Measurement bias

Statistical Control. Statistical quahty control (SQC) is the apphcation of statistical techniques to analytical data. Statistical process control (SPC) is the real-time apphcation of statistics to process or equipment performance. Apphed to QC lab instmmentation or methods, SPC can demonstrate the stabihty and precision of the measurement technique. The SQC of lot data can be used to show the stabihty of the production process. Without such evidence of statistical control, the quahty of the lab data is unknown and can result in production challenging adverse test results. Also, without control, measurement bias cannot be determined and the results derived from different labs cannot be compared (27). 

If the random errors are higher than can be tolerated to meet the goals of the test, the errors can be compensated for with rephcate measurements and a commensurate increase in the laboratory resources. Measurement bias can be identified through submission and analysis of known samples. Establishing and justifying the precision and accuracy reqrtired by the laboratory is a necessary part of estabhshing confidence. 

The result from a measurement on a RM is commonly a difference between the observed value and the certified value. This difference is called measurement bias, and can, appreciating both the uncertainty on the RM as well as the imcertainty added during the measurement, be tested for (statistical) significance. ISO Guide 33... 

Measurement bias is determined by comparing the mean of measurement results obtained for a reference material, using the method being validated, with the assigned property values for that reference material. The number of replicate analyses required (n) depends on the precision of the method (.s ) and the level of bias (8) that needs to be detected [12]. A useful approximation is shown in the following equation ... 

Measurement bias can also be determined by comparing the results obtained using the method of interest with the results obtained using a reference method of known bias. This approach is very similar to comparison with a reference material, with the sample subject to measurement acting as a transient reference material. 

The measurement bias, B, can be calculated as the ratio (often expressed as a percentage) of the difference between the mean of a number of determinations of a test sample, obtained under repeatability conditions, and the true or accepted concentration for that test sample, as shown in the following equation ... 

Measurement bias is often called recovery , which can be expressed as a ratio (R), or a percentage (%R) and can be greater or less than 100%. In some fields of measurement, recovery refers to the amount of added (spiked) analyte recovered during analysis (equation (4.11)). In other fields, recovery is taken as an estimate of the proportion of the total analyte (native plus any added spike) present in a sample that is measured (recovered) by the method. The relationship between recovery and bias is shown in equation (4.12) ... 

Association of Official Analytical Chemists acceptance quality limit American Society for Testing and Materials measurement bias... 

If one or more leaks are considered, the constraint model for the process must be modified to take them into account. Now, the least squares formulation of the problem, when measurement bias are absent, can be stated as... 

In this chapter, the data reconciliation problem for dynamic/quasi-steady-state evolving processes is considered. The problem of measurement bias is extended to consider dynamic situations. Finally in this chapter, an alternative approach for nonlinear dynamic data reconciliation using nonlinear programming techniques will be discussed. 

Note The previous formulations for both normal and abnormal situations are very general and include inputs to the process as well as different types of perturbations (jumps) in normal process behavior. Later on in this chapter we will consider a reduced version of this formulation, since we will be mainly interested in the measurement bias detection and identification problem. 4k... 

Possible measurement bias factors such as droplet deposition in the probe, droplet breakup and coalescence were studied. A simple criterion for minimizing measurement bias was proposed. The system can be used for both water and liquid-metal droplets. 

Data analysis and interpretation (i.e. extrapolation to the target patient population) Different formulas are used to correct duration of the QT interval for heart rate some formulas may overcorrect at high heart rates and undercorrect at low heart rates (e.g. Bazett s formula) consider that with some formulas (e.g. Bazetf s) a QTc increase of 4-5 ms may result from measurement bias Need for an individualized correction formula... 

General information about analytical techniques can be found in another chapter of this volume (Albarede and Beard 2004). Details specibc to Se and/or Cr are covered here. Most of the existing Se and Cr isotope measurements were made by TIMS, but MC-ICP-MS methods are supplanting TIMS methods for Se (Rouxel et al. 2002) and will probably do so for Cr within a few years. With both TIMS and MC-ICP-MS analyses, there is a measurement bias, or discrimination, that must be corrected for (Albarede and Beard 2004). For example, Se is particularly evaporation prone because of its volatility lighter isotopes of Se evaporate from the hlament faster than heavier isotopes, resulting in large changes in measured ratios over time (Johnson et al. 1999). 

We find no measurable bias (under the given experimental conditions)... 

The measurement bias is an estimate of a systematic measurement error. 

The parallel-group, double-blind, placebo-controlled study design represents the golden standard of acute treatment trials of depression, mania and anxiety disorders. This design is intended to limit bias, in particular selection and measurement bias. Trials based on this design are expected to provide information about the effect size of a new compound and its side-effect profile. 

The evaluation of reproducibility results often focuses more on measuring bias in results than on determining differences in precision alone. Statistical equivalence is often used as a measure of acceptable interlaboratory results. An example of reproducibility criteria for an assay method could be that the assay results obtained in multiple laboratories will be statistically equivalent or the mean results will be within 2% of the value obtained by the primary testing laboratory. 

As a consequence, satisfactory performance in IMEP-20 would then mean having a result reported with zeta < 2 and micI < wlab < 0.1-2fref. Laboratories reporting larger uncertainties may not have their experimental procedure under control or may have overestimated some uncertainty components. Laboratories reporting smaller uncertainties are very likely to have either underestimated some of the uncertainty components or not accounted for some uncertainty sources. It has to be emphasized that laboratories with zeta > 3 Prst need to think about the origin of their measurement bias and only subsequently, after corrective measures have been taken, to focus thoroughly on their uncertainty estimation. 

However, many elements of v can be expected to have a worst case at a bound, e.g., measurement bias, flow, concentration, buffering, and reaction rates. For the general case, the vertex assumption cannot be relied upon but provides a basis for useful heuristics in the search for a solution. 

The nominal case was taken as a base flow of 400 m /h with no disturbances and no measurement bias. As the disturbances were infrequent, the economic objective for a given control structure was taken to be minimize the excess reagent used at steady state compared to the reagent required with ideal delay-... 

The steps described in this section may be taken to reduce the role of systematic bias errors (see Section 21.3) in impedance measurements. Bias errors associated with nonstationary effects have greatest impact at low frequencies where each measurement requires a significant amount of time. 

---