Abstracts

INTRODUCTION

Violence against women has been recognized as a public health isse since the 1990s¹1. Hartigan P. PAHO focuses on the problem of violence against women. Rev Panam Salud Publica 1997; 2(4): 290-4., and although there have been advancements in public policies, there is still much to do especially in regard to the prevention of the most severe forms. This type of violence, now called gender-based violence, occurs on a continuum whose extreme situations include torture, mutilations, sexual violence and gender-based murder, defined as femicide²2. Antony C. Compartilhando critérios e opiniões sobre femicidio/feminicidio. In: CLADEM. Contribuições ao Debate sobre a Tipificação Penal Femicídio/Feminicídio. Lima: Cladem [Internet] 2012. Disponível em: http://www.compromissoeatitude.org.br/wp-content/uploads/2013/10/CLADEM_TipificacaoFeminicidio2012.pdf. (Acessado em 21 de julho de 2012).
http://www.compromissoeatitude.org.br/w...

3. Carcedo A (coord.) No olvidamos ni aceptamos: femicídio em Centroamérica, 2000-2006. San José: Associación Cetroamericana de Información y Acción; 2010. ^{- 4}4. Radford J, Russell DEH. Femicide: the politics of woman killing. New York: Twayne; 1992..

The term femicide was first used in 1976 by Diana Russell in an International Tribunal on Crimes against Women⁴4. Radford J, Russell DEH. Femicide: the politics of woman killing. New York: Twayne; 1992.. She conceptualized femicide as murders based on gender, in which women are killed solely because they are women.

In the Americas, there was a discrepancy in the translation/adaptation of the concept. Countries like Mexico and Chile began using the term femicide to characterize any murders of women, and the term feminicide to characterize gender-based murders. On the other hand, activists of Central America have used the term femicide to characterize misogynist murders of women. In any case, both expressions can be used with the same meaning³3. Carcedo A (coord.) No olvidamos ni aceptamos: femicídio em Centroamérica, 2000-2006. San José: Associación Cetroamericana de Información y Acción; 2010. ^{, 5}5. Carcedo A, Sagot M. Femicidio en Costa Rica 1990-1999. Colección Teórica, 1. Washington: Organización Panamericana de la Salud; 2000. ^{, 6}6. Monarrez Fragoso J. Feminicidio serial em Ciudad Juarez: 1993-2001. Debate Fem 2002; 13 (25): 279-305. Disponível em; http://www.debatefeminista.com/descargas.php?archivo=femici779.pdf&id_articulo=779. (Acessado em 21 de julho de 2012).
http://www.debatefeminista.com/descargas... .

The concept of femicide is political and represents a continuous and systemic violation of women's rights, representing a form of domination, exercise of power and control⁷7. Campbell JC, Glass N, Sharps PW, Laughon K, Bloom T. Intimate partner homicide: review and implications of research and policy. Trauma Violence Abuse 2007; 8(3): 246-69. ^{, 8}8. Taylor R, Jasinski JR. Femicide and the feminist perspective. Homicide Stud 2011; 15(4): 341-62.. In recent years, several countries have introduced criminal laws against this type of crime, whether using the term femicide or treating them as feminicides. Among these countries are El Salvador, Costa Rica, Guatemala, Chile, Peru, Mexico and Nicaragua⁹9. Munévar DI. Delito de femicidio: muerte violenta de mujeres por razones de género. Est Socio-Juríd Bogotá (Colombia) 2012; 14(1): 135-75..

Studies show that approximately 60 - 70% of female homicides may be considered femicides, in addition to the fact that more than a third of women murdered are killed by their partners, while only 3% of male homicides are carried out by women, and the majority occurs in situations of self-defense³3. Carcedo A (coord.) No olvidamos ni aceptamos: femicídio em Centroamérica, 2000-2006. San José: Associación Cetroamericana de Información y Acción; 2010. ^{, 5}5. Carcedo A, Sagot M. Femicidio en Costa Rica 1990-1999. Colección Teórica, 1. Washington: Organización Panamericana de la Salud; 2000. ^{, 7}7. Campbell JC, Glass N, Sharps PW, Laughon K, Bloom T. Intimate partner homicide: review and implications of research and policy. Trauma Violence Abuse 2007; 8(3): 246-69..

Violence against women has been more prevalent in situations of gender inequality, in places and scenarios where the patriarchal order is more rigid. Feminist scholars believe that misogyny, sexism and control of women arising from the patriarchal system are predisposing factors to violence and femicide¹⁰10. Wright MW. Necropolitics, narcopolitics, and femicide: gendered violence on the Mexico-U.S. border. Signs (Chic) 2011; 36(3): 707-31..

On the other hand, non-feminist authors consider that femicides are perpetrated regardless of gender perspective, and that they increase in the presence of structural violence¹¹11. Felson RB, Lane KJ. Does violence involving women and intimate partners have a special etiology? Criminology 2010; 48(1): 321-38. ^{, 12}12. Frye V, Wilt S. Femicide and social desorganization. Violence Against Wom 2001; 7(3): 335-51.. Femicide occurs at higher rates in areas where the state is not present. Besides, many offenders have involvement with drug trafficking and other criminal activities³3. Carcedo A (coord.) No olvidamos ni aceptamos: femicídio em Centroamérica, 2000-2006. San José: Associación Cetroamericana de Información y Acción; 2010. ^{, 13}13. Gondolf EW, Shestakov D. Spousal homicide in Russia versus the United States: preliminary findings and implications. J Fam Violence 1997; 12(1): 63-74..

Regional and international differences have been observed in in the rates and types of femicide. In the United States and Canada, the majority of femicides is perpetrated by partners or ex-partners; in Mexico, they have been linked to drug trafficking and urban violence; in Guatemala, femicide rates reached epidemic levels that remain due to impunity for perpetrators¹⁰10. Wright MW. Necropolitics, narcopolitics, and femicide: gendered violence on the Mexico-U.S. border. Signs (Chic) 2011; 36(3): 707-31. ^{, 14}14. Carey D Jr, Torres MG. Precursors to femicide: Guatemalan women in a vortex of violence. Latin Am Res Rev 2010; 45(3): 142-64. .

Brazil, a country of continental dimensions, shows marked differences in the coefficients of female mortality due to aggression. Over the past 30 years, more than 90,000 women were murdered. Data available on the Violence Map shows that female homicides increased from 2.3 per 100,000 in 1980 to 4.4 in 2010¹⁵15. Brasil. Mapa da Violência 2012: Os novos Padrões da Violência Homicida no Brasil. Caderno complementar 1. Homicídio de mulheres no Brasil. 2012. Disponível em: http://www.mapadaviolencia.org.br. (Acessado em 21 de julho de 2012).
http://www.mapadaviolencia.org.br... . Although these data have not been standardized, most deaths correspond to younger women.

In this scenario, Rio Grande do Sul has an average of 3 deaths per 100,000 women. Rio Grande do Sul in a Brazilian region with a predominantly agro-pastoral economic activity, and traditional gaucho culture follows a patriarchal model, anchored in the notion of honor¹⁶16. Peristiany JG, Pitt-Rivers J (eds.). Honor and Grace in Anthropology. Cambridge: Cambridge University Press; 1992.. The principles of the culture of honor reinforce the differences in gender roles, leaving the men to care for the moral and sexual behavior of women. In this context, women belong to men who will correct them through violence when they violate social norms of gender, such as being unfaithful to their husbands, leaving them or even showing to be too independent¹⁷17. Biglia B, San Martin C. Estado de wonderbra: entretejiendo naraciones feministas sobre las violencias de gênero. Barcelona: Virus Editorial; 2007..

This study aimed to analyze the female mortality due to aggression in the state of Rio Grande do Sul, relating it to socioeconomic, demographic as health factors, as well factors related to access to services.

METHODOLOGY

This is an ecological study that related female mortality due to agression according to the microregions of the state of Rio Grande do Sul, Brazil, with demographic, socioeconomic, health and access to services variables for the period from 2003 to 2007. It is part of a larger research called "Femicide and Gender-Based Killings in Rio Grande do Sul"* and it continues the ecological analysis of female mortality due to aggression in the Units of the Brazilian Federation¹⁸18. Meneghel SN, Hirakata VN. Femicides: female homicide in Brazil. Rev Saúde Pública 2011; 45(3): 564-74..

As in the previous study¹⁸18. Meneghel SN, Hirakata VN. Femicides: female homicide in Brazil. Rev Saúde Pública 2011; 45(3): 564-74., the total of female homicides was used as an "indicator or marker of femicide", considering that 60 to 70% of female deaths due to aggression are femicides³3. Carcedo A (coord.) No olvidamos ni aceptamos: femicídio em Centroamérica, 2000-2006. San José: Associación Cetroamericana de Información y Acción; 2010. ^{, 5}5. Carcedo A, Sagot M. Femicidio en Costa Rica 1990-1999. Colección Teórica, 1. Washington: Organización Panamericana de la Salud; 2000. ^{, 7}7. Campbell JC, Glass N, Sharps PW, Laughon K, Bloom T. Intimate partner homicide: review and implications of research and policy. Trauma Violence Abuse 2007; 8(3): 246-69.. Working with the total female deaths due to aggression may overestimate the actual values of the event. However, it is believed that this possible overestimation can compensate the ill-defined diagnoses, in which female homicides were classified as other types of violent deaths (suicides or accidents). This may occur even in Rio Grande do Sul, where the quality of the Mortality Information System (SIM) of the Unified Health System (SUS) is considered good and the coverage is close to 100%¹⁹19. Brasil. Ministério da Saúde. Sistema Nacional de Vigilância em Saúde. Rio Grande do Sul. Brasília: Ministério da Saúde; 2009..

The coefficient of female mortality due to aggression was used as the dependent variable, including those corresponding to the X85 - Y09 range from the International Classification of Diseases (ICD-10). This data was obtained in the SIM, using as the denominator population data provided by DATASUS. The coefficient was calculated from the 35 microregions of Rio Grande do Sul for the period 2003-2007, thus reducing temporal fluctuations common in events of small magnitude. Mortality rates were standardized using the population provided by the World Health Organization (WHO) for the period 2000-2025 as the standard population²⁰20. World Health Organization (WHO). Standard population 2000-2025. Disponível em: http://seer.cancer.gov/stdpopulations/world.who.html. (Acessado em 21 de julho de 2012).
http://seer.cancer.gov/stdpopulations/wo... .

Independent variables were obtained through secondary data published by the Statistics Economics Foundation of Rio Grande do Sul (FEE-RS)²¹21. Fundação de Economia e Estatística (FEE). Série Histórica Nova Tecnologia. Índice de Desenvolvimento Socioeconômico (IDESE) nos municípios do Rio Grande do Sul - 2010. Disponível em: http://www.fee.rs.gov.br/indicadores/indice-de-desenvolvimento-socioeconomico/serie-historica-nova-metodologia/. (Acessado em 26 de junho de 2012).
http://www.fee.rs.gov.br/indicadores/ind... , the Ministry of Health/DATASUS²²22. Brasil. Ministério da Saúde. Departamento de Informática do Sistema Único de Saúde (DATASUS), 2000. Disponível em: http://www2.datasus.gov.br/DATASUS/index.php?area=02. (Acessado em 26 de junho de 2012).
http://www2.datasus.gov.br/DATASUS/index... and the Brazilian Institute of Geography and Statistics (IBGE)²³23. Instituto Brasileiro de Geografia e Estatística (IBGE). Sistema IBGE de Recuperação Eletrônica (SIDRA). 2000. Disponível em: http://www.sidra.ibge.gov.br/. (Acessado em 26 de junho de 2012).
http://www.sidra.ibge.gov.br/... .

The independent variables were composed of 23 indicators, divided into the following classifications: (a) socioeconomic and demographic: Socioeconomic Development Index (SDI) (total, income, education, sanitation and households, health). The SDI is a synthetic index, inspired by the Human Development Index (HDI), which covers a range of social and economic indicators, classified into four thematic areas: education, income, sanitation/households and health, which aims to measure and monitor the level of development of the state²¹21. Fundação de Economia e Estatística (FEE). Série Histórica Nova Tecnologia. Índice de Desenvolvimento Socioeconômico (IDESE) nos municípios do Rio Grande do Sul - 2010. Disponível em: http://www.fee.rs.gov.br/indicadores/indice-de-desenvolvimento-socioeconomico/serie-historica-nova-metodologia/. (Acessado em 26 de junho de 2012).
http://www.fee.rs.gov.br/indicadores/ind... . The other indicators were: literacy (percentage of literate population, percentage of illiterate voters and percentage of female voters); female conjugality (coefficient of married women and coefficient of separated/divorced women); birth rate (number of live births per thousand inhabitants); agglomeration (percentage of household members according to strata: 1 - 3, 3 - 6, and more than 7); women heads of household (percentage); religion (percentage of Catholics and Evangelicals); race/color (percentage of whites and blacks); (b) access: urbanization (percentage of urban population); communication (number of telephone lines in service according to population, number of vehicles registered according to population, distance in kilometers from the capital); (c) health: coefficient of male mortality due to aggression, coefficient of infant mortality, coefficient of mortality by breast cancer and cervical cancer, coefficient of mortality by acquired immunodeficiency syndrome (AIDS) (male and female), proportion of ill-defined deaths (male and female), life expectancy, coverage of the Family Health Strategy (percentage), number of medical consultations/inhabitant and coefficient of hospitalization for alcohol (hospitalizations according to population).

The data were compiled and evaluated using SPSS version 18.0. Standardized coefficients of female mortality due to aggression according to the microregions of the state for the period 2003 - 2007 were related to other variables through Pearson's correlation coefficient, and then a multiple linear regression, with variable selection by backward stepwise method, was performed. Eight variables with p values < 0.20 in the Pearson correlation, and excluding those that showed multicollinearity, were included in the multivariate model: Residents per household, proportion of ill-defined female deaths, tested using the variance inflation factor.

This study was approved by the Research Ethics Committee of the School of Public Health, Department of Health of Rio Grande do Sul State, under protocol no. 473/09. There is no conflict of interest related to this research.

RESULTS

In Rio Grande do Sul, in the period 2003-2007, the average standardized coefficient of female mortality due to aggression was 3.1 ± 1.4 deaths/100,000. Figure 1 shows the average standardized coefficients of female mortality due to aggression under the 35 microregions of the state. The regions showing the highest rates in the period were: Frederico Westphalen (6.2), Ijuí (5.6), Vacaria (5.2), Campanha Meridional (4.7), Passo Fundo (4.2), Cruz Alta (4.1) e Porto Alegre (4.1).

Table 1 presents the 23 independent variables selected in the study, according to the source, year of acquisition the data, mean, standard deviation and maximum and minimum values.

Table 2 presents the bivariate correlations between female mortality due to aggression in the microregions of Rio Grande do Sul in the period 2003-2007 and the independent variables. Among the socioeconomic and demographic variables analyzed, there was an association with income measured by the SDI, number of household members (1 to 3 and 3 to 6), women conjugality and Catholic religion. The variables related to access were not associated and, in relation to health, the coefficient of male mortality due to aggression, the coefficient of hospitalization for alcohol, the coverage of the Family Health Strategy and the proportion of ill-defined deaths in men were significant.

Variables with p < 0.20, excluding those in which there was multicollinearity, were included in the multiple linear regression model: agglomeration (number of household members) and ill-defined deaths in women. The following variables remained in the final multiple regression model, explaining the female mortality due to aggression, the coefficient of mortality due to aggression by males (p = 0.016), hospitalization for alcohol (p = 0.005) and male mortality from ill-defined causes (p = 0.015), which, in the final model, showed a negative association (β = -0.348). (Table 3). Together, these three variables explain mortality due to aggression in women by 39% (adjusted r2 = 0.391).

DISCUSSION

The standardized coefficient of female mortality due to aggression found in Rio Grande do Sul in 2003-2007 (3.10 deaths/100.000) was lower than that found in Brazil in the same period (4.10/100,000)¹⁵15. Brasil. Mapa da Violência 2012: Os novos Padrões da Violência Homicida no Brasil. Caderno complementar 1. Homicídio de mulheres no Brasil. 2012. Disponível em: http://www.mapadaviolencia.org.br. (Acessado em 21 de julho de 2012).
http://www.mapadaviolencia.org.br... . However, in 7 microregions of the state, higher values than those of the country were found.

The microregions that presented the highest rates of female mortality due to aggression are the most economically significant regions in the agriculture of the State of Rio Grande do Sul, such as Passo Fundo (3^rd place), Vacaria (5^th place), Cruz Alta (6^th place) and Frederico Westphalen (7^th place)²⁴24. Rovani FFM, Oliveira LCB, Cassol R. Caracterização das microrregiões do Rio Grande do Sul a partir de técnicas quantitativas e da cartografia temática. Revista Discente Expressões Geográficas 2010; 6: 41-54.. It is known that, in livestock farming regions, there is a high appreciation of the male gender roles, in which the courage, strength and virility of men are exalted and considered essential to work with cattle, which often extends to gender relations and daily life. Moreover, not only Rio Grande do Sul, but other Brazilian rural regions that have the same economic matrix, still strongly maintain traditional patriarchal values that stimulate sexism and the submission of women to men²⁵25. Azevedo MA. Mulheres espancadas: a violência denunciada. São Paulo: Cortez; 1985..

When analyzing the "male aspect" of violence against women, it has been associated to the very construction of masculinity, present in the socialization of boys, in the connection between masculinity and violence emphasized by gender roles and the principles of cultures governed by honor¹⁶16. Peristiany JG, Pitt-Rivers J (eds.). Honor and Grace in Anthropology. Cambridge: Cambridge University Press; 1992..

Femicides have been frequent in situations of gender inequality and discrimination, patterns of the hegemonic masculinity expressed by aggressiveness and by sexism³3. Carcedo A (coord.) No olvidamos ni aceptamos: femicídio em Centroamérica, 2000-2006. San José: Associación Cetroamericana de Información y Acción; 2010. ^{, 9}9. Munévar DI. Delito de femicidio: muerte violenta de mujeres por razones de género. Est Socio-Juríd Bogotá (Colombia) 2012; 14(1): 135-75. ^{, 10}10. Wright MW. Necropolitics, narcopolitics, and femicide: gendered violence on the Mexico-U.S. border. Signs (Chic) 2011; 36(3): 707-31.. In Rio Grande do Sul, as well as in Brazil¹⁸18. Meneghel SN, Hirakata VN. Femicides: female homicide in Brazil. Rev Saúde Pública 2011; 45(3): 564-74., the male and female mortalities due to aggression were associated, showing the greater prevalence of violence against women in places where violence between men is also high. This association between an indicator of structural violence (male homicides) and an indicator of gender-based violence (female deaths due to aggression), corroborates the perception that both types of violence have become inextricably linked, that is, wherever society is more violent, women are more penalized. Other studies have shown that structural violence and social disorganization are factors that increase the vulnerability of women, and in territories disputed by drug traffickers, armed conflicts and human rights violations, gender-based crimes are frequent³3. Carcedo A (coord.) No olvidamos ni aceptamos: femicídio em Centroamérica, 2000-2006. San José: Associación Cetroamericana de Información y Acción; 2010. ^{, 9}9. Munévar DI. Delito de femicidio: muerte violenta de mujeres por razones de género. Est Socio-Juríd Bogotá (Colombia) 2012; 14(1): 135-75. ^{, 14}14. Carey D Jr, Torres MG. Precursors to femicide: Guatemalan women in a vortex of violence. Latin Am Res Rev 2010; 45(3): 142-64. .

We believe, therefore, that these situations are not mutually exclusive and that the structural violence enhances gender-based violence. We then consider that male mortality due to aggression reflects conditions of violence in society, a scenario that exacerbates gender inequality, including femicides, perpetrated by intimate partners and strange men in public places.

In this study, a relationship between the coefficient of hospitalization for alcohol and female deaths due to aggression was found. Several studies have found a higher consumption of licit and illicit drugs in families where there is violence towards women or when the offender uses alcohol and drugs²⁶26. Murphy CM, O' Farrell TJ, Fals-Stewart W, Feehan M. Correlates of intimate partner violence among male alcoholic patients. J Consult Clin Psychol 2001; 69(3): 528-40.

27. Roberts SC. What can alcohol researchers learn from research about the relationship between macro-level gender equality and violence against women? Alcohol Alcohol 2011; 46(2): 95-104. ^{- 28}28. Rabello PM, Caldas Júnior AF. Violência contra a mulher, coesão familiar e drogas. Rev Saúde Pública 2007; 41(6): 970-8., characterizing alcohol consumption as a risk factor for violence against women.

The relationship between alcohol consumption and crime is recognized as a serious social and public health issue, because alcohol, due to pharmacological stimulation, acts as a releasing factor for aggression, and has been used as justification for abusive behavior by both the perpetrators and the professionals and institutions²⁸28. Rabello PM, Caldas Júnior AF. Violência contra a mulher, coesão familiar e drogas. Rev Saúde Pública 2007; 41(6): 970-8. ^{, 29}29. Cook PJ, Moore MJ. Violence reduction through restrictions on alcohol availability. Alc Hlth Res World 1993; 17(2): 151-6..

However, in recent years, researchers have criticized the causal attribution of violence to addictive behaviors such as alcohol and drugs, showing that it can minimize the sociocultural character of violence¹⁷17. Biglia B, San Martin C. Estado de wonderbra: entretejiendo naraciones feministas sobre las violencias de gênero. Barcelona: Virus Editorial; 2007. ^{, 30}30. Angulo-Tuesta AJ. Gênero e violência no âmbito doméstico: a perspectiva dos profissionais de saúde [dissertação de mestrado]. Rio de Janeiro: Fundação Oswaldo Cruz, Escola Nacional de Saúde Pública; 1997.. It should be noted that alcohol is not a determining factor to violence and to assume this causality means to mask the social and structural causes of this phenomenon²⁹29. Cook PJ, Moore MJ. Violence reduction through restrictions on alcohol availability. Alc Hlth Res World 1993; 17(2): 151-6.. In any case, the fact of finding an association between alcoholism and violence does not mean that alcohol is the cause of aggression against women, but that alcohol abuse and aggression may be responding to common determinants³⁰30. Angulo-Tuesta AJ. Gênero e violência no âmbito doméstico: a perspectiva dos profissionais de saúde [dissertação de mestrado]. Rio de Janeiro: Fundação Oswaldo Cruz, Escola Nacional de Saúde Pública; 1997..

One last relationship found in this study was a negative association between deaths from ill-defined causes in men and female mortality due to aggression. This indicates that where there is a higher percentage of ill-defined deaths due to difficult access to health care and poor quality of the information system, there is a higher prevalence of female deaths due to aggression.

CONCLUSION

This investigation is part of a larger research project whose objective was to analyze the female deaths due to aggression under the quantitative perspective, using secondary data. The idea was to verify the existence of associations between female homicides and variables that can signal to higher risk scenarios for the occurrence of femicides.

Structural violence expressed by the high rate of male homicides and hospitalization for alcohol use were the main factors associated with female mortality due to aggression. These findings support the idea that addressing violence cannot occur only based on individual factors of victims and perpetrators.

This is an ecological study, and therefore, subject to ecological fallacies arising from the very design of the investigation, such as limitations on the use of secondary data. In any case, ecological studies allow the exploration of macrostructural aspects in the distribution of determinants of violence, since this is a social phenomenon, and moreover, are exempt from individualist or atomistic³¹31. Grana SJ. Sociostrutural considerations of domestic femicide. J Fam Violence 2001; 16(4): 421-35. ^{, 32}32. Aquino R, Gouveia N, Teixeira MG, Costa MC, Barreto ML. Estudos ecológicos. Desenho de estudos agregados. In: Almeida Filho N, Barreto ML. Epidemiologia & Saúde: fundamentos, métodos, aplicações. Rio de Janeiro: Guanabara Koogan; 2011. p. 175-85. fallacies that affect studies whose unit of observation is individuals.

REFERENCES

Publication Dates

History