As you rightly say, when we look at the data, depending on the source, it can be difficult to interpret because of the way it is recovered. For example, in the USA, they report on infections, some of which are assumed from the evidence they have available. If you look at the reporting of the numbers of pathogens in American produce, such as poultry, they report it in terms of the answer to the question, “Does the food contain more than”—for example—“400 counts of a pathogen per gram of food?” In the UK, the Food Standards Agency reports in terms of “low”, “medium” or “high”. National surveys such as sampling from supermarkets, for example, show that 50% of poultry have very low numbers of pathogens such as a salmonella; only about 5% or 6% have food samples with over 1,000 counts of a pathogen. By those criteria, UK foods appear to be safer—but, I must stress, according to those criteria.
As I say in the written evidence, we now have this vexed question of viable but non-culturable—VBNC—bacteria. When looking at some of the published data, it is very difficult to take that into account, but the work that we and other labs have done is now telling us that we cannot ignore it. We have published our work on chlorine treatment, but we have also looked at what happens when you stress a pathogen such as listeria by depriving it of nutrients. For example, in a factory where you are washing down with tap water, the listeria can still survive, and in those conditions it can become this VBNC form. If all you are doing is regular swabbing and then reporting, you could say, “Our factory is clear of listeria.” In fact, if we used the more modern methods, that might be found to be not true.
We are really talking not just about standards now, but the standards we should adopt in the future, both in the UK and in what we would expect other countries to adopt if we are going to import food from them.