I'll be very quick.
One of the important things about surveillance is that when someone gets ill, they may or may not go to the doctor. If they don't go to the doctor, there's no diagnosis—and the majority never go, and never need to actually. Then those who go to the doctor may or may not get tested. Generally, you're not going to test everybody, and that's quite appropriate clinically. But if they're not getting better, then you'll test; or if they're severely ill or in hospital, you'll test. Then if it's tested, you may or may not pick it up in the laboratory, and then whether it gets typed....
So what we've done in the system—and Frank has described the laboratory system—is to make it better integrated, which increases the chances that if someone is tested, we will actually figure it out and be able to do something and recognize the links. It's far better today than five to ten years ago. If the listeria outbreak had happened ten years ago, we probably would never have actually found it until it was so big that it was just overwhelming.