Digital community safety and crime prevention: insights from a rapid evidence assessment of online moderation practices

Purpose The rise of misinformation, online vigilantism and public distrust in digital spaces has created new challenges for crime prevention and public safety. The Queensland Police Service (QPS) launched the Digital Intelligence and Community Engagement (DICE Prevention) Project to address these issues through the Digital Community Safety Champions initiative, which supports online community leaders such as Facebook community and crime group moderators. This study aims to inform the development of training and resources for digital community leaders by identifying best practices for managing digital safety, misinformation and public communication. Design/methodology/approach A rapid evidence assessment (REA) was conducted, reviewing peer-reviewed and grey literature published between 2013 and up to February 2025. The full review was guided by eight questions that centred on: misinformation management, online moderation, conflict de-escalation, digital literacy, privacy and safety, community engagement and effective communication of crime and safety information. Four of those questions are the focus of this paper, centring on how to moderate and detect misinformation. Findings The REA identified that effective digital community safety depends on clear community guidelines, strong and diverse moderation, user education and trust-building strategies. Integration of AI with human moderation, digital literacy training and proactive engagement were essential. Collaboration with the police emerged as important for combating misinformation and promoting timely, accurate safety messaging. Originality/value This paper provides an applied synthesis of cross-disciplinary evidence on moderation and misinformation management relevant to crime and neighbourhood online groups. Rather than claiming conceptual novelty, this paper consolidates established insights for a policing–community safety context, highlighting where evidence is mixed or context-dependent and identifies priorities for evaluation of real-world interventions (including police–community partnership models).