This can be a controversial subject. What constitutes success? Do we look only at sexual reoffenses, or also at related (compulsive) behaviors like alcoholism, drugs, gambling, etc.? What about other crimes - burglary, assaults, etc.? Do we look at the probationary period only, or is post treatment considered? If post, how long? How do we get data? Self report? Crime reports? Family follow-up?

How hard do we look for failure? Do we rely only on crime reports or do we do a confidential study where we annually polygraph people for ten years following treatment? How do we count treatment drop outs in such a study - as treated offenders, untreated offenders, or as a separate group?

No one expects zero recidivism, so how much does a program have to reduce reoffenses to call itself successful?

Given all the difficulties, here are some things we do know.

Barry Maletzky, MD and Kevin McGovern, Ph.D. of The Sexual Abuse Clinic of Portland Oregon followed about 5000 offenders treated in their clinic and similar clinics between 1973 and 1990 using behavior oriented methods. About 3700 of these were pedophiles -770 were exhibitionists and the remainder were referred for a variety of other paraphilias. Criteria for "success" included:

Using these stringent measures to follow some men for as long as 17 years post treatment, success was achieved with 94.7% of heterosexual and 86.4% of homosexual pedophiles. Rapists showed 73.5% success, exhibitionists and public masturbators about 92% , with men referred for various other paraphilias ranging from 100% for zoophiliacs to 80% for frotteurs. These data do not represent a controlled study, but the sample is large and with success criteria as stringent as they were, the data gives strong indication that treatment is effective for a great many offenders.

A June 1991 report to the State of Washington legislature also supports community treatment as a viable alternative for sex offenders. The report covers 613 probation eligible offenders sentences between January 1985 and July 1986. Three hundred thirteen of these actually received probation sentences while 300 were sent to prison. Both groups were followed. The probationers had significantly lower re-arrest rates and conviction rates in all crime categories. The study concluded that, generally speaking, probationary sentences did not place the community at undue risk and offered a cost - effective alternative to prison.

An Oregon study of sex offender monitoring using polygraphy indicated dramatic success having offenders complete their probationary periods without reoffenses.

In 1999 Margaret Alexander, Ph.D. (Oshkosh, Wis. Correctional Facility) examined no less than 424 studies. After eliminating most of them because they were poorly done she presented an analysis of the remaining 79studies covering 10,988 offenders with some being followed as long as ten years post treatment. (Sexual Abuse; A Journal of Research and Treatment, 11(2) ) Here are some of her findings.

Also in 1999, Grossman, Martis and Fichtner presented an analysis of Medline literature and concluded that offenders treated with anti-androgen and / or cognitive-behavioral therapy showed a robust treatment effect in the neighborhood of 30%. (Psychiatric Services, 50(3) )

Data from a variety of sources show that a "some treatment" is not better than none – treatment dropouts reoffend at the same rate as untreated offenders.

None of this represents true controlled studies. Such experiments are under way in California and Vermont using inpatient populations and preliminary data are promising, but samples are so small that really meaningful data will not be available until the year 2005. Even then we will not necessarily be able to generalize to out patient programs. Controlled out patient studies may never be done because of reluctance to have matched controls at large in the community without any treatment. Even the prison studies in CA and VT may never be completed, due to recent funding cutbacks.

Robert Prentky, Ph.D. (Bridgewater. Mass. Correctional Facility) developed a cost effectiveness model for "success." He suggested comparing the cost of prosecuting a single reoffense, incarcerating the offender, and treating one additional victim to the cost of meaningfully treating an offender during his initial incarceration. According to his figures, the Bridgewater program is cost effective if it reduces reoffenses by 11%, it. When Janice Marques applied his model to California she arrived at a 14% cutoff.

Given available data, it appears that out patient programs do much better than 11 - 14% offense reduction. In fact, it does not seem unreasonable to assume we reduce reoffenses by a third, that we teach offenders some empathy, so that they treat others better in non-sexual ways as well, and that we make a significant contribution to their social functioning (reduce non-sexual crimes, improve employment performance, etc.)

The United States already locks up a greater percentage of its people than any western nation while California, with about 10% of the country's population accounts for about 14% of the prison population. Under theses circumstances, strict conditions of probation, close monitoring and quality treatment paid for by the offenders themselves is clearly the most promising alternative.