Skip to content
Search resources Open menu

Online Offense

Treatment and Prevention for Online Offenders

Published

Snap Shot

  • The question: What does research show about treatment for people convicted of online sexual offenses and about prevention programs for people with sexual interest in children, and how does the law use that evidence?
  • What the research shows: For people convicted of sexual offenses generally, specialized cognitive-behavioral treatment is associated with a modest reduction in sexual recidivism, concentrated in medium- and high-risk men. Prevention programs show that undetected people with sexual interest in children will seek help, and two randomized trials of an anonymous online program found short-term, self-reported reductions in CSAM viewing time or in sexual urges toward children.
  • What it does not show: No controlled study shows that any program lowers official recidivism for people convicted of CSAM-only offenses, whose officially detected contact recidivism was 0.6% over about 4.7 years in one meta-analysis (self-reported CSAM recidivism was 38.3% in four samples, and detection undercounts). No study can say whether a particular person will benefit from treatment.
  • The law: Federal courts may not impose or lengthen prison to provide treatment (Tapia), but rehabilitation remains central to probation, supervised release, and conditions. Courts-martial weigh rehabilitative potential under R.C.M. 1001(b)(5) and 1002(c). In Arizona, amenability is argued under § 13-701(E)(6) and matters most where probation is available.
  • Issues for each side: Prosecutors can show that enrollment in treatment does not establish that offending has stopped, without overstating risk; defense counsel can present documented engagement as amenability, without claiming that any program is proven to work for this group.

At almost every sentencing for an online sexual offense, the court asks whether treatment works, whether this defendant will benefit, and whether the sentence should be built around it. Treatment also shapes supervision conditions, plea negotiations, and, in the military, testimony about rehabilitative potential. A newer question concerns prevention: programs that help people with sexual interest in children before they offend or are detected, and messages that try to deter searches for child sexual abuse material (CSAM).

Statutes and courts use the term “child pornography,” and this paper uses that term when describing the law; elsewhere it uses CSAM. The paper addresses adults. Treatment of adolescents is discussed in the hub’s paper on women and adolescents who commit online sexual offenses.

The general research on whether sex offender treatment reduces recidivism is covered in depth in the Psychosexual Risk Evaluations resource on this site, as are medication and the treatment programs available in military confinement. This paper summarizes that evidence briefly and concentrates on what is specific to online offending, on prevention, and on the law governing how treatment evidence is used.

What Treatment for Online Offending Involves

Most treatment for sexual offending follows the risk-need-responsivity framework. The risk principle directs the most intensive services to the people at highest risk; the need principle directs treatment at dynamic risk factors that are related to reoffending; and the responsivity principle calls for methods suited to the person, which generally means concrete, skills-based cognitive-behavioral work. In a meta-analysis of 23 studies, mostly of weak quality, programs that followed all three principles showed the largest effects (Hanson et al., 2009).

Seto (2025, Chapter 7) describes the programs developed for online offending as sharing a common core: the pathway to offending, emotional and sexual self-regulation, self-regulation of internet use, and social and relationship skills. They differ mainly in intensity, from about ten sessions to weekly sessions for a year or more. He recommends separate streams for people with online-only and contact offenses and notes that no one has yet tested whether the framework improves outcomes for online offending.

The General Outcome Evidence in Brief

In 29 comparisons judged to be of sound quality, sexual recidivism was 10.1% for treated men and 13.7% for untreated men, with larger effects for cognitive-behavioral programs and for medium- and high-risk men, no significant effect for low-risk men, and too little evidence for conclusions about prison programs (Schmucker & Lösel, 2015, 2017). A meta-analysis of 70 studies and 55,604 individuals across several kinds of offending programs found a 32.6% relative reduction in offense-specific recidivism for sexual offense programs; across program types, results were better where a qualified psychologist was consistently involved and staff received clinical supervision (Gannon et al., 2019). An updated meta-analysis of 37 samples (N = 30,394) found a mean effect that the authors described as small (odds ratio 1.54), larger in high- and medium-risk men, larger in specialized treatment, and larger when the study authors were affiliated with the program than when they were independent (Holper et al., 2024). A companion analysis using multimodel inference on 35 studies found that only risk level moderated the treatment effect (Holper et al., 2025).

That literature supports a real but modest effect for treatment matched to risk, but it is drawn mostly from men with contact offenses and observational designs and cannot simply be applied to people convicted of CSAM-only offenses.

Programs for People Convicted of CSAM Offenses

The evidence specific to online offending is thin. A systematic review searched through October 2024 found only nine studies of interventions for people who had committed online sexual offenses involving children: one randomized trial, two non-randomized controlled studies, three qualitative studies, and three descriptive studies, with samples ranging from 1 to 264, all from Europe (Gouveia et al., 2026). No meta-analysis was possible. Five of the six studies that reported reoffending relied on self-report, and dropout ranged from 0% to 80%.

The best-known program is the United Kingdom’s Internet Sex Offender Treatment Programme (i-SOTP), delivered in probation. In a pre-post evaluation of 264 men who completed it, participants improved on most psychological measures, with no comparison group or recidivism outcome (Middleton et al., 2009). Seto (2025, Chapter 7) adds that several measures were not related to sexual recidivism and that about half of the participants with risk scores were low risk, which is not consistent with the risk principle. Its successor, iHorizon, showed large pre-post changes in 92 participants in a 2023 analysis Seto summarizes, again without a comparison group. Inform Plus, a brief psychoeducational program for men under investigation, showed self-reported improvements in distress, self-esteem, and distorted attitudes in 92 men, without a control group or recidivism outcome, in an evaluation by authors affiliated with the program (Gillespie et al., 2018).

The longest follow-up comes from a study of 690 men who had taken part in a community intervention in the United Kingdom for indecent-image offenses, followed for an average of 13 years (Elliott et al., 2019). Sexual reconviction was 12.6% and any reconviction 24.8%; 2.7% of the image-only group were convicted of a later contact offense. Reliable change on the treatment domains did not predict reconviction, and there was no untreated comparison group. The findings cut both ways.

The Low Base Rate and the Risk Principle

Low detected base rates are the central obstacle to evaluating treatment for this group. In a meta-analysis of 30 samples (N = 25,978), officially detected contact sexual recidivism was 0.6% for men with CSAM-exclusive offending, over an average follow-up of about 4.7 years, compared with 8.8% for men with mixed CSAM and contact offending, while self-reported CSAM recidivism was 38.3% (Baskurt et al., 2025). Those figures are discussed in the hub’s paper on recidivism. A program cannot show a large reduction in detected contact offending when that outcome is rare in the samples studied; these are detected rates and minimum estimates.

The low detected base rate also raises the possibility that too much treatment can do harm. Henshaw and colleagues suggested, on the basis of low detected base rates, that many people with CSAM-exclusive offending may not require intensive treatment; that inference depends on detected rates, which undercount, and individual assessment is needed (Henshaw et al., 2020, as cited in Seto, 2025, Chapter 7). In an evaluation of the accredited prison program in England and Wales, treated men had more proven child-image reoffending than matched comparison men (4.4% vs. 2.9%); the figures describe the whole sample, of which only 116 treated men had an image index offense, the design was not randomized, and Seto considers treatment mismatched to risk one possible explanation (Mews et al., 2017, as cited in Seto, 2025, Chapter 7). Seto also notes the opposing consideration: some men with CSAM offenses disclose undetected contact offending, the proportion varies widely by sample and method, and treatment may reduce future undetected offending. The risk principle supports intensive treatment for people at higher assessed risk and lighter or no treatment at the lowest risk, not the assumption that more treatment is always better.

Prevention for People Who Have Not Been Detected

Prevention programs reach people with sexual interest in children who have not offended or have not been caught. The best known is Germany’s Prevention Project Dunkelfeld, which offers confidential treatment to self-referred people and operates under narrow mandatory-reporting rules. Its pilot evaluation compared 53 treated men with 22 men on a waiting list, in a non-randomized design, and reported improvement on some self-reported dynamic risk factors (Beier et al., 2015). None of the reported offending was known to authorities. During therapy, 5 of 25 men with past contact offenses and 29 of 32 men with past CSAM use reported continuing the behavior. A reanalysis comparing change across the two groups found a median effect size of 0.30 on 14 indicators, none statistically significant; the authors noted that the study was underpowered and that an effect could not be ruled out (Mokros & Banse, 2019).

A long-term follow-up reached 56 of 110 eligible men who had completed the program, 1 to 11 years later, with no control group (Beier et al., 2024). Of the men with past contact offenses, 2 (7.7%) reported reoffending. CSAM use fell from 82.1% before treatment to 62.5% after treatment and was 75% at follow-up; among prior users, 89.1% reported using CSAM again, although generally less severe material. A critical commentary noted that 85.7% of the follow-up sample self-reported sexual delinquent behavior, that only one man reported legal consequences, that anonymity prevents checking records, and that iatrogenic effects could not be excluded (König, 2025). A related German study of 165 help-seekers is discussed in the hub’s paper on pedophilia and the diagnostic manuals. In a follow-on study of 131 men with past CSAM use in that voluntary treatment network, only the CASIC predicted self-reported CSAM reuse; the CPORT without the CASIC did not (Pellowski et al., 2026). The study was exploratory, partly scored retrospectively, and conducted by authors employed in the network. Help-seekers also include minors: among 117 who sought help for CSAM use, 97% of them undetected, about half reported a sexual preference for children or early adolescents, in a self-selected, self-reported sample with no comparison group (Westfal, 2026). Adolescents are discussed in the hub’s paper on women and adolescents.

Online tools extend reach: 4,161 users completed the Troubled Desire self-assessment from 2017 to 2020, 11.4% from the United States, and 78.9% reported sexual interest in children (Schuler et al., 2021). The study shows reach, not effectiveness.

The first randomized trial in this area tested Prevent It, an anonymous, therapist-guided internet program of cognitive-behavioral therapy for people already using CSAM, recruited mainly on the darknet (Lätth et al., 2022). The 160 participants were randomized equally to the program or to a psychological placebo. The program produced a significantly larger reduction in self-reported weekly CSAM viewing time, a small effect (d = 0.18). Follow-up was one month, and attrition was heavy: Gouveia et al. (2026) report 29% dropout during treatment and 80% at follow-up, and Seto (2025, Chapter 8), who discloses involvement in evaluating a revised version, notes that almost half the treatment group was lost in the first two sessions.

A second trial tested that revised 9-week program, Prevent It 2.0, in 203 anonymous adults recruited internationally, 92.1% of them men (McMahan et al., 2026). Compared with an attention-controlled waitlist, participants reported a larger reduction in sexual urges toward children (d = −1.01) and about 1.9 fewer hours per week spent viewing children for sexual purposes, including CSAM use (d = −0.52). Because the comparison was a waitlist rather than a placebo treatment, the trial cannot separate the program’s content from the general effects of receiving treatment. All outcomes were self-reported, the endpoint was nine weeks, and no outcome was checked against official records.

Warning Messages and Search Deterrence

In two randomized experiments with a “honeypot” website that purported to offer “barely legal” pornography, used as a legal proxy for CSAM, warning messages reduced attempted click-through. In the first (n = 419), the most effective messages warned that IP addresses can be traced (odds ratio 2.64) and that the material may be illegal (odds ratio 2.99) (Prichard et al., 2022). In the second (n = 474), three quarters of the control group attempted to enter, compared with 35% to 47% of the groups shown deterrent or therapeutic messages (Prichard et al., 2024). The outcome was a single click on legal material; the studies show that brief messages can change immediate behavior, not that they reduce offending.

Medication in Brief

A Cochrane review found seven trials with 138 participants, all more than 20 years old, and concluded that the evidence was of poor quality and did not support firm conclusions (Khan et al., 2015). Since then, a double-blind trial in 52 help-seeking men with pedophilic disorder in Sweden found that the testosterone-lowering drug degarelix reduced a composite risk score more than placebo at two weeks (Landgren et al., 2020). In a secondary report, 15 of 26 men given degarelix (58%) and 3 of 26 given placebo (12%) reported no further sexual interest in children at 10 weeks (Landgren et al., 2022). The outcomes were a proxy score and self-reports over 10 weeks, not offending, and adverse effects were common. The authors of a fluoxetine trial now under way state that no randomized trial of antidepressants of that class for sexual interest in children existed before theirs (Adebahr et al., 2024). Medication is discussed further in the Psychosexual Risk Evaluations resource.

Amenability, Readiness, and Engagement

Amenability to treatment is a clinical judgment, not a test result. The multifactor offender readiness model ties readiness to factors within the person, such as motivation, beliefs about treatment, and emotional and cognitive capacities, and to the setting and program (Ward et al., 2004). The model contains no outcome data. Seto (2025, Chapter 7) notes that motivation ranges from genuine to reluctant and that people who drop out or are removed from programs tend to be at higher risk than completers, regardless of any treatment effect.

An evaluator can describe a person’s assessed risk level and treatment needs, documented engagement and attendance, whether the person accepts responsibility, and whether an appropriate program is available. An evaluator cannot say, on the basis of research, that a particular person will benefit from treatment or that completed treatment has reduced that person’s risk; measured treatment change did not predict reconviction in Elliott et al. (2019).

Limits of the Evidence

Most outcome studies rely on self-report, which anonymous prevention programs cannot check against records. Help-seeking samples differ from people prosecuted in U.S. courts, and nearly all online-specific studies come from Europe; no published outcome evaluation of a U.S. program for people convicted of online offenses was located for this paper. Pre-post designs cannot separate treatment effects from the passage of time, regression to the mean, or the loss of participants who drop out. Effects are larger when evaluators are affiliated with the program (Holper et al., 2024), and follow-up is short in the randomized trials. Seto, whose book is cited here, is a co-author of the 2026 systematic review and of the Prevent It 2.0 trial, and discloses his involvement in that evaluation. No study can predict how a particular person will respond to treatment, and group findings cannot establish that a particular defendant is or is not dangerous.

Questions to Ask About This Research

When the 10.1% versus 13.7% finding (Schmucker & Lösel, 2015) is offered to show that treatment will help a defendant convicted of a CSAM offense, counsel can ask who was studied, since the samples were mostly men with contact offenses and the effect was not significant for low-risk men. When the Prevent It trials (Lätth et al., 2022; McMahan et al., 2026) are cited to show that treatment works for CSAM use, the questions are what was counted and against what: self-reports over one to two months, in anonymous non-forensic samples, against a placebo in one trial and a waitlist in the other. When the 89.1% figure from the Dunkelfeld follow-up (Beier et al., 2024) is offered to show that treatment fails, it is fair to ask whether a study with no control group can show what would have happened without treatment. When the 0.6% contact recidivism rate (Baskurt et al., 2025) is cited to argue that treatment is unnecessary, counsel can ask what it omits: undetected offenses. These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.

Treatment, Prevention, and the Law

Federal Standards

Under 18 U.S.C. § 3553(a)(2)(D), the sentence must provide needed medical care or other correctional treatment in the most effective manner, but § 3582(a) directs courts to recognize that imprisonment is not an appropriate means of promoting correction and rehabilitation. In Tapia v. United States (2011), the Supreme Court held that § 3582(a) bars a court from imposing or lengthening a prison term to promote rehabilitation or to allow the defendant to complete a treatment program. A court may still recommend a program to the Bureau of Prisons, and rehabilitation remains proper in deciding on probation or supervised release and setting conditions. The Court did not decide whether a court may impose less prison to promote rehabilitation. In Pepper v. United States (2011), the Court held that at resentencing a court may consider postsentencing rehabilitation, which may support a downward variance, although a court need not reduce a sentence because of it.

Courts of appeals look to the reasons given. In United States v. Smith (8th Cir. 2022), a supervised-release revocation case, the court found no plain Tapia error where the statutory-maximum term rested on the seriousness of the violations and the danger to children rather than on rehabilitation. In United States v. Crum (2d Cir. 2021), a nonprecedential summary order in a child pornography case, references to treatment supported a facility recommendation rather than the sentence length.

Probation is barred for Class A and B felonies and where a statute precludes it (18 U.S.C. § 3561(a)). Receipt and distribution of child pornography carry five-year mandatory minimum prison terms, so probation is not available; possession, which has no mandatory minimum for a first offense and a 10- or 20-year maximum, is a Class C felony under § 3559(a)(3) and is not statutorily excluded. Treatment conditions are authorized for probation and supervised release (§§ 3563(b)(9), 3583(d)). The 2025 Guidelines recommend, when the offense of conviction is a sex offense, a condition requiring participation in a probation-approved program “for the treatment and monitoring of sex offenders” (U.S.S.G. §§ 5B1.3(d)(7)(A), 5D1.3(b)(3)(G)(i)). The tailoring of supervision conditions is discussed in the hub’s paper on supervision conditions.

Treatment Disclosures and the Fifth Amendment

Treatment programs often require a full sexual history. A probationer’s general duty to answer truthfully does not by itself make admissions compelled (Minnesota v. Murphy, 1984), but in United States v. Antelope (9th Cir. 2005), revoking probation and supervised release because the defendant refused to make sexual-history disclosures without immunity violated the Fifth Amendment. In United States v. Bahr (9th Cir. 2013), the court held that using unconstitutionally compelled statements from a state-ordered treatment program to determine a sentence in a later, unrelated federal proceeding violated the Fifth Amendment, and it vacated the sentence. These cases are discussed further in the hub’s paper on the polygraph.

Military Standards

Under R.C.M. 1001(b)(5), trial counsel may present opinion evidence of the accused’s rehabilitative potential, defined as the potential to be restored, through vocational, correctional, or therapeutic training or other corrective measures, to a useful and constructive place in society. The opinion must be rationally based and relate to the accused’s personal circumstances; the witness’s view of the severity or nature of the offense may not be its principal basis; and the witness may not opine on whether a punitive discharge is appropriate or whether the accused should return to the unit. Under R.C.M. 1002(c), the court-martial must impose punishment that is sufficient, but not greater than necessary, considering among other things the need to protect others from further crimes by the accused, to rehabilitate the accused, and, in appropriate cases, to provide the opportunity for retraining and return to duty. These rules are cited from the 2024 Manual for Courts-Martial; Executive Order 14130 (Dec. 20, 2024), which postdates it, did not amend R.C.M. 1001(b)(5) or 1002(c) (it revised only the concurrency rules in R.C.M. 1002(b)(2)).

As later opinions describe them, United States v. Horner (C.M.A. 1986) requires that a rehabilitative-potential opinion rest on an assessment of the accused’s character and potential rather than on the severity of the offense, and United States v. Ohrt (C.M.A. 1989) held that recommending a punitive discharge, including through euphemisms such as “He should be separated,” is error. United States v. Williams (C.M.A. 1994) treated evidence of future dangerousness as a proper matter under R.C.M. 1001(b)(5), as United States v. George (C.A.A.F. 2000) later applied it. In United States v. Griggs (C.A.A.F. 2005), the court held that the limits in R.C.M. 1001(b)(5)(D) do not apply to defense evidence offered in mitigation, although an accused who opens the door may face rebuttal. Under Mil. R. Evid. 513(d)(7), when the accused offers evidence of mental condition in defense, extenuation, or mitigation, the military judge may order disclosure of the accused’s statements to a psychotherapist as necessary in the interests of justice. Treatment in military confinement is discussed in the Psychosexual Risk Evaluations resource.

Arizona

Arizona’s mitigating circumstances include “any other factor that is relevant to the defendant’s character or background or to the nature or circumstances of the crime and that the court finds to be mitigating” (A.R.S. § 13-701(E)(6)), the catch-all provision under which amenability to treatment is argued. The court may order a presentence mental health examination or diagnostic evaluation (Ariz. R. Crim. P. 26.5), and probation conditions may promote rehabilitation (Ariz. R. Crim. P. 27.1).

Amenability evidence matters most where probation is possible. A completed sexual exploitation offense involving a minor under 15 is a first-degree dangerous crime against children sentenced under § 13-705(F), and § 13-705(K) makes a person so sentenced ineligible for probation. A preparatory offense is a second-degree dangerous crime against children (§ 13-705(R)), a class 3 felony under § 13-705(M) for which a prison term is prescribed only “if the person is sentenced to a term of imprisonment,” and luring under § 13-705(G) is written the same way for a first offense; § 13-705(K) excepts both subsections from its probation bar, so § 13-705 alone does not foreclose probation for those offenses. Offense-specific statutes may: § 13-3554 (luring) and § 13-1428(B) (sexual extortion) each state that the convicted person is not eligible for suspension of sentence, probation, pardon, or release from confinement except as § 31-233(A) and (B) provide. Pleas to preparatory offenses are therefore the clearer place where amenability evidence can bear on probation; aggravated luring (§ 13-3560) was not checked, and Arizona counsel should confirm eligibility for any particular charge. Where probation is available, probation for chapter 14 and chapter 35.1 offenses may extend to life (§ 13-902(E)), and sex offender treatment is ordinarily imposed as a court-ordered condition of that probation. No published Arizona appellate decision addressing how amenability evidence is weighed in an online offense case was located.

Special Issues in Legal Practice

The principles described above apply with equal force to every party in a legal proceeding. The science operates the same way regardless of who is asking the questions, and the research summarized here does not by itself favor either side. What differs between the parties is not the science but the vantage point — the particular places where these issues tend to surface, and the particular errors each party is best positioned to guard against. The two summaries below identify those recurring issues. They are offered for orientation, not as litigation strategy; in every case the controlling question is what the evidence in that case will, and will not, support, and the expert’s role is to inform the trier of fact rather than to advocate for either side.

Special Issues for Prosecutors

The research supports several points. Treatment effects are modest, concentrated in higher-risk men, and larger when evaluators are affiliated with the program, and no controlled study shows reduced official recidivism after treatment for people convicted of CSAM offenses. In the Dunkelfeld studies, 29 of 32 prior CSAM users reported continuing during therapy and 89.1% reported later use, so enrollment in treatment does not by itself show that offending has stopped. In the longest follow-up of treated CSAM offenders, measured progress in treatment did not predict reconviction. In courts-martial, future dangerousness is a proper subject under R.C.M. 1001(b)(5).

The research and law also counsel caution. Tapia bars arguing for a longer federal prison term so that the defendant can receive treatment. A military witness on rehabilitative potential may not base the opinion principally on the offense or recommend a discharge, directly or by euphemism. Contact-offender recidivism rates should not be applied to a defendant with a CSAM-only history, and the Dunkelfeld follow-up, with no control group, cannot show that treatment failed. Requiring disclosures in treatment and then using them against the person raises the Fifth Amendment problems addressed in Antelope and Bahr.

Special Issues for Defense

The research supports a bounded claim in mitigation. Specialized treatment matched to risk and delivered by qualified, supervised staff is associated with modest reductions in sexual recidivism, people with sexual interest in children will engage in prevention programs, and two Prevent It trials found short-term reductions in self-reported CSAM viewing or sexual urges. Voluntary, documented pretrial treatment is concrete evidence bearing on amenability. Tapia permits a court to consider rehabilitation in choosing probation or supervised release, Pepper permits consideration of postsentencing rehabilitation at resentencing, and Griggs holds that the limits of R.C.M. 1001(b)(5)(D) do not restrict defense mitigation evidence. In Arizona, § 13-701(E)(6) provides the vehicle where probation is available.

The research also counsels caution. No controlled study shows that any program lowers official recidivism for people convicted of CSAM offenses, so claims that a particular program “works” overstate the evidence. The Prevent It results are self-reported and short-term, the larger effects came from the waitlist-controlled trial, the Dunkelfeld outcome claims are contested, and the medication evidence is thin. For a person assessed at low risk, the research supports lighter intervention, so an argument based on low detected base rates and a request for intensive treatment can pull against each other. Offering treatment records in mitigation can open them to disclosure, including under Mil. R. Evid. 513(d)(7).

This paper is for educational purposes only and is not legal advice. The author is a psychologist, not an attorney; summaries of law are general and must be verified against current authority or with a licensed attorney.

References

Adebahr, R., Görts Öberg, K., Rahm, C., Byström, M., Sparre, C., Desai Boström, A. E., Bottai, M., Jokinen, J., & Savard, J. (2024). A randomized controlled add-on trial of fluoxetine and cognitive behavioral therapy for help-seeking men with a sexual interest in children: Presentation of the PARACHUTES trial protocol and initial feasibility data. Frontiers in Psychiatry, 15, Article 1448196. https://doi.org/10.3389/fpsyt.2024.1448196

Baskurt, S., Babchishin, K. M., Hilkes, G., & Seto, M. C. (2025). A meta-analysis of recidivism rates among individuals who commit child sexual exploitation material (CSEM) offending. Aggression and Violent Behavior, 85, Article 102080. https://doi.org/10.1016/j.avb.2025.102080

Beier, K. M., Grundmann, D., Kuhle, L. F., Scherner, G., Konrad, A., & Amelung, T. (2015). The German Dunkelfeld Project: A pilot study to prevent child sexual abuse and the use of child abusive images. The Journal of Sexual Medicine, 12(2), 529–542. https://doi.org/10.1111/jsm.12785

Beier, K. M., Nentzl, J., von Heyden, M., Fishere, M., & Amelung, T. (2024). Preventing child sexual abuse and the use of child sexual abuse materials: Following up on the German Prevention Project Dunkelfeld. Journal of Prevention, 45(6), 881–900. https://doi.org/10.1007/s10935-024-00792-0

Elliott, I. A., Mandeville-Norden, R., Rakestrow-Dickens, J., & Beech, A. R. (2019). Reoffending rates in a U.K. community sample of individuals with convictions for indecent images of children. Law and Human Behavior, 43(4), 369–382. https://doi.org/10.1037/lhb0000328

Gannon, T. A., Olver, M. E., Mallion, J. S., & James, M. (2019). Does specialized psychological treatment for offending reduce recidivism? A meta-analysis examining staff and program variables as predictors of treatment effectiveness. Clinical Psychology Review, 73, Article 101752. https://doi.org/10.1016/j.cpr.2019.101752

Gillespie, S. M., Bailey, A., Squire, T., Carey, M. L., Eldridge, H. J., & Beech, A. R. (2018). An evaluation of a community-based psycho-educational program for users of child sexual exploitation material. Sexual Abuse, 30(2), 169–191. https://doi.org/10.1177/1079063216639591

Gouveia, C., Silva, T., Cunha, O., Seto, M. C., Maimone, S., & Caridade, S. (2026). A systematic review on the effectiveness of interventions for individuals who have committed online sexual crimes involving children. Psychosocial Intervention, 35, Article e260808. https://doi.org/10.5093/pi2025a16

Hanson, R. K., Bourgon, G., Helmus, L., & Hodgson, S. (2009). The principles of effective correctional treatment also apply to sexual offenders: A meta-analysis. Criminal Justice and Behavior, 36(9), 865–891. https://doi.org/10.1177/0093854809338545

Holper, L., Habermeyer, E., & Mokros, A. (2025). A multimodel meta-analysis assessing moderators of sexual recidivism as an indicator of treatment effectiveness in persons with sexual offense histories. Journal of Experimental Criminology, 21(2), 503–526. https://doi.org/10.1007/s11292-023-09600-x

Holper, L., Mokros, A., & Habermeyer, E. (2024). Moderators of sexual recidivism as indicator of treatment effectiveness in persons with sexual offense histories: An updated meta-analysis. Sexual Abuse, 36(3), 255–291. https://doi.org/10.1177/10790632231159071

Khan, O., Ferriter, M., Huband, N., Powney, M. J., Dennis, J. A., & Duggan, C. (2015). Pharmacological interventions for those who have sexually offended or are at risk of offending. Cochrane Database of Systematic Reviews, 2015(2), Article CD007989. https://doi.org/10.1002/14651858.CD007989.pub2

König, A. (2025). The German ‘Dunkelfeld’ approach: When the dark figure of sexual delinquency against minors remains shady. Journal of Prevention, 46(4), 463–466. https://doi.org/10.1007/s10935-025-00852-z

Landgren, V., Malki, K., Bottai, M., Arver, S., & Rahm, C. (2020). Effect of gonadotropin-releasing hormone antagonist on risk of committing child sexual abuse in men with pedophilic disorder: A randomized clinical trial. JAMA Psychiatry, 77(9), 897–905. https://doi.org/10.1001/jamapsychiatry.2020.0440

Landgren, V., Olsson, P., Briken, P., & Rahm, C. (2022). Effects of testosterone suppression on desire, hypersexuality, and sexual interest in children in men with pedophilic disorder. The World Journal of Biological Psychiatry, 23(7), 560–571. https://doi.org/10.1080/15622975.2021.2014683

Lätth, J., Landgren, V., McMahan, A., Sparre, C., Eriksson, J., Malki, K., Söderquist, E., Öberg, K. G., Rozental, A., Andersson, G., Kaldo, V., Långström, N., & Rahm, C. (2022). Effects of internet-delivered cognitive behavioral therapy on use of child sexual abuse material: A randomized placebo-controlled trial on the Darknet. Internet Interventions, 30, Article 100590. https://doi.org/10.1016/j.invent.2022.100590

McMahan, A., Joleby, M., Andersson, G., Letourneau, E. J., Seto, M. C., Luke, T. J., & Rahm, C. (2026). An international randomized waitlist-controlled trial of internet-delivered cognitive behavioral therapy for sexual urges toward children. Journal of Consulting and Clinical Psychology, 94(7), 425–436. https://doi.org/10.1037/ccp0001019

Middleton, D., Mandeville-Norden, R., & Hayes, E. (2009). Does treatment work with internet sex offenders? Emerging findings from the Internet Sex Offender Treatment Programme (i-SOTP). Journal of Sexual Aggression, 15(1), 5–19. https://doi.org/10.1080/13552600802673444

Mokros, A., & Banse, R. (2019). The “Dunkelfeld” project for self-identified pedophiles: A reappraisal of its effectiveness. The Journal of Sexual Medicine, 16(5), 609–613. https://doi.org/10.1016/j.jsxm.2019.02.009

Pellowski, J. S., Bergner-Köther, R., Schmidt, S., Peschka, L., Hock, M., Briken, P., & von Franqué, F. (2026). Toward a new tool for risk-assessment: Recidivism of child sexual abuse materials in men under voluntary treatment. Frontiers in Psychology, 17, Article 1764052. https://doi.org/10.3389/fpsyg.2026.1764052

Prichard, J., Wortley, R., Watters, P., Spiranovic, C., Hunn, C., & Krone, T. (2022). Effects of automated messages on internet users attempting to access “barely legal” pornography. Sexual Abuse, 34(1), 106–124. https://doi.org/10.1177/10790632211013809

Prichard, J., Wortley, R., Watters, P., Spiranovic, C., & Scanlan, J. (2024). The effect of therapeutic and deterrent messages on Internet users attempting to access ‘barely legal’ pornography. Child Abuse & Neglect, 155, Article 106955. https://doi.org/10.1016/j.chiabu.2024.106955

Schmucker, M., & Lösel, F. (2015). The effects of sexual offender treatment on recidivism: An international meta-analysis of sound quality evaluations. Journal of Experimental Criminology, 11(4), 597–630. https://doi.org/10.1007/s11292-015-9241-z

Schmucker, M., & Lösel, F. (2017). Sexual offender treatment for reducing recidivism among convicted sex offenders: A systematic review and meta-analysis. Campbell Systematic Reviews, 13(1), 1–75. https://doi.org/10.4073/csr.2017.8

Schuler, M., Gieseler, H., Schweder, K. W., von Heyden, M., & Beier, K. M. (2021). Characteristics of the users of Troubled Desire, a web-based self-management app for individuals with sexual interest in children: Descriptive analysis of self-assessment data. JMIR Mental Health, 8(2), Article e22277. https://doi.org/10.2196/22277

Seto, M. C. (2025). Online sexual offending: Theory, practice, and policy (2nd ed.). American Psychological Association. https://doi.org/10.1037/0000428-000

Ward, T., Day, A., Howells, K., & Birgden, A. (2004). The multifactor offender readiness model. Aggression and Violent Behavior, 9(6), 645–673. https://doi.org/10.1016/j.avb.2003.08.001

Westfal, V. (2026). Adolescents in the Dunkelfeld: A study of help seeking minors who use child sexual abuse materials. Sexual Abuse, 38(2), 187–211. https://doi.org/10.1177/10790632251351254

Legal Authorities

Minnesota v. Murphy, 465 U.S. 420 (1984).

Pepper v. United States, 562 U.S. 476 (2011).

Tapia v. United States, 564 U.S. 319 (2011).

United States v. Antelope, 395 F.3d 1128 (9th Cir. 2005).

United States v. Bahr, 730 F.3d 963 (9th Cir. 2013).

United States v. Crum, No. 18-169 (2d Cir. Apr. 12, 2021) (summary order) (nonprecedential).

United States v. Smith, 29 F.4th 397 (8th Cir. 2022) (per curiam) (No. 21-2037, Mar. 24, 2022).

United States v. George, 52 M.J. 259 (C.A.A.F. 2000).

United States v. Griggs, 61 M.J. 402 (C.A.A.F. 2005).

United States v. Horner, 22 M.J. 294 (C.M.A. 1986).

United States v. Ohrt, 28 M.J. 301 (C.M.A. 1989).

United States v. Williams, 41 M.J. 134 (C.M.A. 1994).

U.S. Const. amend. V.

18 U.S.C. § 3553(a)(2)(D).

18 U.S.C. § 3559(a).

18 U.S.C. § 3561(a).

18 U.S.C. § 3563(b)(9).

18 U.S.C. § 3582(a).

18 U.S.C. § 3583(d).

U.S. Sentencing Guidelines Manual §§ 5B1.3(d)(7), 5D1.3(b)(3)(G) (U.S. Sentencing Comm’n 2025).

R.C.M. 1001(b)(5), 1002(c) (Manual for Courts-Martial, United States, 2024 ed.).

Mil. R. Evid. 513(d)(7) (Manual for Courts-Martial, United States, 2024 ed.).

Exec. Order No. 14130, 2024 Amendments to the Manual for Courts-Martial, United States, 89 Fed. Reg. 105343 (Dec. 27, 2024).

Ariz. Rev. Stat. Ann. § 13-701(E).

Ariz. Rev. Stat. Ann. § 13-705(F), (G), (K), (M), (R).

Ariz. Rev. Stat. Ann. § 13-1428(B).

Ariz. Rev. Stat. Ann. § 13-3554.

Ariz. Rev. Stat. Ann. § 13-902(E).

Ariz. R. Crim. P. 26.5.

Ariz. R. Crim. P. 27.1.

Related topic papers: Recidivism of People Convicted of Online Sexual Offenses · Risk Assessment Tools for Online Offenders · Supervision Conditions and Internet Restrictions · Pedophilia, Hebephilia, and the Diagnostic Manuals · Problematic Online Sexual Behavior: Diagnosis and Measurement · Culpability and Sentencing Factors in CSAM Cases · The Polygraph in Online Offense Cases · Women and Adolescents Who Commit Online Sexual Offenses · Pretrial Release in Online Cases

Related topics from Arizona Forensic Services: Psychosexual Risk Evaluations · Courts-Martial · Autism Online Offense: Sentencing