Risk Assessment of Online Child Sexual Exploitation Offenders
=============================================================

* Matthew E. Hirschtritt
* Douglas Tucker
* Renée L. Binder

## Abstract

Over the past two decades in the United States, a dramatic increase in access to the Internet has facilitated an increase in the production, viewing, and distribution of child pornography. In this context, forensic mental health professionals may be called on to estimate the risk of future violence, especially of contact sexual offenses, among individuals charged with online sexual offenses. We summarize demographic and clinical characteristics that differentiate online from contact and “mixed” offenders (those who commit both online and contact offenses), offending histories of these three groups, and the current state of knowledge regarding risk of progression from online-only to contact offending. Multiple studies suggest that online, contact, and mixed offenders demonstrate distinct profiles, and wide variations exist in the offense histories of online-offending groups. Longitudinal studies of individuals charged with online offenses are few in number and are mostly limited to detection of formal charges. Nevertheless, available studies suggest that most individuals who are charged with online offenses and who do not have histories of contact offenses are unlikely to engage in future contact offenses. Within the limitations of the current literature, we suggest guidance for the evaluation and treatment of online offenders.

Rates of Internet use in the United States have increased dramatically. From 2000 to 2018, the percentage of U.S. adults who reported using the Internet increased from 52 percent to 88 percent.1 The many benefits of this expanded access, including streamlined commerce and the rapid distribution and sharing of important news and ideas, have been accompanied by problematic uses of this technology, notably the production, distribution, and viewing of child pornography. We use the term “online offender” to refer to an individual with online-only engagement in child pornography or other child sexual exploitation material, in contrast to a “contact offender” (also referred to as in-person or hands-on) or a “mixed offender.”

Given the covert nature of online sexual offending, it is challenging to quantify how much child pornography content is available and is actively being accessed, but recent observational studies of popular file-sharing networks provide a sense of the scale of the concern. For instance, a one-year study of a popular, peer-to-peer, file-sharing network in 2013 revealed that, among millions of unique Internet Protocol addresses, 244,920 of them shared files containing child pornography material, over 80 percent of which had saved or shared fewer than 10 known child pornography files.2 In their 2017 Annual Report,3 the Internet Watch Foundation summarized the results of a combination of expert analyst and public reporting of child exploitation material. In that year alone, researchers discovered 78,589 unique web addresses containing “child sexual abuse imagery, having links to the imagery, or advertising it” (Ref. 3, p 15), in which 57 percent of the images depicted youth 10 years old and younger. Nearly a third of the web addresses with child exploitation material were hosted in North America; the United States ranked second in the number of web addresses containing child exploitation material (The Netherlands ranked first). The authors of the Internet Watch Foundation report note the increasing difficulty in detecting online child exploitation material because of increased use of disguised websites that require viewers to follow a digital pathway to access illegal material. This “dark web” of online child exploitation material has been fueled by the “triple-A engine” of the Internet: accessibility, affordability, and anonymity.4

The extent of online offending can also be tracked using trends in criminal prosecutions. In 2015, the U.S. Department of Justice's Internet Crimes Against Children Task Force Program arrested nearly 8,500 individuals for “technology-enabled crimes against children,” including online offending and child sex trafficking (Ref. 5, p 6). Recent years have seen an increased number of individuals convicted of creating, possessing, or distributing online child exploitation material.6

The effects of online offending on its victims are poorly characterized; however, it has been posited that youth may be traumatized not only by the production of the exploitative content, but then by the repeated, perpetual exposure of their images online.7,8 These youth are powerless to control the distribution of and access to these images, potentially leading to re-victimization long after the initial production of the digital material.

The relationship between online offending and contact offending, however, is unclear. As many as 17 to 27 percent of females and 3 to 5 percent of males 15 to 17 years old report being the victim of sexual assault.9 The psychological impact of contact offenses can be profound, including increasing the risk of suicide and non-suicidal self-injury.10

In this context, forensic mental health professionals may be tasked with evaluating individuals charged with online offenses and estimating the risk of engaging in contact offenses. Therefore, in this report we summarize characteristics that may distinguish online from contact and mixed offenders, summarize crossover rates from online to contact offending, and apply these findings to forensic and clinical practice.

## Online, Contact, and Mixed Offenders

### General Considerations

Little is known of the demographic, psychosocial, and Internet-use patterns of online offenders, in part because many online offenders evade detection by law enforcement.11 Other important limitations in the literature about online and contact offenders include lack of differentiation between online-only, contact, and mixed offenders; small sample sizes; and reliance on convenience samples from forensic settings instead of the general population.12

Because rapid developments in technology outpace research regarding online offenders, much of the existing literature may already be outdated. Likewise, many studies comparing online and contact offenders fail to distinguish among a wide range of online offending behaviors, including viewing child pornography, engaging in online sexual communication with minors, harassing minors with sexually explicit material, and locating children as potential victims. Moreover, some researchers have posited that there is overlap between contact and online offenders.12

An analysis of media reports involving online offending proposes three groups: travelers who attempt to meet children in person (also known as solicitation or luring offenders); traders who share child exploitation material; and combined traveler-traders. Traders may also have undetected contact offenses.13 Within the limitations of the literature, the available data suggest that there are characteristics that may help distinguish between online, contact, and mixed offenders, and that patterns of crossover from online to contact offending are complex.

### Group Characteristics

In a recent comprehensive meta-analysis that incorporated data from 30 distinct samples, Babchishin *et al.*14 synthesized the existing literature regarding differences among online, contact, and mixed offenders. Individuals in all three groups are more likely to be male, but the groups were significantly different from one another in various demographic, legal, mental health, and relationship-related variables (see Table 1). The prototypical online offender is a white, single male in his 20s or 30s, is well educated, is employed, has no history of severe mental illness or significant childhood adversity, otherwise functions well in society, has low antisocial traits, and demonstrates sexual deviancy. In contrast, a contact offender is more likely to be an older, partnered male with ready access to children, strong antisocial traits, a criminal offending history, and low economic status, who is unemployed, poorly educated, and has a history of severe mental illness. Mixed offenders are more like contact offenders but have greater familiarity with the Internet and less physical access to children. In addition, mixed offenders may be more likely than online offenders to exhibit traits typical of contact offenders, such as antisocial traits and illicit drug use. These general characteristics should be interpreted with caution, however, given the high probability of biased sampling and altered patterns of Internet access in recent years.

View this table:
[Table 1](http://jaapl.org/content/early/2019/04/15/JAAPL.003830-19/T1)

Table 1 
Comparison of Characteristics Associated With Online, Contact, and Mixed Offenders

Although not specifically referenced in the meta-analysis by Babchishin *et al.*,14 evidence from a small number of studies suggests that online solicitors, when compared with child pornography offenders, have lower relationship stability; deviant sexual preferences, sex drive, and preoccupation15; and higher rates of prior sexual offense-related arrests and substance use.16 Furthermore, based on an analysis of chat logs, emails, and social network posts generated by 200 online solicitors, DeHart *et al.*17 proposed four groups: cybersex-only (online chatting and “coached” masturbation without intent to meet in person), schedulers (arranging in-person encounters), mixed cybersex and schedulers, and buyers (negotiating terms of in-person encounters for a pimp). In addition to their distinct offending behavior patterns, these online solicitor groups were differentiated by demographic characteristics and sexual interests; use of this classification scheme in subsequent studies may help identify online solicitors at highest risk for contact offenses.

Based on limited empiric data, multiple authors have attempted to subtype online offenders based on their motivations (e.g., latent curiosity, pedophilia), types of engagement (e.g., production, distribution, viewing), and patterns of engagement (e.g., daily, one-time). For instance, Merdian *et al.*18 classified online offenders in three dimensions: fantasy or contact-driven, motivation, and level of social engagement. Contact-driven behaviors include efforts to establish an in-person encounter with a minor, whereas fantasy-driven behaviors include trading child pornography and engaging in sexual chats with no efforts toward an in-person encounter. Merdian *et al.*18 further divided motivations for fantasy-driven behavior into two groups. They suggest that generally deviant or pedophilic motivations are amenable to assessment of fantasies and treatment; however, the presence of strong financial or other material motivations should prompt a general criminal assessment and associated treatment. In contrast to fantasy-driven offenders, contact-driven offenders should be referred to traditional sex offender assessment and treatment.18

### The Motivation-Facilitation Model

Despite the relative lack of empiric evidence to support these and other typologies (summarized in Henshaw *et al.*12), they can be used to describe the heterogeneity of online offenders. For the purposes of evaluation and risk assessment, it is important to highlight the factors that may distinguish online-only from contact offenders. In this context, the motivation-facilitation model (MFM) of sexual offending offers a useful framework for understanding the risk factors for sexual offending.19 Specifically, the MFM consists of primary sexual motivations (consisting of paraphilias, high or poorly regulated sex drive, and significant effort devoted to obtaining new sexual partners), which are facilitated by both static (e.g., antisocial traits, criminal history) and dynamic (e.g., substance intoxication, negative affect) characteristics. Finally, motivations and facilitating factors interact with situational factors, such as access to children, which together lead to increased risk for contact offenses.

The strength of the MFM is that it was derived from large community samples of self-identified, hypersexual men as well as clinical and forensic studies of contact offenders. The MFM does not, however, account for other non-sexual motivations, including revenge, anger, need for control, or emotional affinity for children. It also does not explain the behavior of individuals who sexually offend against both minors and adults. Likewise, it does not account for protective factors or victim-specific factors. Furthermore, the MFM, like other models of sexual offending, is based on cross-sectional data; therefore, the causal pathway between motivation and sexual offending remains theoretical. The assumptions of the MFM must still be tested in longitudinal and experimental studies before it can be used as a conceptual framework to estimate risk of contact sexual offending. In the meantime, it provides a conceptual model to help understand factors that may increase the risk of crossover from online to contact offending.

## Prior Contacts Among Online Offenders

In a meta-analysis of rates of previous contact offenses among adults with an index online sexual offense, Seto *et al.*20 examined all published data related to this topic through July 2009. Among 24 studies from France, Australia, the United States, the United Kingdom, Canada, Switzerland, Germany, and New Zealand, with sample sizes ranging from 30 to 870 subjects and a total of 4,697 online offenders, 17.3 percent had a previous contact offense. The majority (75%) of these studies used data only from charges, convictions, or arrests for contact offenses; the remainder used either a combination of these official records with self-report (12.5%) or self-report alone (12.5%). When the authors restricted their analysis to studies that only used official records, the overall rate of previous contact offenses among online offenders decreased to 12.2 percent. Using only data from self-report, the rate of previous contact offenses among online offenders increased to 55.1 percent. In a notable outlier in this meta-analysis,21 researchers administered structured interviews to 155 online offenders enrolled in an intensive, residential, sex offender-specific voluntary treatment program at a medium-security federal prison. At the time of sentencing, 26 percent had an officially documented contact offense; by the end of the treatment period, however, 85 percent had disclosed a history of contact offending. Although it is possible that respondents falsely claimed that they had previously engaged in contact offending to artificially inflate their treatment progress,22 this study does suggest that a large percentage of online offenders may, on closer examination, be mixed offenders.

Using a study design similar to that of Bourke and Hernandez,21 Bourke *et al.*23 interviewed 127 men under investigation for online offending charges, none of whom had a criminal record of contact offending. Only 4.7 percent disclosed a prior contact offense initially, but an additional 52.8 percent (or 57.5% of the total sample) subsequently disclosed a prior contact offense during a polygraph examination. Similarly, among 251 cases of online sexual exploitation of children, Owens *et al.*24 found that 32 percent had engaged in at least one contact offense before or after the index online offense. About half of these cases (about 16% of the total sample) started with an index offense of online offending only, but further investigation of criminal histories, physical evidence, and sentencing information revealed a history of contact offenses.

In summary, the available evidence suggests that individuals with an index online offense may have undetected prior contact offenses. Therefore, online offenders should be carefully evaluated for a contact offending history, preferably through both review of official records and interviews with or without the use of polygraph. As we describe below, a history of contact offending carries important prognostic implications for subsequent offending.

## From Online to Contact Offending

In addition to assessing for a history of contact offenses among online offenders, forensic mental health professionals may be asked to estimate the risk of subsequent sexual offending among these individuals. The existing data on this topic suggest that most online offenders without a history of contact offenses are unlikely to cross over into contact offenses within one to five years after their index offense. Mixed offenders, however, are more likely than online offenders to crossover into contact offending following an online offense. As we describe below, these conclusions should be interpreted with multiple caveats given the limitations of the current literature on this topic.

In the same meta-analysis referenced above, Seto *et al.*20 identified nine studies that quantified rates of reoffending among online offenders from the United Kingdom, the United States, and Canada. Among 2,630 online offenders, only 4.6 percent had any sexual re-offense in follow-up periods ranging from 18 months to six years. Furthermore, only 2.0 percent had a subsequent contact offense, and 3.4 percent had a subsequent online offense. The authors did not examine predictors of re-offense, and all nine of these studies used official records to determine re-offense. The study's reliance on official records may have led to an underestimate of the recidivism rate.

More recent studies generally confirm these findings regarding low rates of officially recorded recidivism among online offenders and provide evidence to support static and dynamic factors that may elevate this risk among certain offenders. Studies based in Canada,25,–,27 Australia,28 the United Kingdom,29 Switzerland,30 and the United States31 have found that 0 to 9 percent of online offenders, followed for one to nine years after their index offense, demonstrate repeat online offenses. Rates of subsequent contact offenses in the follow-up period were lower, ranging from 0 to 6 percent. Mixed offenders were more likely to engage in subsequent contact offenses compared with online offenders.

Eke *et al.*25 found that only 3.9 percent of 541 online offenders had an official record of a contact offense within approximately four years following their index online offense. Recidivism was associated with younger age at first offense, positive prior criminal history, any prior nonviolent history, higher number of prior violent offenses and prior contact sex offenses, and higher severity of prior violent offenses. Perhaps not surprisingly, these characteristics are also associated with the prototypical contact offender. Seto and Eke26 followed 266 online offenders over five years and found that, regardless of their history of contact offending, only 3 percent of the sample had a subsequent contact offense. Using a similar study design, this same research group27 examined fixed five-year recidivism among 80 online and mixed offenders. Rates of sexual recidivism and online offenses were 9.8 percent and 4.9 percent, respectively, among online offenders and 31.6 percent and 21.1 percent among mixed offenders. Put another way, the five-year contact recidivism rate was 2.2 percent for online offenders and 7.5 percent for mixed offenders.

In contrast to these findings from official records, data from the German Prevention Project Dunkelfeld (PPD) provide a rare insight into self-reported behavior among male online offenders.32 The PPD was initiated in 2005 as a public-outreach program designed to provide free, confidential support for adults with pedophilic and hebephilic interests; it aims to prevent any form of sexual offending and reduce the risk of re-offending among sexual offenders. In a recent pilot study using data from a small subset of PPD participants, the authors found that 90.6 percent of online offenders in the treatment arm of the trial endorsed ongoing online behaviors during a one-year treatment period, none of which were reported to the authorities.33 These preliminary results suggest that actual rates of continued online offending following an index online offense may be significantly higher than studies using official records indicate.

In summary, these data suggest a relatively low risk of subsequent contact offending among online offenders in studies using official records with follow-up periods ranging from one to nine years. When anonymous self-report is used, however, rates of subsequent online offenses increase dramatically, again supporting the contention that a significant proportion of online offenses likely go undetected. In addition, mixed offenders are nearly as likely as contact offenders to engage in subsequent contact offenses following an index online offense. This latter finding again points to the importance of assessing for a history of contact offending among online offenders.

## Assessment of Online Offenders

Synthesizing the available data regarding the demographic, clinical, and forensic characteristics and rates and predictors of recidivism of online offenders is challenging. There are multiple limitations of the current literature, including relatively small sample sizes, reliance on official records instead of self-report, relatively short follow-up periods, heterogeneous categorization of online offenders, and varying definitions of recidivism. Furthermore, data about online solicitors is limited because researchers often classify these individuals as either online or contact offenders, depending on whether they have physical contact with a minor. Nevertheless, online solicitors may have a distinct offending and risk profile.

Many of the retrospective and longitudinal studies are based outside of the United States, where different legal regulations may influence research findings and recommendations. For example, in many European countries, the Combating Paedophile Information Networks in Europe (COPINE) Scale has been used to classify the severity of sexual exploitation images. The COPINE Scale is a collection typology that ranks the volume and characteristics of an individual's sexual exploitation collection on a scale of 1 to 10.34 The COPINE scale has been used in samples from the United Kingdom35 and the Netherlands36 to compare characteristics of online and contact offenders. Forensic mental health professionals in the United States are rarely provided access to evaluees' sexual exploitation material, so this scale cannot be applied realistically to estimate their recidivism risk.

Other tools have been developed to systematically approach evaluation and assessment. As recently reviewed by Garrington *et al.*,37 there are at least two English-language, online offending–specific evaluation tools currently available. The seven-item Child Pornography Offender Risk Tool (CPORT)26,27 was developed and validated in two samples of Canadian men 18 years or older who were convicted of an online offense and followed over five years. The CPORT includes four static predictors (younger age at time of offense; presence of any criminal history; prior or index-offense-related failure of probation, parole, or conditional release; and prior or index-offense–related contact offending history). Three predictors are related to the content of the online sexual exploitation material: indication of pedophilic or hebephilic interests; more boy than girl sexual exploitation content; and higher ratio of boy-to-girl content “in nudity or other child content.” Combining results from the development and validation samples, the CPORT demonstrated good predictive accuracy for five-year sexual recidivism (area under the curve [AUC] = .72) and online recidivism (AUC = .74).27 The scoring manual and updated reference material for the CPORT are available online.38

A second risk-assessment tool is the Kent Internet Risk Assessment Tool (KIRAT),39 developed in the United Kingdom among 170 mixed or high-risk and 204 online or low-risk offenders. The authors narrowed 166 candidate variables to nine items, organized into four evaluation steps: previous behaviors (e.g., convictions for sexual offenses), access to children, current behavioral facilitators (e.g., grooming behaviors), and other factors (e.g., domestic abuse charges). In a retrospective, cross-sectional analysis, the KIRAT demonstrated an AUC of .91 discriminating between high- and low-risk offenders. As of this writing, there is no publicly available scoring manual for the KIRAT; however, instrument items and scoring instructions are described in the validation study.39

The COPINE and KIRAT have limited use among forensic mental health professionals in the United States in part because of the required scoring elements related to the sexual exploitation material itself. In addition, as Garrington *et al.*37 note, these scales were developed and validated using relatively small sample sizes and official records, and only the COPINE was developed using longitudinal data. More commonly used actuarial scales (e.g., the Static-99/R40 and the Risk Matrix 2000/R41) and dynamic scales (e.g., the Stable-2007 and the Acute 200742) were developed specifically for contact offenders and, to date, have limited applicability to online offenders. Nonetheless, these scales may be used to help supplement a forensic evaluation of an online offender, which should include, at a minimum, a history of contact offenses, any other criminal history, and whether the individual meets criteria for a paraphilic disorder.

In their opinions delivered to the court, forensic mental health professionals may reference the available data that suggest that most online offenders who have not committed contact offenses are at low risk of future contact offenses, noting the limitations discussed above. The German PPD study32,33 suggests that most online offenders may continue to engage in online offending undetected, but even that study suggests that the crossover from online to contact offending occurs infrequently. When tasked with assessing the risk of recidivism among individuals charged with online offending, forensic mental health professionals can apply validated measures such as the CPORT or KIRAT when sufficient information is available. In addition to estimating risk, these measures have the added benefit of potentially identifying dynamic or modifiable factors that, if addressed, may reduce the risk of sexual recidivism.

## Treatment of Online Offenders

Online offenders may come to clinical attention for a variety of reasons. At the start of treatment, clinicians should explain any mandated reporting requirements that apply. For example, in California, the production, distribution, or promotion of child pornography in any form is considered sexual exploitation and is subject to mandated reporting.43

Evidence supporting behavioral and pharmacologic approaches to treatment of online offenders have been reviewed elsewhere.44,–,46 In brief, given the limited understanding of online offenders compared with contact offenders, current treatment paradigms are frequently based on open-label trials or case series with relatively small sample sizes and short follow-up periods. Furthermore, as Ly *et al.*45 note, published treatment studies for online offenders do not differentiate between viewers or collectors and distributors, which may represent distinct groups with different treatment needs.

Preliminary evidence supports the use of cognitive-behavioral therapy for online offenders, which typically includes appreciating the impact of online offending on its victims, practicing interpersonal skills, differentiating between emotional and physical intimacy, and acquiring prosocial behaviors.45 Behavioral approaches may include aversion therapy.47 Psychotherapeutic modalities may focus on improved self-esteem among online offenders.48 Evidence supporting the use of pharmacologic approaches, including androgen-deprivation therapy and selective serotonin reuptake inhibitors, is reviewed elsewhere.49 Ly *et al.*45 note that there are currently no known pharmacologic trials that exclusively include online offenders; when online offenders are included in trials, they often are underrepresented compared with contact or mixed offenders. Only three pharmacologic trials have included online offenders to date,50,–,52 with unclear applicability specific to online offenders.

Numerous community-based programs are available that focus on correcting cognitive distortions, improving victim empathy, practicing problem-solving strategies, building healthy relationships, and constructing a prosocial lifestyle.45 These programs, including the Internet Sex Offender Treatment Programme (i-SOPT),53 are currently only available outside of the United States.

In summary, there is a robust literature regarding treatment and management of contact offenders, but there are far fewer empirically supported treatments that are tailored to online or mixed offenders. After determining mandatory-reporting requirements, clinicians may offer individual and group cognitive-behavioral therapy or purely behavioral psychotherapeutic approaches to reduce maladaptive behaviors and coping strategies and to enhance prosocial and adaptive skills. Given the paucity of supporting evidence regarding online offenders, pharmacologic approaches should be reserved for treatment-refractory cases or based on patient preference. Certain groups of low-level offenders who are at low risk of recidivism may require no specific treatment. Hanson *et al.*54 and Day *et al.*55 suggest frameworks that forensic mental health experts may apply to stratify offenders into relative risk levels to prioritize treatment for those at higher risk levels.

## Conclusions and Future Directions

Forensic mental health professionals who are retained in cases of online offenders to assess the risk of contact offenses face a conundrum: how does one assess risk of harm to minors with limited empirical data? The studies we have reviewed here suggest that online offenders may demonstrate different demographic characteristics and criminal-offense histories when compared with contact offenders. The former tend to have greater formal education and to be younger, of higher socioeconomic status, and employed. In addition, the limited available data suggest that most online offenders without a history of past contact offenses are unlikely to engage in future contact offenses. Nonetheless, some online offenders may simultaneously be at risk of recidivism for contact offenses or, as demonstrated by the German PPD study, for subsequent online offenses. When forming an opinion regarding the sexual recidivism risk of an online offender, a forensic mental health professional should take into account both static and dynamic factors; the use of novel scales such as the CPORT and KIRAT are helpful in this regard. The literature is limited by factors such as sampling from forensic populations and historical cohorts that may not fully represent the current demographics of Internet users. Because many people who view, distribute, or even produce online sexual exploitation material are never arrested, the current literature likely underestimates the extent of online sexual exploitation in the United States.

More research is needed to stratify the relative risk of online-only offenders in comparison to contact and mixed offenders. For instance, additional longitudinal studies that use anonymous reports among those who engage in online offending, like the German PPD study, may yield valid data about previous and subsequent contact and online offending patterns. Such studies will hopefully allow clinicians and law-enforcement agents to estimate risk based on offender characteristics, leading to fairer sentencing and more effective treatments.

## Footnotes

*   Disclosures of financial or other potential conflicts of interest: None.

*   © 2019 American Academy of Psychiatry and the Law

## References

1.  1.Pew Research Center. Internet/broadband fact sheet. [http://www.pewinternet.org/fact-sheet/internet-broadband](http://www.pewinternet.org/fact-sheet/internet-broadband). Accessed June 13, 2018
    
    

2.  2.Wolak J, Liberatore M, Levine BN: Measuring a year of child pornography trafficking by U.S. computers on a peer-to-peer network. Child Abuse Negl 38:347–56, 2014
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1016/j.chiabu.2013.10.018&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=24252746&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 

3.  3.Internet Watch Foundation. Annual report, 2017. [https://annualreport.iwf.org.uk/pdf/IWF\_2017\_Annual\_Report.pdf](https://annualreport.iwf.org.uk/pdf/IWF_2017_Annual_Report.pdf). Accessed June 12, 2018
    
    

4.  4.Cooper A: Sexuality and the internet: surfing into the new millennium. Cyberpsychol Behav 1:187–93, 1998
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1089/cpb.1998.1.187&link_type=DOI) 

5.  5.U.S. Department of Justice. The national strategy for child exploitation prevention and interdiction: a report to Congress. [https://www.justice.gov/psc/file/842411/download](https://www.justice.gov/psc/file/842411/download). Accessed January 4, 2018
    
    

6.  6.U.S. Sentencing Commission. 2012 report to Congress: federal child pornography offenses. [https://www.ussc.gov/research/congressional-reports/2012-report-congress-federal-child-pornography-offenses](https://www.ussc.gov/research/congressional-reports/2012-report-congress-federal-child-pornography-offenses). Accessed January 4, 2018
    
    

7.  7.von Weiler J, Haardt-Becker A, Schulte S: Care and treatment of child victims of child pornographic exploitation (CPE) in Germany. J Sex Aggress 16:211–22, 2010
    
    

8.  8.Leonard MM: “I did what I was directed to do but he didn't touch me”: the impact of being a victim of internet offending. J Sex Aggress 16:249–56, 2010
    
    

9.  9.Finkelhor D, Shattuck A, Turner HA, Hamby SL: The lifetime prevalence of child sexual abuse and sexual assault assessed in late adolescence. J Adolesc Health 55:329–33, 2014
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1016/j.jadohealth.2013.12.026&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=24582321&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 
    
    [Web of Science](http://jaapl.org/lookup/external-ref?access_num=000341138400006&link_type=ISI) 

10. 10.Maniglio R: The role of child sexual abuse in the etiology of suicide and non-suicidal self-injury. Acta Psychiatr Scand 124:30–41, 2011
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1111/j.1600-0447.2010.01612.x&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=20946202&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 
    
    [Web of Science](http://jaapl.org/lookup/external-ref?access_num=000292149500005&link_type=ISI) 

11. 11.Neutze J, Grundmann D, Scherner G, Beier KM: Undetected and detected child sexual abuse and child pornography offenders. Int'l J L & Psychiatry 35:168–75, 2012
    
    

12. 12.Henshaw M, Ogloff JRP, Clough JA: Looking beyond the screen: a critical review of the literature on the online child pornography offender. Sex Abuse 29:416–45, 2017
    
    

13. 13.Alexy EM, Burgess AW, Baker T: Internet offenders: traders, travelers, and combination trader-travelers. J Interpers Violence 20:804–12, 2005
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1177/0886260505276091&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=15914702&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 
    
    [Web of Science](http://jaapl.org/lookup/external-ref?access_num=000231093300004&link_type=ISI) 

14. 14.Babchishin KM, Hanson RK, VanZuylen H: Online child pornography offenders are different: a meta-analysis of the characteristics of online and offline sex offenders against children. Arch Sex Behav 44:45–66, 2015
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1007/s10508-014-0270-x&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=24627189&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 

15. 15.Seto MC, Wood JM, Babchishin KM, Flynn S: Online solicitation offenders are different from child pornography offenders and lower risk contact sexual offenders. Law & Hum Behav 36:320–30, 2012
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1037/h0093925&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=22849417&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 

16. 16.Navarro JN, Jasinski JL: Demographic and motivation differences among online sex offenders by type of offense: an exploration of routine activities theories. J Child Sex Abuse 24:753–71, 2015
    
    

17. 17.DeHart D, Dwyer G, Seto MC, et al: Internet sexual solicitation of children: a proposed typology of offenders based on their chats, e-mails, and social network posts. J Sex Aggress 23:77–89, 2017
    
    

18. 18.Merdian HL, Curtis C, Thakker J, et al: The three dimensions of online child pornography offending. J Sex Aggress 19:121–32, 2013
    
    

19. 19.Seto MC: The motivation-facilitation model of sexual offending. Sex Abuse 31:3–24, 2019
    
    

20. 20.Seto MC, Hanson RK, Babchishin KM: Contact sexual offending by men with online sexual offenses. Sex Abuse 23:124–45, 2011
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1177/1079063210369013&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=21173158&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 

21. 21.Bourke ML, Hernandez AE: The ‘Butner study’ redux: a report of the incidence of hands-on child victimization by child pornography offenders. J Fam Violence 24:183, 2008
    
    

22. 22.1.  Dubin LA, 
    2.  Horowitz E
    
    Wollert R, Skelton A: Egregious flaws discredit the Butner redux study, in Caught in the Web of the Criminal Justice System: Autism, Developmental Disabilities, and Sex Offenses. Edited by Dubin LA, Horowitz E. London: Jessica Kingsley Publishers, 2017, pp 185–214
    
    

23. 23.Bourke ML, Fragomeli L, Detar PJ, et al: The use of tactical polygraph with sex offenders. J Sex Aggress 21:354–67, 2015
    
    

24. 24.Owens JN, Eakin JD, Hoffer T, et al: Investigative aspects of crossover offending from a sample of FBI online child sexual exploitation cases. Aggress Violent Behav 30:3–14, 2016
    
    

25. 25.Eke AW, Seto MC, Williams J: Examining the criminal history and future offending of child pornography offenders: an extended prospective follow-up study. Law & Hum Behav 35:466–78, 2011
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1007/s10979-010-9252-2&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=21088873&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 

26. 26.Seto MC, Eke AW: Predicting recidivism among adult male child pornography offenders: development of the Child Pornography Offender Risk Tool (CPORT). Law Hum Behav 39:416–29, 2015
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1037/lhb0000128&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=25844514&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 

27. 27.Eke AW, Helmus LM, Seto MC. A validation study of the Child Pornography Offender Risk Tool (CPORT). Sex Abuse, published online March 29, 2018. Available at: [https://journals.sagepub.com/doi/abs/10.1177/1079063218762434#](https://journals.sagepub.com/doi/abs/10.1177/1079063218762434#). Accessed March 22, 2019
    
    

28. 28.Krone T, Smith RG, Cartwright J, et al: Online child sexual exploitation offenders: a study of Australian law enforcement data. [http://crg.aic.gov.au/reports/1617/58-1213-FinalReport.pdf](http://crg.aic.gov.au/reports/1617/58-1213-FinalReport.pdf). Accessed June 8, 2018
    
    

29. 29.Osborn J, Elliott IA, Middleton D, Beech A: The use of actuarial risk assessment measures with UK internet child pornography offenders. J Aggress Conflict Peace Res 2:16–24, 2010
    
    

30. 30.Goller A, Jones R, Dittmann V, et al: Criminal recidivism of illegal pornography offenders in the overall population: a national cohort study of 4612 offenders in Switzerland. Adv Applied Sociol 6:9, 2016
    
    

31. 31.Faust E, Bickart W, Renaud C, Camp S: Child pornography possessors and child contact sex offenders: a multilevel comparison of demographic characteristics and rates of recidivism. Sex Abuse 27:460–78, 2015
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1177/1079063214521469&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=24556314&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 

32. 32.Beier KM, Neutze J, Mundt IA, et al: Encouraging self-identified pedophiles and hebephiles to seek professional help: first results of the Prevention Project Dunkelfeld (PPD). Child Abuse Negl 33:545–49, 2009
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1016/j.chiabu.2009.04.002&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=19765825&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 
    
    [Web of Science](http://jaapl.org/lookup/external-ref?access_num=000272262200007&link_type=ISI) 

33. 33.Beier KM, Grundmann D, Kuhle LF, et al: The German Dunkelfeld Project: a pilot study to prevent child sexual abuse and the use of child abusive images. J Sex Med 12:529–42, 2015
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1111/jsm.12785&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=25471337&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 

34. 34.Merdian HL, Thakker J, Wilson N, Boer DP: Assessing the internal structure of the COPINE scale. Psychol Crime & L 19:21–34, 2013
    
    

35. 35.Webb L, Craissati J, Keen S: Characteristics of internet child pornography offenders: a comparison with child molesters. Sex Abuse 19:449–65, 2007
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1177/107906320701900408&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=18008163&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 

36. 36.Reijnen L, Bulten E, Nijman H: Demographic and personality characteristics of internet child pornography downloaders in comparison to other offenders. J Child Sex Abus 18:611–22, 2009
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1080/10538710903317232&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=20183421&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 

37. 37.Garrington C, Chamberlain P, Rickwood D, Boer DP: Risk assessment of online child abuse material (CAM) offenders: a review of existing tools. J Crim Psychol 8:150–61, 2018
    
    

38. 38.Seto MC, Eke AW, Helmus LM. Child Pornography Offender Risk Tool (CPORT). [https://www.researchgate.net/project/Child-Pornography-Offender-Risk-Tool-CPORT](https://www.researchgate.net/project/Child-Pornography-Offender-Risk-Tool-CPORT). Accessed June 14, 2018
    
    

39. 39.Long MA, Alison L, Tejeiro R, et al: KIRAT: law enforcement's prioritization tool for investigating indecent image offenders. Psychol Pub Pol'y & L 22:12–21, 2016
    
    

40. 40.Hanson RK, Thornton D: Improving risk assessments for sex offenders: a comparison of three actuarial scales. Law & Hum Behav 24:119–36, 2000
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1023/A:1005482921333&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=10693322&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 
    
    [Web of Science](http://jaapl.org/lookup/external-ref?access_num=000085237200007&link_type=ISI) 

41. 41.Kingston DA, Yates PM, Firestone P, et al: Long-term predictive validity of the Risk Matrix 2000: a comparison with the Static-99 and the Sex Offender Risk Appraisal Guide. Sex Abuse 20:466–84, 2008
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1177/1079063208325206&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=18840901&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 

42. 42.Hanson RK, Harris AJR, Scott T, Helmus L. Assessing the risk of sexual offenders on community supervision: The Dynamic Supervision Project. [https://www.publicsafety.gc.ca/cnt/rsrcs/pblctns/ssssng-rsk-sxl-ffndrs/index-en.aspx](https://www.publicsafety.gc.ca/cnt/rsrcs/pblctns/ssssng-rsk-sxl-ffndrs/index-en.aspx). Accessed June 14, 2018
    
    

43. 43.Child Abuse and Neglect Reporting Act, A.B. 1775, 2014 Leg. (Cal. 2014).
    
    

44. 44.Babchishin KM, Merdian HL, Bartels RM, Perkins D: Child sexual exploitation materials offenders: a review. Eur Psychologist 23:130–43, 2018
    
    

45. 45.Ly T, Dwyer RG, Fedoroff JP: Characteristics and treatment of internet child pornography offenders. Behav Sci & L 36:216–34, 2018
    
    

46. 46.Seto MC, Ahmed AG: Treatment and management of child pornography use. Psychiatr Clin North Am 37:207–14, 2014
    
    

47. 47.Campbell-Fuller N, Craig LA: The use of olfactory aversion and directed masturbation in modifying deviant sexual interest: a case study. J Sex Aggress 15:179–91, 2009
    
    

48. 48.Chirban JT: Integrative strategies for treating internet sexuality: a case study of paraphilias. Clin Case Studies 5:126–41, 2006
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1177/1534650103261214&link_type=DOI) 

49. 49.Holoyda BJ, Kellaher DC. The biological treatment of paraphilic disorders: an updated review. Curr Psychiatry Rep 18:19, 2016
    
    

50. 50.Amelung T, Kuhle LF, Konrad A, et al: Androgen deprivation therapy of self-identifying, help-seeking pedophiles in the Dunkelfeld. Int'l J L & Psychiatry 35:176–84, 2012
    
    

51. 51.Briken P, Nika E, Berner W: Treatment of paraphilia with luteinizing hormone-releasing hormone agonists. J Sex Marital Ther 27:45–55, 2001
    
    [CrossRef](http://jaapl.org/lookup/external-ref?access_num=10.1080/00926230152035840&link_type=DOI) 
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=11224953&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 
    
    [Web of Science](http://jaapl.org/lookup/external-ref?access_num=000166306000005&link_type=ISI) 

52. 52.Greenberg DM, Bradford JM, Curry S, O'Rourke A: A comparison of treatment of paraphilias with three serotonin reuptake inhibitors: a retrospective study. Bull Am Acad Psychiatry Law 24:525–32, 1996
    
    [PubMed](http://jaapl.org/lookup/external-ref?access_num=9001750&link_type=MED&atom=%2Fjaapl%2Fearly%2F2019%2F04%2F15%2FJAAPL.003830-19.atom) 
    
    [Web of Science](http://jaapl.org/lookup/external-ref?access_num=A1996WB05000008&link_type=ISI) 

53. 53.Middleton D, Mandeville-Norden R, Hayes E: Does treatment work with internet sex offenders? Emerging findings from the Internet Sex Offender Treatment Programme (i-SOTP). J Sex Aggress 15:5–19, 2009
    
    

54. 54.Hanson RK, Bourgon G, McGrath R, et al. A five-level risk and needs system: maximizing assessment results in corrections through the development of a common language. [https://csgjusticecenter.org/nrrc/publications/a-five-level-risk-and-needs-system-maximizing-assessment-results-in-corrections-through-the-development-of-a-common-language](https://csgjusticecenter.org/nrrc/publications/a-five-level-risk-and-needs-system-maximizing-assessment-results-in-corrections-through-the-development-of-a-common-language). Accessed August 22, 2018
    
    

55. 55.Day A, Ross S, Casey S, et al. The intensity and timing of sex offender treatment. Sex Abuse, published online December 6, 2017. Available at: [https://journals.sagepub.com/doi/abs/10.1177/1079063217745069](https://journals.sagepub.com/doi/abs/10.1177/1079063217745069). Accessed March 22, 2019