The psychological subtype of intimate partner violence and its effect on mental health: protocol for a systematic review and meta-analysis

Intimate partner violence (IPV) is a global health problem characterized as any behaviour within an intimate relationship that causes physical, psychological or sexual harm [1]. At present, it is well-documented that IPV can cause extensive mental health consequences among its victims [2‐5]. IPV can be characterized as an interpersonal trauma, and symptoms of posttraumatic stress disorder (PTSD) have been identified in 31–84.4% of women exposed to IPV [2]; along with other comorbid symptoms such as depression, anxiety, suicidality, substance abuse and sleep disturbances [2, 5, 6].

The subtype of psychological violence (compared to physical and sexual violence) is estimated to be the most common form of IPV in both the USA [7] and Europe [8], affecting between 35 and 49% of men and women. This has led legislators in some European countries to criminalize psychological violence as an independent offence, making it equally punishable as physical violence (e.g. Norway [9] and England [10]). Although some researchers have argued that psychological violence in itself cannot be classified as a trauma, as it does not meet the first criterion of diagnosing PTSD (i.e. threat to life or physical integrity [11, 12]), a more recent review on IPV and mental health argues that psychological violence can independently cause PTSD, depression and anxiety [13].

Despite both legal recognition of psychological violence and documentation of its effect on mental health, the conceptualization of the phenomenon is ambiguous in both research and clinical practice. Acts of psychological violence are distributed along a continuum starting from what is commonly termed psychological aggression (e.g. yelling and insults) and ending with more severe abuse, often labelled coercion (e.g. threats and isolation). How we interpret psychological aggression and how we distinguish it from more severe abuse depends, among other things, on the context in which it occurs, when it occurred in a sequence, how it was interpreted and whether it was perceived as abusive [11].

Another challenge is that psychological violence is often characterized in different ways. For example, the conceptualization of “coercive control” can generally be understood in two ways: firstly, as an overall attempt to control one’s partner, in which IPV is a way of achieving control; secondly, as a subtype of IPV which is similar to—or a part of—the concept of psychological violence. The former is described in a structural perspective as more severe and gender-asymmetrical and is generally understood to be a representation of gender inequality [14‐18]. The latter reflects a continuum of IPV from psychological aggression to more controlling behaviours constituting an assault [11]. Moreover, these are theoretical distinctions that have proven difficult for researchers and practitioners to conceptualize and apply.

The distinction between psychological violence and coercion is evident from the WHO, who define psychological violence (i.e. emotional or psychological abuse) such as “insults, belittling, constant humiliation, intimidation (e.g. destroying things), threats of harm, threats to take away children”, while controlling behaviour is defined as “isolating a person from family and friends; monitoring their movements; and restricting access to financial resources, employment, education or medical care” [1] (p. 1). The specific acts of violence separate these two definitions, while the overall psychological harm combines them. Likewise, the European Institute of Gender Equality (EIGE) states an overall definition of psychological violence, which can be understood from the perspective of caused harm: “Any act or behaviour which causes psychological harm to the partner or former partner. Psychological violence can take the form of, among others, coercion, defamation, a verbal insult or harassment”¹⁹ (p. 45).

These variations in terminology are inevitably reflected in the psychometric instruments developed to assess psychological violence (Table 1; identified by the Centers for Disease Control and Prevention and the National Unit [20] against IPV in Denmark, LUV [21]). Indeed, the varying use of subscales (e.g. #7 and #9) and differing definitions of psychological violence make the findings from these studies difficult to compare and stresses the need to evaluate how the effect on mental health is influenced by such variations. For example, some instruments assess psychological aggression (e.g. #10), while others measure more severe controlling behaviours (e.g. #4). Furthermore, other instruments focus less on the act of violence and more on the effect on the victim (e.g. #12). When measuring the effect of psychological violence on mental health, the difference between acts and effects can be crucial. Evidence demonstrates that some acts defined as psychologically violent (i.e. threats to kill/harm) have been found to significantly load on a physical violence factor, most likely due to the aspect of physical threat, which makes the psychological and physical aspect hard to differentiate [11]. Furthermore, a majority of these assessment tools make use of frequency scores that do not differentiate between the severity of items. As such, items of psychological aggression (i.e. being called “ugly” and “worthless”) are equated to severe controlling behaviour and threats (i.e. threats to kill or take away children). This exemplifies how important these distinctions are if we wish to understand the independent effect of psychological violence on mental health. Examining mental health while evaluating the instruments used to measure the phenomenon will help us understand whether severity, frequency or actual “type” of psychological violence produces the most harm on the victim’s mental health [11].

In addition to problems with the conceptualization, a number of methodological challenges further characterize the field, e.g. sampling, design, scoring and gender bias [11]. Focusing on psychological aggression in lesbian, gay and bisexual individuals, Mason et al. [22] highlight the need for future research to clearly and consistently define psychological violence and separate it from other types of violence seeing that a more consistent definition will facilitate better comparisons across studies. The research group further stresses that scoring methods (e.g. frequency vs. dichotomous scoring) influence the magnitude of the effect size, which makes results difficult to compare. This challenge is further complicated by the use of self-administered questionnaires that may lack systematic development [22]. Moreover, Follingstad [11] emphasizes the need to differentiate between samples (i.e. dating relationships and marital or long-term cohabiting relationships), seeing that dating relationships are characterized by quantitatively and qualitatively less psychological violence. Finally, the majority of assessment tools are developed to specifically measure female victimization of psychological violence, despite male victimization being reported at equally high rates in some studies [7]. Although fewer studies have focused on the effects on mental health among male victims of psychological violence, studies indicate that they too present symptoms of anxiety, depression and sleep disturbances [13, 23].

The aim of the present systematic review is to build on existing knowledge [13] concerning the effect of psychological violence on mental health, while evaluating the psychometric instruments used to assess psychological violence about how they conceptualize the phenomenon. To this day, most studies on IPV and mental health have pooled scores of physical, psychological and sexual IPV in their reporting [2, 4], making a distinction of the individual effect of psychological violence difficult. When directly examining the effect of psychological violence, the lack of a clear and consistent definition of psychological violence has made results difficult to compare [13, 22]. Consequently, important information is lost. The present study therefore aims to consolidate knowledge on psychological violence by evaluating whether frequency, severity or actual “type” of psychological aggression is associated with the most harm on mental health [11]. Methodological challenges will be considered by conducting quality assessments of all included studies, and results will control for the presence of physical and sexual violence (i.e. severity, duration and frequency). When possible, mediating and moderating factors will be evaluated, as well as potential gender differences. Based on the results, a discussion on assessment tools and methodological challenges will provide the grounds for recommendations concerning future research.

To sum up, the aim of this systematic review and meta-analysis is fourfold: (1) to estimate the individual effect of psychological violence on mental health (e.g. PTSD, depression and anxiety); (2) to estimate whether frequency, severity or actual “type” of violence is associated with the most harm; (3) to investigate gender, sampling and cultural differences through moderation analyses; and (4) to discuss the somewhat vague terminology and methodological challenges.

The present review will build on existing knowledge by statistically synthesizing results on the effect of psychological violence on mental health. The review will help consolidate knowledge on psychological violence by evaluating whether frequency, severity or actual “type” of psychological violence produces the most harm. This will strengthen our knowledge on psychological violence, and how best to assess and conceptualize the phenomenon.

While conducting the systematic review, we expect to encounter several limitations. The varying terminology and definitions of psychological violence will make results difficult to compare, just as the many and varying psychometrics tools will (Table 1). This is further complicated by the fact that many studies are likely to not even use validated instruments but instead base their results on a few self-constructed items. The review will aim to overcome these limitations by thorough quality assessment of the included studies with help from the Quality Assessment Tool for Quantitative Studies [25] and by performing relevant subgroup analyses. This does not only apply to psychological violence, but also mental health definitions and the psychometric instruments used to assess symptomatology. By critically examining the applied definitions and terminology as well as methodological challenges (e.g. sampling, design, scoring and gender bias) the review will serve as a status quo of the field and make grounds for future recommendations.

Finally, the systematic review is expected to have several clinical implications. We expect the meta-analysis to deepen our understanding of the different subtypes of psychological violence and how they independently interact with mental health outcomes. Likewise, we expect to give clarity on psychological violence and whether it should be understood as a traumatic event equal to other types of abuse (e.g. physical or sexual violence). Developing both preventive efforts and treatment programmes such information is important if we wish to directly target the needs of those affected by psychological violence and raise awareness to encourage both victims and perpetrators to seek help.

In research, this systematic review is expected to inspire researchers to clearly and consistently define psychological violence while carefully considering the psychometrics used to measure the phenomenon, as well as other methodological challenges. Moreover, we expect to get clarity on any potential gender bias.

On a final note, this work will hopefully inspire others to conduct similar research on children who grow up as either witnesses or victims of psychological violence.