Creme Global

The Real World

- By:

Courtesy of Creme Global

The Real World

Food Safety Assessment

As outlined in our Ideal World Article, it is not feasible or even possible to have a complete food consumption data set which records every single ingredient or chemical. It is especially not possible to have an up-to-date one, since food consumption habits and compositions are changing constantly and food surveys are carried out infrequently due to the cost and time required to perform them.

Fortunately, all is not lost. Effective food safety assessment tools can combine the separate data sets which do exist in a meaningful way to allow accurate estimation of exposure to the population of the chemicals of interest. These assessments are based on the latest available information regarding market shares of brands, pesticide residue levels from routine screening programmes, additive concentrations, changes in the legal framework, and other relevant information sources.

Effective safety assessment tools allow the inclusion of uncertainty and variability in the analysis. This is very important since in most cases there is a significant degree of uncertainty and/or variability in the input data. Uncertainty and variability often drive exposure calculations, but do not present a problem as long as they are treated properly and quantified correctly.

To understand the exposure situation fully, error bars or confidence intervals on all output statistics from an assessment should be provided. This gives the decision maker the mean value (expected value) of the key exposure results for the population along with the error bars, i.e. the range of values the key result can feasibly take.

For example, with this method, the assessor would not simply quote the 95th percentile exposure level of the population, but they would quote the 95th percentile of the exposure in the population, along with confidence intervals outlining the range which that percentile could reasonably take based on the uncertainty and variability in the input data.

This methodology is standard in many areas of risk analysis and emerged originally from research in physics. In these areas, results are always quoted with confidence intervals so as to provide information to the reader on not just the expected value of the result, but also to show how meaningful the result is.

CREMe specialises in handling these issues for our clients, with our state-of-the-art software service which utilises techniques that are used by the major financial institutions to manage financial risk. We will provide the rapid and reliable results (with confidence intervals) that you need.

Customer comments

No comments were found for The Real World. Be the first to comment!