Artigos

Conteúdo (article):

Análise de casos de feminicídio em Campinas, SP Brasil entre 2018 e 2019 por meio do modelo ecológico da violência
Analysis of femicide cases in Campinas, SP Brazil from 2018 to 2019 through the ecological model of violence
Monica Caicedo-Roa
Graduate Program in Public Health, Epidemiology Concentration Area. School of Medical Sciences. University of Campinas. Campinas, SP, Brazil. monicacaicedoroa@gmail.com
https://orcid.org/0000-0001-7313-8260

Ricardo Carlos Cordeiro
Department of Public Health. School of Medical Sciences. University of Campinas. Campinas, SP, Brazil. rica@unicamp.br
https://orcid.org/0000-0002-0437-1066

Resumo
A violência letal contra a mulher é um fenômeno complexo e multidimensional no qual interseccionam e confluem um amplo número de fatores para que aconteça um feminicídio em um momento e lugar específico. Identificaram-se os principais fatores que contribuíram na ocorrência de feminicídios na cidade de Campinas no período de janeiro de 2018 a dezembro de 2019. Foram realizadas entrevistas com membros da família, amigos, vizinhos, testemunhas e agentes de saúde sobre 24 casos de feminicídio utilizando a técnica da autopsia verbal. As autopsias foram complementadas, quando possível, com informações veiculadas na mídia e informes de autopsia clínica. Para o processo de análise de dados foram realizadas narrativas dos casos resgatando os aspectos mais importantes das autopsias verbais e organizando os fatores encontrados nos quatro níveis do modelo ecológico da violência utilizado pela Organização Mundial da Saúde: individual, relacional, comunitário e social. Estruturou-se a análise em categorias seguindo uma abordagem dedutiva. Partindo de casos particulares delimitados no tempo (2018 e 2019) e no espaço (a cidade de Campinas) espera-se compreender o fenômeno da violência letal contra as mulheres em sua dimensão mais ampla.
Palavras chave: Violência de Gênero; Feminicídio; Análise de Gênero na Saúde; Perspectiva de Gênero.
Abstract
Lethal violence against women is a complex and multidimensional phenomenon in which a wide number of factors intersect and converge to make a femicide happen at a specific time and place. The main factors that contributed to the occurrence of femicides in the city of Campinas were identified from January 2018 to December 2019. Interviews were conducted with family members, friends, neighbors, witnesses, and health agents about 24 femicides using the verbal autopsy technique. The autopsies were supplemented, when possible, with information from the media and clinical autopsy reports. For the data analysis process, narratives of the cases were carried out, recovering the most important aspects of the verbal autopsies and organizing the factors found in the four levels of the ecological model of violence used by the World Health Organization: individual, relational, community, and social. The analysis was structured in categories following a deductive approach. Starting from particular cases delimited in time (2018 and 2019) and in space (municipality of Campinas) it is expected to understand the phenomenon of femicide in its broadest dimension.
Key words: Gender-Based Violence; Femicide; Gender Analysis in Health; Gender Perspective.
Introduction
Femicides are multicausal, frequent, and complex events. Every day, an average of 137 deaths of women are accounted for by a member of their family worldwide1. Most cases occur in domestic spaces and are the responsibility of intimate partners or male relatives2. Brazil is considered a country with high levels of violence against women. In 2019, an average of 13 femicides were reported daily3. During the last five years, the country has registered an increase in femicide figures, particularly during the COVID-19 pandemic4.
Multiple variables participate in femicide, and not all have been fully explored. Conceptual models help understand the phenomenon of lethal violence against women. The World Health Organization (WHO) implemented the ecological model of violence in the World Report on Violence and Health5,6. The model, introduced in the 1970s, has been applied to the analysis of child abuse, youth violence, intimate partner violence, and abuse of older adults. The model explores the relationship between individual and contextual factors considering violence as the result of several layers of influence on behavior at four levels of analysis5:
Individual level: refers to the characteristics of the person that increase the possibility of being a victim or perpetrator of violence. It includes biological, historical, demographic, personal, and behavioral factors (personality, low school performance, substance abuse, and history of a violent relationship, among others).
Relational level: encompasses close social relationships (partners, intimate partners, and family members) that increase the risk of violence.
Community level: integrates the contexts in which relationships are embedded (schools, workplaces, and neighborhoods) and identifies characteristics of these scenarios.
Social level: considers the broader factors that influence a climate favorable to violence or reduce inhibitions against it, such as norms that support violence to resolve conflicts. It also includes health, educational, economic, and social policies that keep levels of inequality high.
The ecological framework considers the different causes of violence and the interaction of family, community, and social factors, which are also influenced by cultural and economic values. Notably, this model cannot contemplate all the elements of the reality it intends to model. It is an analytical strategy for understanding the complexity of the phenomenon of lethal violence against women. The choice of this model responds to recognizing the participation of a broad set of factors that intersect and converge so that femicide occurs in a specific temporal and spatial context.
This study aims to analyze, through the ecological model of violence, the cases of femicide that occurred in 2018 and 2019 in the municipality of Campinas, SP, Brazil.
Method
The study was conducted in the municipality of Campinas, with approximately 1.2 million inhabitants, located approximately 100 kilometers northwest of the municipality of São Paulo. Case research enables the analysis of single events to understand the overall context through a deductive approach. It is understood that an individual can be seen as a manifestation of a totality built from social relationships in a specific historical context7. Through a partnership with the municipal Department of Health, death certificates and addresses of all residents who died by homicide (n=38) from January 1, 2018, to December 31, 2019, were obtained. Among the cases, by consensus of the field group, those where the component of gender inequality characterizing femicides was identified were selected.
As this study analyzed cases of femicide that had already occurred, the data were obtained by performing verbal autopsies with family members, friends, acquaintances, and neighbors in the areas where the deceased women lived, interacted, and worked8. A verbal autopsy is an indirect method for estimating a cause of mortality. It collects, codes, and systematizes mortality information in epidemiological and public health surveillance studies8. The verbal autopsy’s objective was to complement the information and expand the understanding of the characteristics of deaths by femicide. Previous studies on femicides used this technique9,10.
The information was collected through fieldwork, moving the research team to the living, working, and interaction surroundings of the deceased women, which implied an approximation of the living conditions of the researched subjects. Autopsies were performed 15 days after death to respect the mourning period of family members and acquaintances. The cases’ more personal and subjective character was not captured; instead, an attempt was made to approximate the victim’s living conditions as closely as possible.
Using verbal autopsies, information from the media, and clinical autopsies as a reference, narratives were elaborated, retrieving the most significant elements of the victims’ lives and deaths for each level of the ecological model. Subsequently, they were organized by frequencies and categories.
The Research Ethics Committee of the School of Medical Sciences of the University of Campinas (Unicamp) approved the study under CAAE: 04005118.9.0000.5404 and 29654720.6.0000.5404. Consent forms were provided to participants. The names used are fictitious, and any information that could reveal the identity of the victim or perpetrator of the femicide has been removed.
Results
In total, 24 cases of women living in Campinas who died by femicide in 2018 and 2019 were analyzed. Table 1 describes the sociodemographic characteristics of the victims, the probable trigger, and the mechanism used to cause the death. Table 2 shows the elements and categories composing each level. These were schematically organized in Figure 1. In this analysis, the relational level proved to be the most relevant for femicide, followed by the individual, social, and community levels.
The death of a woman is not an isolated event. Due to the relevance of their social role, death considerably affects the lives of people who depend on them. In the 24 cases of femicide analyzed, in addition to the deaths of women, three deaths of abusers occurred. They committed suicide after the femicide or died during the aggression. Furthermore, 45 children lost their mothers.
Discussion
The ecological model of violence makes it possible to identify the relevant factors for femicide on a case-by-case basis. Some factors may be cross-sectional, appear at more than one level, and have greater or lesser importance depending on the case. The factors intersect to increase the risk of femicide. Violence against women must be understood as a macro phenomenon inserted in a global, community, and relational context and ends in individual death. The main elements of each level are discussed below.
Relational level
This level is crucial because violent relationships and inequality of power between genders can be located there. The relational level integrates two individualities: the woman and the perpetrator of the femicide. Only in their interaction does the relational level emerge. This level is formed differently from the simple junction of individual levels and operates with its dynamics. The most important categories at this level were:
Violent relationship
Worldwide, 30% (95% CI 27.8–32.2) of women in a relationship reported having experienced physical and/or sexual violence by an intimate partner throughout their lives11,12, with a lifetime prevalence of violence ranging 15–71%13. Brazilian studies reported prevalence values from 27.4 to 57.6%14,15. Women who experience domestic violence experience multiple episodes and various types of abuse.
Most women analyzed in this study had a relationship with a violent man (79%), supporting the very high prevalence of exposure to violence among victims of femicide16. Violence is a resource partners use to ensure obedience, maintain control, and favor the continuity of relationships. It usually starts in low grades and increases in severity and frequency. Violence behaves cyclically17: periods of apparent calm and well-being alternate with stressful situations that trigger aggression. In the repetition of the cycle of violence, the aggressions become more substantial and can end in femicide.
A study in 10 countries, including Brazil, showed that when a woman has been a victim of violence inflicted by her partner, it is very likely that an act of serious violence will subsequently occur. Most violent acts are not isolated incidents, following a pattern of continued abuse13. Studies with women victims of attempted femicide estimated that 67% had a history of abuse before the attempted murder18.
History of an abusive relationship
The history of a violent relationship appeared in 25% of the cases. Previous theories suggested that women victims of violence tend to get involved in new violent relationships. Thus, the history of a violent relationship is a risk factor for femicide16.
Short relationship
The known pattern of violent relationships describes that relationships become increasingly violent until they reach femicide19. However, four of the femicide cases studied contradict the pattern. In Valentina’s case, the relationship lasted four months, Valeriana and Orquídea’s granddaughter three months, and in Carina’s case, it was only a month. The fact that these femicides took place in such short-term relationships is a matter of concern because it reduces the possibility of women seeking support and ending the relationship safely, demanding greater agility of institutions to protect the victims’ lives.
Relationship with a married man
Two cases of femicide occurred as a result of relationships with married men. Manifestations of intense violence were seen before the discovery of male infidelity. Maintaining a relationship with a married man can also be a factor in femicide, especially when pregnancy is suspected or confirmed.
Behavioral issues of the abuser
Death threat
Being threatened with death is a risk factor for femicide (OR=7.36; 95% CI 2.99–18.11)20. Rating scales identified this threat as a sign of risk20-23. Six women in the research were threatened with death, and two filed a police report, but the restraining order was offered in none of the cases. In Ramona’s case, she and her family were threatened with death if they reported the attacks.
Problematic consumption of psychoactive substances (PAS) by the abuser
Studies on the pattern of alcohol and drug use in murders and attempted murders of women by their intimate partners have shown a strong relationship between substance use and violence24. A systematic review found that substance abuse by the abuser, including alcohol and other drugs, increases the risk of femicide by 85% (OR 1.85; 95% CI 1.19–2.86)20. Alcohol acts by disinhibiting behavior. People under the influence of psychoactive substances may show aggressive behavior.
Controlling behavior of the partner
Controlling behavior includes a wide variety of restrictive attitudes, such as:
- Limit the places that the woman frequents and the time they stay in them, such as healthcare services or visits to family/friends;
- Limit access to social networks;
- Limit or prevent activities such as working or studying;
- Decide on the clothes, makeup, food, and drink she can wear or ingest;
- Send controlling messages by cell phone or email insistently;
- Restrict access to money, bank account, or other property owned by the woman or couple.
The behavior of this type is usually accompanied by justifications such as jealousy, distrust of the woman’s word, or an apparent desire to protect her. With the massive use of electronic devices, other forms of control over women are becoming more frequent25. Alana and Samanta were watched through their phones. Alana received calls incessantly and was asked to sleep with the phone camera focusing on her image. In Samanta’s case, the attacker asked for the device location to be constantly sent. A partner exhibiting a controlling behavior is a risk factor for femicide (OR 5.60; 95% CI 4.41–7.13)20.
Threat or attempted suicide by the abuser
In three cases, the abusers threatened to commit suicide before the end of the relationship (cases of Orquídea, Taciana, and Alana). Psychological pressure is used as a way to keep women in relationships. Faced with the threat of suicide by an author of violence, women are recommended to avoid contact with the individual and that they notify the suicide prevention services to manage the crisis. Thus, the abuser can receive support in the face of the suicidal threat, and the women are not exposed to situations of potentially lethal physical confrontation for them.
Idea of betrayal or infidelity
Some femicides are justified by the abusers as a response to betrayal. In Franciele’s case, after ending the relationship, the husband kills her, attacks the alleged lover with two shots, and later commits suicide for not accepting the end of the relationship and believing that he had been betrayed. Alana received 16 gunshots because her partner found messages with hearts on her phone. Valeriana was brutally beaten for having sex with other partners.
The idea of betrayal to justify killing a woman is associated with the conception of ownership over their bodies, actions, and feelings. Theories such as sexual property have suggested that the probability of violence increases when the man believes he has a right over the woman and her reproductive capacities. In this context, the loss of such control triggers aggression20. Other theories focus on the male idea of the right to authority, where violence is seen as an instrument to limit female independence, reinforced by the idea that the woman belongs to him.
Termination of relationship
The end of a violent relationship is a trigger for femicide (OR 2.33; 95% CI 1.64–3.30)20,26. Nicolaidis et al. conducted a study with women who had survived femicide attempts, finding that in 73% of cases, the woman had tried to leave the relationship18. In this research, six of the women died after trying or ending a violent relationship. The murders occurred shortly after the end. The first year after separation is the period of greatest risk26, the first three months being critical.
However, ending the relationship does not guarantee the cessation of violence. Aggressions and threats persist. There are multiple reasons women do not want to end relationships: fear of more severe abuse or death, lack of economic support, concern for children, emotional dependence, lack of support networks, hope for change from the abuser, and associated prejudice with the separation, among others27.
The following items stand out among the most common predictors of termination: the nature of the violence, the woman’s life history, social and psychological factors, external resources, and coping strategies28. The end of an abusive relationship is not usually a linear process, usually with many ups and downs. Sometimes victims intend to leave relationships for reasons other than violence, such as alcohol or drug abuse by partners, financial problems, or infidelity rather than recognizing the risk of their death18.
Generally, the quality of life of abused women improves after separation, but permanence in a relationship lasts an average of 11 years29. The process of breaking up an abusive relationship is not a simple matter. Women in this situation should be supported with available legal resources. Terminations of violent relationships must be assisted by competent domestic violence personnel; otherwise, they may increase the risk of death.
The restraining order is a resource available to women who end a relationship. It notifies the abuser about the protection of the State and limits the possibilities of contact under penalty of imprisonment. Brazilian law also provides that divorce applications do not require the willingness of both parties, which facilitates the termination of legally established ties with the abuser.
Poor support network
Women who experience violence have precarious support networks. 2002 WHO study reported that approximately 80% of the women interviewed in Brazil had talked about physical abuse with someone. However, many women avoid talking about the experience of violence, and when they do, they usually talk to family, friends, or religious leaders13.
Traditionalist conceptions about the role of women, chauvinist attitudes, distrust of women’s words when they denounce violence, tendency to blame the victim or justify the abuser, and feelings of shame and guilt prevent women from exposing situations of violence and seeking help. The dichotomy between public and private makes women vulnerable, placing them in environments where it is more challenging to offer protection.
Victims of moderate and severe physical violence seek institutions or authorities but do so when they can no longer tolerate the aggressions or are seriously injured13. In the face of risk situations for physical integrity, it is recommended to have an escape route planned to protect themselves and subsequently report it to the authorities. Women victims of domestic violence who do not find support networks in the community, school, legislation, or society become victims of femicide.
Support networks are vital for adolescents because they lack sufficient personal and economic resources to face violence. Adolescents like Miranda and Jeannete, aged 13 and 19, had poor support networks. Young women are increasingly experiencing situations of violence by intimate partners, showing that violence starts early13,30.
A study with participants from five US cities found that almost 50% of women who survived femicide attempts underestimated the dangerousness of the situation, did not recognize that their lives were at risk, and were surprised by the abuser’s attack18. The fact that women do not recognize that they are at risk of death and that perpetrators do not have a single pattern reinforces the need to train people who carry out interventions or have contact with women victims of violence—such as healthcare workers, lawyers, psychologists, social workers, and others—to be aware of the seriousness of the situation and the risk of death for the woman.
Individual level
Female gender identity, origin outside São Paulo, problematic consumption of PAS, and low socioeconomic and schooling level were the relevant factors for femicide at this level.
Gender
One essential but controversial and complex point is identifying the gender component in femicide cases. In the research carried out by the Gender Nucleus of the Public Ministry of São Paulo, this question was posed31. Cases are only classified as femicides when there is a relationship or evidence of domestic violence between the victim and the abuser; other murders are not typified as such, which implies under-classification and underreporting of cases. Women are primarily victims of violence in relationships, and while they can resort to violent behavior against abusers, their position in the relationship itself is disadvantageous. Women sometimes use violence to defend themselves but rarely initiate physical aggression13.
Economic conditions
Women with low economic conditions suffer violence more frequently32. High levels of poverty can increase tension and conflict in relationships. Couples dissatisfied with their finances are more likely to resort to violence32. On the other hand, the abuser being employed has been identified as a protective factor against femicide (OR 0.50; 95% CI 0.36–0.70)20.
Economic deprivation and marginalization caused by gender inequities influence femicides. Women who lack financial independence may find it more difficult to leave abusive relationships19. Women with better financial resources are more likely to leave their aggressive partners28. Economic independence makes them less willing to maintain violent relationships, given the low perception of retribution and the high emotional cost.
Women in situations of violence spend their energy on safety for themselves and their children. Violence prevents them from engaging in work activities, causes absenteeism, and causes significant difficulties in maintaining or changing jobs. In addition to providing economic income, working increases women’s self-esteem and freedom. Having one’s own money provides security and a feeling of self-sufficiency33.
Origin
Migrant women face factors that increase their vulnerability, such as isolation, distance, separation from the family, racism, xenophobia, and precarious and informal work relationships34. Migration studies with a gender focus have shown the relevance of incorporating differences by sex in the analysis of migratory flows35. Migrant women are more exposed to structural inequities, lower economic resources, lack of employment options, and divorce36.
PAS consumption by the victim
The victim’s use of PAS increases the risk of femicide (OR 2.56; 95% CI 1.78–3.67). Studies have indicated the consumption of substances by women as a factor that favors their victimization20.
Educational conditions
There is an inverse relationship between education and violence, suggesting that education has a protective effect. Women with higher schooling level are more likely to choose their partners, decide on marriage, and have greater autonomy and control over resources within the marriage13,15. Better educational conditions are closely related to better economic positions, affecting violence exposure.
Femicides of pregnant women, with suspected pregnancy, and postpartum women
Socially, the role of the mother is one of the most valued. The prevalence of violence in pregnancy is estimated to reach up to 45%37-39. Of the autopsies performed, one was of a woman murdered during pregnancy and another during the puerperium. An international study of women assaulted during pregnancy reported that in 90% of cases, the attacks came from the biological father of the unborn child13. Pregnancy has been identified as a risk condition for femicide40. Men who abuse women during pregnancy are more at risk of committing femicide (OR 3.93; 95% CI 2.99–5.18)20.
Social level
At the social level, the misogynist and chauvinist culture stood out, in addition to the high social tolerance of violent behavior against women. Latin American societies such as Brazil have been influenced mainly by traditionalist and religious thought models.
Misogyny and Chauvinism
Misogyny is a term that describes hatred, contempt, or discriminatory prejudice against women. It is a concept of social character manifested in relationships. It can be present as exclusion, hostility, discrimination, subordination, violence, and the maintenance of male privilege over females. Misogyny is typical of environments where women are perceived as violating patriarchal norms and repressed through hostile reactions. Thus, misogyny refers to control over women, which perpetuates their subordination41.
The hatred and control over women are based on religious, naturalistic/biological, and cultural beliefs that pass through different generations in education and socialization practices. Ideas such as women’s lesser intelligence and their propensity for evil are ingrained in some cultures. Women in patriarchal and misogynistic societies cannot exercise their freedom, self-determine their existence, and enjoy fundamental rights, being punished and attacked for their status as women.
Over the last few years, the issue of violence against women has been visible. Issues such as dignity, integrity, respect, well-being, and valuing the feminine are increasingly present in political, academic, social, and family debates as a response against the system of values and beliefs that shape unequal relations between genders.
Men with traditionalist thoughts about gender roles are likelier to engage in violence against their partners to ensure obedience and submission. In affective relationships, the control of women begins with the control and devaluation of the feminine. Violence often escalates into sexual violence; women are required to have unwanted sexual contact or perform sexual acts without their approval. Then, physical aggression appears. All behaviors respond to the purpose of guaranteeing female obedience and submission for the benefit of the male counterpart42.
Violence and lack of social cohesion
Social organization refers to the ability of community residents to regulate behavior. Communities with social isolation find it more challenging to establish agreements on the behavior of residents and non-residents. The lack of ties can generate an apathetic attitude towards situations of domestic violence, e.g., avoiding calling the police or confronting the abuser, making violent men feel free to attack their partners with impunity.
Limiting a woman’s contact with other people is part of the control exercised by partners, reducing the likelihood of exposing the violent behavior. In turn, isolation increases female dependence on men. Women with strong social support are more likely to find help leaving an abusive partner. Social disapproval restrains aggressive male behavior32.
Community level
The cases studied show that femicides affect both women with favorable and precarious economic conditions. Victims of domestic violence rarely rely on their neighbors to help resolve personal issues. Individuals with medium and high incomes are less likely to intervene in situations of domestic violence than individuals with lower incomes43. At this level, the following stand out:
Relationship with justice institutions
A national survey showed that among women who were victims of serious aggression, 22.2% sought an official body such as women’s police stations, the military police, or hotlines30. Reporting domestic violence is rare. In Ramona’s case, the abuser threatened to kill her and her family if a complaint was made.
The social response to domestic violence reduces femicides. Studies have addressed this issue by testing the exposure reduction hypothesis, which advocates that services or interventions that limit contact between the victim and the abusive partner reduce the likelihood of abuse and violence44. Among the interventions are the restraining orders and arrest of male abusers. Men may take revengeful attitudes towards imposed measures, particularly when they feel the measures prevent them from exercising their authority. Thus, they can react violently when restraining orders violate their “right” to domination or control.
Women avoid seeking assistance and legal protection services due to the perception of the complexity of the legal process. This situation occurs particularly with poor women with low education and low economic possibilities to assume the costs of legal processes. Clear police procedures for domestic violence situations and trained units are needed. Traditional police officers may show minor sensitivity to situations of domestic violence or consider it trivial, especially when faced with the ambivalence of denunciators, who denounce the aggressions but do not definitively end the abusive relationships.
Drug trafficking and organized crime
Murders related to human trafficking, drug trafficking, and organized crime affect women for reasons of gender. Gang research has shown that while male gang members experience higher victimization rates, female members are exposed to sexual violence by both opposing gangs and male members in their gangs. Women linked to male gang members are at greater risk of serious violence than women without such ties1,45.
Final considerations
Analyzing the cases of femicide in Campinas through the ecological model of violence gave us an expanded understanding of the phenomenon of lethal violence against women. The most relevant level for femicide is the relational one, in which four categories stand out: the relationship with a violent man, behavioral issues of the abuser partner, the end of the relationship, and a deficient support network. For the individual level, gender, economic conditions, origin, and problematic PAS consumption are relevant. On the other hand, at the social level, misogyny and chauvinism are relevant. Finally, at the community level, the precarious economic conditions in the community, the problematic relationship with the justice institutions, and the community violence derived from drug trafficking and organized crime are present. This analysis shows that several elements contribute to the occurrence of femicide.
Women seeking care services or reporting violence need to be correctly oriented. It is essential to assess the risk of femicide to offer legal and social measures to preserve their integrity, guarantee their security, and improve their quality of life. The reporting is an initial step towards interdicting the cycle of violence. However, the actions cannot be limited to the promotion of the complaint. They must be accompanied by a structure of a material, psychological, and security support for women and their dependents.
If the denunciators are not welcomed and oriented, they will return to situations of violence due to economic and emotional dependence or lack of personal, community, and social resources. Women go through a period that requires strengthening internal mechanisms and seeking external help to end the situation of vulnerability.
Economic issues appeared at both the individual and community levels and undoubtedly contributed to the occurrence of femicide cases. Societies structured in economic, racial, and gender inequality environments have a higher incidence of femicide.
Collaborators
MCR and RC: participated in all phases of the manuscript elaboration, in the conception of the research design, analysis of the text contents, the writing of the article, and its critical review.
Acknowledgments
To Professor Lourdes Maria Bandeira (Department of Sociology at UnB) in memoriam for her considerations and contributions throughout the research.
Funding
The research was financed through a scholarship granted by the Coordination for the Improvement of Higher Education Personnel (Capes) code 001. The São Paulo Research Foundation (FAPESP) partially financed the fieldwork 2018/07162-0.

References
1. United Nations Office on Drugs and Crime. Global Study on Homicide. Gender-related Killing of Women and Girls. Viena; 2018.
2. World Health Organization, Pan American Health Organization. Understanding and addressing violence against women. Femicide; 2012.
3. Instituto de Pesquisa Econômica Aplicada, Fórum Brasileiro de Segurança Pública. Atlas da Violência. Brasília; 2019.
4. Fórum Brasileiro de Segurança pública. Anuário Brasileiro de Segurança Pública; 2021.
5. Organização Mundial da Saúde, Krug EG. Relatório mundial sobre violência e saúde. Genebra; 2002.
6. Dahlberg LL, Krug EG. Violência: um problema global de saúde pública. Cien Saude Colet. 2006;11:1163-78.
7. Heloani R, Capitão CG. Formulações gerais sobre o objeto de pesquisa em Psicologia: estudo de caso. In: Heloani R, Brolesi de Souza RM, Jarros Rodrigues RR, editores. Sociedade em transformação: estudo das relações entre trabalho, saúde e subjetividade. Londrina: EDUEL; 2007. p. 17-35.
8. World Health Organization. Verbal autopsy standards: ascertaining and attributing cause of death; 2007.
9. Caicedo-Roa M, Cordeiro RC, Martins ACA, Faria PH de. Femicídios na cidade de Campinas, São Paulo, Brasil. Cad Saude Publica 2019;35(6) e00110718.
10. Caicedo-Roa M, Nascimento JL do, Bandeira LM, Cordeiro RC. Queima às bruxas: feminismo e feminicídios íntimos por queimadura em uma metrópole brasileira. Cien Saude Colet 2022 Feb;27(2):525–34.
11. World Health Organization, Department of Reproductive Health and Research, London School of Hygiene and Tropical Medicine, South African Medical Research Council. Global and regional estimates of violence against women: prevalence and health effects of intimate partner violence and non-partner sexual violence; 2013.
12. United Nations. The world’s women 2015: trends and statistics. New York. United Nations, Department of Economic and Social Affairs. Statistics Division. 2015.
13. Organización Mundial de la Salud. Estudio multipaís de la OMS sobre salud de la mujer y violencia domestica contra la mujer. Ginebra; 2002.
14. da Silva M, Neto G, Figueiroa J, Cabral Filho J. Violence against women: prevalence and associated factors in patients attending a public healthcare service in the Northeast of Brazil. Cad Saude Publica 2010; 26(2):264-272.
15. Santos IB, Leite FMC, Amorim MHC, Maciel PMA, Gigante DP. Violência contra a mulher na vida: estudo entre usuárias da Atenção Primária. Cien Saude Colet 2020;25(5):1935-46.
16. Campbell J, 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.
17. Lucena K, Deininger L, Coelho H, Monteiro A, Vianna R, Nascimento J. Analysis of the cycle of domestic violence against women. Journal of Human Growth and Development. 2016;26:139-46.
18. Nicolaidis C, Curry MA, Ulrich Y, Sharps P, McFarlane J, Campbell D, Gary F, Laughon K, Glass N, Campbell J. Could we have known? A qualitative analysis of data from women who survived an attempted homicide by an intimate partner. J Gen Intern Med. 2003 Oct;18(10):788-94.
19. Reckdenwald A, Parker KF. Understanding the Change in Male and Female Intimate Partner Homicide Over Time: A Policy-and Theory-Relevant Investigation. Feminist Criminology. 2011;7(3):167-95.
20. Spencer CM, Stith SM. Risk Factors for Male Perpetration and Female Victimization of Intimate Partner Homicide: A Meta-Analysis. Trauma Violence Abuse. 2020;21(3):527-40.
21. Campbell JC, Webster DW, Glass N. The Danger Assessment Validation of a Lethality Risk Assessment Instrument for Intimate Partner Femicide. J Interpers Violence. 2009;24(4):653-74.
22. Conselho Nacional do Ministério Público. Formulário de avaliação de risco FRIDA: um instrumento para o enfrentamento da violência doméstica contra a mulher. 1 ed. Brasília. 2019. p. 55.
23. Echeburúa E, Fernández-Montalvo J, de Corral P, López-Goñi JJ. Assessing risk markers in intimate partner femicide and severe violence: A new assessment instrument. J Interpers Violence. 2009;24(6):925-39.
24. Sharps P, Campbell JC, Campbell D, Gary F, Webster D. Risky Mix: Drinking, Drug Use, and Homicide. NIJ Journal. 2003;196546.
25. Campeiz AB, Carlos DM, Campeiz AF, Silva JLd, Freitas LA, Ferriani MdGC. Violence in intimate relationships from the point of view of adolescents: perspectives of the Complexity Paradigm. Rev esc enferm USP. 2020;54(e03575).
26. Campbell JC, Webster D, Koziol-McLain J, Block C, Campbell D, Curry MA, Gary F, Glass N, McFarlane J, Sachs C, Sharps P, Ulrich Y, Wilt SA, Manganello J, Xu X, Schollenberger J, Frye V, Laughon K. Risk factors for femicide in abusive relationships: results from a multisite case control study. Am J Public Health. 2003 Jul;93(7):1089-97.
27. Ellsberg M, Peña R, Herrera A, Liljestrand J, Winkvist A. Candies in hell: women’s experiences of violence in Nicaragua. Soc Sci Med. 2000;51(11):1595-610.
28. Anderson DK, Saunders DG. Leaving An Abusive Partner: An Empirical Review of Predictors, the Process of Leaving, and Psychological Well-Being. Trauma Violence Abuse. 2003;4(2):163-91.
29. Alsaker K, Moen BE, Kristoffersen K. Health-Related Quality of Life Among Abused Women One Year After Leaving a Violent Partner. Social Indicators Research. 2008;86(3):497-509.
30. Fórum Brasileiro de Segurança Pública. Visível e Invisível: a vitimização de mulheres no Brasil; 2019.
31. Núcleo de Gênero Ministério Público do Estado de São Paulo. Raio X do feminicídio em São Paulo É possível evitar a morte; 2018.
32. Van Wyk JA, Benson ML, Fox GL, DeMaris A. Detangling Individual-, Partner-, and Community-level Correlates of Partner Violence. Crime & Delinquency. 2003;49(3):412-38.
33. Alsaker K, Moen BE, Baste V, Morken T. How Has Living with Intimate Partner Violence Affected the Work Situation? A Qualitative Study among Abused Women in Norway. J Fam Violence. 2016;31:479-87.
34. Bertoldo J. Migração com rosto feminino: múltiplas vulnerabilidades, trabalho doméstico e desafios de políticas e direitos. Revista Katálysis. 2018;21(2):313-23.
35. Peres RG, Baeninger R, organizadores. Migração Feminina: um debate teórico e metodológico no âmbito dos estudos de gênero. Anais do XVIII Encontro Nacional de Estudos Populacionais; 2012. nov 19-23, Águas de Lindoia, Brasil. Associação Brasileira de Estudos Populacionais; 2012.
36. Sabri B, Campbell JC, Messing JT. Intimate Partner Homicides in the United States, 2003-2013: A Comparison of Immigrants and Nonimmigrant Victims. J Interpers Violence. 2021;36:4735-57.
37. Campbell JC, Lewandowski LA. Mental and physical health effects of intimate partner violence on women and children. Psychiatr Clin North Am. 1997;20(2):353-74.
38. Gharacheh, M., Azadi, S., Mohammadi, N., Montazeri, S. & Khalajinia, Z. Domestic Violence During Pregnancy and Women\'s Health-Related Quality of Life. Global journal of health Science. 2016; 8: 27-34.
39. Conceição, H. N., Coelho, S. F. & Madeiro, A. P. Prevalência e fatores associados à violência por parceiro íntimo na gestação em Caxias, Maranhão, 2019-2020. Epidemiol Serv Saude. 2020; 30(2): e2020848.
40. Campbell, J. C. & Lewandowski, L. A. Mental and physical health effects of intimate partner violence on women and children. Psychiatr Clin North Am. 1997; 20, 353-374.
41. Manne K. Down Girl: The Logic of Misogyny: Oxford Scholarship Online; 2017. [cited 2021 dec 21]; Available from: https://oxford.universitypressscholarship.com/view/10.1093/oso/9780190604981.001.0001/oso-9780190604981?gclid=CjwKCAjwur-SBhB6EiwA5sKtjmT4Aropde8bXsgqIFuoFO9QqELmU1y09BMow04aDq7TAEBkT__MZxoCAA8QAvD_BwE
42. Conopoima Y. El femicidio como resultado de la educación patriarcal. Universidad y Sociedad. 2019(4):118-23.
43. Edwards KM, Mattingly MJ, Dixon KJ, Banyard VL. Community Matters: Intimate Partner Violence Among Rural Young Adults. Am J Community Psychol. 2014;53(1-2):198-207.
44. Dugan L, Rosenfeld R, Nagin DS. Exposure Reduction or Retaliation? The Effects of Domestic Violence Resources on Intimate-Partner Homicide. Law Society Review. 2003;37(1):169-98.
45. Manjoo R. Report of the Special Rapporteur on violence against women, its causes and consequences: United Nations; 2012.