Criminal Police Perceptions of the Internet-Delivered Child Sexual Exploitation and Abuse Prevention Program Prevent It 2.0 in Germany,Portugal,and Sweden

Participants

Participants were recruited using a purposive sampling approach to ensure the inclusion of criminal police personnel with current experience working on CSEA cases. A total of 26 active criminal police investigators and supervisors from Germany, Portugal, and Sweden took part in the study. By including both investigators and supervisors we ensured a comprehensive picture of police perspectives: While investigators can contribute firsthand experience working in CSEA investigations, supervisors can offer their perspectives on organizational aspects. The participant group comprised 17 women and 9 men between 28 and 61 years (M = 43.57; SD = 9.06). All participants actively worked with CSEA cases, most of them focusing exclusively on offline CSEA (n = 10) or exclusively on online CSEA (n = 9), while 7 participants worked with both types of cases. A characterization of study participants is presented in Table 1.

OPEN IN VIEWER

Table 1:

Sociodemographic Characteristics of Focus Group Participants

Participant	Country	Role	Gender	Age range in years a	Area of investigation	Experience with CSEA cases in years
DE-I1	Germany	Investigator	Male	30-39	Both	7
DE-I2	Germany	Investigator	Female	40-49	Online	16
DE-I3	Germany	Investigator	Female	30-39	Both	7
DE-I4	Germany	Investigator	Female	20-29	Both	4
DE-I5	Germany	Investigator	Male	60-69	Both	9
DE-S1	Germany	Supervisor	Female	40-49	Offline	23
DE-S2	Germany	Supervisor	Female	40-49	Both	4
DE-S3	Germany	Supervisor	Female	30-39	Online	12
DE-S4	Germany	Supervisor	Female	40-49	Online	15
PT-I1	Portugal	Investigator	Male	50-59	Offline	18
PT-I2	Portugal	Investigator	Female	40-49	Offline	20
PT-I3	Portugal	Investigator	Female	50-59	Offline	18
PT-I4	Portugal	Investigator	Male	50-59	Offline	23
PT-I5	Portugal	Investigator	Female	50-59	Offline	14
PT-S1	Portugal	Supervisor	Female	50-59	Both	1
PT-S2	Portugal	Supervisor	Male	50-59	Both	10
PT-S3	Portugal	Supervisor	Female	40-49	Offline	17
PT-S4	Portugal	Supervisor	Male	50-59	Offline	16
SE-I1	Sweden	Investigator	Female	20-29	Online	7
SE-I2	Sweden	Supervisor b	Female	40-49	Online	10
SE-I3	Sweden	Investigator	Female	20-29	Online	6
SE-I4	Sweden	Investigator	Male	30-39	Offline	6
SE-S1	Sweden	Supervisor	Male	30-39	Offline	6
SE-S2	Sweden	Supervisor	Male	40-49	Online	15
SE-S3	Sweden	Supervisor	Female	40-49	Online	12
SE-S4	Sweden	Supervisor	Female	30-39	Online	9

Note. CSEA = Child sexual exploitation and abuse; DE = Germany; PT = Portugal; SE = Sweden.

a

Age ranges were used instead of specific ages to minimize risk of identification due to the small participant group. ^bThis Swedish participant, although working as a supervisor, joined the investigator focus group, a fact revealed only during the interview and therefore accepted.

Dataset Generation

Data were generated between October and November 2022—before the start of recruitment for the two Prevent It 2.0 trials—at criminal police departments in Portugal and Sweden, as well as at the University Medical Center in Hamburg-Eppendorf in Germany. The duration of the focus groups was between 1.5 and slightly over 2 hours. Each focus group was facilitated in the corresponding native language by one (MJ, Sweden) or two (JH, AD, Germany; VA, HG, Portugal) researchers from the PRIORITY consortium who had at least a Master’s degree and had completed training with a qualitative research expert. All were also involved in the Prevent It 2.0 trial in some capacity, which was known to study participants. Before each focus group, participants received information detailing the study’s aim, the voluntary nature of participation, their right to withdraw at any time, and the confidentiality of their responses. Written consent was obtained from all participants, and no compensation was provided for participation.

A semi-structured interview guide with open-ended initial questions was used to facilitate the discussion and an interactive dialogue between participants while allowing flexibility for narrative-generating follow-up questions to explore emerging topics. The interview guide consisted of three sections: (a) experiences and challenges associated with working with CSEA cases (Tselenti et al., 2025), (b) perceptions of individuals with a sexual interest in children (Hillert et al., 2025), and (c) perceptions of CSEA prevention, including a specific focus on the Prevent It 2.0 intervention (this study). Participants were introduced to the Prevent It 2.0 program via a single-page leaflet outlining its aim, target group, format, and content, and were then invited to perform a SWOT analysis (Benzaghta et al., 2021) of the program. To stimulate a meaningful discussion, they were given time to read and ask questions before identifying and writing down encouraging (internal strengths and external opportunities) and discouraging (internal weaknesses and external threats) aspects of Prevent It 2.0 that could impact whether it would achieve its objective to prevent CSEA. Once participants finished writing their reflections, facilitators invited them to share and discuss the aspects they identified with each other. The written notes served as prompts to support reflection and facilitate discussion and were not collected or analyzed. The analysis instead focused on the co-constructed interaction between participants, as our interest was the collective meaning-making process rather than individual pre-discussion views.

The focus group discussions were recorded, and all potentially identifying information in the audio was anonymized before the recordings were transcribed by a professional transcription service. After checking the transcriptions in the original language, they were translated into English by independent translation companies. The final transcripts contained no identifying information, and participants were indicated by a system of letters and numbers.

Data Analysis

The focus group transcripts were managed using the NVivo R1 software and analyzed following Braun and Clarke’s (2022) reflexive thematic analysis to identify shared meaning within the dataset. While data generation was informed by the SWOT framework, the analytic process was primarily inductive and reflexive. Together, this approach can be described as a framework-informed reflexive thematic analysis, indicating that while data generation was initially structured, coding and theme creation remained data-driven and prioritized data-based meaning. This involved active engagement and reflexive interpretation of the data to develop the themes and subthemes based on conceptual significance rather than the SWOT framework.

The data analysis began with becoming familiar with the data, followed by systematic coding of relevant parts. Initial coding was conducted by VA, followed by independent interpretative engagement by the first author, LH. The codes were then grouped into potential themes, representing overarching patterns, and subthemes, capturing distinct aspects within each theme. In line with Braun and Clarke (2022), no numeric threshold was applied to determine themes. Instead, themes were generated to convey patterns of shared meaning relevant to the study aim, with attention to how meaning was constructed in the interaction and recognizing that a concept expressed by a single participant can offer valuable insights. Themes and subthemes were refined through regular discussions between LH and senior author AD to enhance the quality of the analysis and ensure a rich and nuanced interpretation of the data. Finally, each theme was clearly labeled and defined to develop the narrative presented in this paper.

Although focus groups were conducted separately with investigators and supervisors, the data were analyzed collectively. This approach reflects the aim of identifying shared meaning across the data rather than comparing role-specific perspectives, while ensuring that both frontline and organizational perspectives were represented in the data.

Reflexivity

All researchers involved in this study were engaged in Prevent It 2.0 in various roles, which inevitably influenced both dataset generation and analysis. While this involvement brought valuable contextual knowledge, it also called for continuous critical awareness of how researchers’ roles and assumptions might shape participant responses and data interpretation. Participants may have been inclined to provide responses aligning with what they assumed facilitators valued. To mitigate this, we actively encouraged constructive feedback, inviting participants to identify and discuss weaknesses and threats of the program. To further maintain analytical rigor, the team used regular reflexive discussions throughout the research process to ensure transparency, self-examination, and thoughtful interpretation.

In addition, the integration of the SWOT framework in the interview guide shaped the focus group discussions and directed participants toward identifying strengths, weaknesses, opportunities, and threats of Prevent It 2.0, aligning with the study’s aim to explore perceptions of these aspects. This enabled a focus on these perspectives while allowing attention to be paid to the way participants interpreted these categories. Nonetheless, integrating the framework in the interview guide may have limited exploration of interrelated or cross-category factors. The group format, together with facilitator follow-up questions, helped mitigate this by encouraging participants to build on and challenge each other’s contributions and elaborate on underlying relationships between factors. In addition, reflexive attention during the analysis ensured that themes were developed from the data rather than being limited by the SWOT structure, allowing for nuanced insights.

Theme 1: The Format Reduces Barriers but Imposes Demands on Participants

Criminal police officers perceived not only strengths but also weaknesses and threats in Prevent It 2.0’s anonymous internet-delivered format. On the one hand, they recognized its potential to lower barriers for accessing and engaging with prevention offers; on the other hand, they identified format-specific challenges and demands on participants.

Theme 2: The Framework Can Foster Credibility, but May Cause Disruptions

Criminal police officers revealed a conflicted understanding of the organizational framework in which the intervention is embedded, framing it as both a strength and a source for a potential threat.

Theme 3: Expected Increase in Awareness and Support Amid Concerns About Participants’ Motives

Criminal police officers expressed diverse expectations about the impact of Prevent It 2.0. On the one hand, they expressed optimism that it would support prevention efforts, pointing to expected future benefits rather than only describing a current internal strength or external opportunity. On the other hand, they identified potential external threats related to concerns that participants might “misuse” the program.

Theme 4: Officers are Willing to Collaborate, but Success Depends on Public Reception

Criminal police officers expressed a strong willingness to collaborate with program providers—indicating potential for referrals—but also emphasized that the program’s success is closely tied to how it is received by the public, representing both opportunity and threat.

Strengths: Flexibility, Anonymity, and EU Association Foster Engagement

Officers identified that the flexible, low-threshold, and anonymous design of Prevent It 2.0 is particularly valuable in engaging this hard-to-reach population and overcoming help-seeking barriers. Anonymity was regarded as a crucial strength in engaging individuals at risk of CSEA offending, who are often deterred from seeking help due to concerns about legal consequences or stigma (Insoll et al., 2024; Jahnke, 2018). Consistent with prior research (Borghouts et al., 2021; Woolderink et al., 2015), this study also reinforces the value of privacy guarantees in this context. To effectively reach this security- and privacy-cautious target group, it is valuable to concretely communicate the measures taken to ensure anonymity and data privacy. In addition, the ability to access the program anytime, from any location, was deemed beneficial for engaging participants, which is especially important in rural areas where in-person mental health care is limited or entirely unavailable (Gaebel et al., 2009; Saurman et al., 2015). This reflects previous research highlighting the flexibility of digital mental health interventions (Borghouts et al., 2021; Hillert et al., 2024) and aligns with police diversion research indicating the importance of timely program access to engage those referred by the police (Blais et al., 2022).

In addition, the program’s multinational scope and EU association were considered advantageous as officers felt that these enhanced the credibility, legitimacy, and trust in the intervention. In line with Hensell (2022), EU-funded projects may benefit from being associated with the European Commission’s established authority and strategic engagement, which could increase participant confidence and further reduce participation barriers. While the specific impact of cross-country implementation has not yet been tested for CSEA prevention programs, it likely enables larger sample sizes and can offer insights into program efficacy across different cultural and legal contexts. The evaluation of Prevent It 2.0 across three EU member states will provide valuable insights in this context.

Weaknesses: Inclusivity, Usability, Provider Well-Being, and Ethical Considerations

Despite these strengths, officers identified several weaknesses that could limit the program’s impact. One perceived limitation was the exclusion of those under 18. Officers viewed this as a barrier for early intervention, especially in light of evidence that a considerable proportion of CSEA offenses is committed by minors (Letourneau et al., 2017; Mathews et al., 2024). This highlights the need for future research into the feasibility and efficacy of adopting internet-delivered prevention programs for adolescents at risk of CSEA, including potential adaptations to ensure appropriateness.

Consistent with prior studies on mobile and eHealth interventions in forensic contexts (Aarts et al., 2025; Kip & Bouman, 2021), officers also flagged usability as a concern, particularly regarding individuals with lower literacy or cognitive capacities. Research has linked lower education levels to higher dropout rates and poorer outcomes in online and self-help interventions (Martinez et al., 2008; Rozental et al., 2017; Salomonsson et al., 2020). However, findings on cognitive abilities among individuals at risk of committing CSEA remain mixed: Some studies report lower IQ scores among pedophilic individuals who offended (Franke et al., 2019), while others suggest that these differences may be less pronounced in non-forensic samples (Jahnke et al., 2022). This raises the question of whether certain subgroups of at-risk individuals are actually able to navigate and benefit from a written, internet-delivered program. Future studies should investigate how educational and cognitive factors influence engagement and outcomes in internet-delivered CSEA prevention programs, and explore different strategies—such as simplified language, interactive attention-engaging elements, and platform use tutorials (Friedman & Bryen, 2007; Kip & Bouman, 2021; Wild et al., 2019)—to improve usability.

While anonymity was generally considered a strength, officers noted that it restricts providers’ ability to assess or intervene when participants disengage. This reiterates concerns that online settings hinder intervention in cases of disengagement and uncertainty as to whether this stems from technical issues, the treatment, or crises (e.g., suicidality; Rummell & Joyce, 2010). The dual perception of anonymity highlights the complexity of designing interventions for CSEA prevention: The very feature that can help facilitate access (strength) can also create challenges for monitoring and following up with participants (weakness). Officers also highlighted the emotional toll on providers. Working with this population has been associated with compassion fatigue, vicarious trauma, secondary traumatic stress, burnout, and a heightened sense of responsibility felt toward potential victims and society (Baum & Moyal, 2020; Hardeberg Bach & Demuth, 2018; Raymond et al., 2023). The anonymous online format may aggravate this burden by increasing the perceived expectation of constant availability (Feijt et al., 2018), while limiting the ability to intervene or report offenses. It can also complicate legal responsibilities, leaving providers uncertain how to respond when risk is unclear (Harris & Birnbaum, 2015). To address these challenges, clear guidelines and training are essential to help providers assess risk situations and navigate mandatory reporting laws. In addition, providers should receive adequate support, including supervision, access to mental health resources, and guidance for navigating difficult situations (Evans & Ward, 2019). In the case of Prevent It 2.0, all counselors participated in weekly joint supervision with a licensed specialist in forensic psychology.

Finally, officers also questioned the absence of personal interaction between participants and counselors. Although some studies suggest that online formats can increase therapist-client contact frequency, communication, and even perceived relationship quality (Aarts et al., 2025; Feijt et al., 2018; Mishna et al., 2017), others underscore the risk of miscommunication due to the absence of verbal and nonverbal cues (Harris & Birnbaum, 2015; Stoll et al., 2020). Specific training should be provided to actively prepare providers to facilitate meaningful relationships and teach them how to communicate emotions, empathy, genuineness, and openness in online settings. In addition, future studies could explore blended care models, which integrate online and in-person elements, as an approach to balance the benefits and limitations of both modalities (Kip & Bouman, 2021; Kip et al., 2018).

Opportunities: Interdisciplinary Collaboration and Targeted Outreach

Officers identified several external opportunities the Prevent It 2.0 program can leverage to expand CSEA prevention and address gaps. Most notably, they expressed a willingness to support recruitment efforts for the program. This signals an important shift from a shared understanding toward a more active, collaborative role in prevention where law enforcement works alongside researchers and practitioners. Even when treatment becomes available after conviction, CSEA cases typically take several months to be processed (Quas et al., 2024), leaving a critical gap during which those under investigation may experience substantial psychological distress, as well as shame, uncertainty, and isolation, but may also be especially motivated to seek help. Interdisciplinary collaboration could present a key opportunity to reach this population earlier. Research highlights that insufficient training and lack of leadership support can leave officers uncertain how to fit referrals into their daily schedules (Blais et al., 2022; Dickson-Gomez et al., 2022; Hoke et al., 2020), while clear protocols and supervisor endorsement foster collaboration (Eilfgang et al., 2024; Hoke et al., 2020; Shefner et al., 2024). Building on this, our study suggests that sharing flyers, links, or QR codes with individuals suspected of CSEA could be integrated into police workflow when these resources are readily available and distributed in accessible formats. Consistent with prior research, engaging supervisors may further facilitate collaboration and ensure consistent implementation. Program providers should therefore use proactive outreach—such as training workshops or presentations—to distribute information within police departments, clarify referral processes, and secure leadership support.

Consistent with a recent review (Roche et al., 2026), officers also suggested using various strategies to recruit for CSEA prevention. Recruitment opportunities were identified in targeting spaces frequented by at-risk individuals, such as specific darknet fora or pornography sites, which could help reach the secondary prevention potential of the program. Previous research supports that individuals seeking to stop their CSAM use can be reached through the darknet (Nurmi et al., 2024), a strategy successfully employed in the first Prevent It trial (Lätth et al., 2022). In addition, a report by Protect Children Finland (2024) found that 77% of individuals who searched for CSAM on the darknet had previously encountered it on clearnet sites, primarily on pornography platforms and social media. In this context, officers recommended warning banners and help-seeking messages. Recent studies support this approach, showing that warning and referral messages on pornography sites can not only deter individuals from using barely legal pornography but also promote engagement with linked prevention resources (Prichard et al., 2024; Scanlan et al., 2024). Finally, officers suggested outreach campaigns in health care and public, which have successfully been employed in in-person prevention initiatives such as Prevention Project Dunkelfeld (Beier, 2016; Beier et al., 2009), and collaboration with other CSEA prevention stakeholders.

Threats: Legal, Technological, and Sociocultural Challenges

Finally, officers identified potential external threats that could undermine the program’s impact, reputation, and ability to recruit and maintain participants. While the multinational scope was seen as a strength, officers also cautioned that it introduces complexity. Implementing across different legal frameworks and cultural contexts demands careful cultural adaptation and provider training to ensure the intervention’s acceptability across diverse settings (Evlat et al., 2021; Patel et al., 2011). Consistent with this approach, Prevent It 2.0 was culturally adapted (Azevedo et al., 2025), and all counselors completed four days of joint preparatory training.

Although our analysis was conducted across the full dataset and did not aim to assess national differences, one noteworthy nuance emerged: Legal frameworks were identified as a possible threat only in Portugal, suggesting that this may be a context-specific challenge, shaped by the specific legal framework. Portuguese officers pointed out that mandatory reporting laws—such as in Portugal—can create a barrier to participant engagement. This concern is supported by prior research (Kewley et al., 2023; Seto et al., 2024; Stephens et al., 2021) and reflected in the experience from the Portuguese Prevent It 2.0 trial, where recruitment had to be suspended for several months to assess how to navigate the reporting obligations. To facilitate recruitment, questions regarding CSEA behavior had to be dropped from all assessments in the Portuguese arm before resuming the trial (see Maimone et al., 2025, for a full study protocol). In addition, a protocol for handling offense disclosures was put into place together with the University’s legal department and the external study monitor, and a reminder was added at the beginning of each module in the Portuguese arm to remind participants of the obligation to report in case they disclosed any CSEA behavior (i.e., in worksheets or in messages to the counselors). This underscores how legal constraints can complicate implementing and delivering preventive efforts and highlights the importance of tailoring prevention strategies to national legal frameworks.

Another concern centered on “unintended” use of the program by those under legal investigation. Officers noted that although offering treatment during investigation might provide support for a vulnerable population, it could result in some individuals participating primarily to appear favorable in court, rather than from genuine motivation. Indeed, experience from the Prevent It 2.0 trial—particularly in Germany—showed several participants asking for participation certificates to present in court. And while initial evidence suggests only minor differences in treatment progress between mandated and voluntary treatment for men who committed CSEA (von Franqué & Briken, 2021), further research is needed to explore how such motivations influence treatment engagement and outcomes for those currently under investigation. Building on research suggesting the value of tailoring eHealth interventions (Kip & Bouman, 2021), individually adapting programs to first target motivation, for example, using motivational interviewing techniques, may be a promising strategy.

Platform unreliability was identified as another potential threat that could disrupt treatment and cause frustration, echoing prior research on the negative impact of technical issues on program engagement and therapeutic alliance (Borghouts et al., 2021; Lopez et al., 2019). Officers also worried that hackers or data breaches could compromise participant anonymity and weaken trust in the program. This concern is widely recognized in digital mental health literature (Inkster et al., 2023; Lustgarten et al., 2020; Stoll et al., 2020) and seems prevalent with forensic populations (Kip et al., 2018). Although security was a key interest during the development of the Iterapi platform (Vlaescu et al., 2016), where Prevent It 2.0 is hosted, and several efforts were taken to enhance platform security, these concerns emphasize that constant testing and updating of platforms is essential to maintain trust and program integrity. In addition, accessible resources, such as Frequently Asked Questions (FAQs) and responsive technical support, could help participants resolve issues quickly and remain engaged with the program.

The public discourse surrounding individuals who commit CSEA was mentioned as another threat. Consistent with prior research (Grady et al., 2019; Insoll et al., 2024; Jahnke et al., 2015; Levenson & Grady, 2019), officers acknowledged that stigma can be a meaningful factor for recruitment. They noted that negative public attitudes—such as skepticism about the potential of change (Guastaferro et al., 2023; Höing et al., 2016) and a preference for punitive measures (Rosselli & Jeglic, 2017)—may discourage individuals from seeking help. Media coverage, while potentially helpful to bring attention to the program, often uses emotionally negative language that reinforces punitive attitudes (Harper & Hogue, 2015; Stelzmann et al., 2020) and might hinder help-seeking (Levenson & Grady, 2019). It is therefore vital to raise public awareness about the importance of secondary CSEA prevention to promote support. In general, the relationship between stigma and help-seeking is not yet fully understood and internalized stigma may even encourage help-seeking. A recent study found that individuals who experience higher levels of internalized stigma have more positive attitudes toward seeking treatment (Jahnke et al., 2024). Future research is needed to better understand the relationship between stigma and help-seeking behavior.

Finally, officers echoed concerns identified in prior research that an association with the police may deter potential participants out of police distrust or fear of legal consequences (Barberi & Taxman, 2019; Costigan et al., 2022). Future research should examine how this dynamic affects help-seeking behavior and identify strategies for maintaining trust, such as clearly defining the limits of police involvement and structuring collaborations to avoid misunderstandings and mitigate negative impacts on recruitment.

Limitations

The study should be interpreted in light of its limitations. As noted in the Reflexivity section, the involvement of focus group facilitators in Prevent It 2.0 may have limited officers’ willingness to critique the program. To address this, participants were explicitly asked to offer critical perspectives and reflect on weaknesses and threats of the program. Notably, participants followed through, discussing multiple weaknesses and threats, and not just strengths and opportunities. In addition, participants only received a flyer about Prevent It 2.0 and did not see the actual program website, which was still being finalized, and it might be questioned whether this provided them with sufficient understanding to give meaningful feedback. To mitigate this, participants were given time to read and ask questions to the facilitators, as experts on the program, to provide an adequate base for discussion. Finally, the study did not aim to systematically compare national differences, and additional research would be needed to examine national specificities in depth. The observed nuance with the legal framework in Portugal may therefore not be taken as a definitive cross-country comparison but instead as a context-specific insight that warrants further examination.