Acceptance limit, statistical validation

III the equivalence approach, which typically compares a statistical parameters confidence interval versus pre-defined acceptance limits (Schuirmann, 1987 Hartmann et al., 1995 Kringle et al., 2001 Hartmann et al., 1994). This approach assesses whether the true value of the parameter(s) are included in their respective acceptance limits, at each concentration level of the validation standards. The 90% 2-sided Cl of the relative bias is determined at each concentration level and compared to the 2% acceptance limits. For precision measurements, if the upper limit of the 95% Cl of the RSDn> is <3% then the method is acceptable (Bouabidi et al., 2010) or,... 

Compared with conventional impurities measurements, trace analyses caimot be expected to achieve the same linearity and precision values. This is due to the lower signal-to-noise ratios inevitable at low levels. Hence, while the same approaches can be used, greater latitude will be necessary in the acceptance criteria. What must be demonstrated is that the data is statistically valid to show that the levels of toxic analytes are below their specification limits. 

During method validation the parameters, acceptance limits, and frequency of ongoing system suitability tests or quality control checks should be defined. Criteria should be defined to indicate when the method and system are out of statistical control. The goal is to optimize these experiments in such a way that with a minimum number of control analyses the method and the complete analytical system will provide long-term results that will meet the objectives defined in the scope of the method. 

Principles and Characteristics Whereas parameters most relevant to method development are considered to be accuracy, system precision, linearity, range, LOD, LOQ, sensitivity and robustness, method validation parameters are mainly bias, specificity, recovery (and stability of the analyte), repeatability, intermediate precision, reproducibility and ruggedness. However, method development and validation are highly related. Also, validation characteristics are not independent they influence each other. Acceptance criteria for validation parameters should be based on the specification limits of the test procedure. Quantitation and detection limits need a statement of the precision at their concentration levels. Procedures used for validation of qualitative methods are generally less involved than those for quantitative analytical methods. According to Riley [82], who has discussed the various parameters for validation of quantitative analytical methods, the primary statistical parameters that validate an analytical method are accuracy and precision. 

NIRS equations were selected, taking into account the best values of r, SEP and other statistics ratio of performance to deviation and range error ratio. Table 1 summarizes statistics for calibration and validation of the degradability parameters. Prediction of feeds chemical composition was satisfactory. The effective degradation of DM, CP and NDF was also well predicted by the MRS equations, as well as fraction a in the case of DM and CP, and fraction b for DM. In the case of c value and the asymptote, R and r coefficients were below acceptable limits. These results are in accordance with Andres et al. (2005) and Ohlsson et al. (2007), even though current values are higher probably due to the higher variation in the present material. 

The most important aspects of data handling for potency assays and low-precision assays are that the data is handled by validated computer programs and that the acceptance and rejection criteria incorporated are clear and based upon statistical or proven (at validation) limits. 

Retrospective validation involves using the accumulated in-process production and final product testing and control (numerical) data to establish that the product and its manufacturing process are in a state of control. Valid in-process results should be consistent with the drug products final specifications and should be derived from previous acceptable process average and process variability estimates, where possible, and determined by the application of suitable statistical procedures, that is, quality control charting, where appropriate. The retrospective validation option is selected when manufacturing processes for established products are considered to be stable and when, on the basis of economic considerations and resource limitations, prospective qualification and validation experimentation cannot be justified. 

Every model has limitations. Even the most robust and best-validated regression model will not predict the outcome for all catalysts. Therefore, you must define the application domain of the model. Usually, interpolation within the model space will yield acceptable results. Extrapolation is more dangerous, and should be done only in cases where the new catalysts or reaction conditions are sufficiently close to the model. There are several statistical parameters for measuring this closeness, such as the distance to the nearest neighbor within the model space (see the discussion on catalyst diversity in Section 6.3.5). Another approach uses the effective prediction domain (EPD), which defines the prediction boundaries of regression models with correlated variables [105]. 

It is necessary to study stability in solution in the solvent used to prepare sample solutions for injection in order to establish that the sample solution composition, especially the analyte concentration, does not change in the time elapsed between the preparation of the solution and its analysis by HPLC. This is a problem for only a few types of compound (e.g. penicillins in aqueous solution) when the sample solution is analysed immediately after the preparation of the sample solution to be injected. The determination of stability in solution is more of an issue when sample solutions are prepared and then analysed during the course of a long autosampler run. While the acceptance criteria for stabUity in solution may be expressed in rather bland terms by making a statement such as, e.g. the analyte was sufficiently stable in solution in the solvent used for preparing sample solutions for reliable analysis to be carried out , in practice it has to be shown that within the limits of experimental error, the result of the sample solution analysis by the HPLC method is the same for injections at the time for which stability is being validated as for injections immediately subsequent to the sample solution preparation. While this may be done by a subjective assessment of results with confidence limits, strictly speaking a statistical method known as the Student s t-test should be used. 

The statistical model for precision is shown in equation (1). For precision, Y is defined as observed mass. For each of the eight expected masses the precision components are reported as variances for each component. The total variance is the sum of the component variances. The component percent of total is calculated on the variance scale as (Variance component estimate/Total variance) x 100%. The square root of each variance estimate is taken and reported as the standard deviation. Confidence intervals were generated by SAS and reported on the variance scale square roots of the confidence limits were calculated and reported as the confidence interval for the component standard deviation. The coefficient of variation (%CV) is calculated as (Observed component std dev/Observed average) x 100%. Table 10 shows the precision results for the example validation. The acceptance criteria for... 

Another important aspect in validation of a new dissolution method is to investigate how sensitive the dissolution results of the product, for which the method has been developed, are for minute variations in operating conditions. Examples of factors to consider in such a test are temperature of test medium, rotational speed, volume, sampling procedure, medium compositions and testing performed by different operators. Based on such robustness tests of the method, limits can be defined for acceptable variations of test conditions. Statistical design may be useful to apply in situations such as those demonstrated earlier in this chapter. 

By using the 96-well microELIS A format and instituting the use of rephcates of the samples and controls, they enabled statistical evaluation of the validity of the tests. An accepted method of determining the LOD of EIAs is to determine the mean OD value of the replicates of the appropriate negative controls and adding 3 standard deviations of those replicates. The detection limit of the tests developed, depending on the antibody pool used, was determined to be between 100 and 500 ppm, more than adequate for the detection of economic adulteration. 

Acceptance criteria should take into account method performance attributes and the intended use of the methods. For example, in some instances it may be critical that the method precision and sensitivity (i.e., for impurities) are similar to that obtained by the method development laboratory. In such cases, the samples selected for transfer purposes, the statistical tools applied to demonstrate equivalence and the acceptance criteria should be selected carefully to ensure that the method performance is properly evaluated. On the other hand, in some instances the capabilities (e.g., sensitivity) of the development method may exceed the method performance requirements for commercial release testing. For example, while a gas chromatographic (GC) method may be validated to have sensitivity down to 0.002% for a number of residual solvents monitored during development and, if the specifications are set only on total residual solvents with a limit of 0.5%, it may not be necessary to demonstrate sensitivity for individual solvents to 0.002% to qualify the QC laboratory for routine use. The acceptance criteria should be considered on a case-by-case basis for each method for each product and must be established in advance of the formal testing. 

---