, without extraordinary effort, and therefore at reasonable cost. This property is needed for a detection limit to be practically useful while scientifically sound. Representative equipment and analysts must be included in the study that generates the data to calculate the WDE.

Inclusion of Routine Sources of ErrorIf appropriate data are used in calculation, the WDE Practice will realistically account for sources of variation and bias common to the measurement process and routine for sample analysis. These sources include, but are not limited to: 1) intrinsic instrument noise, 2) some typical amount of carryover error, and 3) differences in analysts, sample preparation, and instruments (including signal-processing methods and software versions).

Exclusion of Avoidable Sources of ErrorThe WDE Practice excludes avoidable sources of bias and variation, (that is, those which can reasonably be avoided in routine field measurements). Avoidable sources would include, but are not limited to: 1) inappropriate modifications to the method, the sample, measurement procedure, or measurement equipment, and 2) gross and easily discernible transcription errors (provided there was a way to detect and either correct or eliminate such errors in routine sample testing).

Low Probability of False DetectionConsistent with a measured concentration threshold (YC), the WCL is a true concentration that will provide a high probability (estimated at 99 %) of true non-detection [and thus a low estimated probability of false detection (α) equal to 1 %]. Thus, when a sample with a real concentration of zero is measured, the probability of not detecting the analyte (that is, the probability that the measured value of the blank will be less than the WCL) would be greater than 99 %. To be most useful, this property must be demonstrated for the particular matrix being used, and not just for reagent-grade water.

Low Probability of False Non-detectionWhere appropriate data have been used for calculations, the WDE provides a true concentration at which there is a high estimated probability (at least 95 %) of true detection [and thus a low estimated probability of false non-detection (β) equal to 5 % at the WDE], with a simultaneously low estimated probability of false detection. Thus, when a sample with a true concentration at the WDE is measured, the probability of detection would be estimated to be at least 95 %. To be useful, this property must be demonstrated for the particular matrix being used, and not just for reagent-grade water.

Note 1—The referenced probabilities, α and β, are key parameters for risk-based assessment of a detection limit.

When this Practice is utilized by a laboratory to develop these false-positive- and false-negative-control point estimates, from data representative of routine operations, the laboratory may confidently claim these levels of false-positive and falsenegative control in the future, so long as the data used remain representative of that future operation. The laboratory may also qualify reported data using the appropriate point estimates (for example YC, YD, WCL, WDE) or censor data below the WCL as a valid basis for these data-reporting practices.

The WDE Standard does not provide the basis for any prospective use of the test method by other laboratories for reliable detection of low-level concentrations, even for the same analyte and same media (matrix).

The WDE values from a given laboratory may be used to compare the detection power of different methods for analysis of the same analyte in the same matrix by that laboratory.

The WDE Practice applies to measurement methods for which calibration error (that is, the error in the calibration of the measurement system) is minor relative to the combined other sources of variability. Some examples of other sources and when they may be dominant are:

Sample Preparation (dominant especially when calibration standards do not go through sample-preparation steps).

Differences in analysts where a laboratory has more than one person who may perform each method step.

Instrument differences (measurement equipment), which could take the form of differences in manufacturer, model, hardware, electronics, separation columns, sampling rate, chemical-processing rate, integration time, software algorithms, internal-signal processing and thresholds, effective sample volume, and contamination level.

Reducing calibration error by use of allowable, though more stringent, calibration procedures (for example, multiple concentrations, replication, tight calibration-acceptance criteria, etc.) and through calibration verification (for example, analysis of a traceable standard from a second, independent source, calibration diagnostics) can reduce the magnitude of the calibration error.

Alternative Data-Quality ObjectivesOther values for α, β, confidence, etc. may be chosen as parameters; however, this procedure addresses only those stated here in.

Collectively, the many sources of variation combine to cause within-laboratory measurements at any true concentration to be Normally distributed. The assumption of Normality is important for some of the statistics used; data Normality should be assessed if there is reason to believe this assumption is not valid.

If control of false negatives is not a data-quality objective, the WCL determined through this procedure provides a sound criterion for future determination of false-positive control; in such cases, the laboratory may confidently claim that true values above the WCL have a statistically significant difference from like-matrix zero-concentration samples (for example, from the method blank), but nothing more.

Where as-measured values (for example, not corrected for bias), not true values are of interest, YC and YD may be used as these asmeasured levels of the WCL and WDE.

1.1 This Practice provides a procedure for computinga 99 %/95 % Within-laboratory Detection Estimate (WDE) and theassociated critical level/value (WCL). The WDE is the minimumconcentration, with false positives and false negativeappropriately controlled, such that values above these minimums arereliable detections. The WCL is the point at which only falsepositives are controlled appropriately. A false positive is thereporting of an analyte as present when the analyte is not actuallypresent; false negatives are reports of analyte absence when theanalyte is actually present. This Practice is distinguished fromthe Interlaboratory Detection Estimate (IDE) Practice in that theIDE Standard utilizes data from multiple, independent laboratories,while this Practice is for use by a single laboratory. The IDEwould be utilized where interlaboratory issues are of concern (forexample, limits for published methods); this Practice (and valuesderived from it) are applicable where the results from a singlelaboratory, single operator, single instrument, etc. are involved(for example, in understanding, censoring and reporting data).

1.2 The establishment of a WDE involves determiningthe concentration below which the precision and bias of ananalytical procedure indicates insufficient confidence infalse-positive and false-negative control to assert detection ofthe analyte in the future analysis of an unknown number of samples.Most traditional approaches attempt to determine this detectionlimit‘ by estimating precision at onlya single, arbitrary point. The WDE approach is intended to be amore technically rigorous replacement for other approaches forestimating detection limits. The WDE Practice addresses a number ofcritical issues that are ignored in other approaches.

1.2.1 First, rather than making a single-pointestimate of precision, the WDE protocol requires an estimate ofprecision at multiple points in the analytical range, especially inthe range of the expected detection limit. These estimates are thenused to create an appropriate model of the method‘s precision. This approach is a more credible wayto determine the point where relative precision has become toolarge for reliable detection. This process requires more data thanhas been historically required by single-point approaches or byprocesses for modeling the relationship between standard deviationand concentration.

1.2.2 Second, unlike most other approaches, the WDEprocess accounts for analytical bias at the concentrations ofinterest. The relationship of true concentration to measuredconcentration (that is, the recovery curve) is established andutilized in converting from as-measured to true concentration.

1.2.3 Third, most traditional approaches todetection limits only address the issue of false positives.Although false negatives may not be of concern in some data uses,there are many uses where understanding and/or control of falsenegatives is important. Without the false-negative-controlinformation, data reported with just a critical-level value areincompletely described and the qualities of data at these levelsincompletely disclosed.

1.2.4 Fourth and last, the WDE Standard utilizes astatistical-tolerance interval in calculations, such that futuremeasurements may reasonably be expected to be encompassed by theWDE 90% of the time. Many older approaches have used thestatistical confidence interval, which is not intended to encompassindividual future measurements, and has been misunderstood andmisapplied. Procedures using the confidence interval cannot providethe stated control when the detection-limit value is applied tofuture sample results; such application is the primary use of thesevalues.

1.3 To summarize, the WDE is computed to be thelowest true concentration at which there is 90% confidence that asingle (future) measurement (from the studied laboratory) will havea true detection probability of at least 95% and a truenondetection probability of at least 99 % (when measuring a blanksample). For the laboratory in the study, the critical value is thetrue concentration at which, on average, (with approximately 90%confidence) will not be exceeded by 99% of all measurements ofsamples with true concentration of zero (that is, blanks). Thesevalues are established by modeling the precision and establishingthe recovery/bias over a range of concentrations, as well as byusing a tolerance interval. The complexities of the WDE proceduremay appear daunting, but the additional considerations arenecessary if meaningfully estimates of the actual detectioncapabilities of analytical methods are to be made. The concepts aretractable by degreed chemists, and the use of the available ASTMDQCALC Excel-based software makes the data analysis and limitdeterminations easy.

1.4 A within-laboratory detection estimate is usefulin characterizing the concentration below which a method, for ananalyte, as implemented in a specific laboratory, does not (withhigh confidence) discriminate the presence of the analyte from thatof the absence of an analyte. As such an estimator, the WDEStandard (and the WDE and WCL values produced through itsapplication) are useful where a trace-analysis testing method needsto be used.

1.5 The values stated ininch-pound units are to be regarded as standard. The values givenin parentheses are mathematical conversions to SI units that areprovided for information only and are not consideredstandard.