Objective. To investigate whether sexual assaults are more likely to co-occur with some types of abuse rather than others in violent intimate relationships. Design. Cross-sectional study. Setting. A self-administered questionnaire was sent to all Norwegian women's shelters. Population. Women seeking refuge at Norwegian women's shelters in 2002 and 2003. Methods. Sexual assault and experiences of intimate partner violence were measured using the Severity of Violence against Women Scale (SVAWS) and psychological violence was measured using the Psychological Maltreatment of Women Inventory (PMWI). Main outcome measures. Student's t-test analyses were performed between the mean values of the different acts of reported violence, and linear regression analyses were used to examine the association between sexual violence and the other forms of violence reported. Results. Sexual violence correlated significantly with the other eight categories in SVAWS, and with violence directed at the pregnant woman's abdomen and psychological violence in PMWI. When we adjusted all categories for each other by linear regression analysis, sexual intimate partner violence was significantly associated with hair pulling, arm twisting, spanking or biting, dominance and isolation abuse and violence directed at the pregnant woman's abdomen. Conclusion. Sexual assaults are more likely to co-occur with some types of physical and psychological violence than with others. This knowledge may be important for improving our understanding of sexual violence in intimate partner relationships and in the efforts to detect intimate partner violence. Bruises, loss of hair and bite marks may suggest that sexual acts were committed against the victim's will.