Association between violence against women and intimate partner characteristics: a population-based study conducted in Vitória, Espírito Santo.

Association between violence against women and intimate partner characteristics: a population-based study conducted in Vitória, Espírito Santo*

Objective:

to analyze the association between violence against women and the sociodemographic and behavioral characteristics of their intimate partners.

Method:

cross-sectional study conducted with 1,086 women aged 18 years or older who had an intimate partner in the past 24 months. The World Health Organization instrument on violence against women, validated for use in Brazil, was used to classify psychological, physical and sexual violence. The intimate partners’ sociodemographic and behavioral characteristics were collected. The results were presented as Prevalence Ratios and their respective 95% confidence intervals.

Results:

women whose partners were ≥41 years old had a 19% lower prevalence of psychological and physical violence. Higher prevalence of such abuse was found among women whose partners were black and who consumed alcohol. Psychological violence was 25%, 17%, and 30% higher, respectively, among those whose partners did not have a paid job, smoked or were former smokers. Physical violence was 73% more frequent among participants whose partners belonged to the first tertile of the per capita family income. Psychological, physical or sexual victimization was higher among those who felt fear of their partners (PR: 1.42; 95% CI: 1.24-1.64; PR: 2.52; 95% CI: 2.02-3.13; PR: 1.97; 95% CI: 1.48-2.63) and with controlling partners (PR: 1.48; 95% CI: 1.29-1.70; PR: 1.30; 95% CI: 1.04-1.62; PR: 1.43; 95% CI: 1.09-1.86).

Conclusion:

sociodemographic and behavioral characteristics of partners increase women’s vulnerability to violence.

Highlights:

(1)Higher prevalence of violence against women whose partners are controlling.

(2)Women who have ever felt afraid of their partners had a higher prevalence of violence.

(3)Alcohol use by the partner was associated with a higher frequency of violence.

(4)Condom refusal was associated with higher prevalence of violence.

(5)Encouraging gender equality policies and tackling forms of violence against women.

Introduction

Violence against women is defined by the United Nations as any violent gender-based action that results in or is likely to result in physical, sexual or mental suffering for women1. Thus, intimate partner violence (IPV) is characterized by a common agent who acts in the violence and concomitantly relates to the victim, constituting an intimate partner. Furthermore, IPV is one of the most prevalent forms of violence against women and involves behaviors such as slapping, throwing objects that can hurt, pushing, and kicking2. Global estimates by the World Health Organization (WHO) indicate that 1 in 3 women experiences intimate partner violence in her lifetime, representing approximately 35% of women1. WHO also points out that 38% of murders against women were carried out by intimate partners1.

Women around the world report suffering from violence perpetrated by their partners. According to a cross-sectional study carried out with female primary-care users in Vitória - ES (Brazil) who had an intimate partner, violence perpetrated by partners was associated with lack of occupation, refusal to use condoms during sex, controlling behavior, alcohol consumption, smoking habits and partners with less than 8 years of schooling3. In Wuhan, China, it was also observed that partners who smoked (33%) and consumed alcohol (35%) had a greater relationship with domestic violence than partners who did not smoke (23.89%) and did not drink (24.21%)4. In Europe, the prevalence of women who have suffered intimate partner violence at some point in their lives is 51.7%, and the study also agrees with the profile of intimate aggressors described: under 40 years of age, with no occupation or a lower salary than their partners5.

Violence against women holds short-, medium-, and long-term consequences for the mental and physical health of its victims6. With regard to the health of women in situations of physical and/or sexual violence by an intimate partner, according to WHO, there is a high rate of health problems, such as: they are more than twice as likely to have an abortion and experience depression, as well as 1.5 times more likely to contract HIV when compared to women who do not experience violence1. Exposure to violence can also lead to headaches, sexual disorders, phobias and prolonged fears7
-
8.

According to a cross-sectional study, women who suffer with IPV are more likely to be late for their periodic cytopathological examination (pap smear), with a delay of 64% in victims of sexual violence, whereas in victims of physical violence the delay rate is twice as high9. As it has been shown, intimate partner violence continues to threaten the physical and mental health of women around the world. Despite the increase in surveys on the occurrence of intimate partner violence and its relationship with the demographic, socioeconomic and health profile of women, there is a scarcity of studies investigating the characteristics of intimate partners10
-
11.

A systematic review on the profile of aggressors against women has already highlighted the importance of analyzing the characteristics of aggressors, providing information beyond sociodemographic data and addressing the behaviors and forms of psychological violence adopted by aggressors10
-
11. This is an important gap to be filled, and it can help the effectiveness of interventions regarding this problem. There is no population-based research on this topic conducted in the city where the study took place, and few population-based studies on the same topic in other locations in Brazil have been reported12. This investigation will contribute to a better understanding of the phenomenon.

It is worth noting that this study hypothesizes that partner behaviors, such as control and substance use, predict violence. In this context, this study aimed to analyze the association between violence against women and the sociodemographic and behavioral characteristics of their intimate partners.

Method

Type of study

This is a cross-sectional, analytical, population-based study based on the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

Context

This study was conducted in the city of Vitória, Espírito Santo, Brazil. The city of Vitória, the capital of Espírito Santo state, was expected to hold a population of 369,534 people in 2021, in a territory of 97,123 km2, and a Municipal Human Development Index of 0.845. It has 79 neighborhoods and six Regional Health Areas13. According to data from the 2010 census, the female population represented 49.98% of the total population (163,853 women)14.

Participants

The study population consisted of women aged 18 and over who had or had had an intimate partner in the last 24 months prior to the interview. In this study, an intimate partner was defined as a partner or a former partner and/or current boyfriend, as long as they were having sexual intercourse, regardless of whether they were in a formal marital relationship or not.

Selection criteria

As an exclusion criterion, women who were unable to understand or communicate due to intellectual or sensory deficit and who were, therefore, unable to respond to the research data collection instruments were excluded.

Sample definition

A multi-stage sampling process was adopted. The primary sampling unit was the census tracts of the city of Vitória provided by the 2010 Census, carried out by the Brazilian Institute of Geography and Statistics (IBGE).

To calculate the sample size, the estimated 50% prevalence of intimate partner violence in the study population was considered so as to maximize the sample, with a confidence level of 95% and an acceptable error of 5%. For studying associations between risk factors and outcomes, we specified a 95% confidence level and 80% power. Also, an exposed/unexposed ratio of 1:1 was considered. To this value, we added 10% for losses and 30% for confounding factors, thus reaching a necessary sample size of 1,100 women.

Study variables

The independent variables relating to the sociodemographic characteristics of the women’s partners were age group in complete years; ethnicity/skin color; years of schooling; per capita family income in tertiles; and having a job. The behavioral variables analyzed were the use of illicit drugs; tobacco use; drinking alcohol; feeling afraid; being jealous; controlling; and refusing to use condoms.

The three types of violence against women (psychological, sexual, and physical) practiced by intimate partners throughout their lives were the dependent variables in this study. Each one was considered present when the woman answered “yes” to one of the items for each type of violence (psychological, physical or sexual).

Instruments used to collect information

An instrument containing the sociodemographic and behavioral characteristics was developed to obtain the independent variables. The ethnicity/skin color variable was self-reported according to IBGE: white, black and brown.

To identify the outcomes, the short version of the original WHO questionnaire on violence against women was used. This questionnaire is validated for Brazil and aims to track the different forms of violence against women in their psychological, physical and sexual domains. This instrument has 13 questions and shows high internal consistency, as presented by Cronbach’s coefficients (an average of 0.88)15.

Data collection

The pilot study took place in December 2021. Data were collected from January to May 2022 by a team of properly trained interviewers, all of whom were females. The research team also included field supervisors who were responsible for monitoring the field researchers regarding adherence to the data collection schedule, identifying and correcting flaws in the data collection process and/or inconsistencies in the data. The data collected in the pilot study did not comprise the final research sample. Field work began after the data obtained in the pilot study had been analyzed. The interviews were conducted at home, in a private place, respecting the participants’ privacy and confidentiality. All these measures were taken to minimize potential bias.

Data processing and analysis

Descriptive analysis was carried out showing the relative frequencies. Heterogeneity and Fisher’s exact tests were used in the bivariate analysis according to assumptions. We assessed multicollinearity between the explanatory variables using the variance inflation factor, particularly due to the potential overlap among socioeconomic characteristics, and found no evidence of it, as observed by the Variance Inflation Factor (VIF) (>5 considered multicollinear). Poisson regression with robust variance was used to estimate the crude and adjusted prevalence ratios (PRs) and their respective 95% confidence intervals (95%CI). Initially, the variables associated with the outcomes under study (p<0.20) were included in the model in order to consider possible confounding factors. The backward selection method was used to exclude variables. In the final model, the 5% significance level (p<0.05) was adopted. Statistical analyses were carried out using the statistical program Stata®, version 15.1.

Ethical aspects

The study was approved by the Research Ethics Committee of the Federal University of Espírito Santo, according to report no. 2.819.597. All the interviewees signed the Informed Consent Form and were informed of the objectives of the study and its possible risks and benefits.

Results

As regards Table 1, it is noteworthy that most partners were aged 41 or over (63.3%); approximately 48% were white; almost 80% had 9 years or more of schooling; 34.0% were in the first family-income tertile, and 70.1% had an occupation. As to behavioral characteristics, approximately 6% used illicit drugs; 16.5% smoked, and almost 51% drank alcohol. According to the women, approximately 27% of their partners were jealous; almost 24% were controlling, and 10.4% reported feeling afraid of their male partners. Regarding the use of condoms, 16.9% had already refused to use them during sexual intercourse.

Table 1Sociodemographic and behavioral characteristics of partners of women living in the city of Vitória, ES, Brazil, 2022 (N = 1,086)

Characteristics %* 95%CI†

Sociodemographic

Age group (N=1,085)

Up to 40 36.7 33.9-39.6

41 years or more 63.3 60.4-66.1

Ethnicity/Skin color (N=1,074)

White 47.5 44.5-50.5

Black 19.9 17.6-22.4

Brown 32.6 29.8-35.5

Years of schooling (N=1,086)

Up to 8 years 20.3 18.0-22.8

9 years or more 79.7 77.2-82.0

Per capita family income (in tertiles) (N=1,067)

1st tertile - the poorest 34.0 31.2-36.9

2nd tertile 33.7 30.9-36.5

3rd tertile - the richest 32.3 29.6-35.2

Employed (N=1,079)

No 29.9 27.3-32.7

Yes 70.1 67.3-72.7

Behavioral

Uses illicit drugs (N=1,077)

No 94.1 92.4-95.3

Yes 5.9 4.7-7.5

Uses tobacco (N=1,079)

No 76.2 73.5-78.6

Yes 16.5 14.4-18.8

Former smoker 7.3 5.9-9.0

Drinks alcohol (N=1,086)

No 49.4 46.4-52.3

Yes 50.6 47.7-53.6

Jealous (N=1,086)

No 73.3 70.6-75.8

Yes 26.7 24.2-29.4

Already felt afraid of her partner (N=1,086)

No 89.6 87.6-91.3

Yes 10.4 8.7-12.4

Controlling (N=1,086)

No 76.2 73.6-78.7

Yes 23.8 21.3-26.4

Refuses condoms

No 83.1 80.8-85.3

Yes 16.9 14.7-19.2

*% = Relative frequency; †95%CI = 95% Confidence interval

It can be seen in Table 2 that there was an association between higher frequency of psychological, physical, and/or sexual violence throughout life and the following variables: partner’s ethnicity/skin color and partner’s years of schooling, per capita family income, drug use, smoking, jealous partners, controlling, feeling afraid of the partner and refusal of condoms during sexual intercourse (p<0.05). Furthermore, it was found that age and alcohol use were related to psychological and physical violence, and the variable employed was related to psychological violence.

Table 2Prevalence of violence (psychological, physical and/or sexual) throughout life according to partners’ demographic, socioeconomic and behavioral characteristics. Vitória, ES, Brazil, 2022

Psychological violence Physical violence Sexual violence

Characteristics P* (95% CI†) P-value P* (95% CI†) P-value P* (95% CI†) P-value

Sociodemographic

Age group (N=1,085)
<0.001‡

0.044‡

0.133‡

Up to 40 52.3 (47.3-57.1)
29.2 (24.9-33.8)
19.6 (16.0-23.8)

41 years or more 41.2 (37.6-44.9)
23.6 (20.6-26.9)
15.9 (13.3-18.8)

Ethnicity/Skin color (N=1,074)
0.001§

<0.001§

<0.001§

White 40.4 (36.2-44.7)
19.2 (16.0-22.9)
11.8 (9.2-14.9)

Black 55.1 (48.4-61.7)
33.2 (27.2-39.8)
23.4 (18.2-29.5)

Brown 46.6 (41.4-51.8)
30.6 (26.0-35.6)
21.4 (17.4-26.0)

Years of schooling (N=1,086)
0.041‡

<0.001‡

0.005

Up to 8 years 51.4 (44.8-57.9)
36.4 (30.3-42.9)
23.6 (18.5-29.7)

9 years or more 43.7 (40.4-47.0)
22.9 (20.2-25.8)
15.6 (13.3-18.2)

Per capita family income (in tertiles) (N=1,067)
<0.001
<0.001
0.001

1st tertile - the poorest 53.4 (48.3-58.5)
37.5 (32.6-42.6)
23.1 (19.1-27.8)

2nd tertile 43.2 (38.1-48.4)
21.5 (17.5-26.0)
16.2 (12.7-20.3)

3rd tertile - the richest 39.1 (34.1-44.4)
18.3 (14.5-22.7)
12.8 (9.6-16.7)

Employed (N=1,079)
0.006§

0.481§

0.517§

No 39.0 (33.8-44.4)
24.2 (19.8-29.1)
16.1 (12.5-20.5)

Yes 48.0 (44.5-51.6)
26.2 (23.2-29.4)
17.7 (15.2-20.6)

Behavioral

Uses illicit drugs (N=1,077)
<0.001‡

<0.001‡

0.002‡

No 43.6 (40.6-46.7)
24.1 (21.5-26.8)
16.3 (14.1-18.7)

Yes 70.3 (58.1-80.2)
46.9 (35.1-59.1)
31.3 (21.1-43.6)

Smokes (N=1,079)
<0.001§

0.016§

0.004§

No 41.2 (37.9-44.6)
23.4 (20.6-26.4)
15.2 (12.9-17.8)

Yes 60.7 (53.3-67.6)
32.6 (26.1-39.8)
24.2 (18.4-31.0)

Former smoker 51.9 (41.0-62.7)
31.7 (22.4-42.7)
24.1 (15.9-34.7)

Drinks alcohol (N=1,086)
<0.001‡

0.005‡

0.098§

No 38.6 (34.6-42.8)
21.8 (18.5-25.5)
15.3 (12.5-18.6)

Yes 51.6 (47.5-55.8)
29.3 (25.6-33.2)
19.1 (16.0-22.6)

Jealous (N=1,086)
<0.001‡

<0.001‡

<0.001‡

No 40.3 (37.0-43.8)
21.6 (18.9-24.6)
14.2 (11.9-16.8)

Yes 58.6 (52.9-64.2)
36.6 (31.2-42.3)
25.5 (20.8-30.9)

Already felt afraid of her partner (N=1,086)
<0.001‡

<0.001‡

<0.001‡

No 41.5 (38.5-44.7)
20.9 (18.4-23.5)
14.2 (12.1-16.5)

Yes 77.0 (68.3-83.8)
66.4 (57.2-74.4)
43.4 (34.5-52.6)

Controlling (N=1,086)
<0.001‡

<0.001‡

<0.001‡

No 38.7 (35.4-42.0)
21.0 (18.4-23.9)
13.7 (11.5-16.2)

Yes 66.3 (60.3-71.8)
40.3 (34.5-46.4)
28.7 (23.5-34.5)

Refuses condoms
<0.001‡

<0.001‡

<0.001‡

No 41.6 (38.5-44.9)
23.0 (20.4-25.9)
13.5 (11.4-15.9)

Yes 62.8 (55.6-69.5)
38.3 (31.5-45.5)
35.5 (28.9-42.7)

*P = Prevalence; 95% †CI = 95% Confidence interval; ‡Fisher’s exact test; §Chi-square test

After adjustments, Table 3 showed that women whose partners were older (aged 41 or over) experienced 19% lower prevalence of psychological violence. In addition, this type of abuse was more frequent among women with partners of black ethnicity/skin color (PR: 1.24; 95% CI: 1.06-1.44) and who did not have a job (PR: 1.25; 95% CI: 1.08-1.45). Partners who smoked (PR: 1.17; 95% CI: 1.01-1.36), former smokers (PR: 1.30; 95% CI: 1.04-1.61) and those who drank alcohol (PR: 1.19; 95% CI: 1.04-1.36) were representative of the characteristics showing a higher prevalence of psychological violence when compared to the reference group. Women who had felt afraid of their partners (PR: 1.42; 95% CI: 1.24-1.64), with controlling partners (PR: 1.48; 95% CI: 1.29-1.70) and with partners who refused to use condoms during sexual intercourse (PR: 1.24; 95% CI: 1.08-1.42) were among those who had experienced more psychological violence, conditioning for possible confounding factors (p<0.05).

Table 3Crude and adjusted analysis of the lifetime prevalence of psychological violence according to partner characteristics. Vitória, ES, Brazil, 2022

Characteristics Crude analysis Adjusted analysis

PR* (95% CI†) P-value PR* (95% CI†) P-value

Age group
<0.001
0.002

Up to 40 years Ref.‡

Ref.‡

41 years or more 0.79 (0.69-0.89)
0.81 (0.71-0.93)

Ethnicity/Skin color
0.001
0.029

White Ref.‡

Ref.‡

Black 1.37 (1.16-1.60)
1.24 (1.06-1.44)

Brown 1.15 (0.99-1.34)
1.09 (0.94-1.26)

Years of schooling
0.033
0.697

Up to 8 years 1.18 (1.01-1.37)
0.97 (0.82-1.14)

9 years or more Ref.‡

Ref.‡

Per capita family income (in tertiles)
<0.001
0.232

1st tertile - the poorest 1.37 (1.16-1.61)
1.15 (0.97-1.36)

2nd tertile 1.10 (0.92-1.32)
1.04 (0.88-1.23)

3rd tertile - the richest Ref.‡

Ref.‡

Employed
0.009
0.004

No 0.81 (0.70-0.95)
1.25 (1.08-1.45)

Yes Ref.‡

Ref.‡

Uses illicit drugs
<0.001
0.854

No Ref.‡

Ref.‡

Yes 1.61 (1.35-1.92)
1.02 (0.82-1.27)

Smokes
<0.001
0.016

No Ref.‡

Ref.‡

Yes 1.47 (1.27-1.70)
1.17 (1.01-1.36)

Former smoker 1.26 (1.00-1.58)
1.30 (1.04-1.61)

Drinks alcohol
<0.001
0.012

No Ref.‡

Ref.‡

Yes 1.34 (1.17-1.53)
1.19 (1.04-1.36)

Jealous
<0.001
0.565

No Ref.‡

Ref.‡

Yes 1.45 (1.28-1.65)
1.04 (0.90-1.22)

Already felt afraid of her partner
<0.001
<0.001

No Ref.‡

Ref.‡

Yes 1.85 (1.64-2.10)
1.42 (1.24-1.64)

Controlling
<0.001
<0.001

No Ref.‡

Ref.‡

Yes 1.71 (1.52-1.94)
1.48 (1.29-1.70)

Refuses condoms
<0.001
0.002

No Ref.‡

Ref.‡

Yes 1.51 (1.32-1.72)
1.24 (1.08-1.42)

*PR = Prevalence ratio; 95% †CI = 95% Confidence interval; ‡Ref.= Reference group

In Table 4, the results of the adjusted analysis show that women with older partners (aged 41 or over) had a 19% lower lifetime prevalence of physical violence when compared to women with younger partners (p<0.05). Women with black partners experienced 33% more physical violence, and the prevalence of that condition was 73% higher among those with poorer partners, that is, those belonging to the first tertile of per capita family income. It was found that women with partners who consumed alcohol showed a prevalence of experiencing physical violence over their lifetime that was 1.27-fold that of women with partners who did not consume alcohol. It is noteworthy that the participants who reported feeling afraid of their partners showed higher prevalence of physical violence when compared to those who did not feel afraid, which was 2.52-fold higher The lifetime prevalence of physical violence was 30% higher among women with a controlling partner than among those without a controlling partner (p<0.05).

Table 4Crude and adjusted analysis of lifetime prevalence of physical violence according to partner characteristics. Vitória, ES, Brazil, 2022

Crude analysis Adjusted analysis

Characteristics PR* (95% CI†) P-value PR* (95% CI†) P-value

Age group
0.040
0.037

Up to 40 years Ref.‡

Ref.‡

41 years or more 0.81 (0.66-0.99)
0.81 (0.67-0.99)

Ethnicity/Skin color
<0.001
0.085

White Ref.‡

Ref.‡

Black 1.73 (1.33-2.24)
1.33 (1.02-1.74)

Brown 1.59 (1.25-2.02)
1.25 (0.98-1.59)

Years of schooling
<0.001
0.590

Up to 8 years 1.59 (1.28-1.97)
1.06 (0.85-1.33)

9 years or more Ref.‡

Ref.‡

Per capita family income (in tertiles)
<0.001
<0.001

1st tertile - the poorest 2.05 (1.58-2.66)
1.73 (1.34-2.24)

2nd tertile 1.17 (0.87-1.58)
1.12 (0.84-1.50)

3rd tertile - the richest Ref.‡

Ref.‡

Employed
0.484

No 0.92 (0.73-1.16)
- -

Yes Ref.‡

- -

Uses illicit drugs
<0.001
0.323

No Ref.‡

Ref.‡

Yes 1.95 (1.47-2.58)
1.18 (0.85-1.63)

Smokes
0.013
0.072

No Ref.‡

Ref.‡

Yes 1.40 (1.09-1.78)
0.94 (0.74-1.21)

Former smoker 1.35 (0.96-1.92)
1.43 (1.03-2.00)

Drinks alcohol
0.005
0.017

No Ref.‡

Ref.‡

Yes 1.34 (1.09-1.65)
1.27 (1.04-1.54)

Jealous
<0.001
0.620

No Ref.‡

Ref.‡

Yes 1.69 (1.38-2.07)
1.06 (0.84-1.33)

Already felt afraid of her partner
<0.001
<0.001

No Ref.‡

Ref.‡

Yes 3.18 (2.66-3.81)
2.52 (2.02-3.13)

Controlling
<0.001
0.019

No Ref.‡

Ref.‡

Yes 1.92 (1.57-2.34)
1.30 (1.04-1.62)

Refuses condoms
<0.001
0.079

No Ref.‡

Ref.‡

Yes 1.66 (1.33-2.07)
1.22 (0.98-1.52)

*PR = Prevalence ratio; 95% †CI = 95% Confidence interval; ‡Ref. = Reference group

Table 5 shows that women with black and brown partners had 75% and 69% higher prevalence of sexual violence over their lifetime when compared to women with white partners. In addition, participants who reported feelings of fear towards their partners (PR: 1.97; 95% IC: 1.48-2.63) and who considered them controlling (PR: 1.43; 95% IC: 1.09-1.86) also showed higher prevalence of such abuse. Refusal to use a condom was associated with 2.18-fold more occurrences of sexual violence committed by an intimate partner (p < 0.05).

Table 5Crude and adjusted analysis of lifetime prevalence of sexual violence according to partner characteristics. Vitória, ES, Brazil, 2022

Crude analysis Adjusted analysis

Characteristics PR* (95% CI†) P-value PR* (95% CI†) P-value

Age group
0.116
0.190

Up to 40 years Ref.‡

Ref.‡

41 years or more 0.81 (0.62-1.05)
0.84 (0.64-1.09)

Ethnicity/Skin color
<0.001
0.001

White Ref.‡

Ref.‡

Black 1.99 (1.41-2.79)
1.75 (1.26-2.43)

Brown 1.82 (1.33-2.49)
1.69 (1.25-2.29)

Years of schooling
0.004
0.750

Up to 8 years 1.52 (1.14-2.01)
1.05 (0.77-1.43)

9 years or more Ref.‡

Ref.‡

Per capita family income (in tertiles)
0.001
0.555

1st tertile - the poorest 1.81 (1.30-2.53)
1.20 (0.85-1.72)

2nd tertile 1.27 (0.88-1.82)
1.08 (0.75-1.54)

3rd tertile - the richest Ref.‡

Ref.‡

Employed
0.519

No 0.91 (0.68-1.22)
- -

Yes Ref.‡

- -

Uses illicit drugs
0.001
0.421

No Ref.‡

Ref.‡

Yes 1.92 (1.30-2.83)
1.20 (0.77-1.86)

Smokes
0.004
0.171

No Ref.‡

Ref.‡

Yes 1.59 (1.17-2.16)
1.18 (0.87-1.59)

Former smoker 1.58 (1.04-2.42)
1.47 (0.95-2.25)

Drinks alcohol
0.099
0.493

No Ref.‡

Ref.‡

Yes 1.25 (0.96-1.62)
1.10 (0.84-1.44)

Jealous
<0.001
0.804

No Ref.‡

Ref.‡

Yes 1.80 (1.39-2.33)
1.04 (0.78-1.38)

Already felt afraid of her partner
<0.001
<0.001

No Ref.‡

Ref.‡

Yes 3.06 (2.35-3.97)
1.97 (1.48-2.63)

Controlling
<0.001
0.009

No Ref.‡

Ref.‡

Yes 2.10 (1.62-2.72)
1.43 (1.09-1.86)

Refuses condoms
<0.001
<0.001

No Ref.‡

Ref.‡

Yes 2.63 (2.04-3.40)
2.18 (1.68-2.83)

*PR = Prevalence ratio; †95%CI = 95% Confidence interval; ‡Ref. = Reference group

Discussion

The study showed an association between the sociodemographic and behavioral characteristics of intimate partners and the occurrence of violence against women throughout life. It was found that women with partners who had a controlling behavior, were jealous, refused to use condoms during sexual intercourse, were smokers or former smokers, used alcohol, did not have a job, belonged to the black ethnicity/skin color group and to the first tertile of per capita family income (the poorest) showed a higher prevalence of violence (psychological, physical, and/or sexual) throughout their lives. The findings are in line with national studies carried out among female primary care users3 and with puerperal women in a habitual risk maternity hospital16.

Women with younger partners, aged up to 40, had a higher prevalence of psychological and physical violence, after adjustments. The results of this study corroborate a national study carried out in 2003 with 251 female health service users in Porto Alegre, Rio Grande do Sul, which showed an association between the occurrence of psychological and physical violence and the partner’s age under 3017. An international study showed that women with older partners (aged 60 or over) were 43% less likely to suffer intimate partner violence than those with younger partners (18-29)4.

The highest prevalence of physical violence was observed among women with partners belonging to the 1st income tertile (the poorest) compared to those belonging to the richest tertile. It is crucial to contextualize that the association between greater poverty and greater recognition of violence against women should not be interpreted as a direct causal relationship, but as a reflection of structural conditions. Contexts of poverty often involve greater financial stress, precarious housing, less access to support networks and protection services, factors that can intensify conflicts and reduce alternatives for non-violent resolution. Furthermore, violence perpetrated by poor men tends to be more visible and less socially tolerated, while similar behaviors in higher socioeconomic groups may be normalized or concealed, affecting the recognition and reporting of violence18.

Other characteristics of the aggressor that were associated with higher probabilities of committing psychological, physical and sexual violence were black ethnicity/skin color. The socioeconomic and racial reality of Brazil shows that black Brazilian men have a lower level of education and occupy less prestigious positions when compared to white men19. Men and women with poorer education are more exposed to experiencing violence, with 91.8% of homicide victims being men, and 75.7% being black men20.

It is worth considering that this result may reflect racial bias in the perception of violent behaviors, where stereotypes associate black men with greater aggressiveness, leading to a faster or more intense recognition of violence when the aggressor is black than when he is white, even in similar situations - a phenomenon that has already been observed in studies on how race influences judgments of violence and culpability in intimate partner cases, for example, research on perceptions of intimate partner violence and jury decisions in racially differentiated scenarios indicates that the aggressor’s race alters how violence is evaluated21.

Furthermore, women with partners who did not work had a higher lifetime prevalence of psychological violence. Men without stable employment tend to exert greater coercive control over their partners, considering the context in which economic stress can be a significant predictor of violence5.

It was evidenced that women who self-reported fear of their partners had a higher prevalence of violence (psychological, physical and sexual) throughout their lives. Violent interpersonal relationships are often marked by fear and controlling behavior between partners1, so violent male behavior capable of justifying the fear suffered by these women is a relevant factor for physical and psychological consequences before and after the occurrence of violence. Such oppressive male behavior is related to the patriarchal culture and its structure in which men and women cannot be separated, thus reproducing the role of women’s subordination to men22.

The different spheres that can trigger fear in women in situations of domestic violence must be considered, and fear appears as an immobilizer and incapacitates reaction and breaking of the cycle8. As already discussed, fear can also be the result of a long series of threats from the aggressor as well as potentially damaging in the short and long term, increasing with the sense of impunity in which the State responds to violence8. It can also be based on threats to the safety of one’s own life and the lives of people close to them, such as children and family22.

Regarding condom use, the study found that women with partners who refused to use condoms had higher prevalence of psychological and sexual violence. In conjunction with the findings in this study, the literature reaffirms the evidence pointed out3
,
16
-
17. The result after adjustments found in this study for sexual violence was the same as that found in a study with female primary health care users in the same city3. The study showed that women with partners who refused to use condoms were nine times more likely to suffer sexual violence even during pregnancy16. The difficulty of negotiating both the use of condoms and other contraceptive methods points to the fragility of women’s decision-making power over their own sexual and reproductive health, evidencing a context in which this power belongs to the partner3
,
16
,
23.

In addition to the role of male domination when it comes to the decision to use condoms, there is also the domination of women’s sexual health. It is known that men who understand that condom use prevents diseases such as HIV also have a greater positive influence on their sexual relations with women, as opposed to men who did not understand sexual protection methods24. Additionally, a study carried out in Haiti showed that women’s refusal to have sexual intercourse was associated with physical violence and concomitant lower condom use rates24.

With regard to male controlling behavior, women who live in violent relationships may experience social distancing from family and friends as they live under constant surveillance and imposition of rules that violate human rights and harm their quality of life. In addition to social control, women have limited power in sexual circumstances, hindering the space for negotiating condom use and putting themselves at risk for violence1
,
23.

In addition to gender inequality and inequality in power relations within the social and relationship structure, social inequality is also an important factor in not breaking the cycle of violence. Factors such as financial dependence25, low level of education23, worries about the future of children and emotional dependence are determining factors in women’s remaining in the cycle of violence8
,
26. Another relevant factor in the vulnerability of remaining in the context of violence is related to having experienced other types of domestic violence during one’s life, as well as a violent family context that can influence the naturalization of violence27.

This study found a higher prevalence of psychological violence against women committed by men who were smokers or former smokers, a result also observed in a study conducted in the city of Shenzhen in China28. This is probably due to factors associated with smoking, such as stress, impulsivity and concomitant use of alcohol and other substances, which are already known as risk factors for violence28
-
29.

In this study, the use of alcohol by the intimate partner was associated with a higher prevalence of psychological and physical violence. Alcohol is a psychoactive substance, and its use affects neural connections and pathways. Also, depending on the amount consumed the effects can be increased, with possible consequences leading to the impairment of attention, cognition and dexterity; increased impulsivity; an aggressive profile; loss of behavioral control, and the exacerbation of reactions30. Therefore, the use of alcohol can increase the risk for violence due to the possibility of conflict and externalization of violent behavior, expressing an impulsive and aggressive side of partners who have a social construction with rooted machismo and patriarchy3
,
5
,
16.

In Brazil, men are the main consumers of excess alcohol when compared to women30. The attributable risk and hospital costs related to violence are high, and the total average cost of interpersonal violence from 2010 to 2018 due to alcohol can be estimated at more than 17 million dollars26. It is understood that the phenomenon under study is multidimensional. However, Brazil is one of the countries that has made slow progress in implementing public policies to tackle the harmful use of alcohol, with no intersectoral contact aimed at reducing the risk of violence31.

The occurrence of violence throughout life can cause several physical, psychological, and social consequences in the short and long term for women with a history, and fear plus financial and emotional dependence can make it difficult to break the cycle of violence1. Health professionals are therefore a critical point of communication for these individuals32. However, in order to provide effective and efficient care for women in situations of violence, it is essential that health professionals develop their perspective on this serious public health problem, which encompasses different dimensions of human life (social, spiritual, physical, mental and biological). It is also necessary that this professional perspective have an interprofessional and intersectoral character so that the main referrals are made according to the real needs of the women being served33. In addition, it is important to highlight the compulsory notification that must be carried out by these professionals in cases or suspicions of violence. This has become a key tool for epidemiological surveillance and the formulation of public prevention and intervention policies34.

The results of this study are of great relevance to clinical nursing and other health professionals, especially in the care of women in gynecological and obstetric consultations, as they reinforce the importance of knowing their partners’ profiles, considering that certain characteristics can lead to greater risk for violence. The findings, therefore, reinforce the need for a more complete and holistic data collection, as well as the need to report this problem and include women in the care and protection network.

It is the first population-based study to describe the profile of the current partner among women living in a capital city in the southeastern region of the country. Violence was measured using a validated instrument and women were guaranteed privacy to answer the entire questionnaire. It adds to the national and international literature on the profile of partners of women with a higher prevalence of violence throughout life, which has several gaps, so the information presented can support the planning and implementation of intersectoral public policies to combat violence against women.

Some limitations need to be mentioned, such as reverse causality, since this is a cross-sectional study and the temporality of the relationship is compromised, making it very important to conduct longitudinal studies. Social desirability bias can lead women who are victims of domestic violence to adjust their accounts to conform to social norms or the researcher’s expectations. This can result in minimizing the severity of violence by omitting episodes or describing the perpetrator in a less negative light, especially when there is fear of judgment or stigmatization. Furthermore, feelings of guilt, emotional or economic dependence can influence more defensive responses towards the perpetrator. Thus, this bias can compromise the accuracy and completeness of data on the violence suffered. For the reasons already mentioned, there may also be residual confounding.

Conclusion

The results showed an association between violence against women and the socioeconomic and behavioral characteristics of their intimate partners. It is known that the phenomenon of violence is complex and multicausal, but this study demonstrates that characteristics, especially behavioral characteristics of intimate partners, have a stronger association with the occurrence of violence. These data present a path for evidence-based intervention measures, since behavior can be modified as long as efficient interventions are applied. The findings highlight the importance of intersectoral action to promote a culture of peace, address violence, social and gender inequality, as well as the need to reduce the use of alcohol, tobacco and other drugs.

References

1World Health OrganizationGlobal and regional estimates of violence against women: prevalence and health effects of intimate partner violence and non-partner sexual violence [Internet] GenevaWorld Health Organization2023 Aug 2651 51 Available from: https://www.who.int/publications/i/item/9789241564625

2Decker M Peitzmeier S Olumide A Acharya R Ojengbede O Covarrubias L Prevalence and health impact of intimate partner violence and non-partner sexual violence among female adolescents aged 15-19 years in vulnerable urban environments: a multi-country study. J Adolesc Health 2014 55 6 S58 S67 10.1016/j.jadohealth.2014.08.022 25454004

3Leite FMC Luis M Amorim MH Maciel ELN Gigante DP Violence against women and its association with the intimate partner’s profile: a study with primary care users Rev Bras Epidemiol 2019 22 e19005610.1590/1980-549720190056 31826112

4Chang X Yang Y Li R The characteristics of husbands and violence against women in Wuhan, China: a cross-sectional study BMC Womens Health 2022 22 1 89 89 10.1186/s12905-022-01650-z PMC8922799 35287639

5Barbier A Chariot P Lefèvre T Intimate partner violence against ever-partnered women in Europe: prevalence and associated factors - results from the violence against women EU - wide survey Front Public Health 2022 10 1033465 1033465 10.3389/fpubh.2022.10334645 PMC9755339 36530735

6World Health OrganizationViolence against women prevalence estimates, 2018 [Internet] GenevaWorld Health Organization2021 2023 Aug 26Available from: https://www.who.int/publications/i/item/9789240022256

7Lourenço LM Costa DP Intimate partner violence and implications for women’s health Gerais Rev Interinst Psicol 2020 13 1 e13010910.36298/gerais2020130109

8Lima C Moreira I Garcia L Violence against women: the need for preventive and educational action to break the domestic violence cycle Rev Duc Altum 2022 13 31103 122 10.22293/2179507x.v13i31.2181

9Leite FMC Amorim MH Gigante DP Implication of violence against women on not performing the cytopathologic test Rev Saude Publica 2018 52 89 89 10.11606/S1518-8787.2018052000496 PMC6474743 30484483

10Rosa AG Boing AF Büchele F Oliveira WF Coelho EBS Marital violence against women from the point of view of the men who were the aggressors Saude Soc 2008 17 3 152 160 10.1590/S0104-12902008000300015

11Silva AC Coelho EB Moretti-Pires RO What we know about men who commit violence against their intimate partners: a systematic review Rev Panam Salud Publica Internet2014 2024 Nov 0235 4 278 283 Available from: https://pubmed.ncbi.nlm.nih.gov/24870007/
24870007

12Lindner SR Coelho EBS Bolsoni CC Rojas PF Boing AF Prevalence of intimate partner physical violence in men and women from Florianópolis, Santa Catarina State, Brazil: a population-based study Cad Saude Publica 2015 31 4 815 826 10.1590/0102-311X00159913 25945990

13Prefeitura Municipal de Vitória (BR)A Cidade [Internet] VitóriaPrefeitura Municipal de Vitória2023 2023 Aug 26Available from: https://www.vitoria.es.gov.br/cidade

14Instituto Brasileiro de Geografia e EstatísticaCenso Demográfico 2010 [Internet] Rio de JaneiroIBGE2023 2023 Aug 26Available from: https://censo2010.ibge.gov.br/

15Schraiber LB Latorre MRDO França I JuniorSegri NJ D’Oliveira AFPL Validity of the WHO VAW study instrument for estimating gender-based violence against women Rev Saude Publica 2010 44 4 658 666 10.1590/S0034-89102010000400009 20676557

16Silva RP Leite FMC Santos ET NetoDeslandes SF Intimate partner violence during pregnancy: a focus on partner characteristics Cien Saude Colet 2022 27 5 1873 1882 10.1590/1413-81232022275.06542021 35544815

17Kronbauer AFD Meneghel SN Profile of gender violence by intimate partners Rev Saude Publica 2005 39 5 695 701 10.1590/S0034-89102005000500001 16254643

18Mossière A Maeder EM Pica E Racial composition of couples in battered spouse syndrome cases: a look at juror perceptions and decisions J Interpers Violence 2018 33 18 2867 2890 10.1177/0886260516632355 26917572

19Silveira L Leão N The impact of occupational segregation by gender and race on income gap in Brazil over three decades (1986-2015) Rev Latinoam Poblac 2020 14 27 41 76 10.31406/relap2020.v14.i12.n27.2

20Instituto de Pesquisa Econômica Aplicada; Fórum Brasileiro de Segurança PúblicaAtlas da violência 2020 [Internet] Brasília-DFIPEA2020 2023 Aug 26114 114 Available from: https://publicacoes.forumseguranca.org.br/server/api/core/bitstreams/c245e130-3df5-4cc6-8a4c-b54c982db89d/content

21Baruwa OJ Biney E Asuma BS Amoateng AY Factors associated with intimate partner violence among ever-partnered women in two Southern African countries: a multilevel analysis BMC Womens Health 2025 25 1 304 304 10.1186/s12905-025-03831-y PMC12231718 40611088

22Frías S Strategies and help-seeking behavior among Mexican women experiencing partner violence Violence Against Women 2013 19 1 24 49 10.1177/1077801212475334 23404245

23Sikweyiya Y Addo-Lartey A Alangea D Dako-Gyeke P Chirwa ED Coker-Appiah D Patriarchy and gender-inequitable attitudes as drivers of intimate partner violence against women in the central region of Ghana BMC Public Health 2020 20 1 1002 1002 10.1186/s12889-020-08825-z PMC7222313 32404153

24Conserve DF Whembolua GL Surkan PJ Attitudes toward intimate partner violence and associations with condom use among men in Haiti: an analysis of the nationally representative Demographic Health Survey J Interpers Violence 2016 31 6 989 1006 10.1177/0886260514564065 PMC5101235 25542524

25Moe AM Silenced voices and structured survival: battered women’s help seeking Violence Against Women 2007 13 7 676 699 10.1177/1077801207302041 17600305

26Freitas M Silva EN Direct and indirect costs attributed to alcohol consumption in Brazil, 2010 to 2018 PLoS One 2022 17 6 e027011510.1371/journal.pone.0270115 PMC9595536 36282815

27Kisely S Najman JM Strathearn L Childhood maltreatment and intimate partner violence: 30-year follow-up of a birth cohort study J Aggress Maltreat Trauma 2024 33 7 844 863 10.1080/10926771.2024.2343941

28Wu F Zhou L Chen C Lin W Liu P Huang W Association between intimate partner violence and prenatal anxiety and depression in pregnant women: a cross-sectional survey during the COVID-19 epidemic in Shenzhen, China BMJ Open 2022 12 5 e05533310.1136/bmjopen-2021-055333 PMC9121109 35589360

29Mouzinho IT Teixeira EC Analysis of the relationship between domestic violence against women and smoking in Brazil Serv Soc Soc 2025 148 3 e662843410.1590/0101-6628.434

30GBD 2020 Alcohol CollaboratorsPopulation-level risks of alcohol consumption by amount, geography, age, sex, and year: a systematic analysis for the Global Burden of Disease Study 2020 Lancet 2022 400 10347 185 235 10.1016/S0140-6736(22)00847-9 PMC9289789 35843246

31Santos LN Oliveira JF Rodrigues AS Oliveira MM Oliveira DS Jesus MEF Violence against women in the media during the Covid-19 pandemic Rev Gaucha Enferm [Internet] 2023 2023 Aug 2644 e2022024910.1590/1983-1447.2023.20220249.en 37672402

32Dheensa S McLindon E Spencer C Pereira S Shrestha S Emsley E Healthcare professionals’ own experiences of domestic violence and abuse: a meta-analysis of prevalence and systematic review of risk markers and consequences Trauma Violence Abuse 2023 24 3 1282 1299 10.1177/15248380211061771 PMC10240650 34978481

33Silva VG Ribeiro PM Violence against women in primary health care nurses’ practice Esc Anna Nery 2020 24 4 e2019037110.1590/2177-9465-EAN-2019-0371

34Vasconcelos NM Bernal RTI Souza JB Bordoni PHC Stein C Coll CVN Underreporting of violence against women: an analysis of two data sources Cien Saude Colet 2024 29 10 e0773202310.1590/1413-812320242910.07732023EN 39292040