Background: Can deliberate interaction between the public and persons affected by leprosy reduce stigmatization? The study described in this paper hypothesises that it can and assesses the effectiveness of a ‘contact intervention’. Methods/Principal Findings: This cluster-randomized controlled intervention study is part of the Stigma Assessment and Reduction of Impact (SARI) project conducted in Cirebon District, Indonesia. Testimonies, participatory videos and comics given or made by people affected by leprosy were used as methods to facilitate a dialogue during so-called ‘contact events’. A mix of seven quantitative and qualitative methods, including two scales to assess aspects of stigma named the SDS and EMIC-CSS, were used to establish a baseline regarding stigma and knowledge of leprosy, monitor the implementation and assess the impact of the contact events. The study sample were community members selected using different sampling methods. The baseline shows a lack of knowledge about leprosy, a high level of stigma and contrasting examples of support. In total, 91 contact events were organised in 62 villages, directly reaching 4,443 community members (mean 49 per event). The interview data showed that knowledge about leprosy increased and that negative attitudes reduced. The adjusted mean total score of the EMIC-CSS reduced by 4.95 points among respondents who had attended a contact event (n = 58; p <0.001, effect size = 0.75) compared to the score at baseline (n = 213); for the SDS this was 3.56 (p <0.001, effect size = 0.81). About 75% of those attending a contact event said they shared the information with others (median 10 persons). Conclusions/Significance: The contact intervention was effective in increasing knowledge and improving public attitudes regarding leprosy. It is relatively easy to replicate elsewhere and does not require expensive technology. More research is needed to improve scalability. The effectiveness of a contact intervention to reduce stigma against other neglected tropical diseases and conditions should be evaluated.