Just as those who over eat want to control their appetite for food rather than simply stop eating so do some who have a drink or drugs problem. In many cases the drink is a symptom not a cause. Remove the cause and for some people at least the drink problem will subside.
I worry also that the AA members are also a subset of those who have a drinking problem, are we sure that they are a comparable subset to those who took other paths? How do the studies account for people who drop out of either AA or some other program and did the studies follow up years later to find out if people had fallen back into addiction?
The article was a bit light on details.
And lastly the article mentions relative efficacy but doesn't say anything about how efficacious any of the programs were leaving me wondering if any of them are worthwhile.