Purpose: There is a critical need to clarify the long-term effects of anti-stigma interventions. The study aimed to assess the long-term effects of repeated filmed social contact or internet-based self-study on mental health-related stigma through a randomised controlled trial with 2-year follow-up.

Methods: We randomly allocated 259 university or college students to a filmed social contact group, an internet-based self-study group, or a control group. The filmed social contact and internet-based self-study groups each received a 30-min initial intervention followed by emailed interventions every 2 months over a 12-month period. The Japanese version of the Reported and Intended Behaviour Scale (RIBS-J) and the Mental Illness and Disorder Understanding Scale (MIDUS) were used to assess behaviour, behavioural intentions (attitudes), and knowledge regarding mental health.

Results: Of the 259 original participants, 187 completed the 24-month follow-up assessment. Mean scores for the RIBS-J future domain and MIDUS peaked at 1 month after initial intervention. Compared with baseline, at 24-month follow-up, we found a significant difference in RIBS-J future domain scores between the filmed social contact and control groups at 24-month follow-up (B = 0.95, 95% CI = 0.01,1.90, p = 0.049), while MIDUS scores in the filmed social contact group (B = - 4.59, 95%CI = - 6.85, - 2.33, p < 0.001) and the internet-based self-study group (B = - 4.51, 95%CI = - 6.86, - 2.15, p < 0.001) significantly decreased compared with the control group.

Conclusion: While outcome scores peaked at 1 month after initial intervention, results suggest that filmed social contact might have a long-term effect on behavioural intentions, and both filmed social contact and internet-based self-study may contribute to improved knowledge of mental health.