[Geographical variability in violence against women in Spain].

OBJECTIVES To describe the prevalence of intimate partner violence and non-partner violence against women in Spain overall and by autonomous regions, as well as the characteristics of this violence, and to study the association between the prevalence of intimate partner violence and the frequency of formal complaints and deaths. METHODS A cross-sectional survey was performed through a self-administered questionnaire in 11,475 women attending primary care from 2006 to 2007. Multistage stratified sampling was carried out in a representative sample in each autonomous region. Data on intimate partner and non-intimate partner violence, formal complaints to the police and deaths due to intimate partner violence were gathered. A descriptive analysis was conducted. Correlations between the prevalence of intimate partner violence and police reports and deaths from this cause were determined through Pearson's correlation coefficient. RESULTS A total of 24.8% (23.4%-26.3%) of the sample reported having experienced intimate partner violence at some time. The highest prevalence of intimate partner violence was found in Ceuta and Melilla (40.2%) and the Balearic Islands (32.5%) and the lowest in Cantabria (18%). Fifteen percent of the sample reported having experienced non-partner violence at some time, and the most frequent type was that committed by a relative. Again, the highest prevalence of non-partner violence was found in Ceuta and Melilla and the Balearic Islands and the lowest in Cantabria. A significant positive linear correlation was found between intimate partner violence in the previous year and the rate of reports of intimate partner violence in the autonomous regions (r=0.398; p=0.005). CONCLUSIONS The prevalence of intimate partner violence varies from one autonomous region to another. The reasons for this variation should be further explored in future studies.