| Popis: |
Analytical results may be affected by a large number of factors inherent to the measurement chemical process (MCP). Ruggedness studies use statistics to establish factors that donor do not affect the stability of results. Nevertheless, such studies are performed during method development or at the beginning of method validation, so they do not consider other factors that can occur in a longer period (e.g., one year). We assess the choice of the response on which the effect of changes in the factors, which is not usually discussed, as, in most cases, instrumental signals or the recovery rates are chosen. As complementary information to that from ruggedness tests, data provided by internal quality-control programmes of routine laboratories allow assessment of the factors that most influence the stability of analytical results. To study the likely influence of factors on the stability of results, calibration plots seem to be more appropriate than simple instrumental signals or recovery rates. Analysis of covariance (ANCOVA) performed on the calibration curves shows if the stability of measurements is affected by the following factors: • the number of concentration levels used for the calibrations; • the use internal standards; • changes in the primary standard solutions; • changes in the chromatographic columns; and, • the active element of nitrogen phosphorus detection (NPD). The variability of results associated with such factors is minimised dramatically by adopting measures relating to internal quality control (IQC) and laboratory management. |
Full Text from ScienceDirect