Developmental and Psychological Profiles, Risk Assessment, and Recidivism in Online Versus Contact Child Sexual Offending: A Systematic Review

Background: Online child sexual offending, particularly offences involving child sexual abuse material (CSAM), represents a growing area of concern. However, uncertainty remains regarding whether individuals who engage in online offences differ meaningfully from those who commit contact child sexual abuse (CSA) in their developmental histories, psychological characteristics, risk profiles, and treatment needs. Methods: A systematic search was conducted in Scopus, Web of Science, and PubMed in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Studies examining developmental, psychological, criminological, risk assessment, recidivism, or treatment-related characteristics of individuals involved in online and/or contact child sexual offending were eligible. Twenty-nine studies were included and synthesised narratively by developmental and clinical domain. Results: Evidence indicated both overlap and meaningful differentiation between online and contact offending. Childhood adversity was reported across offence groups, although sexual victimisation histories tended to be less prevalent among online than contact offenders. Broad personality traits showed limited differentiation, whereas clinician-rated psychopathy and antisocial characteristics were generally more pronounced among contact offenders. Online and mixed offenders were more consistently characterised by atypical sexual interests and broader paraphilic or hypersexual patterns, although substantial within-group heterogeneity was evident. Some studies supported the relevance of CSAM-specific instruments and dynamic risk measures, but the available evidence did not establish their consistent superiority over static tools developed primarily for contact offenders. Recurrence findings from three treatment cohorts were synthesised narratively because differences in outcome definitions, ascertainment methods, denominators, and observation periods prevented meaningful pooling or direct comparison of CSAM-related and contact offending. Limited uncontrolled prospective evidence suggested reductions in offence-supportive cognitions and psychological distress, alongside substantial attrition and continued CSAM-related offending in some samples. Conclusions: Developmental adversity, atypical sexual interests, antisociality, offence history, and self-regulatory difficulties appear to combine differently across online, contact, and mixed offending pathways. Risk assessment findings should be interpreted in relation to the population, outcome, and setting in which each instrument was evaluated. Further validation is needed to establish whether CSAM-specific instruments and dynamic measures improve prediction beyond existing approaches. The evidence base remains constrained by heterogeneous definitions, predominantly male and forensic samples, variable follow up periods, and limited prospective treatment research. Larger longitudinal studies using harmonised definitions and validated outcome measures are required to clarify developmental pathways, transitions between online and contact offending, and treatment effectiveness.