I remember a famous animal study about heroin addiction with mice. Almost all mice that were in a very stressful environment and had easy access to heroin became addict. But if mice were living in a pleasurable environment, most mice would not become addict.

So, even if it looks very likely that the association between a habit and an environment was at work here too, I wonder how much the low relapse rate was linked not just to a change of environment, but also to a change from an incredibly stressful situation (war) to a much more satisfying one.