The broad approach in the application of the OU method, in all cases, was to specify an experimental design for primary sampling and chemical analysis for the selected food/analyte combination, that allows the uncertainty to be estimated by the method of Ramsey (1998).
sampling of foods either directly by, or monitored by, project staff;
chemical analysis by subcontracted labs, to the agreed experimental design, with normal in-house quality control, and all measurements reported in unrounded and untruncated form (e.g. no censorship below nominal detection limits);
calculation of measurement uncertainty and its components, for selected analytes from each case study, using robust analysis of variance;
construction of a loss-function based on estimates of consequence costs and the calculation of optimal uncertainty and minimal expectation of loss;
calculation of optimal division of expenditure between sampling and chemical analysis; and
comparison of the optima found with the original un-optimised values, and consideration of what steps are needed to achieve the required changes.
The general approach to the estimation of uncertainty from physical sample preparation involved specifying an experimental design which allowed for the estimation of uncertainty from physical sample preparation (including systematic error estimation from this source) in addition to primary sampling and chemical analysis. Calculation of measurement uncertainty included individual estimates of random error from primary sampling and both random and systematic errors from sample preparation and chemical analysis using robust analysis of variance. The general approach for the adaptation of OU method for optimisation of sample preparation processes involved the construction of a loss function that allowed for the optimisation of measurement uncertainty (including physical sample preparation) and optimal apportionment of expenditure between primary sampling, sample preparation and chemical analysis.
When anyone measures the concentration of any constituent of food they will not report the 'true' value, but an estimate. It is important therefore to know the range about the measured value of concentration that contains the true value; this is called the uncertainty of the measurement.
To get a reliable value of the uncertainty, consideration must be given to the contribution made to the uncertainty by both the primary sampling of the food, as well from its chemical analysis. Once this reliable value of uncertainty is known, the concentration value can be compared with statutory thresholds, to judge the safety of the food.
There will always be a possibility of misclassifying the food, due to this uncertainty. There may a 'false positive' classification, where the measured concentration is above some threshold, but the true value is below. Alternatively, there may be a 'false negative', where the measured concentration is below the threshold, but the true value is above. In the food sector both of these errors may have financial consequences. The false positive will cause the rejection of a batch of food and a financial loss equal to its commercial value. The false negative may lead to food being sold with contamination present at a concentration above a statutory limit. This may, if detected, lead to potential health implications or financial consequences from either litigation or from loss of corporate reputation of a manufacturer or retailer. There is therefore a balance to be struck between the cost of the sampling and analysis, and of the potential financial outcomes of misclassifying the food.
The aim of this study was to test a new Optimised Uncertainty (OU) method for achieving this balance.
<p>Find more about this project and other FSA food safety-related projects at the <a href="http://www.food.gov.uk/science/research/" target="_blank">Food
Standards Agency Research webpage</a>.