In the late 1970s, William Pithers, a psychologist with the Vermont Department of Corrections, began studying sex-offender treatment programs around the country.

At the time, rapists and child molesters typically underwent individual psychoanalytic therapy in state hospitals and prisons and, in some cases, outpatient clinics. But, as numerous studies concluded, there was virtually no difference in recidivism rates between sex offenders who had received treatment and those who had not.

By 1983, Pithers and four fellow researchers had adapted a cognitive-behavior model for sex offenders, known as "relapse prevention," that was successful in treating alcoholics and substance abusers. Cognitive-behavioral therapy is based on the theory that, to a great degree, our thoughts determine our behavior, and that changing the way we think leads to changes in the way we act. For addicts, that might involve a new appreciation for a healthier lifestyle. For sex offenders, relapse prevention attempts to teach the pedophile how to control thoughts that trigger impulses which, in turn, lead to deviant sexual arousal.

The work of Pithers and his colleagues represented a major shift in sex-offender treatment at a time when child molesters represented one of the fastest growing segments of the U.S. prison population. According to the Department of Justice's Center for Sex Offender Management, the number of people incarcerated for sex crimes involving children increased by more than 300 percent between 1980 and 1994.

Today, more than 90 percent of the 600-plus adult sex-offender treatment programs in the U.S. use relapse-prevention techniques. And while no one claims child molesters can be cured, a growing body of evidence suggests that treatment reduces the risk that a sex offender will strike again.

According to a summary of scientific data by the Association for the Treatment of Sexual Abusers, cognitive-behavioral treatment can reduce recidivism rates among sex offenders by as much as 40 percent.

In Vermont, the results are even more dramatic. According to a 2003 study of the Vermont Treatment Program for Sexual Abusers, or VTPSA, the recidivism rates for offenders who completed treatment was just over 5 percent. That's compared to recidivism rates of more than 30 percent for those who did not complete the program or received no treatment at all.

As a result, the VTPSA is widely considered to be among the most comprehensive and effective treatment programs in the nation. Alisa Klein, public policy consultant for the Association for the Treatment of Sexual Abusers, said that Vermont has been "on the forefront" of sex-offender treatment. "There's been a lot of important work that's been created in Vermont," Klein said.

But the VTPSA will likely be scrutinized and debated next month, when lawmakers convene a series of hearings on Vermont's sex-crime laws. The call to re-examine the laws followed the arrest of Michael Jacques, who is accused of kidnapping Brooke Bennett, his 12-year-old niece. Bennett's body was found in a shallow grave, near Jacques' Randolph home, in early July.

Jacques, 42, has a history of sex offenses dating back to 1985, when, according to court documents, he was charged with lewd and lascivious conduct with a female relative. That case was eventually dismissed. In 1993, Jacques was sentenced to six-to-20-years in prison for kidnapping and raping an 18-year-old woman - a crime that, according to a state judge, "rivals the most brutal and terrifying scenarios seen . . . on television and in the movies."

Jacques served four years of his sentence, during which time he completed the VTPSA. Jacques underwent several years of additional therapy in a VTPSA outpatient program while on probation. In 2006, Jacques was released from state supervision, largely because he was considered a model of treatment success.

The Vermont Department of Corrections employee who supervised Jacques spoke up at a 2004 probationary hearing. Michael Kearney told the judge: "When I make comments about success in sex offender treatment, I have three names, of which Michael Jacques is one."

While it is clear that the VTPSA failed in Jacques' case, clinicians and supporters of the program worry that public outrage will encourage lawmakers to make policy changes that could have a negative impact on sex-offender treatment in Vermont. Tom Powell, a forensic psychologist who, for 18 years, was the clinical director for the Department of Corrections, said he understands the anguish over Brooke Bennett's horrifying end. However, he said, radical changes in state sex-crimes law would represent "a crude response" to a complex problem.

"A lot of public policy is being contemplated right now based on a single bad event, by a single alleged actor whose conduct was reprehensible and deadly," Powell said. "But the question that is so important is, do we engage in wholesale policy changes based on one bad event by one person? Or do we say, wait a minute, let's take a look at the system and not let Nancy Grace and Bill O'Reilly drive our policy, which I believe is happening."

******

Back in 1992, as evidence emerged suggesting that cognitive-behavioral therapy was effective in reducing relapse in sex-offenders, William Pithers told The New York Times, "The most successful candidates for treatment are men who have no other criminal record, have an established network of family and friends, hold a job and who are not so preoccupied by their sex fantasies that they think of them hours a day."

Pithers' comments underscore two important assumptions that govern sex-offender treatment. The first is that not all sex offenders are alike: Some are compulsive predators with no apparent conscience; others are social misfits whose Internet-driven fantasies lead to deviant behavior.

The second assumption: The efficacy of the treatment depends on figuring out which kind of sex-offender you're dealing with.

All cognitive-behavioral models used to treat sex offenders begin with a risk assessment that examines the offender's background (about a third were molested themselves as children), age and criminal history, propensity for violence and the characteristics of the offense. While the state uses four diagnostic tools to establish the risk of relapse, Robert McGrath, VTPSA's clinical director, says none of them can predict which particular offender will fail to respond to treatment.

"The risk tools can't predict the behavior of any one person," McGrath said. "But they can predict, moderately well, the behavior of a group of people."

Vermont sex-offenders typically undergo two risk assessments - one upon incarceration to determine the level of treatment; another, four to six months before release, is used to guide treatment at one of 13 community-based programs operated by the Department of Corrections.

In prison, treatment begins by getting the offender to acknowledge his crime, typically by confronting him with police reports and victim affidavits. If this doesn't happen, the offender is booted from the program and will likely serve his entire sentence.

In the 2003 VTPSA study, conducted by McGrath and program coordinator Georgia Cumming, of 195 adult sex offenders in the study, 90 - more than a third - did not receive treatment, in most cases because they wouldn't admit they'd committed a crime.

"If a person was found guilty, our assumption is they committed the offense," McGrath said. "Traditionally we make an effort to help them work through the denial. But ultimately if someone says, 'I didn't do it, so I can't talk about it and I don't want treatment,' they aren't going to end up in the treatment program."

Unlike psychoanalysis, which can take years to change behavior, the cognitive-behavioral model assumes there is a point when formal therapy will end. The highest risk offenders in VTPSA spend two to three years in therapy; low and moderate risk offenders receive anywhere from six to 14 months of treatment, according to McGrath.

Treatment usually begins several months before the offender is scheduled for release. Over time, the sex-offender therapy has become less confrontational than earlier approaches, and relies on specific "homework" assignments geared toward helping the offender understand the harm his behavior has caused, as well as the thoughts and emotions that led to the crime.

Offenders are asked to diagram their thoughts and feelings at varying points in the sexual assault process. They work up a self-management plan to deal with anger, and they often receive social-skills training, from how to initiate and maintain a simple conversation to learning how to express emotions in intimate relationships.

Upon release, treatment picks up in a community-based clinic under the supervision of a specially trained parole officer. McGrath said community-based treatment could involve group therapy one or more times a week for several years.

Before release, family and friends of the offender are brought into the prison, where they are informed of the offender's patterns of deviant behavior and how to recognize them. That support network is also key to deciding when to end treatment altogether. It's a decision based in part on interviews with the offender, family, friends and employer.

About 70 percent of sex-offender programs, including the VTPSA, also use polygraph examinations to gauge the progress of treatment. Questions include whether the offender is viewing pornography, using drugs or alcohol, or making contact with children without the knowledge of his family and probation officer.

"We look at observable behaviors that were linked to the offense," McGrath said, "as well as what the guy is saying, what the probation and treatment people are saying, what family members and employers are saying, and ask, 'Is all that supported by what the polygraph says?'"

Presumably, in Michael Jacques' case, everything added up. Indeed, when Jacques went before the court, seeking to be released from probation, he was accompanied by his wife. As the judge noted, he owned a home and had a good job.

According to the testimony of probation officer Richard Kearney, Jacques not only met all the requirements of his community treatment; he exceeded them. "If what we're looking to do is make significant life changes so that this doesn't happen again," Kearney said, "Mr. Jacques has made those significant life changes."

******

As it turns out, of course, Jacques had not been a model offender in the VTPSA. Before he was charged with kidnapping Brooke Bennett, police announced that while on probation Jacques had allegedly molested a young girl for five years, beginning when she was nine.

It doesn't take a case like Jacques' to raise the question of whether or not sex offender treatment works. Even clinicians like Robert McGrath acknowledge that parsing the data on recidivism rates offers no definitive answers.

The problem, he and others say, is that the greatest rates of success are among those who are motivated enough to engage in treatment.

"Is it because those who volunteered and made it through had characteristics that already made them successful candidates?" McGrath noted, "Or is it the treatment?"

Alisa Klein, the consultant for the Association for the Treatment of Sexual Abusers, acknowledges that sex-offense treatment is a "fairly new" field in which good data is still being gathered.

"The percentages aren't huge in terms of the reduction of recidivism," Klein said. "But they are big enough that we know we're going in the right direction with specialized treatment of sex offenders."

For that reason, Tom Powell, who retired as clinical director for the Department of Corrections, says lawmakers should carefully consider any policy changes that could impact the VTPSA. "The implication is that the current laws and programs are totally inadequate and we need to start from scratch," he said. "That completely ignores the entire history of the program and its remarkable success at treating people."

McGrath is a little more reluctant to describe the success of the VTPSA as a "remarkable success." However, he too, urges caution.

"Virtually everyone who is incarcerated for sex offenses in Vermont and the U.S. will eventually get out of prison," McGrath said. "The question is, what do we do to reduce the risk of re-offending? Treatment is not a panacea, but it can have an impact on reducing risk and making our communities safer."