Factors associated with domestic violence in women: systematic ecological review

Factores asociados a violencia doméstica en mujeres: revisión ecológica sistemática

Fatores associados à violência doméstica em mulheres: revisão ecológica sistemática

Domestic violence, primarily perpetrated by men and predominantly affecting women, is a global public health concern with serious implications for the physical and mental health of millions of women¹. Understanding the associated patterns is essential for designing effective interventions and policies that reduce its incidence and provide support to victims².

Violence causes physical, psychological, and emotional harm on its victims³, manifesting in various aspects of their lives, in both public and private settings. It does not discriminate by political systems, socioeconomic status, religion, race, or culture, violating fundamental rights³,⁴ inherent to the human condition. It is estimated that approximately 18% of women between the ages of 15 and 49 have experienced physical or sexual abuse by their partner, with this figure rising to nearly 30% over their lives⁵.

The Pan American Health Organization (PAHO)⁶ defines violence against women as any form of gender-based violence that causes physical, sexual, or psychological harm, including threats, coercion, or unjustified restrictions on their freedom, whether in public or private spaces.

Bronfenbrenner's ecological model⁷ analyzes the microsystem, mesosystem, exosystem, and macrosystem levels comprehensively to understand how they contribute to the perpetuation of domestic violence. This approach examines individual and family factors in conjunction with social and cultural contexts, offering a deeper, complete view of the problem.

Heise's⁸ ecological model and Bronfenbrenner's ecological approach address the interaction of multiple levels of influence on human behavior, making both models effectively articulable for a study of domestic violence. According to Bronfenbrenner, interconnected systems, ranging from the microsystem (close relationships such as family and partners) to the macrosystem (cultural norms and values), influence behavior. Heise identifies four levels of domestic violence: individual, relational, community, and social. Therefore, domestic violence can be understood as a phenomenon influenced by social and structural factors that interact at different levels and not merely as an individual problem.

Although there are numerous studies on domestic violence, it is necessary to synthesize and systematically evaluate the available evidence. A systematic review consolidates the findings of various studies, identifies common patterns and gaps in knowledge, and provides a solid foundation for developing intervention strategies and public policies.

Therefore, the question arises: What patterns are associated with domestic violence against women, according to the ecological model, and how these patterns manifest across the different levels—microsystem, mesosystem, exosystem, and macrosystem—based on a systematic review of the existing literature?

The analysis involved an exhaustive literature review that served as an observational, retrospective, secondary research study using a qualitative approach that integrates studies addressing the same question⁹. The PRISMA methodology was used, which includes a checklist and a flow diagram to guide each stage of the process, from study identification to study inclusion in the final analysis.

A clear and specific research question was formulated, and then an exhaustive search was conducted in relevant databases using predefined criteria. The process was documented using the PRISMA flow diagram, which encompasses the identification, selection, eligibility, and inclusion of studies. An initial screening was conducted to identify relevant studies, assessing their quality through a critical review and synthesizing the findings qualitatively. Experts played a key role in guiding the formulation of the research question, defining selection criteria, recommending databases and search terms, and evaluating and interpreting the studies, ensuring the accuracy and depth of the review.

It also ensures a clear, consistent, and transparent presentation of the studies, guaranteeing a thorough selection, evaluation, and synthesis of the evidence and facilitating the reproduction and critical assessment of the results¹⁰. Inclusion criteria were established to include articles in English, Spanish, and Portuguese, as well as qualitative research on domestic violence published between 2013 and 2023. Articles not meeting these criteria were excluded.

A total of 725 articles on patterns associated with domestic violence were reviewed using the ecological model approach. The studies were gathered from various valuable sources, including Scopus, JSTOR, Sociological Abstracts, and PubMed, covering the period from 2013 to 2023. Search terms like “gender violence,” “violence towards women,” “domestic violence,” “violencia de género,” and “violencia doméstica” were used in combination with Boolean operators (AND, OR) to refine results. From these searches, 22 articles that met the inclusion criteria were selected using a purposive non-probabilistic selection method¹¹. The data from this study are stored and accessible in the repository of the University of Magdalena¹².

Since this study is a systematic literature review and does not involve human subjects, informed consent and ethics committee registration were not required.

As detailed in Figures 1, 725 articles were found and examined; 539 were discarded based on title, 125 were duplicates, and 61 did not meet the established inclusion criteria. After abstract review, 22 articles related to domestic violence and associated elements were selected. According to SCImago Journal Rank indicators, nine articles fall in the Q1 quartile, eleven in Q2, and eight in Q3. Patterns linked to domestic violence were identified, of which 32.10% were found in Scopus, 14.08% in JSTOR, 18.40% in PubMed, and 35.50% in Sociological Abstracts.

Figure1
Selection diagram of the manuscripts under study.
Source: Moher et al¹³

As evidenced in Table 1, the general matrix of the systematic literature review, the study with the largest population examined 42,002 women¹⁴, while the smallest involved 200 women¹⁵.

Table 1
General matrix of literature review

This section describes the factors linked to domestic violence from the ecological model conceptualized by Bronfenbrenner. The patterns found within the microsystem include the victim’s higher educational level¹⁴,¹⁶,¹⁷, the aggressor’s lower educational level¹⁸,¹⁹, illiteracy in both¹⁵,²¹,²², and the perpetrator's high level of education²³.

Perpetrator’s alcohol consumption¹⁴-¹⁶ is the most common cause, along with the use of other substances²⁰,²⁴,²⁵, and gambling habits⁹,²⁴. Furthermore, factors such as the victim being younger than the aggressor¹⁹,²²,²⁶, woman’ alcohol abuse²⁷, and victim’s possible mental illness²⁸ increase the likelihood of experiencing violence.

Divorce is a factor associated with domestic violence²⁹. However, the likelihood of experiencing it increases for people who have been married multiple times¹⁴,²⁸,³⁰, are pregnant³¹ or have several children, depending on the number of them²⁵,²⁶,³². Additionally, both a family history of violence in the victim¹⁴,³³ and the aggressor’s own exposure to violence in childhood¹⁵,¹⁹,²⁷ are factors associated with domestic violence.

The dynamics that exist between domestic violence and the traditional role of housewives show a family power relationship²⁵. Mass media channels also play a role in shaping attitudes that contribute to violence against women¹⁷, particularly when women are not involved in health decision-making³⁴. Furthermore, couples married for love can experience domestic violence due to emotional and affective dependence²¹.

As shown in Table 2, factors within the mesosystem includes a lack of life skills, taking refuge in religion, and managing emotions. The ability to communicate effectively and resolve conflicts constructively can decrease the risk of domestic violence¹⁷,²⁸,³¹. Marital dissatisfaction is another factor; however, it does not justify or excuse violence in relationships³⁶. In addition, the influence of in-laws and the children, including those with special needs³⁰, can play an important role.

Within the exosystem, the influence of socioeconomic factors such as living in areas of extreme poverty, facing economic problems influenced by social and cultural factors¹⁵,¹⁶, lack of paid work²⁹,³⁰, low family income¹⁹,²³,³¹, unemployment²⁸,³², having migrant parents¹⁴, and living in rural areas³² contribute to increased risk of domestic violence. Although a partner's criminal record does not guarantee or predict violence, it is associated with a heightened risk¹⁹.

From the macrosystem perspective, factors include the husband's controlling behavior²⁴,²⁵, dominance and instilling fear in the partner²⁸,³³, as well as an extended marital life¹⁵. Additional risk factors are living in an environment where alcohol is consumed³², the perpetrator's low job status²⁴, and the wife's economic, work, and financial contributions³⁰.

Table 2
Patterns associated with domestic violence

In the microsystem, educational level and illiteracy are closely linked to domestic violence. Studies reveal that women with low income or incomplete education are more likely to be victims of domestic violence36. This probability decreases when women earn their own income or if their partner has the same or higher educational level as them37.

Consumption of alcohol and other toxic substances can exacerbate aggressive behavior by reducing self-control and intensifying violence38. Additionally, age disparity between victims and aggressors influences the prevalence of violence, with younger victims facing a higher likelihood of experiencing violence39.

About 89.5% of women with depression have experienced violence40. The intergenerational transmission of violence within the home affects many women who suffer abuse in adulthood, often linked to the violence the aggressor experienced during childhood41. Furthermore, children who have been abused by their parents or family members tend to develop antisocial behaviors42.

Domestic violence and the traditional role of women as primary family caregivers reveal underlying home power dynamics43, particularly the high proportion of women dedicated exclusively to domestic chores. In married couples, emotional dependency can intensify domestic violence, as the tendency to idealize the relationship and accept abusive behavior in the name of love contributes to abuse perpetuation44.

In the mesosystem, factors like a lack of life skills, dependence on religion, and lack of decision-making stand out as relevant themes45. Husbands who share home decision-making responsibilities tend to exhibit less violent behavior, which is linked to women’s financial independence, many of whom are often employed46. Additionally, marital dissatisfaction can contribute to relationship violence47. Abuse of children within marriage is common, with abusers usually being children or stepchildren, parents or stepparents, and siblings or stepsiblings48.

In the exosystem, factors such as lack of paid employment, low income, poverty, unemployment, residence in areas of extreme poverty, economic challenges, migration, and rural living conditions are identified as relevant factors for domestic violence49. Research indicates that people with criminal records for violent behavior are more likely to engage in domestic violence. In particular, men with a history of domestic violence tend to have distorted thoughts that justify the mistreatment of women50.

In the macrosystem, the husband's controlling behavior is identified as an important factor in domestic violence51. This type of behavior can manifest itself through attitudes of domination, extreme jealousy, isolation of the partner from her support networks and control of economic resources. These power and control dynamics not only reflect cultural norms and values in some contexts, but also perpetuate gender inequalities that place women in more vulnerable situations.

The results of this systematic review, based on the ecological models of Heise and Bronfenbrenner, highlight that domestic violence arises from a complex interaction of factors at the individual, family, social, and cultural levels. Identifying patterns across the microsystemic, mesosystemic, exosystemic, and macrosystemic levels reveals the need for multifaceted interventions that address both the immediate symptoms and the long-term effects. This comprehensive approach underscores the importance of designing prevention strategies that consider the complexity of domestic violence, promote healthy relationships, address entrenched inequalities, and adapt to diverse contexts and gender approaches to improve the effectiveness of interventions.

We would like to thank Oskarly Pérez Anaya for his dedication, meticulousness, and commitment.