Relapse prevention methods seek to discourage people from behaving in undesired ways by raising awareness of behavioral chains and triggers, and by helping them to plan safer alternative behaviors that can use when they get into trouble. There is no direct effort to alter the undesired behavior in relapse prevention; only strenuous effort at undermining it.
In contrast to this nuanced approach, aversion therapy methods seek to directly prevent undesired behavior by making it unpleasant to engage in. Aversive methods work by pairing aversive stimulation (punishment) with undesired behavior so as to discourage that behavior from ever being acted on. When aversive therapy works, bad habits become unpleasant for people to pursue and so they naturally stop pursuing them. It is unclear that aversion therapy works in many cases, however, and not at all clear that it works better than non-aversive methods such as relapse prevention.
Aversion therapy can be administered in two different ways. The consequences of a bad habit can be made aversive, and the habit itself can be made aversive.
Making consequences aversive works well for some people, but typically not for people who are having problems resisting bad habits. The main problem is that there is a time delay between the consequence of an action and the taking of that action. In some cases that time delay is measured in years! When consequences of bad behavior are delayed, the short term benefits of that behavior become compelling, and people tend to act out their bad habits. Examples of delayed negative consequences that don't really have much deterrent effect are: the death penalty (which does not reliably lower violent crime rates), and AIDS and Hepatitis (which does not increase safe sex behaviors).
Methods that make acting out a bad habit actually immediately aversive fair better than those that introduce a long delay between action and consequence. It is possible to administer a medication to alcoholics that makes them sick if they drink. When this medication (called Antabuse) is on board, alcoholics are less likely to risk drinking, because they do not want to risk getting sick. Alcoholics can easily defeat this solution if they want to by not taking the medication, however.
A variation of aversion therapy for smoking was described a number of years ago called rapid smoking. Smokers are asked to smoke a vast number of cigarettes in a row, one after another, until they get sick. Research on outcomes from this procedure are mixed, with some claiming the technique has benefit and others finding no specific benefit. Individual results vary, of course. Because of the uncertainty surrounding rapid smoking, and the fact that it is not particularly healthy to smoke so much, this method is not recommended for use in smoking cessation circles.
Variations of shock therapy also fall under the rubric of aversion therapy. In a typical application, a person is encouraged to shock themselves (using a portable battery operated electric shock device, usually attached to the arm or leg) while thinking about engaging in problem behaviors. Various tools, pictures and other props associated with problem behavior may be used as part of the therapy to make the in vitro (imagined) experience more real. Shock levels are set so as to be uncomfortable, even painful, but not intensely so and certainly not damaging or dangerous. Multiple trials associating the shock and the problem behavior are administered. If the therapy works, the shock recipient comes to feel uncomfortable when thinking about engaging in the problem behavior, and desire to do so is lessened or extinguished.
Aversion therapies in general, and shock therapy in particular, are potentially dangerous techniques that should not be practiced on one's own. A professional therapist or doctor, experienced with these techniques should be involved. The equipment used to produce shocks should be professionally manufactured. Most important of all, the patient experiencing shocks or other aversive stimulation should be in control of how much pain he or she experiences at all times. Aversion therapy should never be coercive.
A major reason why aversion therapy is not a mainstream therapy approach today is that it has a large potential for abuse. Apart from the dangers inherent in inducing pain or discomfort (via mechanical, medical or electrical means), a whole other area of concern revolves around whether the "problem behavior" that is to be conditioned away is actually a problem in the first place. Some problems, such as homosexuality, are culturally defined. Conservative religious groups tend to view homosexuality as an abomination, but scientifically informed professions view it as a perfectly normal variation of human sexuality. Whether or not homosexuality is a problem behavior thus reduces to an epistemological question then; a question of how you go about determining what is true and what is false. The authors of this document unequivocally side with the scientific position with regards to homosexuality and urge you to do the same. Homosexuality is a biologically-based and normal variation of human sexuality. It is inappropriate, unethical and inhumane to use aversive conditioning procedures on homosexual human beings for purposes of attempting to change their sexual orientation.