Ensuring the Safety of Food

Tony Hasting looks at the options for quality assurance

ON 21 May, 1961 President John F Kennedy stated that the United States should set as a goal “the landing of a man on the Moon and return him safely to Earth by the end of the decade”. While engineers and scientists strove to successfully meet the challenge, a less glamorous but equally important issue had to be resolved, that of ensuring the safety of the food the astronauts would eat. A novel approach based on the concept of quality assurance was developed by the Pillsbury Company in conjunction with NASA and the US Army Natick Laboratories to ensure that safe food would be available for manned space missions. This quality assurance approach has been adapted and improved over the subsequent years. Termed Hazard Analysis and Critical Control Point (HACCP), it has become the internationally-recognised method of managing food safety and thus protecting consumers.

My experience of investigating contamination problems is that a low-level intermittent contamination is usually the most difficult to deal with. There is also a tendency to look for over complex reasons for such problems, believing that the obvious problems would already have been identified and resolved. One particular example was an aseptic soft drinks line where intermittent contamination problems were occurring. The capping of the bottles took place under sterile air pressure and there was a drain line close to this point to remove water and chemicals from the enclosed sterile zone. To my disbelief, this line was connected directly to an external main drain without any protection against back flow from the drain. The filler supplier claimed that it was impossible for any water to flow back from the main drain, a statement that was shown to be untrue as it was possible to see the backflow, particularly when the adjacent line was discharging and there was a surge of fluid in the line.

Quality control

The food industry has for many years used the traditional quality control approach for food safety primarily based on end-product sampling and testing, particularly for microbiological contamination. In practice, this method has several fundamental limitations:

• It is retrospective and can be expensive to implement, particularly for high capacity process lines.
• For a long shelf-life product, it may take several months for any contamination to become apparent, by which time the product may have been distributed to several countries.
• It measures effects rather than causes.

For a new installation prior to commercial production, the sampling rate is usually based on the minimum acceptable failure rate. With aseptic processes, a typical plant might use 3 x 10,000 packs, ie three independent runs with 10,000 packs each, to confirm that the failure rate was below 1 in 10,000. To maximise sensitivity a microbiological growth medium may be used and processed in the same way as the product. Any failure would occur more rapidly and visual inspection, turbidity, or a pH change resulting in a colour change would be easier to detect¹. After successfully passing the microbiological challenge tests, a number of short production runs may be carried out. Packaged product samples would be taken to the laboratory, product removed from the pack under sterile conditions and analysed for presence of microorganisms. It may also be possible to improve the chances of identifying potential contamination by storing samples at higher temperatures, 30–35^oC, to speed up any microbial growth. This could also be used to handle a larger quantity of samples in for example a temperature-controlled chamber, but may only be effective if the contamination results in gas production and gives a visual indication. Sampling should be focussed on the higher-risk stages of processing, such as startup, intermediate stoppages, and end of production. It may also be appropriate to temporarily increase sampling after a major maintenance programme has taken place.

My experience of investigating contamination problems is that a low-level intermittent contamination is usually the most difficult to deal with