Safety first

By Paul B Young

A number of years ago whilst working as a food safety regulator, I had occasion to defend in court analytical test results demonstrating the presence of the banned antibiotic chloramphenicol in poultry imported into the United Kingdom. The reason for the challenge lay in the fact that the exporter had been diligently carrying out in-house testing for this antibiotic for several years and had never once detected its presence.

Our findings indicated that the chloramphenicol was present at less than two parts per billion (ppb); a low concentration but violative nonetheless. During the hearing, it came to light that the test employed by the exporter had a limit of detection of five parts per million (more than 15,000 times higher than is necessary to afford protection from violative findings in Europe). European Union regulations demand that analytical techniques employed for detection of chloramphenicol must be able to detect the antibiotic at 0.3 ppb or less, so this was an unfortunate case where a food producer was diligently employing an analytical technique completely oblivious to the fact that the test itself was not fit-for-purpose. In other words, the test result did not provide the correct information to allow the producer to make an appropriate decision.

This unfortunate example perfectly illustrates the importance of understanding the purpose of testing before we even begin to evaluate the parameters that need to be characterised (accuracy, precision, sensitivity, specificity, robustness, etc.) alongside those that may exert a significant effect, such as changing matrix. 

It's often stated that 'we can't test our way to safe food,' and indeed this is true.  Testing can only definitively tell us that the piece of food that we just destroyed in the testing process was safe. Rather, effective food safety is built in through a series of systems employed during production processes, designed to minimise the possibility of hazards being introduced. Those systems themselves need to be continually monitored through testing to verify their efficacy. Additionally, regulatory bodies are also required to carry out a significant level of surveillance testing to ensure the systems are being employed effectively across the industry. If the verification tests are not fit-for-purpose then there can be no assurance that the systems are affording any protection.

Traditionally, fitness-for-purpose of individual analytical test methods has been verified through the establishment of reference methods, which have undergone very extensive collaborative studies in the hands of numerous scientists from a large number of laboratories. Whilst this is very time-consuming, such methods can play a valuable role in settlement of disputes arising when there is poor agreement between findings in different laboratories. One reason for the different findings could lie in the fact that scientists may make the assumption that reference methods do not require some level of validation to demonstrate in-house efficacy. More often, however, these discrepancies occur because the development of reference methods cannot keep pace with scientific advances and the reference methods themselves become outmoded and inefficient and are therefore no longer the method of choice for routine use in busy food safety laboratories.

The World Trade Organisation's Technical Barriers to Trade agreement encourages countries to recognise each others' procedures for assessing whether a product conforms. However, this potential for disputed laboratory findings creates a very real challenge for countries to accept the assurances given by their trading partners regarding the safety of foods produced overseas.

In the absence of universally accepted and enforced harmonised criteria to assess the fitness-for-purpose of analytical methods, authorities will always have doubts about the efficacy of tests employed elsewhere and will be forced to carry out high level of verification (re-testing) at import. Conversely, adoption of universally accepted in-house validation procedures permits authorities to move from a regular and frequent inspection based approach to regulating the safety of food imports, to a regular but infrequent audit based approach to ensure the procedures employed overseas offer equivalent guarantees of safety as those employed for domestically produced foods.