Intimate Partner Violence Trends and Associated Factors: Results from a National Representative Survey from Mexico

1. Introduction

Intimate partner violence (IPV) is a public health problem and a human rights violation that causes damage to physical and mental health [1]. Overall, the severity and prevalence of IPV are higher among women than in men. [2,3]. For example, in the U.S., approximately 32.9% of women and 28.1% of men have experienced IPV [4]. In Brazil, 28.0% of women and 24.0% of men reported ever IPV prevalence, while another study conducted in Mexico reported that the prevalence was higher in women (31.0%) than in men (17.0%). In contrast, in Chile, IPV was similar between women (31.0%) and men (30.0%) [5]. Nevertheless, few studies have analyzed IPV trends in women and men using nationally representative samples.

Trends of IPV Violence

Among the few IPV trends studies, a study using data from the United Nations Population Fund (UNFPA) 2021 reported that the IPV prevalence experienced by women in the last 12 months in Rwanda had decreased significantly from 44.3% in 2010 to 20.7% in 2015 [6]. A similar analysis in Mexico found that lifetime IPV prevalence with their current partner slightly decreased among women from 47.0% in 2011 to 43.9% in 2016 [7]. One study from Brazil found that past-year IPV prevalence decreased significantly among Brazilian women from 8.8% to 6.3% [8]. Given that IPV prevalence is high, particularly among low and middle-income countries, there is a need to monitor IPV trends and the associated factors by sex to help policymakers evaluate the efficacy of health programs and policies that aim to reduce all types of violence.

Correlates of IPV Violence

Previous studies suggest that adolescents and younger women had higher rates of IPV than older [8,9]. It has also been documented that having a low socioeconomic status [10,11] and a low educational level [8,9] is associated with a higher prevalence of IPV among men and women [12,13].Previous studies from Brazil, the U.S., Cambodia, China, and Sri Lanka found that having a partner who used alcohol was associated with IPV [10,14,15]. Likewise, a multicountry cross-sectional study in Asian and Pacific countries reported that drug use was positively associated with IPV outcomes [10]. Being married status is another correlate of IPV. A multinational study reported that single women in Bangladesh, Ethiopia, and Tanzania are less likely to experience IPV outcomes than those formally married [16]. However, another study in Zimbabwe found that women who cohabit without marriage have a higher risk of IPV than married women [17]. Several studies of Asia-Pacific countries and Peru found that cohabiting was associated with a higher likelihood of IPV than married women [10,18]. A study in Uganda found a positive association between relationship duration and IPV [9]. Another study in Peru found that women who reported previous partners were more likely to report IPV than those who had not [18] previous partners. Finally, data from the Zimbabwe Demographic and Health Survey (ZDHS) reported that women living in rural areas had a lower risk of IPV than in urban areas [17]. In Mexico, a cross-sectional study found that substance use behaviors were a correlate of IPV, but there is a need to identify more factors and to provide a potential mechanism of IPV. This study will fill this gap by identifying the correlates of IPV that will help to understand the phenomenon in detail better.

Mexican Context

In 1994, Mexico ratified the Global and Regional Frameworks to end violence against women. [19]. In 1999, it was created the National Program against Domestic Violence and the Official Mexican Standard NOM-190-SSA1-1999 for medical care for interfamily violence [20]. This standard NOM is mandatory for institutions, agencies, and personnel that provide public, social, and private health services. In 2002, the National Program for Equal Opportunities and Non-Discrimination against Women was created [21]. In 2007, the General Law on Women's Access to a Life Free of Violence came into effect [22]. This law establishes the legal and administrative guidelines to intervene at all levels of government to guarantee and protect women's right to a life free of violence [22]. Given the importance of Mexico in Latin America, this case study experience may help inform policymakers about the efficacy of these programs. In this context, this study aimed to analyze IPV trends in women and men and identify associated factors from 2011 to 2016 in Mexico.

2. Materials and Methods

2.1. Study Design, Population, and Setting

We used data from the 2011 and 2016 National Survey on Addictions (NSA), a nationally representative cross-sectional survey (https://encuestas.insp.mx/repositorio/encodat.php). The NSA survey used a multistage sampling design to produce national and regional estimates of drug, alcohol, and tobacco use, intimate partner violence (IPV), and other related outcomes. The overall response rate of the NSA 2011 was 75%, and for the NSA 2016 was 74% [23,24]. NSA is a household survey interview. Trained interviewers conducted face-to-face interviews in both waves (2011 and 2016). Questions about IPV were asked using Audio-Computer-Assisted Self-Interviewing (ACASI) only in the 2016 edition. A detailed description of the methodology of the NSA study has been published elsewhere [25,26]. Pooling NSA data resulted in a sample of 45,057 respondents with partners aged 14-65 years (2011=10,193 and 2016=34,864). After excluding missing covariates (n=94) values, the final analytic sample consisted of 44,963 participants (18,642 males and 26,321 females). The Institutional Review Board (IRB) approved the 2011 and 2016 NSA survey protocols at the Mexican National Institute of Public Health. All participants signed written informed consent. In addition, adolescents aged 12 to 17 years provided written assent and their parents' informed consent.

2.2. Measures

Intimate Partner Violence

To assess IPV, we constructed an index of physical, psychological, and sexual IPV by using the seven items. Respondents were asked if, at the time of a conflict with their current partner, they have experimented in the last 12 months any of the following situations (yes/no): "Has your current partner yelled at, insulted, or humiliated you?"; "Has your current partner has threatened to hit you?"; "Has your current partner has hit you?"; "Has your current partner has forced you to have sex?"; "Has your current partner has controlled most of your activities?"; "Has your current partner has expressed his jealousy by hitting you, insulting you, humiliating you or controlling your activities?"; "Has your current partner has threatened or tried to commit suicide." This index has been previously validated in the Mexican Population (alpha=0.8) [27].

Substance Use Behaviors

Drug use and alcohol use among partners were evaluated with similar questions: "Does your current partner use drugs/drink alcoholic beverages? (No, occasionally, almost every weekend, daily or almost daily) [28]. Those who responded "occasionally," "almost every weekend" or "daily or almost daily" were classified as current users.

Sociodemographic Variables

We included the following covariates: sex (female, male), categorical group of age (14-17, 18-28, 29-39, 40-59, and 60 years and over), current employment (yes, no), education level (no formal education, primary, secondary, high school, college or more), monthly family income (≤1minimum salary, >1-2 minimum salaries, >3 minimum salaries and do not know), region of the country (north-west, north-central, north-east, west, Mexico City, central, south-central and south), and place of residence (rural, urban and metropolitan). Additionally, we included some variables related to the partner relationship, such as marital status (married/living with a partner, and never married/divorced/widowed), relationship length (year of living with the partner: 1 year or less, 2 to 5 years and 5 and more years), and previous partners (yes, no).

2.3. Statistical Analyses

First, we estimated the sociodemographic characteristics of the analytical samples. Second, we performed two-sample tests for linear differences in proportions to determine changes in estimates of the psychological, physical, and sexual prevalence of IPV between 2011 and 2016. Third, we assessed changes in all forms of IPV by variables—sociodemographic and behavioral substance use. Fourth, we specified bivariate and adjusted Poisson models that regress the outcome (IPV) on the study variables with models stratified by sex. Models were also adjusted for the survey year using a dummy variable. We also use sampling weights to account for the complex survey designs. We performed all analyses using Stata 16.1.

3. Results

About one-third of the sample were participants between 40-59 years old. The sample characteristics were similar across survey years (2011 and 2016) (Table 1).

3.1. Trends in the Prevalence of IPV

From 2011 to 2016, the prevalence of global IPV did not change (from 15.58% to 14.90% p=0.30). The IPV prevalence is higher among females than males (2016: Males (11.77%), Females (17.80%)). There are also no significant changes in global IPV prevalence by sex (Table 2). When we stratified the IPV variable by type of violence (psychological, physical, and sexual), we also did not find any statistical change in the prevalence of IPV in the study (Table 2).

3.2. Trends in the Prevalence of IPV by Sociodemographic and Substance Use Behaviors Variables

In the stratified trend analysis for males, we found that the prevalence of global IPV decreased from 2011 to 2016 among those aged 29-39 years old (16.55% to 12.68%, p=0.033) and those with >1-2 minimum salaries (16.32% to 10.68% p=0.000). In addition, IPV decreased among those with a partner who did not drink alcohol (11.59% to 9.82% p=0.044) or did not use drugs (13.42% to 11.59% p=0.044). Moreover, IPV decreased among those who reported that they were married or living with a partner (14.87% to 12.14%, p=0.007) and those who had >5 years or more of couple coexistence (14.42% to 11.47%, p=0.008) (Table 3).

In the stratified analysis for females, we found that the prevalence of global IPV violence decreased among those with >1-2 minimum salaries (21.16% to 17.78% p=0.027). Conversely, the IPV violence increased among never married/divorced/widowed (9.20% to 14.62%, p=0.011), those who reported having previous partners (19.44% to 24.74%, p=0.003), and those who had ≤1 year or less of couple coexistence (7.80% to 12.30%, p=0.028) (Table 3).

3.3. Correlates of IPV

In the adjusted models, we observed that the prevalence ratio of IPV was higher among adolescents 14-17 years old compared to those 60 years and over (Adjusted Prevalence Ratio (APR)=1.79, CI 95% 1.35-2.37). In contrast, those with an income of three or more minimum salaries were less likely to report IPV than those with one minimum salary ((APR= 0.82, CI 95% 0.72-0.92). In addition, current partner alcohol use ((APR= 1.68 CI 95% 1.54-1.84) was associated with a higher prevalence of IPV than those who reported that their partner does not drink alcohol. Similarly, current partner drug use was positively associated with IPV (APR=2.80, CI 95% 2.46-3.18). Compared to those who reported being married/living with a partner, being never/married/divorced or widowed was associated with lower IPV prevalence (APR=0.66, CI 95% 0.56-0.78). The prevalence of IPV was lower among those who reported having previous partners (APR=0.60, CI 95%, 0.55-0.66) than those with no previous partners. Conversely, IPV prevalence was higher among those with five or more years of coexistence than those with one year or less (APR=1.81, CI 95% (1.47-2.23). Compared to those who lived in rural areas, living in urban (APR=1.18, CI 95%, 1.05-1.33) or metropolitans (APR=1.16, CI 95% 1.04-1.28) was positively associated with IPV (Table 4).

We found similar results in the adjusted models stratified by sex. Female adolescents between 14 to 17 years old were more likely to report IPV than females aged 60 years and over (APR=1.78, CI 95% 1.29-2.47), while the prevalence of IPV was higher in males aged 18-28 (Males: APR=2.17, CI 95% 1.57-3.00) compared to those aged 60 years and over. In both models, those having an income of three or more minimum salaries were less likely to report IPV compared to those with an income of one minimum salary IPV (Females: APR= 0.95, CI 95% 0.77-1.17; Males: RR=0.82, CI 95% 0.72-0.92) (Table 4).

In addition, partner alcohol use was associated in both stratified models with higher IPV prevalence (Males: [APR=1.68, CI 95% 1.54-1.84]; Females: [APR=1.61, CI 95% 1.40-1.86]. Additionally, IPV prevalence was higher among those who had a partner who used drugs (Males: [RR=2.80, CI 95% 2.46-3.18]. We also found that being never married/divorced/widowed was associated with a lower prevalence of IPV than those who reported being married or living with a partner (Male: APR=0.66 CI 95% 0.56-0.78; Female: APR=0.59 CI 95% 0.45-0.76). Those with previous partners (Male: APR=0.60 CI 95% 0.55-0.66; Female: APR=0.52 CI 95% 0.45-0.60) were less likely than those without previous partners to report self-reported IPV.

Having more than 5 years of couple coexistence was associated with a higher likelihood of self-reported IPV than couples with one year of less coexistence (Male: APR=1.81 CI 95% 1.47-2.23; Female: APR =1.71 CI 95% 1.25-2.34). Compared to those who lived in rural areas, living in urban (Male: APR=1.18 CI 95% 1.05-1.33) or metropolitans (Male: APR=1.16 CI 95% 1.04-1.28; Female: APR=1.20 CI 95% 1.00-1.43) was positively associated with IPV.

4. Discussion

The present study provides relevant information on IPV prevalence and associated factors in a representative sample of women and men in Mexico between 2011 and 2016. This study confirmed that both women and men are victims of IPV, although the prevalence was higher in women. The findings of our study showed that the global prevalence of IPV remained unchanged in the study period (From 17.6% in 2011 to 17.8% in 2016). In contrast, a prior study in Mexico conducted with the National Survey on the Dynamics of Household Relationships (ENDIREH), where the prevalence of IPV among women decreased from 25.6% to 20.7% between 2016 (7) and 2021 [29]. These differences could be related to the different definitions in the surveys. However, both surveys found that IPV is still high among the Mexican female population.

We found that low income was associated with higher IPV prevalence; this association was particularly prominent among women, and this result is consistent with previous research in high- and middle-income countries [30,31]. Therefore, effective intervention approaches must be designed to incorporate this risk factor and ensure a better income for women, progressively ensuring a better socioeconomic level.

We found that people whose partners consume alcohol were more likely to experience IPV, and that is consistent with the existing literature on IPV [9,10,14,15,32]. In addition, IPV outcomes are more common when both members consume alcohol [14]. Although, due to the cross-sectional nature of our study, the temporality of the associations cannot be established, this association could be explained through the model by Øverup et al., [33] which states that alcohol use is a strategy to "cope with some difficult personal or emotional situations" which could potentially trigger violence events. We found a similar association for drug use, which is also consistent with previous studies that found that drug use by a partner increases the risk of IPV [10,15].

Considering marital status, we found that single people (never married, divorced, or widowed) are less likely to report IPV. Previous studies found similar results and have described that women who did not cohabit with their partners experienced a lower level of IPV [10,16,17]. In this sense, being married or living with a partner could establish a relationship of power in which male control with a patriarchal structure could be used for the submission or subjugation of women [34].

In addition, we found that residents from urban and metropolitan settings are more likely to report IPV. Iman'ishimwe et al. found that residents of rural areas have a protective effect on IPV (17). However, another potential explanation could be that people who lived in rural settings underreported a higher proportion of IPV outcomes because they could normalize IPV situations [35].

This study has several limitations. First, its cross-sectional nature prevents the establishment of causal associations [36]. Second, some important variables, such as childhood violence history and family violence history, were not included in the study because there were no available survey questions [16,35]. Third, IPV is a sensitive topic, and due to the survey mode (face-to-face interview and ACASI), it is possible that IPV prevalence could be underestimated. Despite these limitations, this is one of the first studies in low, and middle-income countries (LMICs) that have evaluated trends in IPV and their associated factors using a nationally representative study. Therefore, the findings of this study are relevant for policymakers of Mexico and LMIC countries with similar contexts.

5. Conclusions

This study makes visible the problematic situation of IPV in Mexico. From 2011 to 2016, the global prevalence of IPV remained high. Moreover, IPV prevalence was disproportionally higher among women and people with low income and did not decrease in the study period. Although current programs for violence in Mexico focus on attending to women as victims, comprehensive prevention and future interventions must consider the two directions of IPV (victim and perpetrator). Likewise, it is essential to harmonize interventions to prevent and treat alcohol and drug use and mental health. Therefore, it is necessary to improve and strengthen epidemiological surveillance systems to identify the use of addictive substances in people who report or are reported for IPV and vice versa. It is also necessary to strengthen people's psychosocial skills to achieve gender equality, improve their quality of life, and access a better education with higher income, thus preventing and synergistically reducing the problem of IPV in Mexico.