【正文】 es is inappropriate when performing powder blend/mixture uniformity or dosage form content uniformity determinations. In these cases, the testing is intended to measure variability within the product, and the individual results should be reported.平均法也會隱藏樣品不同部分的差異性。例如：當(dāng)進行粉末混和/混合物的均勻性檢測或制劑含量均勻性檢驗，使用平均法是不合適的。在這些情況下，檢測是為了測量產(chǎn)品內(nèi)的差異性的，因此應(yīng)該使用各單獨測試結(jié)果。It should be noted that a test might consist of replicates to arrive at a result. For instance,an HPLC assay result may be determined by averaging the peak responses from a number of consecutive, replicate injections from the same preparation (usually 2 or 3). The assay result would be calculated using the peak response average. This determination is considered one test and one result. This is a distinct difference from the analysis of different portions from a lot, intended to determine variability within the lot. The use of replicates should be included in the written, approved, test methodology. Unexpected variation in replicate determinations should trigger investigation and documentation requirements (21 CFR ).應(yīng)該注意一個檢驗結(jié)果可能是反復(fù)測定得出的。實例：一個HPLC分析結(jié)果可能是對相同樣品溶液幾次連續(xù)、反復(fù)的進樣（23次）所得峰的響應(yīng)值的平均而確定的。含量測試結(jié)果應(yīng)使用峰響應(yīng)值的平均值來計算。這個確定被認(rèn)為是一次試驗和一個結(jié)果，這與同一批產(chǎn)品的不同部分的分析以判斷一批產(chǎn)品的可變性是明顯不同的。重復(fù)測定應(yīng)該收載在書面的、經(jīng)批準(zhǔn)的檢驗方法里。重復(fù)測定結(jié)果的意外變更應(yīng)引起調(diào)查并進行記錄(21 CFR )。In some cases, a series of assay results may be a part of the test procedure. If some of the results are OOS and some are within specification and all are within the documented variation of the method, the passing results should be given no more credence than the failing results, in the absence of documented evidence that analytical error had occurred. 在某些情況下，一系列的測試結(jié)果可以是分析過程的一部分。如果其中一些結(jié)果是OOS結(jié)果而有一些在標(biāo)準(zhǔn)之內(nèi)而且所有結(jié)果都在該方法的變異范圍之內(nèi)，在沒有能文件證明存在檢驗錯誤的情況下，合格的結(jié)果并不比失敗的結(jié)果有更高的可信度。 Relying on test data averaging in such a case can be particularly misleading. For example, in an assay with a given range of 90 to 110 percent, test results of 89 percent, 89 percent, and 92 percent would produce an average of 90 percent even though two of the assay values represent failing results.依據(jù)對檢驗結(jié)果的平均在某些方面能產(chǎn)生特別的誤導(dǎo)，例如：一個含量標(biāo)準(zhǔn)范圍是90%110%，檢驗結(jié)果是：89%、89%、92%，盡管兩個含量值表明不合格，其平均值卻為得到90%。To use averaged results for assay reporting, all test results should conform to specifications. Although the above average of 90 percent may be useful in terms of an overall assessment of process capabilities, the individual assay results indicate nonconformance because two of the three results are outside of the range. A low assay value should also trigger concerns that the batch was not formulated properly because the batch must be formulated with the intent to provide not less than 100 percent of the labeled or established amount of active ingredient (21 CFR (a)). The above example does not necessarily require the manufacturer to fail the batch, but indicates that an immediate investigation should be conducted for batch disposition decisions.要應(yīng)使用平均結(jié)果作為含量測定結(jié)果，所有的檢驗結(jié)果都必須符合標(biāo)準(zhǔn)。盡管上面的平均值90%可以用于生產(chǎn)能力的全面評估，因為3個結(jié)果中的2個在范圍之外，單獨的含量測定結(jié)果就標(biāo)明產(chǎn)品不符合要求。一個低的含量值應(yīng)引起注意 ：因為一批產(chǎn)品應(yīng)該按照能提供不低于100％標(biāo)示量或確定數(shù)量的原料的標(biāo)準(zhǔn)來制備(21 CFR (a))，因此該批沒有正確配制。上面的例子并不一定要求企業(yè)把此批作為不合格批，但是表明應(yīng)該對此批立即進行調(diào)查以確定對該批的處置方法。4. Outlier Tests逸出值的檢驗The CGMP regulations require that statistically valid quality control criteria include appropriate acceptance and/or rejection levels (167。 (d)). On rare occasions, a value may be obtained that is markedly different from the others in a series obtained using a validated method. Such a value may qualify as a statistical outlier. An outlier may result from a deviation from prescribed test methods, or it may be the result of variability in the sample. It should never be assumed that the reason for an outlier is error in the testing procedure, rather than inherent variability in the sample being tested.CGMP法規(guī)要求統(tǒng)計的有效質(zhì)量控制標(biāo)準(zhǔn)包括合適的可接受的和/或不合格水平(167。 (d))。在極少情況下，用一個用驗證過的方法獲得的一個值會明顯的不同于得到的一系列值，這樣的一個值應(yīng)該作為一個統(tǒng)計的逸出值，一個逸出值可以是規(guī)定檢驗方法的偏差的結(jié)果，或是樣品存在差異的結(jié)果。決不能將出現(xiàn)逸出值的原因假定為檢驗操作的錯誤，而不認(rèn)為被檢驗樣品存在固有的差異。Outlier testing is a statistical procedure for identifying from an array those data that are extreme. The possible use of outlier tests should be determined in advance. This should be written into SOPs for data interpretation and be well documented. The SOPs should include the specific outlier test to be applied with relevant parameters specified in advance. The SOPs should specify the minimum number of results required to obtain a statistically significant assessment from the specified outlier test.逸出值檢驗是從一組數(shù)據(jù)中識別出極端數(shù)據(jù)中的統(tǒng)計方法。逸出值檢驗的可能使用應(yīng)該事先確定。對這中使用應(yīng)該寫入用來解釋數(shù)據(jù)的SOP里并做好記錄。SOP應(yīng)該包括特定的逸出值適用于預(yù)先指定的相關(guān)參數(shù)。該SOP應(yīng)該規(guī)定用于從特定的逸出值測試中獲得統(tǒng)計學(xué)上有意義的評估結(jié)果的最少的檢測結(jié)果數(shù)量。For biological assays having a high variability, an outlier test may be an appropriate statistical analysis to identify those results that are statistically extreme observations. The USP describes outlier tests in the section on Design and Analysis of Biological Assays (USP 23, p. 1705). In these cases, the outlier observation is omitted from calculations. The USP also states that arbitrary rejection or retention of an apparently aberrant response can be a serious source of bias. . .the rejection of observations solely on the basis of their relative magnitudes is a procedure to be used sparingly (USP 23, p. 1705).生物制劑含量測定有更高的變異性，逸出值測試可能是判斷這些極端結(jié)果的適當(dāng)?shù)慕y(tǒng)計分析。USP在生物制劑設(shè)計和分析章節(jié)描述了逸出值。在這些情況下，逸出值在計算時被省略。USP也規(guī)定了“對明顯的異常值的任意拒絕或保留可能是發(fā)生偏差的重要原因. . . . . .僅依靠相對重要性來拒絕觀測值是一個應(yīng)謹(jǐn)慎使用的方法”For validated chemical tests with relatively small variance, and if the sample being tested can be considered homogeneous (for example, an assay of posited dosage form to determine strength), an outlier test is only a statistical analysis of the data obtained from testing and retesting. It will not identify the cause of an extreme observation and ,therefore, should not be used to invalidate the data. An outlier test may be useful as part of the evaluation of the significance of that result for batch evaluation, along with other data.對于驗證過的相對變化較小的化學(xué)測試來說，如果被檢驗的樣品是均質(zhì)的（例如，測定一復(fù)方制劑的濃度），逸出值測試僅僅是對檢驗和再檢驗數(shù)據(jù)的統(tǒng)計分析。不能確定出現(xiàn)極端觀察值的原因，因此，不能用來決定使數(shù)據(jù)無效。逸出值測試與其它數(shù)據(jù)一起