Abstract Background Sexual violence (SV) is an important public health problem which may cause long-lasting health problems. SV in older adults remains neglected in research, policies and practices. Valid SV prevalence estimates and associated risk factors in older adults are currently unavailable. In this study we measured lifetime and past 12-months sexual victimisation in older adults living in Belgium, its correlates, assailant characteristics and the way that victims framed their SV experiences. Methods SV was measured using behaviourally specific questions based on a broad definition of SV. Participants were selected via a cluster random probability sampling with a random route finding approach. Information on sexual victimisation, correlates, assailant characteristics and framing was collected via structured face-to-face interviews with adults aged 70 years and older living in Belgium (community-dwelling, assisted living and nursing homes). Results Among the 513 participants, the lifetime SV prevalence was 44% (55% F, 29% M). Past 12-months prevalence was 8% (9% F, 8% M). Female sex and a higher number of sexual partners were associated with lifetime SV (p < .05), non-heterosexual sexual orientation with past 12-months SV (p < .05). Correlates identified to be linked to elder abuse and neglect in previous studies were not linked with SV in our sample. ‘Someone unknown’ was identified as most common assailant. Conclusions Sexual victimisation appears to be common in older adults in Belgium. Both correlates and assailant characteristics seem to differ from previous studies on elder abuse and neglect. Recognizing older adults as a risk group for sexual victimisation in research, policies and practices is of the utmost importance.