On the west coast of Canada, where Atlantic salmon are not indigenous, there seems to be very little risk of them hybridizing with Pacific salmon. We do get the odd few survivors, but they die very early in development, and never would contribute any of the genetic material to the next generation. So it's not really an issue on that side.
With regard to where you can have hybridization between escaped aquaculture fish and wild populations, there's quite a bit of literature on Atlantic salmon but also other salmonids—Pacific salmon, rainbow trout—that shows that when they do hybridize, they produce a less fit individual. Depending on the numbers of animals that escape, there is the potential, particularly, as we've heard, in small populations, for those introgressions, or hybridization events, to cause genetic damage and fitness damage to those smaller populations.
A number of measures are being considered. Of course, the first thing you need to focus on is excellent physical containment. There have been massive improvements in that in the aquaculture cages. As well, biological containment measures are being developed with the development of sterile triploids, fish that are unable to reproductively interact if they do escape from a net pen.