A study in 1971 revealed that about 20% of the American soldiers in Vietnam were addicted to heroin at some point during their term of service overseas.
Almost half (45%) had used either opium or heroin at least once during their tour of duty.
I've often wondered, if/when faced with the prevailing conditions in Vietnam, if I would have been part of the 45%.
Interestingly, however, when re-examined a year after their return home, only 5% had resumed heroin use.
Prior to this study, the conventional wisdom was that heroin addiction was a disease and typically had very high relapse rates (close to 90%).
For soldiers in Vietnam, heroin use had become associated with being in Vietnam. Back home (in their homes, workplaces or classrooms and associating with their family and friends) they did not have the same cues and opportunities for heroin use.
To a behaviorist, context is a critical component in the formation and shaping of behavior.
They hypothesize that heroin use was a habit, not a disease. Like all habits, it's best broken by avoiding people, places and things that are linked to the habit.
It seems like a pretty sound hypothesis to me.