Measurement analytical variations

Available data should also be examined. For example, historic data are often available and can be used to evaluate process variation for trends and cycles using variographic techniques. We can examine control charts in the lab to measure analytical variation and compare it to process and sampling variation. We can compare our release data to acceptance data from customers, if available. 

A compound or element added to all calibration standards and samples in a constant known amount. Sometimes a major constituent of the samples to be analysed can be used for this purpose. Instead of preparing a conventional calibration curve of instrument response as a function of analyte mass, volume or concentration, a response ratio is computed for each calibration standard and sample, i.e. the instrument response for the analyte is divided by the corresponding response for the fixed amount of added internal standard. Ideally, the latter will be the same for each pair of measurements but variations in experimental conditions may alter the responses of both analyte and internal standard. However, their ratio should be unaffected and should therefore be a more reliable function of... 

In the passive mode, the optical device measures the variation in fluorescence characteristics (intensity, lifetime, polarization) of an intrinsically fluorescent analyte. The optical device can have different optical configurations involving in most cases an optical fiber (passive optode) (Figure 10.44). 

This diagram shows a histogram of measurements with variations between 45.5 and 51 mg/g of a specific analyte. Most of the measurements are between 48 and 50 mg/g. It can be seen, that it is necessary to have a large number of measurements to be able to draw a histogram. 

Despite the fact that direct absorbance/transmittance measurements are well established in analytical chemistry owing to the simplicity of the instrumentation and their broad applicability and versatility towards a large number of analytes, most of the reported miniaturized optical devices are based on the measurement of variations of the real part of the refractive index, such as SPR sensors [84,109-111] or interferometric sensors [94,112]. 

A second absorbance reading measures the variation in absorbance due to the reaction itself. This basic scheme may vary and thus with more sophisticated machines readings are taken more frequently, for example every 12s. Thus the reaction curve may be established and the initial reaction rate measured which, under precise analytical conditions, may be proportional to the concentration of the compound being measured. 

Variability in Absorption Estimates In this study, the occurrence of a negative absorption value for one subject and the absence of a significant vitamin C effect raise some questions about the accuracy of the method However, the expected changes in absorption due to dietary treatments may be masked by the analytical variations associated with absorption measurements and biological variabilities of iron absorption Analytical variations can be introduced at several stages of the analytical procedures incomplete fecal collection, inhomogeneous samples, iron contamination, incomplete colorimetric reaction, non-quantitative recovery after chemical ashing, and variations in isotopic measurements due to ion statistics, memory effects, instrument drift, etc Some of these are not as serious as others, for example, contamination with natural iron woiold not affect the estimate of tracer concentrations provided it occurs before the total iron content is measured ... 

The control chart is the basic analytical tool of SPC and is used for first assessing a process, then for monitoring a process output with respect to on-target control and process variability. A control chart is basically a time plot of a statistic calculated from a variable associated with a process. This variable may either be a process variable, such as temperature or flow rate, or a product variable, such as fill weight or potency. Examples of statistics are an individual measurement, an average of two or more measurements, a percentage of defective output items, a count of defect occurrences in time or space, or a measure of variation such as a range or standard deviation of two or more measurements. 

In laboratory studies of analytical variation, it is estimates of imprecision that are obtained. The more observations, the more certain are the estimates. Commonly the number 20 is given as a reasonable number of observations (e.g., suggested in the NCCLS guideline on the topic) To estimate both the within-run imprecision and the total imprecision, a common approach is to measure duplicate control samples in a series of runs. For example, one may measure a control in duplicate for more than 20 runs, in which case 20 observations are present with respect to both components. One may here notice that the dispersion of the means (x, ) of the duplicates is given as ... 

With regard to the sample(s) with low analyte concentration, one may preferably spike a set of serum samples from various patients with the analyte (e.g., a drug), rather than just one serum sample or a serum pool. For endogenous compounds, ideally a set of patient samples with concentrations in the low range might be used. A pooled SDs estimate can then be derived from repeated measurements of the set of samples (e.g., 10 measurements of each of 10 samples [see the example presented later in this chapter]). Measurements on different days should be carried out, so that SDs reflects the total analytical variation. 

Figure I 9-5 Six Sigma methodologies for measuring process performance. Method of measuring process variation is applicable to analytical testing processes.

The application of sigma metrics for assessing analytical performance depends on measuring process variation and determining "process capability in sigma units. This approach malces use of the information on precision and accuracy that laboratories acquire initially during method vahdation studies and have available on a continuing basis from internal and external quahty control. An important... 

The normal physiological component of variation is calculated from the total variation of measurements in serial specimens from the same patients, after adjusting for analytical variation. Such estimates differ somewhat from study to study, but after an extensive review of the literature, the NCEP panels concerned with lipid and lipoprotein measurement assumed average physiological... 

Serial samples. Using the mean of several serial measurements for clinical decisions averages out the effects of physiological and analytical variation. Measurements should therefore be made in at least two serial samples collected at least 1 week apart with the values averaged. Three serial samples are preferred for triglycerides, HDL cholesterol, and LDL cholesterol measurements, but two serial specimens can be used if necessary. 

Control charts used for monitoring the reproducibility of methods (repeated analyses of one or several reference materials) may be considered as long-term references for analytical measurements since they allow the monitoring of analytical variation with respect to an anchorage point, i.e. the reference material(s). This concerns reproducibility checking but not necessarily trueness whose evaluation relies on relevant CRM analysis. 

Biochemical measurements vary for two reasons. There is analytical variation, and also biological variation. 

Physiological variation is established as follows. The mean (and SD) for measurements in serial samples from the same individual is calculated. The variation can also be expressed in terms of a coefficient of total variation, CVt. CVt includes the contributions of both physiological and analytical variations. Physiological variation (CVp) is estimated by adjusting the coefficient of total variation for the contribution of analytical variation. If analytical variation is small compared to physiological variation, CVt will... 

---