Introduction
Intimate Partner Violence against women (IPV; also known as Gender Violence in the Spanish legal framework) is a public health problem with severe consequences for a high proportion of women around the world (Delegación del Gobierno contra la Violencia de Género (DGVG), 2019; World Health Organization (WHO), 2013). Much research in this field has focused on the damage that violence causes in different life areas, conceptualizing women as passive victims of their circumstances (Arias et al., 2016). Offering a less partial and pathologizing view of women, Gondolf and Fisher (1988) proposed the survivor theory. Their approach recognized that mistreated women are active agents using diverse and innovative coping strategies to guarantee their own safety and to protect their children. In fact, studies have found that women usually seek formal and/or informal support, despite the obstacles of fear, guilt, physical and emotional discomfort, economic constraints, and absence or inadequacy of community resources (Fontanil et al., 2020; Lelaurain et al., 2017).
As a result of Gondolf and Fisher’s contributions, a research approach focusing on women’s strengths emerged. From this perspective, it is assumed that women employ various resources to cope with and overcome the violence. These resources must be promoted since trauma and adversity are change opportunities. Studies have found that women are able to survive and draw positive learnings from their experience, with many of them subsequently becoming involved in activism work and actions to help other women who suffer or have suffered IPV (Crann & Barata, 2016; Fernández-Sánchez & López-Zafra, 2019). After decades of research, it is important to reflect on the extent to which existing knowledge can guide future policies and professional practices. However, as Anderson et al. (2012) point out, even though suffering should not be taken as the central component of the women’s identity, nor should the severity of their experience be minimized. The trauma recovery process is characterized by the co-occurrence of positive and negative experiences, fluctuating between states of suffering and personal growth, or resilience and psychopathology; denial of any of these events implies ignoring a part of the women’s reality (Fontanil et al., 2020; Humphreys, 2003; Rodríguez et al., 2008).
From this perspective, based on the strengths of women to overcome the violent relationship, the pivotal concept of resilience emerges.
Resilience: still unclear
The construct of resilience has become a core element in the analysis of vulnerable populations from a strengths-based approach. Even so, there is still a lack of consensus about how it should be defined and assessed (Bushati, 2020; Chmitorz et al., 2018; Stainton et al., 2018; Ungar, 2019).
In relation to the resilience concept, trait-oriented, outcomes-oriented, and process- oriented definitions coexist in resilience literature. Trait-oriented definitions describe resilience as a personality trait that is an internal, relatively stable, and consistent personal characteristic (Hu et al., 2015). Outcomes-oriented definitions consider that resilience is the ability to maintain a state of equilibrium; the main resilience indicator is the result, consequence, or impact that adversity has on the person’s circumstances (Bonanno, 2004). Finally, process-oriented definitions support the dynamic character of resilience, which operates through a multitude of individual and contextual resources in constant interaction with each other. From this perspective, resilience is characterized by temporal and contextual specificities, that is, the same person can show resilience in certain environments or domains but not in others, or can be resilient at a specific life stage but not in all (Bushati, 2020; Stainton et al., 2018; Ungar, 2019; Ungar & Theron, 2019). Other investigations support an ecological view of resilience, similar to process-oriented definitions (Fontanil & Alcedo, 2018; Fontanil et al., 2020).
Various methods have been used to assess resilience. Some studies have measured resilience by the presence/absence of psychopathology, while others have explored related protective factors (for example, self-efficacy) or used specific resilience scales (Stainton et al., 2018; Ungar & Theron, 2019).
In response to the lack of consensus, Ungar (2008) proposed a socio-ecological model in which resilience is considered as both the capacity to individually and collectively negotiate and access the psychological, social, cultural, and physical resources that sustain well-being, and the ability to experience them in culturally meaningful ways. Resilience is therefore composed of multiple and interrelated dimensions and factors present at different ecological levels. These resilience factors include relationships, defined identity, power/control, social justice, access to material resources, sense of cohesion/belonging, and cultural adherence (Ungar, 2013).
Contextual and dynamic definitions of resilience have the advantage of recognizing the heterogeneity of functioning shown by people who overcome adverse situations and, in the specific case of IPV against women, remove the focus from the negative discourse surrounding women exposed to IPV (Ahmad et al., 2013; Howell et al., 2018). At the same time, ecological definitions highlight those resilience factors that are potentially variable over time versus those that are more stable (Howell et al., 2018; Kuijpers et al., 2012). Process-oriented definitions are useful for analyzing women’s resilience in the different phases of the violent relationship and after the break up, detecting which are the main resilience correlates at each stage (Labronici, 2012). Also, from an ecological perspective, political efforts to provide women with resources to cope with their circumstances are vital (Fontanil & Alcedo, 2018; Fontanil et al., 2020).
Correlates of resilience
Despite these conceptual and methodological issues, research on resilience in recent decades has identified a series of consensual and cross-cutting resilience factors that are present at the individual, relational, and contextual ecological levels. In the individual sphere, a range of resilience-enhancing variables have been reported, including problem-solving skills, agency and self-efficacy, sense of humor, adaptability, and meaning making. In the relational domain, the literature describes a variety of characteristics related to the quality of relationships, for example, security, trust, nurture, care, stability, and acceptance. Finally, the contextual resilience resources include sense of belonging, educational and employment resources, service provision, and policies (Bushati, 2020; Liebenberg et al., 2017; Stainton et al., 2018; Ungar & Theron, 2019).
Until recently, women survivors of IPV tended to be a largely disregarded collective in research on resilience. Over the last number of years however, the study of this construct in the field of IPV has attracted increased research attention (Fernández-Sánchez & López- Zafra, 2019; Howell et al., 2018; López-Fuentes & Calvete, 2015), resulting in the identification of different factors related to resilience in this specific population. In this context, there is a clear need to compile the findings obtained in this field to determine if there is sufficient empirical evidence to enhance resilience in women survivors of IPV and to suggest future research directions.
With this objective, the present study reviews the empirical literature on resilience in women survivors of IPV to (a) analyze the consensus on the definition and assessment of resilience; (b) study the correlates of resilience in women survivors of IPV; and (c) reflect on the applicability of such knowledge in policies and professional practice.
Method
A systematic review was carried out in accordance with PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses; Moher et al., 2009) and the recommendations for the reporting of systematic reviews (Rubio-Aparicio et al., 2018).
Inclusion criteria
The inclusion criteria did not apply any limits to publication date, context, and methodology or experimental design. Exclusion criteria were defined for document type, language, and the publication adaptation to the aims of the systematic review. A full list of the inclusion and exclusion criteria for the identification of relevant literature is shown in Table 1.
Search strategy
Literature searches were performed in PsycINFO, Scopus and all the Web of Science (WoS) bibliographic databases. Figure 1 presents the combinations of keywords entered into WoS and a summary of the study selection process. Searches were repeated in all of the databases using the corresponding keywords in the Spanish language. An additional four relevant articles were found through a manual search in Google Scholar, and another two by scanning references from the selected articles. Both hand-searching (Vassar et al., 2016) and reference list scanning (Liberati et al., 2009) are supplemental approaches that increase the quality of systematic reviews.
a The search keywords used in WoS: TOPIC: ("Intimate partner violence" OR "IPV" OR "Intimate partner aggression" OR "Partner violence " OR "Partner abuse" OR "Partner aggression" OR "Domestic aggression" OR "Domestic violence" OR "Domestic abuse" OR "Gender violence" OR "Domestic aggression" OR "Spouse abuse" OR "Spouse violence" OR "Spouse aggression" OR "Spousal abuse" OR "Spousal violence" OR "Spousal aggression" OR "Violence against women" OR "VAW" OR "battered woman" OR "mistreat*" OR "dating violence") AND TOPIC: ("Resilience" OR "Resilient" OR "Resiliency") AND TOPIC: ("Woman" OR "Women" OR "Female" OR "Females" OR "Girl" OR "Girls") NOT TOPIC: ("homosexual" OR "gay" OR "lesbian" OR "LGBTQI" OR "LGBTQI" OR "transgender" OR "transsexual" OR "same-sex relationship" OR "same sex relationship" OR "same-sex partner" OR "same sex partner"). Refined by: Type of document: (article), Language: (English OR Spanish).
The search yielded 42 articles from 2000 to 2019. Articles included in the review are marked with an asterisk in the references section. To extract the data, a form was developed containing sociodemographic, methodological, and theoretical information. Results obtained were organized according to the following areas: (a) the aims of the systematic review (to analyze the consensus on the definition and assessment of resilience/to study the correlates of resilience in women survivors of IPV); and (b) the design of the selected studies (qualitative/quantitative).
Results
Description of the studies
Most of the reviewed studies were cross-sectional (85.8%) and quantitative (57.1%).
Among the studies that recruited sample from services (81%) or from services and other sources (7.1%), a high proportion recruited either from specific women’s and IPV services (46%) or from unspecific services (40.5%); only 13.5% recruited sample from both types of services.
Resilience definition and assessment
Most of the studies, whether qualitative or quantitative/mixed-method, included at least one definition of resilience, as shown in Table 2. Qualitative inquiries presented process-oriented definitions more frequently than quantitative and mixed-method studies, which often defined resilience from a trait-oriented perspective. The outcomes-oriented approach was the least common in both types of research.
Turning to the assessment approach (Table 3), a large number of qualitative studies used semi-structured interviews and frequently explored contextual factors and resilience changes over time. Quantitative articles usually employed a resilience scale. When Resilience was measured in this way, only The Resilience Research Centre Adult Resilience Measure (RRCARM; Liebenberg et al., 2012) comprehensively explored resilience from an ecological perspective, incorporating social/community inclusion, attachment and family support, and national and cultural identity as part of the assessment (Scrafford et al., 2019). Among the small number of quantitative studies that assessed resilience as an outcome, scores in depression and/or PTSD scales were the most frequent outcome variables (Machisa et al., 2018; Kramer et al., 2015), but studies also measured revictimization by a new partner (Herrero et al., 2018) and quality of life (Kuijpers et al., 2012). Only a few studies assessed resilience through a set of variables; in this case, contextual variables such as social support or marginality were also reported, but not from a dynamic perspective (Alvi et al., 2008; Cesario et al., 2014; McFarlane et al., 2014).
Correlates of resilience in women survivors of IPV
The present systematic review found several factors associated with resilience at the individual, relational, and contextual ecological levels (Table 4).
The individual variables identified include personal characteristics such as problem- focused coping, sense of control, self-efficacy, self-esteem, sense of humor, optimism/hope, proactivity, adaptability, creativity, and spirituality. Studies also explored resilience processes, among these, identity reconstruction, reframing the violent relationship, setting goals, and personal growth.
One of the reviewed articles established physical activity as an enabler of women’s resilience (López-Fuentes & Calvete, 2015). The existing data suggest that resilience is positively correlated with perceived physical health (Humphreys, 2003) and negatively correlated with early labor and delivery (Scrafford et al., 2019), and medication consumption (Fernández-Sánchez & López-Zafra, 2019). In contrast, Ford-Gilboe et al. (2009) found that resilience had a direct effect on mental health but not on physical health.
As regards the correlates of resilience and psychological health, research has shown that higher resilience is positively correlated with psychological health and negatively correlated with anxiety-depression and post-traumatic stress symptoms. Choi et al. (2019) noted that resilience did not predict generalized anxiety scores when controlling for variability in the heterogeneity of women’s past victimization. The authors posited that this was because economic and employment stability, social support, and the history of abuse had a stronger effect than resilience on the anxiety levels of their sample given the participants’ characteristics (young women with children).
As for the relational factors, the ability of women to rebuild and expand their social network was key. Help-seeking and the use of support systems, both formal and informal, were positively correlated with resilience, as they were a way to access instrumental/material, informational, and emotional resources, while providing validation and models of strong and empowered women.
Many of the studies found that having children enhanced resilience. One article, however, suggested the opposite, that is, women without children had higher resilience scores and suffered less violence (Jaramillo-Vélez et al., 2005).
Some of the articles examined the influence of a history of child abuse and adversity on resilience. Herrero et al. (2018) found that exposure to abuse during childhood increased the likelihood of a woman being non-resilient (more likely to have had more than one abusive partner). Other studies observed higher resilience scores in women who had been victimized during their childhood (Roditti et al., 2010; Schultz et al., 2009; Scrafford et al., 2019), but only Scrafford et al. (2019) reported statistically significant results.
Finally, among the contextual elements, the most frequently studied factor was formal support. As stated above, this systematic review acknowledged that formal support is positively related to resilience, as it is a way for women to access the resources they need in their day-to-day lives and generates a sense of community belonging.
Figure 2 contains a summary of the main contextual, relational, and individual factors associated with resilience in women survivors of IPV.
Discussion
The global aim of the present study was to analyze if the empirical evidence on resilience in women survivors of IPV is useful for guiding future research and policies, shedding light on what supports are needed by the women to restore control over their lives. In this regard, the findings from this systematic review offer helpful insights.
Although the conceptual confusion around the definition and assessment of Resilience was also reflected in IPV research, studies have identified several correlates of the construct in the individual, relational, and contextual dimensions. This result is in line with what has been found in studies in other populations affected by extreme stress, trauma, violence, or low socioeconomic conditions, such as immigrants, refugees, indigenous, and LGBTQ populations (Liebenberg et al., 2017).
The appraised articles found moderate to high resilience scores in women survivors of IPV, which could be explained by the high proportion of studies that recruited sample from support services. Participants’ resilience may have been reinforced by the process of help- seeking (Crowe et al., 2016) or by the support they received from services (WHO, 2018).
Research must consider this distinction, given that not every woman who suffers from IPV has access to the same quantity and quality of resilience resources.
The individual variables and processes found by this systematic review (e.g., reframing the relationship, identity reconstruction, setting goals, self-efficacy, etc.) are in line with the results from classical research on the process of leaving an abusive relationship (Burke et al., 2004; Enander & Holmberg, 2008; Landenburger, 1989; Merritt-Gray & Wuest, 1995; Moss et al., 1997). Sense of control stands out as one of the principal sources of resilience. Since trauma destroys a woman’s control over her own life, restoring power to the survivor is a key part of the recovery process (Herman, 1992). To resist and survive violence, women adapt to the circumstances, fluctuating between a powerless/victim identity and a powerful/survivor identity in a nonlinear process (Van Schalkwyk et al., 2014). Sense of control influences the help-seeking process in a complex way: perceived low control is related to the need for help but actual help-seeking behavior seems to be associated with a prior higher sense of control (Katerndahl et al., 2019). Once the violent relationship has ended, levels of coercive control decrease and women’s space for action expands, especially with respect to psychological well-being and relationships, but it is important to clarify that through the recovery process women’s space for action is sometimes constrained by structural barriers (Sharp-Jeffs et al., 2018).
In the relational and contextual dimensions, help-seeking and support obtained from informal and formal networks are essential components, since they promote access to informational, instrumental, material, and emotional resources, as established in other studies (DGVG, 2019; Lelaurain et al., 2017).
Previous research has shown that one of the main motivators for help-seeking is the presence of children (Fontanil et al., 2020; Katerndahl et al., 2019; Wood et al., 2019). However, other studies have found that having children can be an obstacle in the process of breaking up the violent relationship (DGVG, 2019; Domenech del Río & Sirvent-García del Valle, 2016; Lelaurain et al., 2017). As recognized by the Council of Europe in Resolution 1714 (2010), every child witnessing violence against their mother is a victim of a form of psychological abuse with potentially severe consequences. The present systematic review shows that women become aware of the effects of violence on their children and seek help for their well-being. Empirical evidence points to an urgency for institutional holistic responses to IPV, and for collaboration between child welfare and IPV agencies to address women and children’s needs (Langenderfer-Magruder et al., 2019; Mennicke et al., 2018).
Finally, employment, education, housing, and a safe environment are facilitators of resilience in women survivors of IPV, while other macrosystemic variables, such as gender roles, social attitudes, and different structural barriers, negatively influence the process of leaving the violence. These results are consistent with previous research in the field, which endorses the adoption of policies to provide women and children with instrumental, material, and informational support and to fight gender discrimination and social attitudes that are favorable to violence (DGVG, 2019; Fontanil et al., 2020; Lelaurain et al., 2017; WHO, 2018). Political actions taken at the macrosystemic level not only enhance the contextual resilience but also reinforce resilience in the individual and relational spheres, since all ecological systems are interrelated (Bronfenbrenner, 1979).
Limitations
This is the first systemic review to explore factors associated with resilience in women survivors of IPV. It has identified an exhaustive list of existing studies, providing a useful summary of their content. This review, however, is not without its limitations.
The first limitation relates to conceptual and methodological issues in some of the reviewed articles: for example, the use of convenience, nonrepresentative, and small-sized samples; the use of retrospective assessment (risk of recall bias); and in some cases, the use of non-standardized IPV assessment and data collection based exclusively on women’s reports with no triangulation (risk of social desirability and/or recall bias).
In addition, the risk of bias in the articles included in this systematic review was not evaluated. Future research should assess the quality and risk of bias in the selected studies, as recommended by Moher et al. (2009) and Ruiz-Aparicio et al. (2018). However, it is important to point out that the researchers who carried out this systematic review carefully appraised the bibliographical records from the databases to ensure a high level of quality.
Further, hand-searching and reference list scanning were carried out to minimize bias and increase the quality of the systematic review (Vassar et al., 2016). It would also have been useful to perform the searches in additional databases to reduce the risk of publication bias. Finally, by selecting empirical evidence published only in English or Spanish, information contained in studies written in other languages was overlooked.
Further research
Future quantitative studies with large sample sizes should test integrative resilience models in order to obtain a consensual, operative, multidimensional, contextual, and process- oriented definition, as recommended by experts in the field (Bushati, 2020; Chmitorz et al., 2018; Stainton et al., 2018; Ungar, 2019; Ungar & Theron, 2019). Further, with recent research supporting the idea of resilience as a phenomenon that extends beyond the appearance of the stressor event (Chmitorz et al., 2018), longitudinal investigations monitoring resilience in different phases and after the break up are needed. Qualitative studies could be useful in this regard.
Research should assess IPV by using standardized instruments or comprehensive interviews carried out by experts in the field, properly distinguishing coercive control situations from situational couple violence (Johnson, 2008). Studies should also record and consider the type of IPV (psychological/emotional, physical, sexual), the duration of the violence, and the time elapsed since the end of the relationship. Some authors highlight the need to explore the violent behaviors of women to better understand the context of the violence (Herrero et al., 2018; Howell et al., 2018; Kuijpers et al., 2012). In this respect, further research could triangulate the information with data provided by other sources.
Research on resilience needs to delve into the mechanisms through which vulnerability or protective factors exert their influence on groups with specific risk conditions. The particularities of the adversity and how these interact with the personal and environmental characteristics of the people concerned make it difficult to extrapolate findings from one sample to another. Therefore, it is important to consider the heterogeneity of adverse events that may be experienced by women. Future research should adopt an ecological perspective and analyze macrosystemic factors and policies, considering the accessibility and availability of resources at each ecological level (Bushati, 2020; Fontanil et al., 2020; Ungar, 2019; Ungar & Theron, 2019). It would also be interesting to incorporate an ethnographic focus to explore the relationship between race or ethnicity and resilience.
Finally, the feminist perspective could enrich studies by drawing attention to women’s voices. Consulting women and working with their narratives and ideas is an appropriate way to build guidelines for professional interventions and policy practices (Crawford et al., 2009; Davis, 2002; Fontanil et al., 2020; Leung et al., 2019).
In summary, it has been detected that research on this field have had a stronger focus on individual resilience factors when compared with the study of contextual variables, therefore it becomes necessary to deeply explore resilience from a contextual perspective that includes ecological mesosystemic, exosystemic, and macrosystemic factors. Furthermore, it is essential to analyze the influence of multiple heterogeneous women´s vulnerability conditions that could potentially affect resilience, such as race, ethnicity, and disability. Finally, there is also a lack of long-term studies following women after the end of the relationship. It could be beneficial to full this gap by analyzing women´s resilience in the post-separation phase, especially when they must keep in contact with their former partner because of the presence of children in common.
Conclusions
The study contributes significantly to the body of literature by synthesizing existing empirical evidence on the factors of resilience in women survivors of IPV. To our knowledge, this is the first systematic review on this topic with this specific population. Drawn on the reviewed articles, this study provides support for multidimensional, contextual, and process-oriented explanations of resilience in women survivors of IPV. Individual, but also relational and contextual factors are core in women´s resilience and should be considered for guiding future research and policies.
Given the gravity of IPV, recognized as a serious public health issue affecting a high proportion of women around the world, governments and institutions should take actions to support women and children and to prevent future IPV. Evidence-based policies and professional practices are necessary to facilitate women’s access to education, employment, and housing, which would promote their independence and sense of control, as well as protect them from revictimization. If violence undermines women and children’s psychological and physical well-being, then society should ensure survivors have access to professional health care. It is also important to design prevention programs and implement them on an ongoing basis and at all ecological levels, since violence is a multicausal and multidimensional phenomenon.