Background Intimate partner violence (IPV) is a complex phenomenon and some research suggests that there are qualitatively distinct IPV types. However, little is known about the risk factors associated with different IPV types.

Methods Data from Violence against women: an European Union (EU)-wide survey, conducted by the EU Agency for Fundamental Rights was used. Latent class analysis (LCA) was employed to identify distinct IPV patterns based on the intensity of eight forms of violence by current partners (n=30 675). Multilevel multinomial logistic regression was used to examine individual and country-level risk factors associated with the outcome IPV patterns.

Results A five-class solution was selected based on the LCA results. Two classes encompassed severe coercive IPV: the intimate terrorism class (1.5%) also comprised extensive physical violence whereas the high coercive control class (2.0%) did not. The partner’s alcohol abuse, violent behaviour outside the relationship and the woman’s abuse in childhood were the main individual factors positively associated with IPV. The country’s gender equality levels were negatively associated with the odds of experiencing intimate terrorism (adjusted OR, aOR 0.35, 95% CI 0.21 to 0.56) and high coercive control (aOR 0.63, 95% CI 0.47 to 0.85) versus no IPV. Although the effects of most individual risk factors were found universally for all IPV patterns, the strongest associations were typically revealed for the intimate terrorism pattern.

Conclusion The results support the importance of coercive control as a factor differentiating between IPV types and also highlight the need to consider IPV typologies in research. Policy implications of the findings are discussed.