Artigos

Conteúdo (article):

PREVENÇÃO E ENFRENTAMENTO À VIOLÊNCIA DOMÉSTICA CONTRA A MULHER: ANÁLISE DAS VARIÁVEIS DO VIOLENTÔMETRO
PREVENTING AND COPING WITH DOMESTIC VIOLENCE AGAINST WOMEN: ANALYSIS OF VIOLENCEMETER VARIABLES

Lucilla Vieira Carneiro Gomes
PhD student of the Postgraduate Program in Decision Models and Health
Institution: Federal University of Paraíba (UFPB)
E-mail: lucilla.vc@hotmail.com
ORCID: https://orcid.org/0000-0002-6988-2297

Kerle Dayana Tavares de Lucena
PhD in Decision Models and Health from the Federal University of Paraíba.
Institution: Professor and the State University of Health Sciences of Alagoas
E-mail: kerledayana@gmail.com
ORCID: https://orcid.org/0000-0001-9918-306X

Hemílio Fernandes Campos Coêlho
PhD in Statistics from the Federal University of Pernambuco.
Institution: Professor of the Postgraduate Program in Decision Models and Health Federal University of Paraíba (UFPB)
E-mail: hemilio.coelho@academico.ufpb.br
ORCID: https://orcid.org/0000-0002-7140-3590

Juliana Sampaio
PhD in Public Health from the Oswaldo Cruz Foundation.
Institution: Professor of the Postgraduate Program in Decision Models and Health Federal University of Paraíba (UFPB)
E-mail: julianasmp@hotmail.com
ORCID: https://orcid.org/0000-0003-0439-5057

Vitória Polliany de Oliveira Silva
PhD student of the Postgraduate Program in Decision Models and Health
Institution: Federal University of Paraíba (UFPB)
E-mail: vitoriapolliany1@gmail.com
ORCID: https://orcid.org/0000-0003-1363-6940


RESUMO:
Objetivou-se realizar a análise inferencial do Violentômetro identificando as variáveis mais expressivas para a ocorrência de violência doméstica contra a mulher. Trata-se de uma pesquisa do tipo aplicada, de base populacional, de corte transversal e natureza quantitativa, desenvolvida em 52 Unidades Básicas de Saúde. A amostra foi constituída por 563 mulheres. Foram utilizados os seguintes critérios de elegibilidade: mulheres a partir de 18 anos de idade que buscaram atendimento nas Unidades Básicas de Saúde durante o período da coleta de dados e concordaram em participar do estudo. Para a análise dos dados quantitativos, foi utilizada a estatística descritiva e inferencial, bem como um modelo de regressão logística e o de classificação binária WoE. O modelo utilizado demonstrou que as variáveis do Violentômetro com maior peso para identificar da ocorrência da violência doméstica foram: ridicularizar/ofender, intimidar/ameaçar, desqualificar, humilhar em público, chantagear, piadas ofensivas, ciúmes, machucar, empurrar, dar tapas, destruir bens pessoais, dar tapinhas/pancadinhas e ameaçar com objetos. Assim, o Violentômetro apresentou-se como uma potente ferramenta no enfrentamento à violência doméstica contra a mulher.

Palavras-chave: Violência contra a mulher; Saúde Pública; Modelos Estatísticos.

ABSTRACT:
This study aimed to carry out the inferential analysis of the Violencemeter, identifying the most significant variables for the occurrence of domestic violence against women. This is an applied research, population-based, cross-sectional and quantitative in nature, developed in 52 Basic Health Units. The sample consisted of 563 women. The following eligibility criteria were used: women over 18 years of age who sought care at Basic Health Units during the data collection period and agreed to participate in the study. For the analysis of quantitative data, descriptive and inferential statistics were used, as well as a logistic regression model and the WoE binary classification model. The model used demonstrated that the Violencemeter variables with the greatest weight in identifying the occurrence of domestic violence were: ridiculing/offending, intimidating/threatening, disqualifying, humiliating in public, blackmailing, offensive jokes, jealousy, hurting, pushing, slapping, destroying personal property, slapping/tapping and threatening with objects. Thus, the Violencemeter presented itself as a powerful tool in combating domestic violence against women.
Keywords: Violence against women; Public health; Statistical Models.

INTRODUCTION
Recognized as a violation of human rights, domestic violence against women (DVAW) has political, economic, social, psychological and cultural implications, which requires the construction of strategies that have as horizon the strengthening of female citizenship. In this direction, it is essential to highlight the centrality of the role of the State, with the participation of civil society, in the process of building public policies aimed at its prevention and overcoming the multiple violence against women1.
Among many types of violence, violence against women is defined by the United Nations as any act of gender-based violence that results or may result in physical, sexual or mental harm or suffering to women, including threats of such acts, coercion or arbitrary deprivation of liberty, whether in public or private life2. The concept of domestic violence against women is quite broad and includes several types of violence, which can be psychological, sexual, physical, moral and patrimonial3. This violence causes high social and economic costs for women, family and society, as well as serious physical, sexual and reproductive health problems in the short and long term4.
It is known that the home is the prevailing place for the practice of DVAW, given that interference from other people is protected5. This naturalization and privatization of violence, often legitimized by a patriarchal order of family organization, hinders a rupturing attitude on the part of women in this situation. During the covid-19 pandemic, with the implementation of social restriction measures and the confinement of women in situations of violence with their aggressors full-time, there was an increase in cases of DVAW. In the face of isolation, situations of violence were exacerbated, becoming even more frequent in several homes. Insecurity, lack of autonomy and aggression were proportionally high in conjugal relations6.
In Brazil, according to the 2023 Brazilian Forum of Public Security, a survey found that 71.9% of feminicide victims were between 18 and 44 years old, and 7 out of 10 women were killed in their own homes7. Despite the continuous growth of violence against women, the resources invested by the Federal Government to tackle violence were drastically reduced in the period between 2019 and 2022. The technical note released by the Institute of Socioeconomic Studies (INESC) showed that, in 2022, there was a lower budget allocation to deal with violence against women. Thus, the Federal Government left 70% of the available resource to deal with violence against women in 2020, the worst year of the pandemic, even with the suspension of the legal rules and the relaxation of the rules for contracts and tenders arising from the decree of public calamity. This percentage means an amount of 93.6 million BRL, which did not reach the states and municipalities to finance the women’s service network8.
On the healthcare network, it is important to work with links between primary, secondary and tertiary health care services, Reference Centers of Social Assistance and Specialized Police Stations in the care of women establishing co-responsibility for care, strengthened by matrix support and constant communication between sectors9.
Although prevention and response to violence against women require a multi-sectoral approach, the health sector has an important role to play. It is necessary to sensitize and qualify health professionals to respond to women’s needs in a holistic and empathic way, seek to prevent the recurrence of DVAW through early identification, providing appropriate referral and support4.
To identify situations of violence against women, Rosas and Lopéz conducted a study in 2009 on the dynamics of marital relations in the student community at the Instituto Politécnico Nacional (IPN) in Mexico. From this, the Violencemeter was developed, a kind of violence scale, used as a tool to help women know how to recognize the signs of domestic violence that may be suffering. The research covered three fundamental points in the relationships established by IPN students: new dynamics, gender roles and different types and manifestations of violence10.
Considering the relevance of this tool and in view that domestic violence against women can and should be prevented from early identification of related factors, the present study aims to perform inferential analysis of Violencemeter data identifying the most expressive variables for the occurrence of domestic violence against women.

METHOD

This is an applied research, population-based, cross-sectional and quantitative. The study was conducted in the city of João Pessoa, capital of the state of Paraíba, northeastern region of Brazil. The municipality has 65 neighborhoods distributed, unequally, in five Health Districts (HD), which are responsible for providing assistance with actions and services to health care in the city of João Pessoa. These districts organize a total of 203 family health teams (FHT)11.
The research was developed upon the agreement with the districts’ and health units’ managers and, subsequently, the invitation and training of female higher-level FHS professionals to apply an instrument with women assisted in the units. There was the selection of 52 Basic Health Units (BHUs) for the application of the instrument. For the selection of the BHUs, a random draw was carried out considering the registration of all BHUs in the municipality of João Pessoa-PB, according to each health district. Thus, the selection of the BHU was random, stratified by health district. In each selected BHU, information was collected from women who agreed to participate in the survey. Thus, the randomization of the BHU was performed to ensure geographical dispersion, generating reliable results within what was stipulated for the estimation of margin of error and calculation of confidence interval.
Therefore, the data collection instrument was composed of 23 items, separated into two sections. Section I addressed issues about the sociodemographic and economic data of the research participants, and Section II addressed issues about DVAW, including the Violencemeter. The instrument was applied through Google Forms to women who sought any type of care in the Basic Health Unit (BHU) from January to March 2023.
It is noteworthy that the following eligibility criteria were used to select the sample: women from 18 years of age who sought care at the BHUs in João Pessoa during the research period and agreed to participate in the study, by signing the Informed Consent Form (ICF). As exclusion criteria, the following were highlighted: women who presented cognitive deficit and communication limitation (changes in the process of expression development and verbal reception) based on the records of the patient’s medical record.
Thus, randomly, women were invited to participate in the survey during the BHU visit. The researcher and professionals (nurses and dentists) trained for this purpose applied the instrument.
The sampling planning step intended to obtain the data of this survey was the sampling stratified by health districts. The sample was selected according to “optimal allocation method”, through an indicator related to women’s care, which is the average percentage of spontaneous demand care in this municipality12. Considering the fixed selection cost for all elements of the target population, it was necessary to consider the following notation:
N→ Total number of services provided by the 5 health districts of João Pessoa. Therefore, we have N=81,550;
H→ Number of health districts. In this case, we have H=5;
N_h→ Number of services of the district h;
W_h=N_h⁄N→ Percentage of services provided by district h in relation to total services provided;
n_h→ Number of women selected in district h;
p_h→ Average percentage of spontaneous demand serviced in the districth;
e→ Margin of error considered in estimating percentages. For this research, a margin of error equal to 2% was defined in the measurement of percentage measures of the research, either way;
z→ Tabulated value of the normal distribution considering the confidence level. In this work it was decided to use a confidence level of 95%, thus z=1.96.
Therefore, we have that the sample size was calculated as follows:
n=A/B,
in which:
A=[∑_(h=1)^H▒〖(N_h/N) √(p_h (1-p_h ) )〗]^2 e B=e^2/z^2 +1/N ∑_(h=1)^H▒(N_h/N) p_h (1-p_h )

Finally, since the sample size is calculated for the entire population, the sample size for each district according to the optimal allocation is given by the following expression:
n_h=n×(N_h √(p_h (1-p_h ) ))/(∑_(h=1)^H▒〖N_h √(p_h (1-p_h ) )〗),

After this calculation of the overall sample size, the sample size was allocated proportionally according to the number of services in each district. In order to avoid losses, 30% were added to each calculated sample size. Thus, Chart 1 below presents a summary of the information considered:
Chart 1

For data analysis, variables were coded and categorized in Excel to allow a transfer to the statistical software R version 4.3.0. After encoding the variables of the database, a detailed analysis of the database was made, using the double-entry validation technique in order to detect inconsistencies in the database, which, when identified, were corrected by the revision of forms and typing. Then, the data were imported into the R software, which was used to perform statistical calculations.
After the descriptive analysis, the Odds Ratio (OR) was calculated, accompanied by the calculation of the Confidence Interval (CI). The logistic regression model was used to identify which variables were more significant for the occurrence of DVAW. The following statistical measures were used for model validity: accuracy, sensitivity and specificity, in addition to the ROC curve graph. The influence was also evaluated through the Information Value (IV) measure, using the Weight of Evidence (WoE) in software R, aiming to investigate the impact of each variable according to the weight of evidence.
In the meantime, the ROC (Receiver Operating Characteristic) curve was used, a statistical estimator used in the performance analysis of classification models. It consists of a graphical representation of the performance of a quantitative data model according to its sensitivity rate (true positive fraction) and false positive fraction (1-specificity), according to different test values13.
The most widely used effect dimension indicator for ROC curves is the area under the curve AUC. The AUC is the result of integrating all points during the curve path, and simultaneously computes sensitivity and specificity, being an estimator of the overall accuracy behavior of the test. The AUC provides an estimate of the probability of correct classification of a random subject (test accuracy); an AUC of 0.7 reflects a 70% chance of correct classification of the case. In general, the AUC values are interpreted as: 0.5-0.6 (very bad), 0.6-0.7 (bad), 0.7-0.8 (poor), 0.8-0.9 (good), > 0.9 (excellent)14.
Moreover, the study follows the Resolution n. 466/2012 of the National Health Council (CNS), which regulates the foundation of ethics in research involving human beings in Brazil15, having been approved by the Research Ethics Committee (REC) of the Center for Medical Sciences/CCM, of the Federal University of Paraíba – UFPB under the Certificate of Presentation for Ethical Assessment (CAAE) no 61355522.0.0000.8069 and opinion number: 5.672.371.
Thus, the present research is part of the first phase of the doctoral thesis of the main author, which seeks to build a decision model for identification of domestic violence in the city of João Pessoa-PB.

RESULTS

The questions adapted from the Violencemeter were divided into three domains. The Caution domain (items 1 to 11 of the Violencemeter) consists of the following forms of violence: V1) offensive jokes, V2) blackmail, V3) lie/cheat, V4) ignore, V5) jealous, V6) blame, V7) disqualify, V8) ridicule/offend, V9) humiliate in public, V10) intimidate/threaten and V11) control/prohibit. The Reaction Domain (items 12 to 18 of the Violencemeter) presents the following forms of violence: V12) destroy personal property, V13) hurt, V14) slap/beating, V15) play with a punch, V16) pinch/scratch, V17) push and V18) slap. The Real Danger Domain (items 19 to 27 of the Violencemeter) displays the following forms of violence: V19) kicking, V20) confining/arresting, V21) threatening with objects, V22) threatening with weapons, V23) threatening with death, V24) forcing a sexual relationship, V25) sexual abuse, V26) rape and V27) mutilate, as shown in Figure 1 below.
Figure 1

The binary logistic regression statistical technique was applied to the construction of the model, in order to determine which independent variables are important to identify the outcome of the occurrence of domestic and family violence. The results of the logistic regression model showed evidence, at 95%, from the estimates of the coefficients, that the variables referring to the marital status separated/divorced, the presence of children and actions taken against women in the real danger domain of the Violencemeter can be classified as more significant for the occurrence of domestic violence against women (Table 1).
Table 1

The analysis and interpretation of the odds ratio (OR) show that, in this study, a woman in the separated/divorced marital situation (p-value = 0.000) presented 3.172 times more chance of experiencing domestic and family violence when compared to a woman in the married/stable union situation (p-value = 0.001). Moreover, a woman with children (p-value = 0.000) has 3.879 times more chance of suffering domestic violence when compared to a woman without children. In addition, it is important to highlight that the items that make up the Care and Protection domains of the Violencemeter did not present significant values when compared to the other variables.
The ROC curve showed that the model provided a good result for sensitivity (89.47%) and low specificity (54.58%), which suggests evidence that the model was adequate. While the model’s classification table provided an accuracy of 74.6%, suggesting that the model has good precision to identify the occurrence of domestic violence against women. The area of the ROC curve was = 0.808, setting a good rate of model success, considering that it is an area close to 1, as can be seen in Figure 2.
Figure 2

Subsequently, the WoE model was used considering the occurrence of domestic violence against women as an outcome, since it the parameter used were the items of the Violencemeter, according to the responses of the study participants. Given the above, the results of the WoE for the Caution domain of the Violencemeter showed that the variables with greater weight for the occurrence of DVAW were the following: ridicule/offend (V8), intimidate/threaten (V10), disqualify (V7), humiliate in public (V9), blackmail (V2), offensive jokes (V1) and jealousy (V5). These seven variables were classified as being too strong (above 0.5) to identify the occurrence of domestic violence.
The results of WoE for the Reaction Domain of the Violencemeter showed that the most significant expressions of violence were: hurt (V13), push (V17), slap (V18), destroy personal belongings (V12) and slaps/beatings (V14). These five variables were classified as being too strong (above 0.5) to identify the occurrence of domestic violence. The results of the WoE for the Real Danger Domain of the Violencemeter showed that the variable with the greatest weight for the occurrence of DVAW was threatening with objects (V21), being classified as very strong (above 0.5) to identify domestic violence. These results can be seen in Figure 3.
Figure 3

Therefore, the WoE model pointed out the variables of each domain of the Violencemeter that showed very strong influence (above 0.5) for the outcome of this study. These results confirmed the most significant variables for the occurrence of domestic violence against women.

DISCUSSION

An epidemiological study conducted with 991 women in the city of Vitória - ES showed high prevalence of intimate partner violence among separated and divorced women. It is noteworthy that physical violence perpetrated by the partner was almost 2.27 times more prevalent in divorced and separated women, when compared to married ones. Similarly, the prevalence of this condition was 70% higher among separated or divorced women when compared to married women (PR: 1.70; 95%CI: 1.37-2.10)17.
It is known that DVAW consists of a cycle, which starts slowly and almost imperceptibly, and can reach dangerous and challenging dimensions. This cycle presents phases that can go from a tension in the relationship, characterized by mild conflicts to acute episodes of violence marked by aggressions18.
From this perspective, violence against women has three distinct phases: it begins with (1) the construction of tension, reaching (2) maximum tension, ending with (3) reconciliation19. In the first phase, the partner is irritated and tense, uttering threats, swearing and offenses, as well as feeling of possession and jealousy about the partner20. In the second phase, the man attacks the woman, whether physically, psychologically, morally, sexually or financially. Most often, it is at this stage that the woman seeks help21. In the third phase, the partner seeks to obtain forgiveness from his partner who usually decides to give a new chance. This cycle becomes vicious and, over time, the phases become more violent, which may result in the feminicide if not interrupted22.
A research conducted in Brazil systematized some serious forms of physical violence or threat to the physical integrity of women, highlighting the extreme vulnerability to which divorced women are subjected. Different studies have already shown that the termination of the relationship and the attempt to separate are risk factors for intimate feminicide23. In contrast, a study conducted in the city of Cajazeiras-PB showed that violent acts committed against women participants in the study were generally committed at the victim’s home by her spouse24.
It should be noted that virtually all indicators related to violence against women showed growth between the years 2021 and 2022: increase of 3.3% in the rate of threat records and 0.6% in the rate of intentional bodily injuries in the context of domestic violence. Sexual harassment crimes increased by 6.6% and 17.8%, respectively25.
Therefore, professionals working at all levels of health services, as well as the entire support network for women in situations of violence, need to be prepared for the embracement and proper notification of these cases. Thus, all opportunities for contact between professionals and women are of extreme importance to establish bonds and trust26.
Data from a national survey revealed that among women victims of severe assault, only 22.2% sought official agencies to file the denunciation23. Many women avoid seeking assistance and protection services because of the perceived complexity of the legal process. In the meantime, the denunciation is an initial step towards the interdiction of the cycle of violence. However, actions cannot be limited to the promotion of reporting and must be accompanied by a whole structure of embracement, guidance, material, psychological and security support for women and their dependents27.
Furthermore, according to the Institute for Economic and Applied Research (IPEA), a woman with children is more likely to experience domestic violence, especially when the abuser is not the biological father28. In this perspective, a national survey evidenced that the condition of being a mother demonstrated an association with violence perpetrated by intimate partner17.
Verbal aggression associated with psychological violence is often not perceived by women as an act of violence. Thus, the suffering of women in situations of violence is not yet seen as an element that deserves intervention, unless there is some objective anatomopathological basis to justify it. Therefore, it is important to analyze the occurrence of psychological violence and the ways to prevent it, considering that it is considered the starting point that triggers all other forms of violence29.
Domestic violence against women may be committed by female or male subjects, however, the aggressor is predominantly male and the violence is committed by former companions, but it can also be perpetrated by children, companions, grandchildren of the victims and several others21. Data from a survey conducted in Brazil showed that among 13 studies analyzed on the theme of DVAW, 12 evidenced gender violence perpetrated by men against women. Jealousy was identified as one of the variables related to violence against women, being presented as a trigger for verbal conflicts in love relationships and used as justification for assault29.
Thus, jealousy is learned culturally as proof of love, romanticized as a demonstration of affection, care and attention. Thus, control by partners is seen in the relationship as something positive at first, but gradually this behavior varies to more aggressive demonstrations of possession through offenses, blackmail and threats that involve even physical violence30. Therefore, jealousy reinforced by unequal gender relations has led to the failure of numerous relationships, which often end in fights and discussions, expressing themselves even in physical violence and bodily injuries31.
Physical violence is defined as minor assaults (pushing, slapping, throwing objects, twisting the arm and pulling the hair) and serious assaults (punching, hitting, kicking, throwing against a wall, burning, using a knife or firearm)32. According to an epidemiological study conducted in the Northeast region of Brazil, it is possible to identify that there is a higher prevalence of DVAW cases when referring to physical aggression, corresponding to 47% of the cases studied, followed by psychological and sexual violence33.
In this sense, it is evident that, in a scenario of male domination, the woman often feeds a condition of subordination as to the deliberations and violence situations, which includes conjugal violence. In this context, when a woman decides to break up with her husband, men react violently, including by using physical force and death threats. Thus, the attributes of masculinity reveal its construction anchored in female antagonism, especially through the normalization of practices based on virility, heteronormativity, provision of material and moral support for the family, exacerbated sexuality and use of force35.
It is important to note that, since this is a survey conducted with a sample of women who are assisted in the Primary Health Care (PHC) of the municipality under study, this may bring some limitations in terms of generalizing the results presented regarding the characterization of the profile of women who suffer domestic violence, as well as the variables with greater weight for the occurrence of DVAW. However, this is a local reality of this sample and further studies will be needed to minimize the consequences of these barriers.
Furthermore, it is important to emphasize the original aspects of the research where, through the knowledge and use of the Violencemeter instrument, through an application that will be created in the next stage of the research, women will know how to identify the types of violence that they often experience in their daily lives, but do not perceive as domestic violence. In addition, this instrument will also provide knowledge to PHC professionals, contributing to the identification and coping of DVAW. Therefore, the use of the Violencemeter is expected to facilitate this identification of domestic violence by women and PHC professionals who will perform the first embracement of these cases, articulating the care of DVAW in daily health practices.

CONCLUSIONS

The results of this study allow the identification of the most significant expressions of domestic violence in the Violencemeter, according to the responses of the participants. It is evident that all conjugal situations exhibited emphasis in relation to cases of domestic violence, especially for separated/divorced women, whether these cases are less or more serious. In addition, women who claimed to have children also presented evidence of domestic violence.
Furthermore, the Violencemeter presented itself as a powerful tool to cope with DVAW, where the model used demonstrated that the variables with greater weight to identify the occurrence of domestic violence were the following: ridicule/offend (V8), intimidate/threaten (V10), disqualify (V7), humiliate in public (V9), blackmail (V2), offensive jokes (V1), jealousy (V5), hurt (V13), push (V17), slap (V18), destroy personal property (12), taps/punches (14) and threaten with objects (V21).
Therefore, it is noticed that the Violencemeter classifies the types of DVAW in blocks of intensity, becoming an important instrument for prevention and assistance policies, besides contributing to women to perceive themselves in situations of violence, considering that many do not see themselves in this condition.
In view of the findings, the bottleneck of domestic violence against women becomes even more evident, a fact that requires a stronger and more decisive action by the intersectoral assistance network. The importance of addressing and preventing domestic violence against women is highlighted, as well as the challenge of comprehensive care for women victims of violence.
A limitation of the research concerns its development in the local reality. However, the data presented will allow building a decision model to identify DVAW in the municipality where the study was conducted. Thus, it will be possible to create indicators for monitoring and identifying domestic violence against women in order to contribute to the improvement of local public policies and the capacity to respond to the problem.
Therefore, the knowledge of the different manifestations of violence against women are fundamental to combat this problem, in view that domestic violence against women is feasible to prevent and that knowledge about its related factors is essential for coping and creating relevant public policies. In this sense, it is necessary to promote actions and new studies that aim to eliminate the exposure of women to situations of violence, evaluating the factors that involve this phenomenon, in different contexts of violence.
Therefore, this study can contribute as a warning to women who cannot identify that certain attitudes experienced in daily life constitute domestic violence at various levels of intensity and that, normally, this violence tends to increase, may culminate with feminicide. In addition, the problem investigated provides a better performance of PHC professionals in the perspective of identifying and denaturizing violence against women.
Acknowledgments: We would like to thank the Coordination of Higher Level Personnel Improvement (CAPES).

References

1 Almeida SS. Essa violência mal-dita. In: Almeida SS, organizadora. Violência de Gênero e Políticas Públicas. Rio de Janeiro: Editora UFRJ, 2007.
2 Organização Pan-Americana da Saúde (OPAS). Folha informativa - Violência contra as mulheres. Brasília (DF); 2017. Disponível em: https://www.paho.org/bra/index.php?option=com_content&view=article&id=5669:folhainformativa-violencia-contra-as-mulheres&Itemid=820.
3 Brasil. Ministério da Saúde. Secretaria Nacional de Enfrentamento à Violência contra as Mulheres. Secretaria de Políticas para as Mulheres. Política Nacional de enfrentamento à Violência contra as mulheres. Brasília, 2011.
4 World Health Organization (WHO). Fichas técnicas - Violência contra mulher. Genebra; 2021. Disponível em: https://www.who.int/en/news-room/fact-sheets/detail/violence-against-women.
5 Machado DF, Almeida MASD, Dias A. Bernardes JM, Castanheira ERL. Violência contra a mulher: o que acontece quando a Delegacia de Defesa da Mulher está fechada? Ciência & Saúde Coletiva 2020; 25(2):483-494.
6 Hisasue T, Kruse M, Raitanen J, Paavilainen E, Rissane, P. Quality of life, psychological distress and violence among women in close relationships: a population-based study in Finland. BMC Womens Health. 2020; 20(1):85.
7 Fórum Brasileiro de Segurança Pública (FBSP). 17 º Anuário Brasileiro de Segurança Pública, 2023. Disponível em: https://forumseguranca.org.br/wp-content/uploads/2023/07/anuario-2023.pdf.

8 Instituto de Estudos Socioeconômicos (INESC). Nota técnica: análise do orçamento das políticas públicas para as mulheres – 2019 a 2021. Brasília, 2023.

9 Mota AR, Machado JC, Santos NA, Simões AV, Pires VMMM, Rodrigues VP. Care practices of the nurse to women in conjugal violence situation. Revista de Pesquisa Cuidado é Fundamental Online 2020; 12(1):840- 849.
10 Rosas MAT, Lopéz SO. Género y amor: Principales aliados de la violencia en las relaciones de pareja que establecen estudiantes del IPN. Programa Institucional de Gestión con Perspectiva de Género. Instituto Politécnico Nacional, México, 2012.
11 João Pessoa. Secretaria Municipal de Saúde. Rede de Serviços de saúde, 2022.
Disponível em: http://www.joaopessoa.pb.gov.br/secretarias/saude/.

12 Bolfarine H, Bussab WO. Elementos de Amostragem. São Paulo: Edgard Blucher, 2005.

13 Polo TCF, Miot HA. Aplicações da curva ROC em estudos clínicos e experimentais. J. Vasc. Bras. 2020; 19:e20200186.

14 Hoo ZH, Candlish J, Teare D. What is an ROC curve? Emerg. Med. J. 2017; 34(6):357-9.
15 Brasil. Ministério da Saúde. Conselho Nacional de Saúde. Resolução n. 466, de 12 de dezembro de 2012. Brasília, 2012.
16 Tribunal de Justiça da Paraíba (TJPB). Reamcav e Ministério da Mulher discutem avanços no enfrentamento à violência contra a mulher na PB. Disponível em: https://www.tjpb.jus.br/noticia/reamcav-e-ministerio-da-mulher-discutem-avancos-no-enfrentamento-a-violencia-contra-a-mulher.
17 Santos IB, Leite FMC, Amorim MHC, Maciel PMA, Gigante DP. Violência contra a mulher na vida: estudo entre usuárias da Atenção Primária. Ciência & Saúde Coletiva 2020; 25(5):1935-1946.

18 Lucena KDT, Deininger LSC, Coelho HFC, Monteiro ACC, Vianna RPT, Nascimento JA. Análise do ciclo da violência doméstica contra a mulher. J. Hum. Growth Dev. 2016; 26(2):139-146.

19 Bianchini A. Os ciclos da violência doméstica contra a mulher. 2011. Disponível em: https://professoraalice.jusbrasil.com.br/artigos/121813937/os-ciclos-da-violenciadomestica-contra-a-mulher.
20 Nogueira JB. A violência doméstica e familiar contra a mulher e a ineficácia de medidas protetivas de urgência previstas na Lei 11.340/06 (Lei Maria da Penha). [Monografia] Universidade Federal do Rio Grande do Sul. Departamento de Ciências Penais. Porto Alegre, 2018.

21 Emerick LB. Violência doméstica contra a mulher: análise dos impactos da violência doméstica na sociedade brasileira. Virtua Jus, Belo Horizonte, 7(12):191-211, 2022.

22 Senado Federal. Panorama da violência contra as mulheres no Brasil: indicadores nacionais e estaduais. n. 1, Brasília: Senado Federal, Observatório da Mulher contra a Violência, 2016.

23 Bueno S, Martins J, Brandão J, Sobral I, Lagreca A. Visível e Invisível: A Vitimização de Mulheres no Brasil - 4ª edição, 2023.

24 Sá BA. Corpos que transitam sob ameaças a (in)visibilidade da violência doméstica contra a mulher. [Monografia] Universidade Federal de Campina Grande. Curso de Graduação em Enfermagem da Unidade Acadêmica de Enfermagem, 2018.

25 Fórum Brasileiro de Segurança Pública (FBSP). Violência doméstica durante a pandemia de Covid-19. Nota Técnica, 2022.

26 Carneiro JB, Gomes NP, Almeida, LCG, Romano CMC, Silva AF, Webler N, Maurício MDALLD. Condições que interferem no cuidado às mulheres em situação de violência conjugal. Escola Anna Nery 2021; 25(5):1–8.
27 Caicedo-roa M, Cordeiro RC. Análise de casos de feminicídio em Campinas, SP, Brasil, entre 2018 e 2019 por meio do modelo ecológico da violência. Ciência & Saúde Coletiva 2023; 28(1):23-36.
28 Brasil. Governo Federal. Instituto de Pesquisa Econômica Aplicada (IPEA). Atlas da Violência 2021. Brasília (DF): IPEA, 2021.

29 Devides MB. C. Variáveis associadas ao ciúme: um estudo de revisão bibliográfica. Oswaldo M. Rodrigues Jr., Carla Zeglio, Vera Lucia Vaccari, Giovanna Eleutério Levatti (Orgs.) Estudos em Sexualidade – volume 2; São Paulo: Instituto Paulista de Sexualidade, 2020.

30 Romeiro NL, Silva FCG, Brisola ACC. A página arrumando letras como um espaço para a desconstrução da dominação do patriarcado. RDBCI: Revista Digital de Biblioteconomia e Ciência da Informação 2018;16(3):317-337.
31 Turatti MC, Lucas MG. Compreendendo o ciúme na relação conjugal: Um olhar sistêmico. Unoesc & Ciência-ACBS 2016; 7(2):145-152.

32 Straus MA. Dominance and symmetry in partner violence by male and female university in 32 nations. Children and Youth Services Review 2008; 30:252-275.

33 Araújo BF, Pereira FMF, Freitas PVL, Saturnino VS, Santos EVL. Análise da Prevalência dos Tipos de Violência Contra a Mulher na Região Nordeste. Journal of Medicine and Health Promotion 2019; 4(1):1086-1095, 2019.

34 Burckhart TR. Gênero, dominação masculina e feminismo: por uma teoria feminista do direito, RDD 2017; 26(47):p. 205-224.
35 Paixão GPN, Pereira A, Gomes NP, Sousa AR, Estrela FM, Silva Filho URP, Araújo IB. Naturalização, reciprocidade e marcas da violência conjugal: percepções de homens processados criminalmente. Rev Bras Enferm. 2017; 71(1):190-196.