Sex Offender Treatment

Hollida Wakefield and Ralph Underwager^*

ABSTRACT: For those convicted of sex crimes, probation with mandated treatment along with some jail time is a common disposition.  The major goal of treatment for sex offenders is the prevention of sexual offenses in the future.  However, until recently there has been little evidence that treatment reduces recidivism.  The type of treatment which is most likely to succeed is an individually-tailored approach that includes careful assessment and uses a broad mix of cognitive-behavioral techniques to support individual behavior change.  There is little evidence for the effectiveness of many commonly-used treatment approaches.
  

Psychotherapy is often ordered for child sexual abuse perpetrators.  Of possible treatment modalities, the most common approach has been group therapy that relies heavily upon punitive and hostile confrontation and a nonsystematic blend of psychoanalytic concepts and traditional talking therapy.  There is often little or no effort to provide a theoretical base for the program.  The result is a procedure that is essentially highly moralistic and reflects the judgmental emotional response of the society rather than an empirically-based healing technique.

Treating people with disordered behavior patterns as morally defective and requiring a change in moral commitments has a long history (Siegler & Osmond, 1974).  However, moral treatments, such as those currently vended for perpetrators of child sexual abuse, should be labeled for what they are.  It is professionally irresponsible to call a procedure therapy, implying it is value free, when, in fact, it is based upon moral values and pursues goals defined moralistically.

Psychotherapy is a venture much studied and researched and there is an extensive literature on psychotherapy processes and outcomes.  There is sufficient information to have some understanding of what may actually work to change behavior. The scientific knowledge available permits more than an educated guess or a trial and error methodology.  Unfortunately, many current treatment programs for sexual abusers fall to use techniques known through research to be effective.

The effectiveness of therapeutic treatment is often measured by its contribution to restoration of emotional health and normal functioning along with the subjective sense of well-being of the individual.  Normal behavior may be defined either by reference to the applicable social norms or by statistical frequency.  With treatment for a person guilty of child sexual abuse the essential goal is the prevention of sexual offenses in the future.  Subjective well-being or conformity to generally accepted norms in other areas is not sufficient to measure treatment outcomes.

In providing treatment to persons accused and/or convicted of child sexual abuse, the situation can be complicated by a number of factors.  When a person is accused of sexual abuse, the accusation is either true or it is not true and the accused may admit or deny the accusation.  But there also may be plea bargaining, dropping of charges for insufficient evidence, dismissal by stipulation in family court, admission of guilt, admission of a mistake by the social service agency, and acquittal by the criminal court along with a finding of abuse by family court.  An accused person may be required to enter a treatment program as a sexual abuser long before there is a determination by the justice system about the accusation.  Often this is a requirement laid down by child protection as a precondition for a parent to have contact with his or her children.  For many parents this is a highly coercive demand.  It is difficult to imagine a more powerful club to hold over parents who love their children.

Such events confront the therapist with a complex situation.  Determining what is to be treated may not be easy.  There are many permutations of the interaction of truth or falsity, denial and admission, and substantiated or unsubstantiated allegations in people entering sex offender treatment programs.  The most difficult is the situation of a person accused who in reality did not do it, denies it, but the accusation is substantiated or the court rules that the abuse had occurred.

Furby, Weinrott, and Blackshaw (1989) note that for those convicted of sexual crimes, probation with mandated treatment and perhaps some jail time is the most common disposition.  Also, a person accused of sexual abuse may be offered a choice of therapy in place of punishment.  The offer may be made in criminal court or in juvenile and family court.  In a divorce or custody battle the parent initiating the allegation may offer to restore visitation if the accused parent admits guilt and successfully completes the treatment program.  In criminal court, the defendant may be offered a plea bargain in which he is put on probation in exchange for entering a treatment program.  He therefore avoids risking criminal conviction and years in prison.  The deal offered is that entering sexual offender treatment will mean avoiding highly aversive consequences such as imprisonment, loss of relationships with children, loss of career or job, financial ruin through an expensive trial, and embarrassing publicity.

The subtleties of such a deal are that apparently everybody wins.  But there is a negative effect upon the process of therapy, as Langevin (1983) points out:

Often the imposition of external force to be treated is unsatisfactory and a poorer treatment outcome can be anticipated ... Court orders for treatment as opposed to jail or in addition to jail make it hard to enact any worthwhile treatment program because treatment becomes a sentence rather than a therapy (p.64).

When a person is sentenced to sex offender treatment as part of a plea bargain or sentencing, the therapy itself may become punishment.  The person is ordered to attend treatment with an indeterminate sentence and usually cannot select the therapist or the program; therapy programs must be approved by the agencies in control.  The distortions and dangers inherent in this situation are reflected in the formation of a group of psychologists in Washington who have initiated a class action lawsuit against the state and against prosecutors who have limited sexual offender treatment programs to a small group approved by the system (Deatherage, 1990).

The therapist who provides such court-ordered treatment for sexual offenders must make regular reports to parole officers, judges, and child protection workers.  The therapist is given the power to judge when the treatment has been successfully completed and discharge is granted.  This puts the therapist in the role of the jailer.  Also, any information given by the sex offender about other victims or offenses must be reported by the therapist.  Langevin and Lang (1985) comment that a therapist who serves both as helper-therapist and as informer for the law becomes a "double agent."

This difficulty is illustrated by a recent Minnesota case.  The client, who was in a sex offender treatment program following his conviction for rape, was asked as part of the treatment to write detailed accounts of other times in which he had sexually assaulted women.  He complied and produced a written account of several other incidents.  This was then given to the police by the probation officer.  The man was later convicted on the basis of this written account and sentenced to nine years in prison (Zack, 1990).

These circumstances also provide a subtle opportunity for any hostility or pathology in a therapist to affect the therapist's behavior and the process of therapy.  The research in sensitivity groups (Lieberman, Yalom, & Miles, 1973) demonstrates the damage a hostile therapist can do to vulnerable group members.  A hostile therapist can cause serious emotional harm to patients.  The seductiveness of the powerful level of control available to a therapist can cause the therapy to be destructive and damaging.  Countertransference by a therapist must be actively considered, examined, and dealt with when found to be present.

If an admission of guilt is required before being admitted into a program, additional complications and potential hindrances to successful treatment are generated.  There are no empirical data to demonstrate that a threshold admission of guilt has any relationship to outcomes.  It may, however, prevent both actual perpetrators and innocent people from being able to progress in resolution of their individual situations.  It may increase the likelihood of error in the justice system.

When an accused person who is actually innocent enters treatment with the hope of eventually having a relationship his children or of getting some benefit from therapy, it can be disastrous.  Successful completion of treatment is often defined by the requirement that the accused admit guilt.  It cannot be a general, bland admission, "Yes, I am an abuser," but often must be specific, detailed, given regularly in group, and may include an admission and apology to the victim.

Patton (1990) observes that if the court sustains a dependency petition, it can require that the parents cooperate in psychotherapy and can rule that failure to admit the abuse is sufficient evidence to warrant continued foster care or termination of parental rights.  He states, "There is really no dispute regarding the coercive nature of requiring parents to confess in court-ordered therapy as a condition for regaining child custody" (p.515).  At the same time, in almost all jurisdictions, the prosecutor in any pending criminal case has access to statements made during the therapy ordered by the family court.  Therefore, an accused person is put into a situation where if he remains silent in therapy, he may lose his child, but if he confesses the abuse, he provides the prosecutor with damaging evidence.  He must waive his fifth amendment rights in order to protect his fundamental right to have a relationship with his child.
  

Treatment and Recidivism

The most important goal of treatment for sex offenders is that they refrain from committing sex offenses in the future. This goal is more important than emotional health or adjustment, self-esteem, feelings of well-being, self-actualization, reported satisfaction with therapy, or improvement as measured by psychological tests.

Until recently there has been little evidence that treatment reduces recidivism. In a review of the research on the treatment of sexual abusers, Finkelhor (1986) concludes:

Unfortunately, the available studies tell us very little about what is perhaps the most important question: Does treatment reduce recidivism?... The recidivism rates for the treated groups are not consistently better than the nontreated groups ... So it cannot be said that ... (anybody's) recidivism study provides strong evidence in favor of the positive effects of treatment (pp.136-137).

A more recent review of sex offender recidivism (for all sex offenses, not just child sexual abuse) by Furby et al. (1989) reaches a similar conclusion.  These reviewers critically examined 42 empirical studies of sex offender recidivism and report that the recidivism rates ranged from 0% to over 50%.  They found little consensus about the continuance of sexual offenses following treatment and conclude that there is no evidence that clinical treatment effectively reduces recidivism.  Also, there are no data at present for assessing the relative effectiveness of treatment for different types of offenders.

However, the authors report that the methodological weaknesses and lack of uniformity in the recidivism studies make it difficult to discern patterns or draw conclusions from them.  In addition, they note that treatment models have been evolving constantly and many of those in the studies they reviewed are now considered obsolete.  Pithers and Cumming (1989) state that Furby et al. believe that outcome data from specialized treatment programs for sexual offenders will demonstrate therapeutic efficacy.  In fact, initial data on the Vermont cognitive-behavioral relapse prevention program (Pithers & Cummings, 1989, discussed below) is promising.

A recent study by Hanson, Steffy, & Gauthier (1990) on recidivism examined offenders from 3 to 23 years after treatment.  The treatment was a short-term, multimodal program and recidivism was assessed through records of reconvictions.  The researchers report that 44.3% of their total sample of 106 child molesters were reconvicted with 9.4% of the total sample being reconvicted between 10 and 23 years after being released.  Incest perpetrators were reconvicted at the slowest rate (21%), homosexual pedophiles at the highest rate (66.7%), with heterosexual pedophiles and undifferentiated offenders showing an intermediate rate (42.2% & 36.36%).  This study demonstrates the importance of extending the follow-up period when examining recidivism.

Sexual offenders now account for an average of 10% of the prison population of the United States, with some jurisdictions reporting rates as high as 21% (Borzecki & Wormith, 1987).  In 1988, sex offenders constituted the largest single group of inmates in Minnesota (Prince, 1988).  The difficulty in treating persistent and dangerous offenders against children is illustrated by Crawford (1981) who concludes that only castration has been found to be successful in preventing recidivism for this population.  Recidivism rates, although variable, suggest that incarceration alone is not sufficient.  The necessity to do something besides warehouse sex offenders in prison until they are released, unchanged, led people to conclude that treatment must be offered.  Therefore, treatment programs developed to fill this need.
  

Traditional Offenders Program

Most treatment programs for sex offenders insist at the onset that the perpetrator admit guilt as a condition of acceptance into the program.  Therefore, if an innocent person plea bargains and agrees to treatment, the person must admit guilt to be admitted into the program.  If guilt is not admitted, the probation is violated and the person may be sent to jail.  This requirement is widespread, although there are a few treatment programs which do not require this threshold admission of guilt (Blush & Ross, 1986; Brown, undated; Krop 1986; Lampel, 1986; Langevin, 1989; Simkins, Ward, Bowman, & Rinck, 1990; Underwager & Wakefield, undated).

Group therapy is seen as the most appropriate form of treatment in the United States (Borzecki & Wormith, 1987).  The rationale for this is the argument that sex offenders require group therapy because effective confrontation of manipulative behavior can only be done by other individuals who have been through the same dynamic.  It is assumed that all sexual abusers are skilled at manipulation and will demonstrate manipulativeness.  It is also assumed they must stop being manipulative.  Groups are seen as necessary and appropriate for all sexual offenders, regardless of their individual personalities and the factors underlying their abusive behavior.

The expression of feelings is absolutely required.  Cognitive, learning theory-based approaches therefore may be seen as a way of allowing the accused to avoid dealing with feelings of remorse, guilt, or shame which are considered to be essential parts of treatment.  Common treatment goals include bringing the perpetrator to the point where he admits all of his abusive behaviors, expresses guilt and remorse for them, and is willing to admit and apologize to the victim.  These are moralistic goals with no demonstrated relationship to outcomes.

The therapist must be authoritarian and allied with the justice system.  Sgroi is often cited to support the demand for therapy to be highly authoritarian.  Sgroi (1982) believes that effective treatment can only be accomplished in an authoritative fashion and from a position of power.  She believes that anything else invites the abuser to misuse power to suppress the allegation and undermine the child's credibility.  The necessary submission to authority in treatment is demonstrated by compliance with the demands of the program for completion of all assignments, attendance, and sobriety.  This usually includes meeting stages of performance that show progress through the treatment.

There is no evidence for the effectiveness of this type of treatment to cure sexual abuse and prevent recidivism.  There is no support for the assertion that such therapy is the only right way to treat sex offenders.  Nevertheless, this approach is the type of intervention generally insisted upon by the system.

Quinsey, in 1977, commented on the research supporting the use of group therapy for child molesters:

Group therapy remains the most widely used treatment for child molesters.  However ... the therapy approaches described in the literature appear to be based on contradictory premises.  Furthermore, few data have been reported to indicate that changes occur within these groups, and no studies have been conducted that compare group therapy to other types of treatment.  An additional difficulty is that the description of the treatment method itself in these studies is at such a general level that replication of them would appear to be impossible (p. 213).

This conclusion remains true today.  There is no empirical support for the belief that groups are more effective in confronting attempted deception or manipulation.  A study by McCaghy (1967) showed that child abusers in therapy readily adopt the language and rhetoric of the therapist to describe and account for their abuse.  Those that were in many therapy sessions changed their explanations of their behavior to include descriptions of their early childhood, exposure of personal weaknesses, and use of mental health terminology.  Often child abusers report the sexual preferences they know therapists want to hear in order to obtain an early release or to meet the therapist's expectations.

Marcus (1970) and Marcus & Conway (1971) report on a group therapy treatment program they ran in the Canadian prison system.  They state that groups construct a dynamic defense against the therapist to prevent knowledge and identification of group members who deteriorate or really don't change.

Incarceration is almost uniformly discouraged as a form of treatment for the perpetrator in cases of intrafamilial child sexual abuse (Costell, 1980; Quinsey, 1977; Giarretto, 1976).  Recurrence of incestuous activity is unlikely after disclosure (Cormier, Kennedy, & Sangowicz, 1962; Lang, Pugh, & Langevin, 1988; Pithers & Cumming, 1989).  On the other hand, several authors (Fitch, 1962; Hanson, Steffy, & Gauthier, 1990; Lang et al., 1988; Mohr, Turner, & Jerry, 1964; Quinsey, 1977) report recidivism to be higher for homosexual offenders and pedophiles (usually extrafamilial abuse).  Group therapy is not effective.  Langevin (1983) states:

Collectively group therapy studies of pedophilia has (sic) been poorly delineated without reference to the direction of treatment or theoretical characteristics of the pedophile which are the targets of treatment.  Follow ups were short and assessments so general that the effectiveness of this procedure is uncertain.  The poor outcome of group therapy with exhibitionism (Chapter 10) which could be traced might serve as a guide to the use of this approach (p. 292).

There is a lack of credible research on the treatment for child sexual abusers (Borzecki & Wormith, 1987; Furby et al., 1989).  There are few reports that compare different treatments, use control groups, have adequate outcome measures, or include follow-up data.  Many are nothing more than case reports that do not even follow accepted single case design standards.  The case studies are unsystematic, uncontrolled and so confounded that no variables can be seen to be operative in treatment outcomes.  There is no report that meets all of the criteria for well-designed and scientifically credible research.
  

Treatment Directions With Research Support

The therapy modalities that have been used in treating child sexual abusers include behavior therapy with many classical and operant conditioning techniques, hypnotism, psychoanalysis, traditional talking psychotherapy, group therapy, chemical interventions, castration, electroconvulsive therapy (ECT), and psychosurgery.  Langevin (1983) reviews and evaluates each of these modalities.  Although aware of the criticisms that have been leveled against behavior therapy, he believes that the principles of behavior therapy should be followed.  He also recommends using the assessment methods developed by behavior therapists.

Langevin and Lang (1985) maintain that "sexual preference is a powerful and persistent feature of human behavior and there is no evidence that therapy in any form can change it" (p.409).  Therefore the goal of therapy must be to help the pedophile manage his urges for sexual contact with children.  A key factor in the success in any treatment of pedophiles will be motivating them to change; most pedophiles are resistant to giving up a sexual behavior pattern which they perceive as positive and rewarding.  Pedophiles often initially deny, minimize, or rationalize their abusive behaviors and the first goal of treatment must be to get the offenders to admit their past behaviors and overcome their rationalizations.

We recommend an individually-tailored approach that includes careful assessment of the situation along with the capacities, personality, and behaviors of the individual and a therapy program that uses a broad mix of learning theory-based treatment techniques to support individual behavior change.  Different treatment interventions must be planned for different types of child molesters.  Such an approach has the best research support.

The implications of the research lead Quinsey (1977) to say "... treatment programs should be individualized" (p.216). Lampel (1986) reports on the success of individualized treatment approaches.  Giarretto (1976) emphasizes individualized treatment.  Dixen and Jenkins (1981) recommend an individualized multi-component therapy approach.  Langevin (1983) sees individualized behavior therapy techniques as the treatment of choice.  Borzecki and Wormith (1987) state that "Individually tailored treatment is commonly lacking in the voluntary American programmes; here is an obvious source of concern, since treatment needs vary tremendously across offenders" (pp. 34-45).

Although there are little outcome data on treatment for sexual offenders, the approach that is most supported by what data are available is cognitive-behavioral (Anderson & Shafer, 1979; Dixen & Jenkins, 1981; Langevin, 1983; Quinsey, 1977).  A large number of specific techniques and methods are included in the therapy possibilities that learning theory and a cognitive-behavioral strategy generate.  The therapist can construct a highly individualized and flexible treatment approach which can be changed and refined as treatment progresses.  An important component of a behavioral therapy is social skills training to redress the weakness and inadequacies of child molesters in adult interactions (Dwyer & Amberson, 1985; Langevin, 1983; Overholser & Beck, 1986; Quinsey, 1977).  Dwyer (1990) reports a significant reduction in paraphilic fantasies following a treatment approach which includes a broad range of techniques including cognitive restructuring, social skills training, psychodynamic treatment, and family therapy.  Annon (1989) reports a 85% to 95% success rate with an individualized treatment approach which uses behavioral/cognitive methods.

While not directly dealing with child sexual abusers, there is good research dealing with rehabilitation of offenders generally.  For some time rehabilitation has been seen negatively because the idea spread that it didn't work.  Now there is evidence about the factors present when therapy doesn't work and factors present when it does work.  This knowledge is applicable to the treatment for sexual abusers.

Lipton (1986) identified recurrent problems that can lead to the failure of any rehabilitation effort.  They include 1) hostility to change, 2) a coercive correctional system, 3) lack of any theoretical base for the treatment program, 4) failure to implement the program fully, and 5) inability to relate to the world beyond the institution.  Gendreau (1986), after a decade of research, adds that unsuccessful programs use approaches that are inappropriate for the offender, rigid, imposed from the top down, and use only negative reinforcers.  He described what makes a rehabilitation program work:

Effective programs tend to follow a social learning, cognitive behavior theory type of approach, as opposed to a psychodynamic model. ... But they are more flexible and less mechanistic than early behavior modification and contingency management approaches ... They maintain authority not by bashing heads, but by setting limits and enforcing probation orders and other rules.  They adopt a problem solving approach, with positive modeling, and make extensive use of community resources.  They build on the quality of interpersonal relationships, and they try to mediate between the needs of the client and what exists in the real world (p.14).

More recently, Gendreau reports that his research indicates "appropriate" treatment programs reduce recidivism rates by 53%.  The most successful programs are those that employ behavioral modification techniques that reward pro-social behavior and target antisocial attitudes and values that fuel criminal behavior.  Effective rehabilitation programs teach offenders skills they can use to keep from reoffending (reported by Freiberg, 1990).

There is sufficient research to conclude that the most effective treatment for child sexual abusers is individualized, uses cognitive-behavioral techniques, and is adaptive and flexible.  While more research is needed, the clinician who must provide treatment can do a better job by following these directions.

Treatment for persons accused of child sexual abuse must also consider the situation of an innocent person who is accused.  Although denial, minimization, and rationalization are found in actual sex offenders, there is always the possibility that a person maintaining denial is innocent.  This is apt to be more likely when there is a plea bargain or no adjudication.  However, even people who continue to deny abuse after being criminally convicted may, in fact, be innocent.  There is no way to know how often this happens, but it does happen, and clinicians providing treatment should not automatically assume that all persons who deny abuse are unmotivated and defensive abusers.
  

Relapse Prevention

Relapse prevention is a self-control program designed to teach individuals who are trying to change their behavior how to anticipate and cope with the problem of relapse.  It developed within the area of addictive disorders but has been expanded to sex offenders.  It is based on social learning theory and combines behavioral and cognitive interventions.  There is an emphasis on self-management.  It is not an isolated treatment; relapse prevention was developed as a maintenance strategy and is intended to preserve gains in whatever treatment preceded it (George & Marlatt, 1989; Laws, 1989).

The relapse prevention program is individually developed following a careful assessment of the individual.  Cognitive-behavioral techniques such as aversive conditioning, cognitive restructuring, thought stopping, covert sensitization, satiation, contracts, covert reinforcement, modeling, role playing, social skills training, and relaxation training are used.  Offenders learn to identify and anticipate high risk situations, control their urges, develop more effective coping skills, maintain a more balanced lifestyle and gain a sense of control and self-efficacy.  Through this process, it is hoped that they will be less apt to relapse and recommit a sexual offense.

Three treatment programs using this approach are described in Laws' book, Relapse Prevention With Sex Offenders (1989).  These programs are The Sex Offender Treatment and Evaluation Project at Atascadero State Hospital in California, The Center for Prevention of Child Molestation at the Florida Mental Health Institute in Tampa, and The Vermont Treatment Program for Sexual Aggressors.  Laws points out that these programs are "highly unrepresentative" and their generalizability is limited.  However, they do represent what can be done with adequate resources and heavy reliance on the concepts of relapse prevention.

Only the Vermont program (Pithers & Cumming, 1989) reports follow-up data as the other programs are too new.  On a six-year follow-up on recidivism for 167 offenders who attended the program in 1982, there was a 4% recidivism rate for the group as a whole.  This differed for rapist and pedophiles, with a rate of 3% for pedophiles (4 out of 147) and 15% for rapists (3 out of 20).

Pithers and Cumming then compared this to the recidivism on a five-year follow-up on sex offenders who were treated with a standard peer-group milieu therapy at Atascadero Hospital (this was the old treatment program, before relapse prevention was instituted).  There was no significant difference in recidivism between the Atascadero rapists and the Vermont rapists (26% versus 15%), but there was a highly significant difference for the pedophiles.  While only 3% of the Vermont pedophiles relapsed, 18% of the Atascadero pedophiles reoffended.

These results suggest that an individually tailored cognitive-behavioral treatment approach for child sexual abusers has the best chance of succeeding in reducing recidivism.  The relapse prevention model appears to be a promising treatment approach, although more data are needed in order to generalize the results of this outcome study.
  

Conclusions

There is a need for effective treatment programs for those who sexually abuse children.  Many commonly used treatment programs are not supported by research evidence and have no demonstrated efficacy.  There is sufficient research to have confidence in programs that are individualized, cognitive-behavioral, flexible, and related to the real situation of the abuser.
  

References

Anderson, L. M., & Shafer, 0. (1979). The character-disordered family: A community treatment model for family sexual abuse. American Journal of Orthopsychiatry, 49, 436-445.

Annon, J. S. (1989, April). Assessing and treating sex offenders: Training and videotape demonstration. Presentation made at the Fifth Annual Symposium in Forensic Psychology of the American College of Forensic Psychology, San Diego.

Borzecki, M., & Wormith, J. S. (1987). A survey of treatment programmes for sex offenders in North America. Canadian Psychology, 28(1), 30-44.

Blush, G. J., & Ross, K. L. (1986). Sexual allegations in divorce: The SAID syndrome. Conciliation Courts Review, 25(1). Also personal communication.

Brown, J. (undated). Treatment program description.

Cormier, B. M., Kennedy, M., & Sangowicz, J. M. (1%2). Psychodynamics of father-daughter incest. Canadian Psychiatric Association Journal, 7, 203-217.

Costell, R. M. (1980). The nature and treatment of male sex offenders. In National Center on Child Abuse, Sexual abuse of children: Selected readings. (DHHS Publication No. 78-3061.) Washington DC: Government Printing Office.

Crawford, D. A. (1981). Treatment approaches with pedophiles. In M. Cook & K. Howells (Eds.), Adult Sexual Interest in Children () (pp. 181-250).

Deatherage, E. (1990). Personal communication.

Dixen, J., & Jenkins, J. O. (1981). Incestuous child sexual abuse: A review of treatment strategies. Clinical Psychology Review, 12, 211-222.

Dwyer, S. M. (1990). Reduction of sex offender paraphilic fantasies: 6 month and 1 year follow-up. Journal of Psychology & Human Sexuality, 3(1), 57-65.

Dwyer, S. M., & Amberson, J. I. (1985). Sex offender treatment program: A follow-up study. The American Journal of Social Psychiatry, 4(fall), 56-60.

Finkelhor, D. (1986). A Sourcebook on Child Sexual Abuse ()(). Beverly Hills, CA: Sage Publications.

Fitch, J. H. (1962). Men convicted of sexual offences against children. British Journal of Criminology, 3, 18-27.

Freiberg, P. (1990, September). Rehabilitation is effective if done well, studies say. The APA Monitor, 21(9), pp. 1 & 17.

Furby, L., Weinrott, M. R., & Blackshaw, L. (1989). Sex offender recidivism: A review, Psychological Bulletin, 105, 30-30.

Gendreau, P. (1986). Paper presented at the National Center for Institutions and Alternatives Conference. June, 1986. Reported in J. Marvis, Rehabilitation: Can it work now? The APA Monitor, 17, September, 1986, p.14.

George, W. H., & Marlatt, GA. (1989). Introduction. In D. R. Laws (Ed.). Relapse Prevention with Sex Offenders () (pp.1-31). New York: The Guilford Press.

Giarretto, H. (1976). Humanistic treatment of father-daughter incest. In E. Ray & C. H. Kempe (Eds.), Child Abuse and Neglect: The Family and the Community () () (pp. 143-158). Cambridge, MA: Ballinger Publishing Company.

Hanson, R. K., Steffy, R. A., & Gauthier, R. (1990, August 13). Recidivism of child molesters 20 years (4ter treatment. Presented at the 98th Annual Convention of the American Psychological Association at Boston.

Krop, H. (1986). Personal communication.

Lampel. A. K. (1986). Post-divorce therapy with highly conflicted families. The Independent Practitioner, 6, 22-26.

Lang, R. A., Pugh, G. M., & Langevin, R. (1988). Treatment of incest and pedophilic offenders: A pilot study. Behavioral Sciences & the Law, 6, 239-255.

Langevin, R. (1983). Sexual Strands: Understanding and Treating Sexual Anomalies in Men (). Hillsdale, NJ: Lawrence Erlbaum Associates.

Langevin, R., & Lang, R. A. (1985). Psychological treatment of pedophiles. Behavioral Sciences & the Law, 3, 403A19.

Laws, D. R. (Ed.) (1989). Relapse Prevention with Sex Offenders (). New York: The Guilford Press.

Lieberman, M. A., Yalom, I. D., & Miles, M. B. (1973). Encounter Groups: First Facts (). New York: Basic Books.

Lipton, D. (1986). Paper presented at the National Center for Institutions and Alternatives Conference. June, 1986. Reported in J. Marvis, Rehabilitation: Can it work now? The APA Monitor, 17, September, 1986, p. 14.

Marcus, A. M. (1970). Encounters with the dangerous sexual offender. Canada's Mental Health, 18, 9-14.

Marcus, A. M., & Conway, C. (1971). A Canadian group approach study of dangerous sexual offenders. International Journal of Offender Therapy, 5, 59-66.

McCaghy, C. H. (1967). Child molesters: A study of their careers as deviants. In M. B. Clinard & R. Quinsey (Eds.). Criminal Behavior Systems: A Typology () (pp.143-161). New York: Holt, Rinehart, & Winston.

Mohr, T. W., Turner, R. E., & Jerry, M. B. (1964). Pedophilia and Exhibitionism. Toronto: University of Toronto Press.

Overholser, J. C., & Beck, S. (1986). Multimethod assessment of rapists, child molesters, and three control groups on behavioral and psychological measures. Journal of Consulting and Clinical Psychology, 54, 682-687.

Patton, W. W. (1990). The world where parallel lines converge: The privilege against self-incrimination in concurrent civil and criminal child abuse proceedings. Georgia Law Review, 24(3), 473-524.

Pithers, W. D., & Cumming, G. F. (1989). Can relapses be prevented? Initial outcome data from the Vermont treatment program for sexual aggressors. In D. R. Laws (Ed.), Relapse Prevention with Sex Offenders () (pp. 313-325). New York: The Guilford Press.

Price, P. (1988, August 11). Sex offenders need tougher sentences, experts tell panel. Minneapolis Star Tribune, p. 5B.

Quinsey. V. L. (1977). The assessment and treatment of child molesters: A review. Canadian Psychological Review, 18, 204-220.

Siegler, M., & Osmond, H. (1974). Models of Madness, Models of Medicine ()(). New York: Macmillan Publishing Company.

Simkins, L., Ward, W., Bowman, S., & Rinck, C. M. (1990). Characteristics predictive of child sex abusers' response to treatment: An exploratory study. Journal of Psychology & Human Sexuality, 3(1), 19-55.

Sgroi, S. M. (1982). Handbook of Clinical Intervention in Child Sexual Abuse (). Lexington, MA: Lexington Books.

Underwager, R., & Wakefield, H. (undated). Treatment program description.

Zack, M. (1990, January 23). Rapist who confessed in treatment convicted. Minneapolis Star Tribune.