However, several important issues remain unresolved, including the most effective treatment program for a given type of sex offender and the amount by which the most effective program actually reduces recidivism. The evaluations of treatment programs which have been conducted to date have included case studies and uncontrolled treatment studies, but not studies with randomly assigned control groups and long-term followup periods. However, the available research has shown that offense history is strongly related to recidivism. In addition, sexual arousal patterns as measured in the laboratory are related both to prior history and to subsequent recidivism. Patients' sexual arousal patterns can be readily changed using certain behavioral methods. These changes are related to recidivism in the short-term. Patients who recidivate appear to revert to their original sexual preference patterns. Changes in sexual knowledge and in social skills can result from behavioral methods, but it is unclear whether initial deficits in these areas are related to offense history of whether improvements are predictive of lower recidivism rates. Group and individual psychotherapies have not been shown to produce short or long-term changes with sex offenders, but these methods have not been rigorously evaluated. Thirty-three references are listed. (Author summary modified)