Intimate Partner Violence (IPV) and Witnessing Domestic Violence (WDV): A Comparison of Italian and International Evidence

1. Introduction

Intimate Partner Violence (IPV) and Witnessing Domestic Violence (WDV) which derives directly from IPV, are a major problem worldwide, having severe repercussions on the physical and mental health of those involved, and also has economic implications. Italian and international news platforms carry almost daily reports of incidents concerning domestic violence, intimate partner violence, and violence by men against women. Many countries conduct surveys on these topics but scientific research with qualitative and quantitative studies to establish the possible correlations between IPV, mental disorders, and risk factors, is mainly done in the UK and the United States of America. In 2011 the WHO developed guidelines for health professionals on how to respond adequately to intimate partner violence and sexual violence against women.

This review aims at describing the phenomenon of IPV in terms of scientific definitions, international epidemiological data and psychiatric populations, risk factors, and general and mental health consequences, using the available literature. We carried out a free search on Google, and a specific search of Pubmed and PsycInfo databases. Up-to-date epidemiological data were obtained via Google and from national statistic websites such as the Italian Statistics Institution (ISTAT) and the UK Home Office. We conducted a separate search for information regarding WDV.

2. Materials and Methods

A) Intimate Partner Violence and Domestic Violence

Searching Medline via Pubmed using the string below, we found 757 publications on IPV and mental disorders published over the past 17 years:

(("Mental Disorders"[Majr]) AND "Crime Victims"[Majr]) AND (("Domestic Violence"[Majr]) OR "Intimate Partner Violence"[Majr])

and we found 338 papers in PsycInfo using this search string:

MAINSUBJECT.EXACT.EXPLODE("Victimization") AND MAINSUBJECT.EXACT.EXPLODE("Mental Disorders") AND(MAINSUBJECT.EXACT.EXPLODE("Intimate Partner Violence") ORMAINSUBJECT.EXACT.EXPLODE("Domestic Violence"))

We interrogated Google and national websites such as ISTAT and the UK Home Office to obtain up-to-date epidemiological data on domestic violence.

B) Witnessing Domestic Violence

We also conducted a search on Medline via PubMed for Witnessing Domestic Violence, keeping in mind that it derives directly from IPV. We used the string below and found 334 publications published over the last 17 years:

("Mental Disorders"[Majr]) AND (Witnessing Domestic Violence OR children witnessing violence)

We searched PsycInfo with the string below and found 205 papers

MAINSUBJECT.EXACT.EXPLODE("Victimization") AND (Witnessing Domestic Violence OR children witnessing violence)

3. Results

3.1. IPV Definitions

Terminologically speaking, intimate partner violence (IPV) lies within the broader spectrum of "possible violence" that occurs within the household. This violence not only involves spouses or partners but can also include violent acts perpetrated by a parent toward a child or by other relatives toward family members. The definition of domestic violence provided by the UK Home Office is "any incident of controlling, coercive, threatening, violent or abusive behaviour between people aged 16 and over who are, or have been, intimate partners or family members, regardless of gender or sexual orientation." Abuse can be "psychological, physical, sexual, financial, emotional".¹

IPV specifically includes physical, sexual, emotional violence, and controlling behaviours (e.g., stalking) by a current or past partner, excluding other forms of family violence that do not manifest in a partner relationship. An "intimate partner" is defined as a person with whom one has a close personal relationship characterized by emotional connection, regular contact, physical and sexual contact, identity as a couple, and familiarity with each other's lives.

The clinical-legal issue of IPV can be examined from three angles²: physical violence, sexual violence, and emotional (or psychological) violence. Physical violence is defined as the intentional use of physical force that can cause death, disability, injury or pain, including behaviours such as scratching, pushing, choking, slapping, hitting, and the use of weapons. Sexual assault is defined as a sexual act committed or attempted without free consent or against a person incapable of giving consent, including forced or substance-facilitated penetration, unwanted touching, and compulsion to engage in sexual acts with a third party.

Emotional (or psychological) violence consists of the use of verbal and nonverbal communications with the intent to mentally or emotionally harm a person and/or exert control over them. This form of violence can be expressive (e.g., name calling, humiliating), coercive (limiting access to transportation, money, friends), excessive monitoring, threats of physical or sexual harm, and control of reproductive or sexual health. Emotional violence often coexists with other forms of IPV, may precede physical and sexual violence, and has a significant impact on overall IPV.

Controlling behaviours also include stalking, which involves repeated and unwanted attention and contact that causes fear or concern for one's own safety or that of others.

3.2. International Prevalence Data

"Violence against women is nowadays widely recognized as a serious human rights abuse and, increasingly, as a major public health problem with substantial consequences for women's physical, mental, sexual and reproductive health".³ This "awareness was reinforced by agreements at international conferences in the 1990s that identified the paucity of adequate information on the prevalence, nature, causes and consequences of violence around the world as a serious obstacle to a broader recognition of the scope and seriousness of the problem and the development of effective intervention strategies. Since that time, international research has produced more evidence of the prevalence of violence against women, particularly physical violence perpetrated by male intimate partners".³ A review of more than 50 population-based studies in 35 countries conducted before 1999 showed that "between 10 and 52 percent of women worldwide report having been physically abused by an intimate partner in their lifetime" and "between 10 and 30 percent have experienced sexual violence by an intimate partner".^4,5 Due to differences in the research designs and methods of these studies, however, data comparison was impossible.

In an attempt to find a solution to this problem, a WHO multi-country study on women's health and domestic violence against women was developed, with the objectives of: (a) estimating the prevalence of different forms of violence against women, with particular emphasis on physical, sexual, and emotional violence by intimate partners; (b) establishing the association between intimate partner violence and a range of health outcomes; (c) identifying factors that can both protect and put women at risk of intimate partner violence; and (d) documenting strategies and services that women can use to cope with this type of violence.³ In this regard, population surveys on women's health and experiences of different forms of violence were conducted between 2000 and 2003 in 15 sites located in 10 countries. 24,097 women responded to a standardized structured questionnaire that investigated physical violence, sexual violence, and controlling behaviour by a partner over a lifetime and in the previous 12 months. Between 15% and 71% of women with partners reported physical or sexual violence, or both, by a partner, over their lifetime. Between 4% and 54% of women reported physical or sexual violence, or both, by a partner in the previous 12 months.

The lowest prevalence rates of partner violence over a lifetime and in the previous 12 months were found in metropolitan Japan and Serbia-Montenegro: this finding suggests that abuse rates only partly reflect the country's degree of economic development.

The U.S. National Violence Against Women (NVAW) Survey, sponsored by the U.S. National Institute of Justice and the Centres for Disease Control and Prevention (CDC), found that intimate partner violence is very prevalent in the U.S. population: about 25% of women surveyed and 7.6% of men surveyed said they had been sexually assaulted or physically assaulted by a current or former spouse, cohabitant, or partner in their lifetime; 1.5% of women surveyed and 0.9% of men surveyed said they had been sexually or physically assaulted by a partner in the past 12 months.⁶ The most recent data - again from a survey conducted in America in 2011, the National Intimate Partner and Sexual Violence Survey (NISVS) - indicate that more than 10 million women and men in the United States experience physical violence each year from a current or past intimate partner.

The Crime Survey for England and Wales for the year ending March 2023 estimated that 5.1% of people aged 16 to 59 years had experienced domestic abuse in the last year. It recorded a 5% decrease in domestic abuse-related offences compared with the year ending December 2022. In the year to December 2023, stalking and harassment accounted for nearly a third (32%) of all police recorded violence; this is a 6% decrease from the year ending December 2022. Most recently, conduct crimes such as stalking and harassment, which were until May 2023 recorded in addition to other crimes, are now recorded as the sole offence if the conduct crime is deemed to be the more serious offence. As a result, offences that often occur alongside a conduct crime will no longer be recorded.⁷

Stöckl et al. (2013) ⁸ published an authoritative reference on intimate partner homicide prevalence data; the authors ran a systematic search of 5 databases (Medline, Global Health, Embase, Social Policy, and Web of Science) which provided 2167 abstracts. The authors selected 118 articles with 1,122 prevalence estimates of partner homicides. The data were obtained from 66 countries. 13.5% (IQR 9.2-18.2) of homicides were committed by an intimate partner, and this proportion was 6 times higher for female homicides than for male homicides (38.6%, 30.8-45.3 Vs. 6.3%, 3.1-6.3). The average rates of partner homicide, overall (men and women) and women-only, were higher in high-income countries (overall, 14.9%, 9.2-18.2; women's homicides, 41.2%, 30.8-44.5) and Southeast Asia (18.8%, 11.3-18.8; 58.8%, 58.8-58.8). At least one in seven murders, globally, and more than one-third of murders of women, are perpetrated by an intimate partner. Such violence usually represents the culmination of a long history of abuse.

South Africa has high rates of IPV, with epidemiological research suggesting that up to half of South African women will be a victim of violence in an intimate relationship in their lifetimes. A significantly higher proportion of women compared to men reported being a victim of physical IPV in their lifetime (54% vs. 40%) or in the past 3 months (47% vs. 31%).^9-15

Canadian General Social Survey (GSS) data on victimization from 2014 show how estimates of domestic violence declined over the course of the decade.¹⁶ In 2014 there were about 19.2 million Canadians in the provinces with a former or current spouse or partner; of these, about 760,000 (4%) reported being physically or sexually abused by their partner in the previous 5 years. This figure was significantly lower than the prevalence of domestic violence reported in 2004 (7%) and 2009 (6%).

3.3. Italian Prevalence Data

National data¹⁷ are sourced from the Women's Safety Survey covering the five-year period 2010-2014. Violence against women is a widespread phenomenon. 6,788,000 women have experienced some form of physical or sexual violence in their lifetime: 20.2% have experienced physical violence, 21% sexual violence, and 5.4% more serious forms of sexual violence such as rape and attempted rape. 652,000 women have experienced rape and 746,000 are victims of attempted rape. Current or former partners commit the most serious violence with 62.7% of rapes being committed by a current or former partner. In contrast, perpetrators of sexual harassment are strangers in most cases (76.8%). There are important signs of improvement from the previous survey: physical or sexual violence has decreased from 13.3% to 11.3% in the last 5 years (2010-2014), compared to the 5 years prior to 2006. This is due to more information, field work, but above all, a greater ability of women to prevent and combat the phenomenon and a social climate that increasingly condemns violence. Both physical and sexual violence have diminished, both by partners and ex-partners (physical from 5.1% to 4%, sexual from 2.8% to 2%) and non-partners (from 9% to 7.7%). Psychological violence by current partners is also in sharp decline (from 42.3% to 26.4%), especially when not accompanied by physical and sexual violence. Women are increasingly able to leave violent relationships or prevent them, and they are also more aware of the problem. They consider violence in the relationship as a crime more often than previously (from 14.3 % to 29.6 % for partner violence) and report it more frequently to law enforcement agencies (from 6.7 % to 11.8 %). They tend to confide more about their concerns regarding violent situations (from 67.8% to 75.9%) and seek help from specialized services, anti-violence centres and counters (from 2.4% to 4.9%). The trend for violence by non-partners is the same.

However, negative elements are also reported: there is no change in the hard core of violence, rapes and attempted rapes (31.2% for both 2006 and 2014). The violence is more serious with increases in the number of cases that caused injuries (26.3% to 40.2% by partners) and the number of women who feared for their lives (18.8% in 2006 to 34.5% in 2014).

3,753,000 women (16.1% of the female population) have experienced stalking in their lifetime; in 1,524,000 cases the offender was an ex-partner, in 2,229,000 persons other than an ex-partner.¹⁷

The Covid-19 pandemic and the measures taken to contain its spread (e.g. confinement at home), as well as the unfolding socio-economic consequences of the crisis triggered by the health emergency, have accentuated the risk of violent behaviour.

In Italy in 2020, more than 15,000 women started a personalized path-out-of-violence in the anti-violence centres and in more than 70% of the cases the violence did not originate with the pandemic. The data show that the mobility-restrictive measures taken to contain the pandemic have amplified women's fear for their own safety. Indeed, in the first nine months of 2020 there was an increase in reports of violence in which the victim felt her life and/or that of her children or close family was in danger (3,583 compared to 2,663 in 2019). In contrast, the easing of the constraints in the same perod of 2021 led to a decrease in reports of violence in which the victim perceived imminent danger (2,457 in 2021). The awareness campaign to help victims of violence feel less alone during the pandemic also led to less severe violence reported to the toll-free anti-stalking and anti-violence number in 2021 compared to 2020. Finally, in terms of homicide rates, of the women killed in 2020, 57.7 % were in a relationship and were murdered by their partner.¹⁸

3.4. Prevalence Data of IPV in Special Populations: Psychiatric Diagnoses and Patients with Severe Mental Illness (SMI)

The studies conducted by King's Women's Mental Health (KWMH) of King's College London, formed by Prof. Louise Howard in 2008 and now led by Siam Oran, achieve considerable impact, with its research on perinatal mental health, domestic abuse, and modern slavery being recognized in the Research Excellence Framework (REF) 2014 and 2021. A study conducted by the group (Trevillion et al., 2012)¹⁹ analysed 42 studies on the prevalence of IPV in mental disorder patient populations in a systematic review and meta-analysis, reported the following findings:

• In the sample affected by depressive disorder (7 studies), the median value of lifetime IPV is 45.8%, with pooled odds ratio of 2.77 (95% CI 1.96-3.92); on the other hand, the median value of prevalence in the last year (7 studies) is 35.3% with pooled odds ratio of 3.31 (95% CI 2.35-4.68).
• In the sample with anxiety disorder (5 studies), the median value of lifetime IPV stands at 27.6% with pooled odds ratio of 4.08 (95% CI 2.39-6.97), while that in the last year (4 studies) is 28.4% with pooled odds ratio of 2.29 (95% CI 1.31-4.02). In women with PTSD (4 studies) the median value of lifetime prevalence of IPV is 61.0% with pooled odds ratio 7.34 (95% CI 4.50-11.98), while it is known from a single study that the prevalence of physical IPV in the last year in women with PTSD is 27.0% with OR 3.62 (95% CI 2.32-5.67) compared with the group of women without PTSD.
• Special mention should be made of the group of patients with common mental disorders (CMD), i.e., depressive and/or anxiety disorders identified but not disaggregated, a diagnostic set in which the diagnoses of adjustment disorder with anxiety, adjustment disorder with depressed mood, adjustment disorder with mixed anxiety and depressed mood could be included. In that sample, traceable to 3 studies, the median value of lifetime prevalence of IPV is 48.0% with a higher likelihood in the CMD group to experience lifetime IPV than women without mental disorder. A UK national survey of 7047 people reported increased odds of IPV in the past year in women with CMD (OR: 4.4 95% CI: 3.32-5.82) compared with women without CMD with a prevalence estimate of 15.2%.^20-22

Following Trevillion et al.’s systematic review of 42 studies (without control groups) showing a lifetime prevalence of IPV of 33% among women with psychiatric pathology followed by community-based outpatient clinics, Oram et al. (2013) explored the UK national territory with respect to the IPV phenomenon in the SMI population, starting with a comparison with a control group.²³ In particular, they referred to two specific papers on the topic:

• In the first paper, Khalifeh et al. (2014),²⁴ the aim was to compare prevalence and impact of IPV in the population of patients with SMI versus the general population. 303 randomly recruited psychiatric patients in contact with community services for more than 1 year were interviewed using the British Crime Survey domestic/sexual violence questionnaire. Prevalence and correlates of violence in this sample were compared with data from 22,606 general population controls concurrently participating in the 2011-2012 National Crime Survey. Lifetime domestic violence was 69% in women with SMI vs. 33% in female controls. 49% in men with SMI compared with 17% in male controls. Domestic violence in the past year was found to be 27% in women with SMI vs. 9% in female controls. 13% in men with SMI compared with 5% in control men. Lifetime sexual violence was found to be 61% in women with SMI compared to 21% in female controls, while 23% in men with SMI compared to 3% in male controls. Sexual violence in the past year was found to be 10% in women with SMI compared to 2% in female controls. Family (non-partner) violence included a larger proportion of all domestic violence in the SMI group than in the control victims (63% v. 35% p<0.01). Compared with non-SMI women, women with SMI were more likely to report adverse psychological/social consequences (91% v. 64% p<0.001) and attempted suicides (53% v. 3% p<0.001) as a result of serious sexual assaults suffered in adulthood, but an equal likelihood of reporting illness or physical injury (49% v. 40% p=0.35) as a result of serious sexual assaults. Finally, women with SMI who had experienced IPV were more likely than controls to be able to disclose their experiences of violence to health professionals (43% v. 15% p<0.001) and the police (37% v. 16% p<0.001); while an equal proportion between the two groups were able to confide in informal networks. In summary, this study shows that people diagnosed as SMI who are followed by psychiatric services have odds 2 to 4 times higher for all subtypes of violence (emotional, physical, sexual) than the general population and odds 6 to 8 times higher for sexual assaults; 50% of women who experience severe sexual assaults attempt suicide. These data suggest that clinicians should investigate, in their practice, not only the experiences of physical violence, but also those of emotional and sexual violence; all the more so on the basis of the studies, found in the literature, which show that emotional abuse has a greater impact on health than physical abuse.²⁵
• The second paper, also by Khalifeh et al.²⁶ analysed data on 23,222 adults participating in the British Crime Survey 2010-2011. After an initial anamnestic information-gathering interview conducted by a trained interviewer, each respondent was asked to fill out a self-administered form pertaining to experiences of emotional, physical, and sexual violence experienced by a partner or ex-partner or family members in the previous year. The main inclusion criterion was the presence of a chronic mental illness, defined as "any enduring mental illness condition such as depression that has lasted at least 12 months or longer and limits daily activities." The presence of emotional, physical, sexual violence was defined by positivity to at least one item in the respective group of questions. Further investigations in terms of secondary outcomes, such as physical or psychological consequences of IPV and seeking help from specific representatives, were requested in case of positivity to IPV. Among women, the prevalence of IPV in the previous year was 20% (89/442) and 5.3% (789/12309) for women with chronic mental illness and no mental illness, respectively. Among men with chronic mental illness, the prevalence of IPV was 6.9% (21/271) compared with 3.1% (356/10221). Comparing the group with mental illness and the group without, the former - if a victim of IPV - was found to have greater negative consequences in terms of emotional/psychological health (53% v. 30%; OR adjusted for socio-demographic variables: 2.2 CI: 1.3-3.8) with particularly high odds in regard to attempted suicide as a result of IPV (13% v. 2%, aOR: 5.4 CI: 2.3-12.9). The two groups equally experienced the physical consequences of IPV intended as illness or physical injury (24%, P=0.97). Victims with and without chronic mental illness equally sought help from any source, but victims with mental illness were less likely - compared to victims without mental illness - to seek help from informal networks (OR adjusted for socio-demographic variables and health problems: 0.47 CI: 0.27-0.83) and more likely to seek help from health professionals (aOR: 2.7 CI: 1.3-5.1).

These results are confirmed by a study in Italy²⁷ conducted on a sample of women participating in a service for the treatment of anxiety disorders and disorders depressants: 36 (24%) were victims of IPV; 35 with emotional abuse, 23 with physical abuse, and 7 with sexual abuse. In the “abused” group, 80% had psychic and physical health consequences, and 53% requested help from relatives/friends and/or healthcare staff.

3.5. Risk Factors Related to IPV: Alcohol and Substances

Although WHO no longer recommends "universal screening" for IPV, it still advocates the importance of asking about conditions that may contribute to or be worsened by IPV.²⁸ Common mental disorders, including the use of alcohol and illicit substances, may increase the risk of IPV victimization and thus deserve thorough clinical attention. According to Schumacher and colleagues,²⁹ the main risk factors for the perpetration of physical IPV include verbal abuse, stress, marital dissatisfaction, anger, and depression; for victimization, on the other hand, child abuse, depression, low education, and violent behaviour toward a partner. A common risk factor for both roles is substance use.

Alcohol is a key factor in aggressive behaviour between intimate partners. Research shows that IPV episodes are more frequent and severe when the perpetrator has consumed alcohol.^30-33 The pharmacological effects of alcohol, such as reduced cognitive and problem-solving skills, increase the likelihood of aggressive behaviour.³⁴

A study by Cunradi et al. (2011) on blue-collar couples found that the husband's problematic alcohol use is associated with unidirectional male-to-female and bidirectional violence, but not with unidirectional female-to-male violence.³⁵ In contrast, other studies indicate that both husband's and wife's problematic alcohol use are associated with two-way violence.³⁶

Two systematic reviews were found showing that opioid use among women is strongly associated with IPV victimization.^37,38 Cannabis use has been correlated with increased victimization and perpetration of both physical and psychological IPV.³⁹ In addition, methamphetamine use has been associated with an increased likelihood of physical IPV victimization and perpetration.

A study by Kraanen et al. (2014) examined the relationships between substance use disorders and IPV in a sample of individuals in substance abuse treatment.⁴⁰ Results indicate that personality-related factors, as well as post-traumatic stress disorders (PTSD), may contribute to the high prevalence of IPV among substance users.

In summary, alcohol and illicit substances play a significant role in perpetuating and experiencing IPV, with complex effects affecting both the severity and frequency of violent episodes.

Risk Factors Related to IPV: Childhood Abuse

The literature shows that being abused in childhood increases the risk of being abused in adulthood.^41-43 For example, an Australian study (IVAWS) shows that the risk of experiencing intimate partner violence (IPV) is one and a half times higher for those who were abused as children (78%) than for those who were not (49%).⁴⁴

Female survivors of IPV are more likely than non-abused women to report physical abuse, sexual abuse, neglect, and witnessing violence (exposure to violence between parents) during childhood.^45-48 These maltreatment experiences tend to co-occur and have lasting effects on mental health, including increased symptoms of post-traumatic stress disorder (PTSD) in both childhood and adulthood.⁴⁹ In addition, experiencing IPV can cause or exacerbate PTSD symptoms.^50-53

Women who experienced childhood abuse and physical or sexual violence in adulthood report more severe PTSD symptoms than those who experienced only some form of victimization in adulthood.⁵⁴ Gobin et al. (2013)⁵⁵ investigated the possible mediation of IPV in the relationship between childhood abuse and PTSD symptoms in adulthood. Contrary to expectations, IPV did not mediate this association. Wuest et al. (2009), Becker et al. (2010) and Trickett et al. (2011) examined the impact of stress and sexual abuse in survivors of IPV. ^56-58 Childhood maltreatment experiences have a direct and lasting effect on PTSD symptoms in adulthood, outweighing the effects of recent IPV experiences. This can be explained by a theoretical framework of personality development, where childhood maltreatment, especially "polyvictimization" (experience of multiple forms of child maltreatment and family violence) is associated with dysregulations in interpersonal functioning and emotional regulation.^59,60 This pattern of chronic dysregulation in affective arousal may explain how childhood maltreatment is directly related to posttraumatic symptoms in adulthood.⁶¹

3.6. General Health Consequences

IPV is not limited to acts of violence but has numerous clinical consequences that affect both physical (illness or injuries) and mental health (anxiety, depression, PTSD). The most extreme consequence of IPV is the death of the victim.

According to Campbell (2002),⁶² the health effects of IPV lead to increased demand for health care and high costs. Population and clinical studies show that abused women have worse mental and physical health, more injuries and higher utilization of medical services than non-abused women. A Canadian study (Ratner et al., 1993) on abused wives showed that this category sought emergency room care and consulted specialists about three times more often than non-abused women.⁶³ In addition, the use of medical services increased with the severity of the physical abuse. Abused women generate 92% higher annual healthcare costs than their non-abused counterparts, with mental health services being the largest cost item. IPV can cause chronic health problems such as chronic pain, gastrointestinal symptoms, heart disease and hypertension. The effects of abuse during pregnancy, such as preterm delivery and other complications, are still being studied. ⁶⁴

Gynaecological problems, such as pelvic inflammatory diseases, sexually transmitted diseases, vaginal bleeding and chronic pelvic pain, are common among IPV victims. Michele C. Black of the Atlanta CDC highlighted that millions of women in the United States suffer IPV injuries each year, with more than 2 million injuries and 1,300 deaths annually.⁶⁵ The United States has the highest intimate partner homicide rate among the world's 25 wealthiest countries, with about 1,500 homicides each year. Between 42% and 66% of women killed by their partners had sought medical care in the previous 12 months.^66,67

Black and colleagues also show the biological effects of IPV, including neural, neuroendocrine, and immune responses to acute and chronic stress, according to McEwen's model of homeostasis, allostasis, and allostatic load.⁶⁸ Chronic and acute stress resulting from IPV can increase secretion of stress hormones and cause prolonged physiological responses, with lasting effects even after the abuse ends.

Mental Health Consequences (PTSD, Anxiety, Depression)

We found only one Italian research group (Romito et al., 2005) that investigated the outcomes of IPV.⁶⁹ A cross-sectional study of women attending six general practice clinics in the province of Belluno showed that episodes of violence were strongly associated with psychological distress, psychoactive substance use, and poor health. Psychological abuse, without physical or sexual violence, was also associated with impaired health status. Women who had been victims of violence in the past year were six times more likely to be depressed and four times more likely to use psychotropic drugs than nonvictims. Comparing these findings with those of a meta-analysis by Golding,⁷⁰ who found the average prevalence of depression among abused women to be 47.6%, Romito and colleagues found 53.3% of abused women had elevated GHQ (psychological distress indicator) scores. Although physical and sexual violence is more prominent, psychological violence has a significant impact on health; 38% of women experiencing psychological abuse only had elevated GHQ scores, compared to 13% of non-abused women. A similar study in France showed similar trends.⁷¹

In the United States, it was shown that PTSD symptoms were predicted by both the intensity of physical aggression and partner dominance/isolation tactics.⁷² Victims often report that psychological violence can be worse than physical aggression.⁷³ An Italian researcher, G. Creazzo, collected similar experiences in her book on surviving sexual violence.⁷⁴

A study by Pico-Alfonso and colleagues examined the impact of physical, psychological and sexual violence on women's mental health.⁷⁵ Abused women showed a higher incidence and greater severity of anxiety-depressive symptoms, PTSD and suicidal thoughts than non-abused women. The concomitance of sexual violence was associated with a greater severity of depressive symptoms and a higher incidence of suicide attempts. In conclusion, psychological violence is as destructive to mental health as physical violence, with independent effects on anxiety-depressive symptoms and contributing to PTSD and PTSD/depression comorbidity.^76-79 Therefore, psychological IPV should be considered by professionals as a significant form of violence.

Domestic violence is also associated with suicidal behaviour, sleep and eating disorders, social dysfunction, exacerbation of psychotic symptoms, and alcohol and substance abuse.⁸⁰ Female victims of IPV are up to six times more likely than non abused women to develop alcohol and substance dependence. The nature of PTSD in these women is often classified as "complex traumatic stress syndrome," which includes changes in attitudes about themselves, the perpetrator, relationships, and their beliefs. In addition, studies have shown a correlation between PTSD and difficulties in positive emotion regulation. There is evidence to support a bidirectional causal link between domestic violence and psychiatric disorders: psychiatric disorders can make a woman more vulnerable to domestic violence, and domestic violence can cause mental health damage.^81-85

3.7. Witnessing Domestic Violence

On a global scale, attention to intra-family violence began after it had been agreed that the rights of the child must be guaranteed, starting from the "Geneva Declaration on the Rights of the Child" of 1924 to the "Convention on the Rights of the Child" of 1989.

In Italy, where the Civil Code that came into force in 1942 still recognises paternal monocratic power, a definition of children’s rights was obtained only after a long legal, social, and psychological journey. Indeed, the concept of the family as an authoritarian institution was only abandoned in 1975 and the first definition of WDV was coined in 2005.

Gracia and Herrero⁸⁶ found that the existence of legislation protecting children from violence, a low infant death rate and personal education is significantly linked to a low level of acceptance of violence.

3.7.1. Children and IPV in Italy: Background and Data

Robust data on the prevalence of children’s exposure to IPV in Italy is limited. As mentioned above, WDV was first defined in 2005 and updated in 2017 as "the experience by the child and adolescent of any form of maltreatment carried out through acts of physical, verbal, psychological, sexual, economic violence and persecutory acts (so-called stalking) on reference figures or other emotionally significant figures, adults or minors”.⁸⁷

However, the most tragic situations are those in which children directly witness violence and where they can also be involved indirectly.^88-93

According to the survey carried out by CISMAI, Terre des Hommes for the Authority for Childhood and Adolescence (AGIA)⁹⁴ in 2015, witnessed violence is the second most widespread form of maltreatment in Italy. It was estimated that between 2009-2014 as many as 427,000 children witnessed domestic violence and that only 7% of women who have suffered abuse have reported it. This survey shows that the majority of the minors in the care of Social Services for maltreatment are in Northern Italy and are predominantly female. Children are usually taken into care between the ages of 11 and 17, which reflects the complexity of detecting situations of fragility in the early years of a child's life. The data collected indicate that usually a minor is taken into charge following only one case of obvious maltreatment; in Italy the most common form of maltreatment is neglect.

In a population survey in 2007 on women who had suffered physical or sexual violence carried out by Italy's National Institute of Statistics, 62.4% stated that their children had witnessed abuse.⁹⁵A more recent ISTAT survey (2015) shows a strong increase in IPV and WDV: the percentage of violent episodes to which children were exposed rose to 65.2% compared to 60.3% in 2007. ⁹⁶

3.7.2. Difficulties in Defining and Measuring Children’s Exposure to IPV

The findings outlined above highlight a serious public health issue; the scale of the problem may be significantly underestimated given the emphasis on quantifying only those children who directly witness abuse.

This measure is problematic for a number of reasons: first, children may also be adversely affected by the effects the violence has on parental mental health, parenting, and on the stability of housing. This broader concept of exposure is not reflected in the way abuse is defined in prevalence studies.

Secondly, the term “witnessing” appears to intend only those acts of abuse that are “observable”.⁹⁷ Research and service delivery tend to treat exposure to domestic violence and abuse (DVA) as a “homogenous unitary”. This is not helped by a paucity of multi-dimensional measures of children’s exposure. Finally, the focus on witnessing abuse conflates the type of abuse to which children are exposed with the means through which they become aware of its presence.⁹⁸

Whilst the focus on witnessing, as opposed to experiencing, domestic violence is by no means a problem unique to Italian prevalence studies, it is nevertheless a hindrance to understanding the scale of the problem, and therefore in planning policies to address children’s needs.⁹⁹

3.7.3. What Is Needed to Improve Estimates of Children’s Exposure to IPV

Several measures could be adopted to improve estimates of the dimensions and nature of WDV:

1)

Recognizing children as direct victims of IPV may change the focus of policy makers and increase the pull for robust data on WDV.

2)

In line with the requirements of the Council of Europe Convention on preventing and combating violence against women and domestic violence, regular population-based surveys are needed to assess the prevalence of WDV.

3)

Multi-dimensional measures of exposure should be developed.

4)

There should be efforts to collect information about the contexts in which children are exposed to IPV. Most of the current understanding relates to children who have experienced IPV between cis-gender parents, with less focus in the context of gender-diverse caregiver relationships.

5) 

Efforts should be made to harmonize measurements both within and across countries.

4. Discussion:

What can be done to prevent domestic violence? According to the Italian National Institute of Statistics, preventing violence means addressing its cultural roots and causes. Policy strategies should aim at education, awareness raising and the realization of equal opportunities in every sphere of public and private life.

Adequate risk assessment is fundamental to violence prevention. Risk assessment is a decision-making process that determines the best course of action by estimating, identifying, qualifying, or quantifying risk, with the goal of reducing harm to victims of intimate partner violence and their children.¹⁰⁰ There are three main approaches to risk assessment:

1. Unstructured clinical decision: an informal approach used by professionals (such as police, social workers and health care providers) that relies on the subjective judgment and experience of the professional.

2. Actuarial approach: a method that predicts specific violent behaviours using evidence-based risk factors, providing an accurate estimate of the likelihood of a repeat attack.

3. Structured professional judgment: the practitioner follows specific risk assessment guidelines that reflect current theoretical, professional, and empirical knowledge about violence, considering a minimum set of risk factors and recommendations for information gathering.

Several tools exist to assess the risk of partner violence, including the Canadian SARA (Spousal Assault Risk Assessment Guide) checklist, and actuarial scales such as ODARA (Ontario Domestic Assault Risk Assessment), DVI (Domestic Violence Inventory) and IRAD (Idaho Risk Assessment of Dangerousness Tool).¹⁰¹

Prevention actions promoted by the Italian Department of Equal Opportunity include:

- Increasing public awareness of the roots, causes and consequences of male violence against women.

- Training public and private sector practitioners on the phenomenology, interception, emergence, intake, assessment and management of cases of violence, including those involving migrant, refugee and asylum-seeking women.

- Strengthen preventive efforts against recidivism through re-education pathways for males.

- Raise awareness in the private sector and the media about the influence of communication and advertising regarding gender stereotypes and sexism, and their impact on male violence against women.

5. Conclusions

Intimate partner violence, as presented according to national and international prevalence data, risk factors and the burden of secondary effects on physical and mental health, assumes an impact and a relevance that needs to be taken into account at multiple levels:

• Health care
  Gynaecology and psychiatry departments and outpatient clinics must be called upon to carry out prevention, interfacing as early as possible with women living in situations of risk who reach out for assistance. If violence has already been committed, these contexts must be prepared to take the victims in and provide shelter, and clinical and legal support, returning the possibility of a wide-ranging path of care, including both clinical and legal aspects, also through collaboration with anti-violence centres, that receive a demand for listening and psychological care from battered women. For broad-based prevention work, it may be useful to use validated screening tools. WHO has developed some guidelines and clinical guidelines on IPV for health workers;¹⁰²
• Political-institutional
  Work must be done at the legislative level to ensure that the appropriate penalties for domestic violence are in place and applied; actions must be taken at institutional to ensure that the centres providing shelter, treatment and support for victims of domestic violence are adequately funded;
• Epidemiological research
  National statistics institutions and academic institutes should provide continuously updated data on the phenomenon to guide the thinking of all stakeholders. With regard to raising awareness on the issue, the Italian Society of Psychiatry's "Women's Mental Health" coordination in the document "Violence Against Women and Mental Health" recommends support programs to improve the training of psychiatrists to recognize and treat victims of violence.¹⁰³

In conclusion, it can be said that the phenomenon of intimate partner violence, domestic violence and child witnessing violence is still widespread today and, often, still poorly recognized and intercepted. It is, therefore, essential for institutions and health professionals to act preventively and promptly, with networking based on an integrated approach aimed at fostering person-focused care and based on treatment guidelines for IPV.^104,105 This article integrates and expands on the finding of our previous search, published in Italian on the Rivista di Psichiatria in 2019¹⁰⁶.