Understanding the motives behind femicides is crucial to design effective prevention strategies and to support women's self-determination, free from threats to their mental and physical integrity. We conducted a retrospective, multicenter forensic study involving 27 Italian Institutes of Legal Medicine, analyzing 1238 female homicides (1950-2023). Cases were classified as femicide or non-femicide female homicide according to the medico-legal definition of femicide as the murder due to the failure to recognize women's right to self-determination. Motives were categorized into 12 groups, and relevant anamnestic and circumstantial data were collected. Of the 1238 cases, 410 were identified as femicides, 395 as non-femicides, and 433 were excluded for insufficient information. Femicides were most frequently driven by jealousy/rejection (n = 185; 45.1 %) and separation/divorce (n = 144; 35.1 %), often in the context of intimate partner violence or inability to accept the end of a relationship. Non-femicides were predominantly associated with the perpetrator's psychiatric disorder and/or drug addiction (n = 126; 31.9 %), violent aggression without gender-related motive (n = 69; 17.5 %), victim's illness (n = 63; 15.9 %), or economic reasons (n = 58; 14.7 %). In the Italian context, most femicides stem from relational dynamics - particularly jealousy, rejection, and separation - whereas non-femicides are more often linked to psychiatric illness, substance abuse, or non-gender-related aggression. Incorporating these findings into clinical risk assessment protocols, especially in emergency and primary care settings, may help identify women at high risk and guide targeted prevention strategies.

Femicides are grievous, but little is known about risk factors and preventive measures. We present the results of a study conducted across 27 Italian Forensic Institutes. We analyzed 1,238 cases of female homicide and, adopting the definition of femicide as the murder due to the failure to recognize women's right to self-determination, we identified 410 cases as femicides and 395 as non-femicides Current partners were identified as aggressors in a much larger proportion of femicide cases (241 femicides vs. 145 non-femicides; odds ratio (OR) of femicide 2.46, 95 % CI 1.85-3.27), such association being more pronounced for ex-partners (102 vs. 11; OR 11.56, 95 % CI 6.10-21.92). Sharp weapons showed a higher frequency in femicides (168 vs. 140; OR 1.26, 95 % CI 0.95-1.68). Femicides were more often associated with bodies found in vehicles (31 vs. 9; OR 3.50, 95 % CI 1.64-7.45) and outdoor (68 vs. 43; OR 1.62, 95 % CI 1.08-2.45). There was an indication of femicides being more frequently associated with overkilling (87/323 vs. 71/324; OR 1.24, 95 % CI 0.88-1.76) and even more with lesions located in erogenous zones (94/316 vs. 70/325; OR 1.38, 95 % CI: 0.97-1.95). However, the strongest (though statistically imprecise) association emerged for overkilling in erogenous zones (12/398 vs. 6/389; OR 1.95, 95 % CI: 0.72-5.25). The number of lesions showed a nonlinear association with femicide likelihood. These findings offer forensic indicators that could contribute in predicting and potentially prevent femicide occurrence in a Western population such as the Italian one.