Psychosocial profile of male perpetrators of domestic violence: A population-based study.

BACKGROUND: Domestic violence can result from societal, community, and relationship factors pertaining to both the perpetrator and the victims. The male perpetrators of domestic violence have rarely been the subjects of a study. AIM: To understand the factors influencing a man's risk in perpetrating domestic violence in a community setting.
MATERIALS AND METHODS: This observational, exploratory study was conducted on 50 male perpetrators of domestic violence in an urban slum. A semi-structured questionnaire which incorporated Adverse Childhood Experience International Questionnaire, Alcohol Use Disorders Identification Test, Hamilton Rating Scale for Anxiety, and Becks Inventory for Depression was used for assessment.
RESULTS: All the respondents reported both verbal and physical violence with 26% reporting complicated physical violence. Alcohol dependence was diagnosed in 77% of respondents. Twenty-four percent had depression and 34% had anxiety. The mean number of ACEs suffered by the respondents was 4 and the range was 0-8. Substance abuse in the family (86%) and witnessing verbal/psychological abuse (82%) were the most common ACE found. Complicated domestic violence was statistically significantly associated with the presence of alcohol dependence, witnessing complicated physical abuse between the parents and parents separated due to death or divorce. Certain adversities, namely verbal punishment, physical punishment, bullying, community violence, caretaker of sibling, and household work, were strongly associated with abusers who suffered from psychiatric comorbidities.
CONCLUSION: The presence and severity of alcohol dependence and the number of childhood adversities are strongly associated with complicated type of domestic violence. Those who were victims of a large number of ACEs also suffered from anxiety, depression, and alcohol dependence. Copyright:

Domestic violence is defined as any incident or pattern of incidents of controlling, coercive or threatening behavior, and violence or abuse between those aged 16 or over who are or have been intimate partners or family members regardless of gender or sexuality. Controlling behavior can be defined as range of acts designed to make a person subordinate and/or dependent by isolating them from sources of support, exploiting their resources and capacities for personal gain, depriving them of the means needed for independence, resistance and escape, and regulating their everyday behavior. Coercive behavior is an act or a pattern of acts of assault, threats, humiliation, and intimidation or other abuse that is used to harm, punish, or frighten their victim, and includes so-called “honor-”based violence, female genital mutilation, and forced marriage.[1] The term intimate partner violence (IPV) refers to domestic violence in context of intimate relations only. It excludes child or elderly abuse or abuse of any other household member. This encompasses, but is not limited to physical, sexual, and emotional abuse. Examples of physical violence are slapping, hitting, kicking, and beating. Sexual abuse includes forced sexual intercourse and all other forms of sexual coercion. Behaviors such as belittling, insulting, constant humiliation, intimidation (e.g. destroying things), threats of harm, and threats to take away children are some common forms of emotional abuse.[2] IPV occurs in all settings and among all socioeconomic, religious, and cultural groups. A WHO study involving 24,000 women in 10 countries reported that 13%–61% women experienced physical violence by a partner, 6%–59% faced sexual violence, and 20%–75% suffered at least one emotionally abusive act from a partner in their lifetime.[2] The ecological model proposes that such behaviors are a result of factors operating at different levels, namely the societal and community factors such as low social and economic status of women, week legal and community sanctions against domestic violence, gender inequitable social norms along with relationship factors such as male dominance, conflict, and dissatisfaction in the marriage, economic stress, and individual factors pertaining to both the perpetrator and the victims.[2] Adult IPV is linked with adverse childhood adversity, especially exposure to violence, including physical abuse and witnessing IPV and alcohol intake of the perpetrator.[3]

Exposure to adverse conditions in childhood is widespread globally with many calling it “hidden health crisis” with “far reaching consequences.”[4] In 17,000 subjects, 64% had encountered at least one adverse childhood experience (ACE), 40% suffered from two or more ACEs, 12.5% had a ACE score of at least four, 30.1% reported physical abuse, 23.5% were exposed to family alcohol abuse, 19.9% reported sexual abuse, 12.5% witnessed their mothers being battered, and 11% suffered emotional abuse.[5] An estimated 54% of men and 40% of women are victims of violence committed by a parent or caregiver before 18 years of age.[5] In a community-based study, 14%–18% of adults reported witnessing and exposure to interparental violence in childhood.[6] Heavy drinking is also linked with an increased risk of IPV.[2] Couples with at least one partner with problem drinking have an elevated risk of IPV.[78]

Violence in any form has serious health implications. IPV affects physical and mental health through direct pathways, such as injury, and indirect pathways, such as chronic health problems that arise from prolonged stress. Physical consequences include abdominal/thoracic injuries, bruises and welts, chronic pain syndromes, disability, fibromyalgia, fractures, gastrointestinal disorders, irritable bowel syndrome, lacerations and abrasions, and ocular damage.[9] The psychological and behavioral squeals of IPV are alcohol and drug abuse, depression and anxiety, eating and sleep disorders, feelings of shame and guilt, phobias and panic disorder, physical inactivity, poor self-esteem, posttraumatic stress disorder, psychosomatic disorders, smoking, suicidal behavior, and self-harm and unsafe sexual behaviors.[10]

Perhaps because women experience injury and other evident problems from Male-to-female interpersonal violence (MFPV), women have been more readily identifiable and easier to study. There are scarce data on the male perpetrators of domestic violence.[11] It is critical to understand the possible factors influencing a man's risk in engaging in an abusive behavior with the partner in an intimate relationship. The national crime records bureau highlighted that every day in India, an average of 285 women were subjected to cruelty by their husband or relatives of husband (Sec 498A IPC) during 2018.[12] The National Family Health Survey 4 done in 2015–2016 observed that 33% of ever-married women had experienced spousal physical, sexual, or emotional violence in their lifetime, and 26% had experienced at least one of these forms of violence in the past 12.[13] The prevalence of physical, emotional, and sexual violence faced by ever-married women in their lifetime was 30%, 14%, and 7%, respectively. The prevalence of physical, emotional, and sexual violence against ever-married in the past 1 year was 23%, 11%, and 5%, respectively.[13]

An extensive population survey reported that partner violence is very common in India, approximately a third of men observed physical violence between parents, while they were growing up, with the man's father being the sole perpetrator in the majority of the violent families (80%).[14] Very few studies have been done to understand the etiology and thus prevent the domestic violence. The rationale of the study was to generate the psychosocial profile of the men who have been reported by their spouses as perpetrators of violence in the marital relationship. It is important that we attempt to study the genesis of their maladaptive behaviors, much of which may originate in their own exposure to violence in childhood. The present study was undertaken in this context to understand the factors influencing a man's risk in perpetrating domestic violence in a community setting.

MATERIALS AND METHODS

This study was conducted in an unauthorized slum in the industrial township of Chinchwad, Pune. This community has existed along the railway lines since 1978 and has around 500 houses with a population of about 5000. This population-based observational, cross-sectional study was conducted from July 2015 to September 2017. Institutional ethical committee approval was obtained before the commencement of the study. The researcher was provided with access into this community by a social worker provided by a nongovernmental organization working in that community. After an initial screening and applying the inclusion and exclusion criteria, a sample size of 50 was selected. The screening and recruitment of the sample, male spouses who abuse their wives, was done by purposive sampling.

The inclusion criteria

Male spouses who abuse their wives and consenting for the study.

Men above the age of 18 years.

The exclusion criteria

Nonconsenting men

Current involvement in police and legal matters because domestic violence and current substance abuse-related states (such as intoxication) that would make it difficult to conduct assessment.

Tools

For conducting the interview, a semi-structured questionnaire using socioculturally sensitive probes was designed to elicit detailed sociodemographic information and to assess the presence or absence of ACEs while incorporating ACE-International Questionnaire (IQ).[15] Ability to fulfill basic needs of food, shelter, and other daily necessities was assessed by direct questioning. We asked the respondents to report their own perception of adequacy of fulfilling their basic needs of food, shelter, and other daily necessities. Problem drinking, anxiety, and depression were measured by administering the Alcohol Use Disorders Identification Test (AUDIT),[1617] Hamilton Rating Scale for Anxiety (HAM-A),[18] and Becks Inventory for Depression.[19]

Method

The community-level social worker along with the researcher visited homes in the community in a consecutive fashion to screen for families suffering from domestic violence. After taking informed consent and assuring them of full confidentiality, the women were asked (in the absence of their husbands) if they were having or have ever suffered from any kind of domestic violence. Of the women whose spouses fulfilled the inclusion criteria, the researcher enquired for a suitable time their partners would be available. The researcher then revisited the house and introduced himself as a resident in psychiatry conducting a study about past and present adverse experiences in males of this community. The short time period available for this dissertation allowed an initial screening of only 150 houses, which identified 62 males fulfilling the inclusion criteria; 12 of them either refused to give consent or were unavailable. Consequently, a sample size of fifty domestic abusers was collected. Initially, the purpose of the study was explained to the participants and written informed consent was obtained. Thereafter, they were interviewed individually for the purpose data collection. The average length of each interview was about 45 min. Screening tools were administered after the interview was completed.

Statistical analysis

First, the distribution proportion (in percentages) of the various relevant variables in the study population was tabulated. This was followed by computation of the quantity of ACEs. It was then relevant to examine if the independent variables, namely, the sociodemographic characteristics of the participants and their exposure to childhood adversities has any correlation with each other and if they have any association with the dependent variable, i.e. domestic violence. The outcome variable/dependent variable of domestic abusers to be grouped separately based on two different criteria each when analyzed for association with the exposure variables would provide different clinically relevant information. These two outcome variables were named as “outcome variable 1 and 2.”

In outcome variable 1, the domestic abusers were categorized into four dimensions as under:

Only domestic abusers

Alcohol-dependent abusers

Abusers with depression or anxiety but no alcohol dependence

Alcohol-dependent abusers with depression or anxiety.

The relevance of dividing the domestic abusers into these four dimensions for association analysis is that it could help us understand if the exposure variables are directly leading to domestic violence or are associated with psychiatric conditions such as anxiety, depression, and alcohol dependence which could in turn increase the violence.

In outcome variable 2, the domestic abusers were categorized into two dimensions based on the WHO definition of simple and complicated violence as follows:

Simple domestic violence which includes slap, kick, punch, or beating

Complicated domestic violence which includes beating with an object such as stick, stone, bottle, and knife.

Both of these categories were analyzed with the selected independent variables to understand and interpret the statistically significant associations between them if any.

All statistical analyses were done in SPSS version 21.0 for windows (SPSS Inc., Chicago, IL, USA). Statistical significance was set at P ≤ 0.05. Either Chi-square test (with Yates correction) or if the conditions were not met Fisher's exact test was used for the analysis.

RESULTS

Sociodemographic profile of the study subjects is given in Table 1. In our study group, 44% of the respondents were daily wage workers, 20% had their own business, 18% had a job which provided a monthly salary, and 18% were unemployed. The mean age at first employment was 14.30 years (range 8–30 years) with 12% first employed below the age of 10 years (minimum being 8 years), 70% were employed below the age of 17 years, and 9% worked for the first time after the age of 17 years with the maximum being 30 years. The literacy rates of our sample in 84%. Majority (78%) of the respondents live in their own house, while 22% live in rented houses. Only 7% had pukka houses, while 93% lived in kaccha houses. However, 92% had TV, 86% had gas stove, less than half (42%) had a fridge, 82% owned a mobile phone, whereas 14% owned a vehicle. Only 24% lived in this locality since childhood whereas 76% spent their childhood somewhere else.

Table 1

Sociodemographic characteristics of the male domestic violence perpetrators (n=50)

Variables	Total (%)
Age (years)
18-29	8 (16)
30-44	22 (44)
45-59	16 (32)
>60	4 (8)
Education
Illiterate	8 (16)
Primary (1-4)	6 (12)
Up to SSC (5-10)	30 (60)
Up to HSC (11-12)	5 (10)
More the HSC	1 (2)
Work pattern
Regular	22 (44)
Irregular	19 (38)
Socioeconomic status
Lower	5 (10)
Lower middle class	30 (60)
Middle class	15 (30)
Type of family
Nuclear	40 (80)
Joint	10 (20)

SSC – Secondary School Certificate; HSC – Higher Secondary Certificate

Objective and accurate measures of the socioeconomic status are hard to establish. For this study, it was relevant to document the respondent's perception of their economic status that would add up to their sense of vulnerability or strength. We asked the respondents to report their own perception of adequacy of fulfilling their basic needs of food, shelter, and other daily necessities. Ten percent of the respondents belong to the lower section of the society as they reported less than adequate means, 30% were from the lower middle class with just about sufficient conditions, while 60% reported adequate more than adequate resources for livelihood. The mean age of first marriage among the sample group is 21.18 with the earliest age being 10 years and the oldest being 30. The mean duration of marriage was 21.38 years, ranging from 1 to 43 years. While 68% reported that their marriage was arranged by their family with their consent, 28% mentioned that their family members got them married without their consent; 4% had married without the involvement of their family members.

Common substance of abuse reported was tobacco by 48 and alcohol by 47 subjects. Based on the diagnostic criteria of International Statistical Classification of Diseases-Diagnostic Criteria for Research,[20] 23% were found to fulfill the criteria of harmful use and 70% fulfilled the criteria of alcohol dependence syndrome. Among the responders with alcohol dependence, 86% had a continuous pattern of alcohol consumption and 14% had binge pattern of use. The mean duration of alcohol consumption is 17.2 years, with 40% consuming alcohol for a period of 10 years or less, 26% between 11 and 20 years, 19% between 21 and 30 years, and 15% more than 30 years. Six percent reported hospitalizations due to withdrawal state without delirium in the past and 8% for delirium induced by alcohol withdrawal. Psychiatric comorbidities among those who had alcohol-related problems (n = 47) included anxiety disorder (34%) and depressive disorder (25%), while 13% harbored suspicions of infidelity. Scores of the respondents on AUDIT, HAM A, and Beck Depression Inventory are shown in Table 2.

Table 2

Scores of the male domestic violence perpetrators (n=50) on the alcohol use disorders identification test Hamilton anxiety scale and becks depression inventory

Scale	Score	n (%)
Alcohol related problems based on AUDIT	No alcohol consumption	3 (6)
	low level of alcohol related problems: - 1-7	11 (22)
	Medium level of alcohol related problems: - 8-15	12 (24)
	High level of alcohol related problems: - >16	24 (48)
Hamilton anxiety scale	Normal range	31 (66)
	Mild	11 (23)
	Moderate	5 (11)
	Severe	0 (0)
Becks depression inventory	Normal range	38 (76)
	Borderline clinical depression	6 (12)
	Moderate depression	5 (11)
	Severe depression	1 (3)

AUDIT – Alcohol Use Disorders Identification Test

All the 50 male abusers perpetrated verbal as well as physical violence; 26% perpetrated complicated physical violence, while 74% simple violence only. The frequency of verbal/psychological abuse was almost every day in 17%, at least 2–3 times a week in 49%, sometimes (3–4 times a month) in 16%, and 4% rarely (once in 6 months). Frequency of simple physical abuse was almost every day by 13%, 2–3 times a week by 41%, 3–4 times a month by 27%, and once in 6 month by 19% of the participants. Frequency of complicated physical abuse was reported as 2–3 times a week by 54%, 3–4 times a month by 23%, and once in 6 months by 23% of the perpetrators. Rank order of childhood adversities is given in Table 3. Quantum of adversities of childhood experiences is shown in Table 4.

Table 3

Rank order of adversities according to adverse childhood experience international questionnaire the male domestic violence perpetrators (n=50)

Adversities (ACE-IQ)	Percentage
Substance use in family members	86
Witnessing verbal/psychological abuse	82
Verbal/psychological punishment	60
Witnessing simple physical violence	44
Simple physical punishment	32
Community violence	28
Death of caregiver	20
Witnessing complicated physical violence	20
Could not express feelings to anyone	20
Did not have understanding parents	18
Did not have secure attachment figure	14
Bullying by friends	12
Complicated physical abuse	10
Bullying by teachers	8
Collective violence	6
Parents separated	4
Mental illness in family other than substance use	4
Inappropriate sexual touch	4
Incarceration of family members	2
Additional childhood adversities
Child labor	54
Household work	34
Home alone for long hours	18
Basic needs were not met	18
Caretaker of siblings	14
Lack of free play	4

ACE-IQ – Adverse childhood experience international questionnaire

Table 4

Adverse childhood experience (WHO IQ) score of the male domestic violence perpetrators (n=50)

ACE score	Count (%)
0	3 (6)
1	4 (8)
2	6 (12)
3	7 (14)
4	13 (26)
5	9 (18)
6	4 (8)
7	2 (4)
8	2 (4)
Additional ACE score	Counts (%)
0	10 (20)
1	16 (32)
2	17 (34)
3	7 (14)

ACE – Adverse childhood experience; IQ – international questionnaire

Quantum of adversities of childhood experiences

The total ACE mean score of the study group is 3.76, median is 4, and range is 0–8. At least 1 ACE was reported by 94% of the study group, 74% reported three or more adversities, while 34% reported at least five adversities. For the additionally documented ACEs, the maximum score was 3. The mean score was 1.42 with the median being 1. Eighty percent of participants suffered from one or more of these adversities [Table 4]. Among the respondents who were not working regularly, 77.8% were alcohol dependent as opposed to 57% among the respondents who worked regularly. Moreover, 30.6% of the irregular at work respondents were the domestic violence perpetrators who had alcohol dependence and other psychiatric comorbidities. This explains the statistically significant association between occupation status and outcome variable 1 (P = 0.03) [Table 5]. There was no statistical significant difference found between the occupation status of the respondents and the type of domestic violence as categorized by outcome variable 2.

Table 5

Association of sociodemographic characteristics and marital status of the male domestic violence perpetrators (n=50) with outcome variable 1

Variables	Row (%)				Statistical test
	Only abusers	Abusers with alcohol dependence	Abusers with other psychiatric comorbidities	Abusers with any psychiatric comorbidities
Sociodemographic variables
Age
≤42	4 (15.4)	12 (46.2)	4 (15.4)	6 (23.1)	Fisher’s exact, P=0.9
>42	3 (12.5)	13 (54.2)	3 (12.5)	5 (20.8)
Education (years)
≤6	4 (16)	11 (44)	4 (16)	6 (24)	Fisher’s exact, P=0.86
>6	3 (12)	14 (56)	3 (12)	5 (20)
Occupation
Irregular to work	3 (8.3)	17 (47.2)	5 (13.9)	11 (30.6)	Fisher’s exact, P=0.03
Regular to work	4 (28.6)	8 (57.1)	2 (14.3)	0
Marital status variables
Age at marriage (years)
</21	2 (7.4)	15 (55.6)	3 (11.1)	7 (25.9)	Fisher’s exact, P=0.43
>21	5 (21.7)	10 (43.5)	4 (17.4)	4 (17.4)
Duration of marriage (years)
</21	3 (12.5)	10 (41.7)	5 (20.8)	6 (25)	Fisher’s exact, P=0.54
>21	4 (15.4)	15 (57.7)	2 (7.7)	5 (19.2)
Type of marriage
Marriage with agreement	5 (14.7)	20 (58.8)	4 (11.8)	5 (14.7)	Fisher’s exact, P=0.19
Marriage without agreement	2 (12.5)	5 (31.3)	3 (18.8)	6 (37.5)

A large number (88.2%) of the participants who were married with their agreement perpetrated only simple domestic violence, while 56.3% of those who were married without their agreement committed complicated violence. This led to a significant association between the type of marriage and outcome variable 2 (P = 0.02) [Table 6]. The duration of marriage and the age at which the respondents got married did not have any statistically significant association with both the outcome variable categories.

Table 6

Association of sociodemographic and marital status of the male domestic violence perpetrators (n=50) with outcome variable 2

Variables	Row (%)		Statistical test
	Simple domestic violence	Complicated domestic violence
Sociodemographic variables
Age
≤42	19 (73.1)	7 (26.9)	Fisher’s exact, P=0.9
>42	18 (75)	6 (25)
Education (years)
≤6	19 (76)	6 (24)	Fisher’s exact, P=0.9
>6	18 (72)	7 (28)
Occupation
Irregular to work	25 (69.4)	11 (30.6)	Fisher’s exact, P=0.30
Regular to work	12 (85.7)	2 (14.3)
Variables related to marital status
Age at marriage (years)
≤21	18 (66.7)	9 (33.3)	Chi-square, P=0.33
>21	19 (82.6)	4 (17.4)
Duration of marriage (years)
≤21	17 (70.8)	7 (29.2)	Chi-square, P=0.75
>21	20 (76.9)	6 (23.1)
Type of marriage
With agreement	30 (88.2)	4 (11.8)	Chi-square, P=0.02
Without agreement	7 (43.8)	9 (56.3)
Variables related to alcohol use
Alcohol dependence
No	14 (37.8)	0	Chi square, P=0.01
Yes	23 (62.2)	13 (100)
Audit score
0	3 (8)	0	Fisher’s exact, P=0.04
1-7	11 (29)	0
8-15	9 (24)	3 (23)
≥16	14 (37)	10 (76)
Suspicion of infidelity (paranoia)
No	34 (91)	5 (38)	Chi-square, P=0.001
Yes	3 (9)	8 (62)
Course of alcohol consumption
Continuous	33	10	Fisher’s exact, P=0.25
Episodic	4	3
Duration of alcohol consumption (years)
1-10	16	3	Fisher’s exact, P=0.172
11-20	8	4
21-30	4	5
>30	6	1

All the perpetrators of complicated domestic violence were alcohol dependent. The association between outcome variable 2 and the presence of alcohol dependence was found to be highly significant (P = 0.01).

These results were echoed in the findings of the association between AUDIT score and outcome variable 2, which was also statistically significant (P = 0.04). An AUDIT score of ≥16 was obtained by 77% of the complicated domestic violence perpetrators compared to only 38% of simple domestic violence perpetrators.

The results of analysis of adverse childhood experience-related independent variables [Tables 7 and 8]

Perpetrators whose parents were separated during childhood had a statistically significant association with outcome variable 1 (P = 0.08). It is clinically relevant to find that three-fourth (75%) of those who suffered from this childhood adversity were abusers suffering from depression or anxiety.

The reporting of any kind of verbal or psychological punishment during childhood had a statistically significant association. It is relevant to find that only 5% who reported no verbal/psychological abuse suffered from anxiety or depression, but 56% of who reported verbal/psychological abuse suffered from those psychiatric comorbidities.

Experiencing physical punishment in childhood was also found to be statistically significant association (P = 0.02); this shows that a higher proportion of respondents who were physically punished in their childhood were the abusers with alcohol dependence and other psychiatric comorbidities. Respondents who said yes to bullying had a statistically significant higher proportion (71%) of psychiatric comorbidities than the ones who did not (26%) (P = 0.01).

All the respondents who had seen community violence in the childhood turned out to be dependent to alcohol as appose 60% of those who did not see community violence, perhaps due to this reason a P = 0.05 for the variable community violence can be taken as significant. Respondents who had to take care of their siblings in their childhood had a significant association (P = 0.001), but the subgroup was too small to draw any conclusions. However if looked together with the variable of having to do house hold work (which also has a P value of than 0.001) it is evident that those perpetrators who had to bear the burden of household responsibilities including taking care of their siblings had a much higher proportion of anxiety or depression in them.

All of the 18 adversity variables were again run with the outcome variable 2 to see if any of them have an association with the type of domestic violence perpetrated. Five variables showed a statistically significant association. The proportion of complicated domestic violence among respondents who had their parents separated in their childhood is 59% compared to 19% of those who did not suffer from that adversity. This explains the highly significant association between outcome variable 2 and separated parents.

Witnessing verbal/psychological abuse in childhood was further categorized into dimensions based on the frequency of abuse witnessed. It was found that the respondents who witnessed verbal abuse regularly (every day basis) were perpetrating complicated violence with a higher proportion then those who witnessed verbal abuse rarely (less than once a month) (P = 0.001). Similarly, the respondents who witnessed complicated physical violence in their parents were more likely to execute complicated physical domestic violence (P = 0.01). The findings among those who suffered from physical punishment in their childhood was similar to those witnessed physical violence among their parents. The significant P = 0.015 shows that those who were punished with sticks, belt, etc., were more likely to commit such complicated violence.

Basic needs not fulfilled in childhood of the respondents were also shown to have a significant association with the type of domestic abuse (P = 0.006). A large number of respondents whose basic needs were not full filled perpetrated complicated domestic violence (66%).

The results of dose-response analysis of adverse childhood experience [Tables 7 and 8]

We wanted to see if the dose for both ACE-IQ and additional ACE had any associations with our outcome variables. For establishing a gradient of dose, ACE-IQ was divided into three ordinal categories, namely 0–2, 2–5, and 6–8. Since the range of additional ACE is only 3, it could only be divided into two groups, i.e. 0–1 and 2–3. The association between type of domestic violence (outcome variable 2) and dose of both ACE-IQ and additional ACE have not shown to be statistically significant [Tables 7 and 8]. Graphs was constructed to analysis the association between the dose of adverse childhood. experiences (WHO International Questionnaire) and additional adverse childhood experiences the outcome variable 1. Figures 1 and 2 clearly show that as the dose of adverse childhood experience International Questionnaire increases from low to medium to high and in the case of additional adverse childhood experience from low to high, the percentage of respondents who had no alcohol dependence and no anxiety/depression decreases and the respondents who had anxiety/depression along with alcohol dependence increases steadily. Figure 3 shows a steady decrease in simple domestic violence as the dose of adverse childhood experience increases and a steady increase in the percentage of perpetrators of complicated domestic violence as the dose increased from low to medium to high. This trend was not found for the additional adverse childhood experiences[Figure 4], whereas the dose increased both the proportion of simple and complicated domestic violence went down. However, it seems relevant to note that the associated Figures 3 and 4 show a steady decrease in simple domestic violence as the dose of ACE increases and a steady increase in the percentage of perpetrators of complicated domestic violence as the dose increased from low to medium to high.

Table 7

Variables related to adverse childhood experience by outcome variable 1

Childhood adversities	Row (%)				Statistical test
	Only abusers	Abusers who are alcohol dependent	Abusers with other psychiatric comorbidities	Abusers with psychiatric comorbidities
Substance abuse in family
No	2 (26)	5 (71)	0	0	Fisher’s exact, P=0.20
Yes	5 (12)	20 (46)	7 (16)	11 (26)
Other mental illness in family
No	7 (14)	23 (48)	7 (14)	11 (23)	Could not commute
Yes	0	2 (100)	0	0
Incarceration of family members
No	7 (14)	24 (49)	7 (14)	11 (24)	Could not commute
Yes	0	1 (100)	0	0
Separation of parents
No	6 (16)	23 (60)	3 (8)	6 (16)	Fisher’s exact, P=0.08
Yes	1 (8)	2 (17)	4 (33)	5 (42)
Could not express emotionally
No	7 (17)	19 (48)	6 (15)	8 (20)	Fisher’s exact, P=0.63
Yes	0	6 (60)	1 (10)	3 (30)
Witnessing verbal/psychological abuse
No	2 (22)	4 (44)	1 (11)	2 (22)	Fisher’s exact, P=0.94
Yes	5 (12)	21 (51)	6 (15)	9 (22)
Witnessing total physical abuse
No	3 (16)	10 (56)	3 (17)	2 (11)	Fisher’s exact, P=0.58
Yes	4 (12)	15 (47)	4 (12)	9 (28)
Verbal/psychological punishment
No	3 (16)	16 (80)	0	1 (5)	Fisher’s exact, P=0.001
Yes	4 (13)	9 (30)	7 (23)	10 (33)
Physical punishment
No	4 (14)	19 (65)	1 (3)	5 (17)	Fisher’s exact, P=0.02
Yes	3 (13)	6 (29)	6 (29)	6 (29)
Sexual abuse
No	7 (15)	25 (52)	7 (15)	9 (19)	Could not commute
Yes	0	0	0	2 (100)
Bullying
No	7 (17)	23 (55)	3 (7)	9 (21)	Fisher’s exact, P=0.01
Yes	0	2 (28)	3 (43)	2 (28)
Community violence
No	7 (20)	15 (41)	7 (19)	7 (19)	Fisher’s exact, P=0.05
Yes	0	10 (72)	0	4 (28)
Collective violence
No	7 (15)	24 (51)	7 (15)	9 (19)	Fisher’s exact, P=0.3
Yes	0	1 (33)	0	2 (66)
Long hours alone
No	7 (17)	20 (49)	7 (17)	7 (17)	Fisher’s exact, P=0.16
Yes	0	5 (55)	0	4 (45)
Care taker of siblings
No	7 (16)	25 (58)	3 (7)	8 (19)	Fisher’s exact, P=0.001
Yes	0	0	4 (57)	3 (43)
Household work
No	7 (21)	2 (63)	1 (30)	4 (12)	Fisher’s exact, P=0.001
Yes	0	4 (23)	6 (35)	7 (42)
Lack of play
No	7 (15)	24 (50)	7 (15)	10 (20)	Fisher’s exact, P=0.99
Yes	0	1 (50)	0	1 (50)
Child labour
No	3 (13)	13 (57)	1 (4)	6 (26)	Fisher’s exact, P=0.33
Yes	4 (15)	12 (45)	6 (22)	5 (18)
Basic needs not met
No	7 (17)	19 (46)	7 (17)	8 (20)	Fisher’s exact, P=0.30
Yes	0	6 (66)	0	3 (33)
ACE score
0-2	3 (23)	9 (68)	0	1 (8)	Fisher’s exact, P=0.01
3-6	4 (14)	15 (52)	5 (17)	5 (17)
5-8	0	1 (13)	2 (25)	5 (62)
Additional adverse child- hood experience score
0-1	7 (28)	17 (66)	1 (3)	1 (3)	Fisher’s exact, P=0.001
2-3	0	8 (33)	6 (25)	10 (42)

ACE – Adverse childhood experience

Table 8

Variables related to adverse childhood experience by outcome variable 2

Variables related to adverse childhood experience	Row (%)		Statistical test
	Simple	Complicated
Substance abuse in family
No	5 (71)	2 (29)	Fisher’s exact, P=0.9
Yes	32 (74)	11 (26)
Other mental illness in family
No	36 (75)	12 (25)	Fisher’s exact, P=0.45
Yes	1 (50)	1 (50)
Incarceration of family members
No	36 (73)	13 (27)	Fisher’s exact, P=0.43
Yes	1 (100)	0
Separation of parents
No	32 (84)	6 (16)	Chi-square, P=0.007
Yes	5 (42)	7 (59)
Could not express emotionally
No	30 (75)	10 (25)	Fisher’s exact, P=0.70
Yes	7 (70)	3 (30)
Witnessing verbal/psychological abuse
No	6 (67)	3 (33)	Fisher’s exact, P=0.001
Rarely	7 (100)	0
Sometimes	14 (100)	0
Often	5 (84)	1 (14)
Regularly	5 (36)	9 (64)
Witnessing physical abuse
No	14 (78)	4 (22)	Fisher’s exact, P=0.01
Simple	20 (87)	3 (13)
Complicated	3 (33)	6 (66)
Verbal psychological punishment
NO	16 (80)	4 (20)	Fisher’s exact, P=0.86
Rarely	7 (78)	2 (22)
Sometimes	3 (75)	1 (25)
Often	8 (66)	4 (33)
Regularly	3 (60)	2 (40)
Physical punishment
No	22 (76)	7 (24)	Fisher’s exact, P=0.015
Simple	14 (88)	2 (12)
Complicated	1 (20)	4 (80)
Sexual abuse
No	36 (75)	12 (25)	Fisher’s exact, P=0.45
Yes	1 (50)	1 (50)
Bullying
No	29 (69)	13 (31)	Fisher’s exact, P=0.09
Yes	8 (100)	0
Community violence
No	28 (78)	8 (22)	Chi-square, P=0.474
Yes	9 (64)	5 (36)
Collective violence
No	36 (76)	11 (24)	Fisher’s exact, P=0.16
Yes	1 (33)	2 (66)
Long hour alone
No	32 (78)	9 (22)	Chi-square, P=0.21
Yes	5 (56)	4 (44)
Care taker of sibling
No	32 (74)	11 (26)	Fisher’s exact, P=0.99
Yes	5 (72)	2 (28)
House hold work
No	27 (81)	6 (19)	Chi-square, P=0.09
Yes	10 (59)	7 (41)
Lack of any play
No	36 (75)	12 (25)	Fisher’s exact, P=0.99
Yes	1 (50)	1 (50)
Child labor
No	14 (60)	9 (40)	Chi-square, P=0.062
Yes	23 (85)	4 (15)
Basic needs not fulfilled
No	34 (83)	7 (18)	Chi-square, P=0.006
Yes	3 (33)	6 (66)
ACE score
0-2	11 (84)	2 (16)	Fisher’s exact, P=0.24
3-5	22 (76)	7 (24)
6-8	4 (50)	4 (50)
Additional ACE score
0-1	22 (85)	4 (15)	Fisher’s exact, P=0.109
2-3	15 (63)	9 (38)

ACE – Adverse childhood experience

Figure 1

Graphs showing the association between the dose of adverse childhood experiences and outcome variable 1

Figure 2

Graph showing the association between the dose of additional adverse childhood experiences and the outcome variable 1

Figure 3

Graphs showing the association between the dose of adverse childhood experiences (WHO-International Questionnaire) with the outcome variable 2

Figure 4

Graph showing the association between the dose of additional adverse childhood experiences with the outcome variable 2

DISCUSSION

From a scientific perspective, while the negative consequences of domestic violence on the victim are well established, the risk factors which could make a person more likely to be a perpetrator are downplayed and trivialized. In both the developing world and the developed, the perpetrators of domestic violence are considered by many as “bad people” who are just that way. The reality is far from this. Many factors such as alcohol dependence, psychiatric comorbidities, adverse, and traumatic childhood experiences, individually or cumulatively can predispose people to become perpetrators of domestic violence. Toward this purpose, this study has documented the range and depth of ACEs in our study population. These data may help fulfill the compelling need to generate strong evidence. Standard methodology such as the WHO ACE-IQ has been, to our knowledge, rarely used in India. In this exploratory study, we have interviewed the study population using an interactive interview that resulted in ethnographic narratives of their life since childhood. Retrospective design of the study brings in some uncertainty, particularly the reliability of recall of childhood experiences. An earlier study also used a retrospective method to assess if childhood adversity increases the risk of adult IPV perpetration and found that for men with high-level childhood adversity, there was an 8.8% increased risk of perpetrating IPV compared to a 2.3% increased risk among men with low-level adversity. Retrospective ascertainment of adverse experiences appeared to offer no disadvantage.[21]

The sociodemographic data provide contextual backdrop to the study group. The male perpetrators for domestic violence from the study group live in a small unauthorized community along the railway lines in a city; the mean age of the respondents was 42 with more than half of them belonging to the third or fourth decade of their life. The literacy rate of our sample is 84% which is in keeping with the literacy rate of Maharashtra (82.34%) and Pune (86.15%). Even though the literacy rate was 84%, only 12% had an education degree higher than 10th pass. It is probably not so surprising to find that a vast number of men in the study group were daily wage laborers who had to go job hunting on a daily basis with no guarantee that they will get work the next day. Two-thirds of the group belong to lower and middle class with just about or lower means of livelihood by the standards and cost of living in this industrial city. It is important to bear in mind that we have documented this profile of the respondents who are in their late thirties and early forties. The adversity data of these respondents pertain to their childhood before the age of 18 years. Some of these respondents were living in famine struck interiors of the state before they were forced to migrate to the city. Thus, this group's profile when they were young may have been different. Some resources have increased, e.g. literacy and employment, while other liabilities may have increased due to rapid urbanization, e.g. overall cost of living and property/housing expenses having escalated. A more accurate computation of the shift of this profile would be important to determine if the change has been positive or adverse.

While 78% of the study group owned their own house, the majority were not made up entirely of brick and cement. Most of the houses had a steel roof, while some even used kuccha material for their walls. It can be extrapolated form the descriptive statistics that at least two-third of the 50 families belonged to the lower economic strata. Thus, the study group belongs to this context of meager economic resources and the hardships that come with it. It would thus not seem illogical to hypothesize that the families' socioeconomic status could contribute to adverse experiences of childhood. Yet, there is not a single question in its present form to tap into at least some of the economic hardships that translate into ACEs.

Data on adverse childhood experiences

ACE-IQ tool recommends asking 29 questions divided into a total of thirteen categories. All but one category was applicable in our study population. In the category of physical neglect, none of the respondents answered yes to the questions mentioned in the ACE-IQ tool. Discontinuation of schooling even when the child wished to study due to monetary reason or migration was a theme which occurred in our study group. From a child's perspective, it may be an adverse experience and a small addition of discontinuation of schooling by parents for a variety of reasons would not be misplaced.

A large majority reported living with a family member who was alcoholic or used drugs, but none of them reported physical neglect due to intoxication. Alcohol or drug abuse among parents or guardians is one of the factors in emotional maltreatment, neglect, physical abuse, and sexual abuse of a child.[22] It would be too simplistic to say that just the presence or absence of substance abuse in family members constitutes an adversity, probably it would be better to ask questions pertaining to the impact it has on the child to see if he was adversely effected or not. The same can be said for the question on any mental illness in the family which in our case was just 4% of the population.

In every community, a large portion of children and young people are subject to serious verbal, physical, and sexual abuse.[23] In our study, also, 60% of the study group experienced some sort of abuse. In our country, most parents have a sense of righteousness in exercising their control over the child and still believe that physical punishment is the only form of discipline. However, research shows that children who have been spanked or hit by their parents are vulnerable to develop aggressive and even antisocial behaviors.[24]

It is striking that the seemingly harmless physical maltreatment of children, which is often trivialized by parents, teachers, and society at large, causes noteworthy psychological consequences. It is not about the physical wounds, but the emotional impact on the child's psyche during the early developmental years. Also, important for a healthy development of the psyche of the child is a secure and loving environment in the house; needless to say that parents who quarrel regularly and in front of their kids can adversely affect the child. Sixty percent of the study group witnessed their parents beating each other in their childhood; more than half of those parents used to hit their spouses with some sort of objects such as a stick, a belt, slippers, or even a stone. Living in such an environment and growing up with a threat perception perhaps made one in five respondents to not able to express their private emotions to their parents. Fourteen percent said that they had absolutely no elderly figure they could get securely attached to in their childhood.

Bullying by teachers is a pattern of conduct, rooted in a power differential, which threatens, harms, humiliates, induces fear, or causes students substantial emotional distress.[25] Students who are bullied by teachers typically experience confusion, anger, fear, self-doubt, and profound concerns about their academic and social competencies. Not knowing why, he or she has been targeted, or what one must do to end the bullying, may well be among the most personally distressing aspects of being singled out and treated unfairly.[25] Bullying by teachers (8% in our study) is as bad as bullying by other kids. The ACE-IQ tools question on bullying should probably include bullying by teachers and/or kids.

Additional adverse childhood experiences

Parents in the lower socioeconomic strata are often daily wage earners, which is also the case in our study. Often both the parents go to work, leaving young children to be on their own home for long periods of time. If these children are the elder ones in the family, they are expected to take care of their younger siblings and do household work. In our study group, a small percentage of respondents spent their childhood in their native village where they were expected to accompany their parents in the farms and do hard vigorous labor. A debatably worse situation for more than half of the study group was having work outside their house in shops and factories for money to support their parents. A large portion (24%) of the participants in the study group spent their childhood in this community. The resources available to them earlier were lesser then they are now; one respondent mentioned having to try to sleep in a house with a leaking roof on a rainy day while another mentioned having to accommodate four people in one bed. A respondent remembered having to eat uncooked corn for days at a stretch. Being deprived of the very basic need to food, water, and shelter would surely constitute as an adversity. Due to all of these economic hardships, ACE additional to the ones in WHO list that were significant included deprivation of basic needs, spending long hours alone, needing to do household work, needing to take care of their siblings, having to work for money as a child and lack of free play as a child.

Quantum of adverse childhood experience

Unlike a large majority of studies on adversities, we chose to explore, document, and quantify a large majority of ACEs and see their relationship with domestic violence. ACEs are reported to be very common and tend to occur in clusters.[26] Our study findings are similar since 94% report at least one ACE out of which 8% report having only one adversity, whereas 26% report having at least four adversities. “ACE scores” probably capture the cumulative (neuro) developmental consequences of traumatic stress in a child. When a child is wounded, the pain and negative longterm effects reverberate as an echo of the lives of people they grew up with. Then when they grow up they are at risk of taking on the same characteristics and behaviors – thereby sustaining the cycle of abuse, neglect, violence and substance abuse, and mental illness.[26]

Profile on domestic violence

All our fifty respondents were spouses of women who reported an abusive experience in their marital relationship. All of these respondents were not just verbal abusers but also perpetrated physical violence on their wives. When talking about IPV, it is important to distinguish severe from mild forms of partner violence, to study whether the correlates of severe IPV can be distinguished from correlates of mild violence.[27] Various methods have been used to classify physical violence in the past; the most common method used by most researchers is to divide based of potential for injury to the victim. Moderate or simple male violence consisted of at least one of the following acts: pushed, shoved, or grabbed and/or slapped. Severe or complicated male violence consisted of kicked or hit or tried to hit with something; beat up; choked; burned or scalded; threatened with a knife or gun; and/or used a knife or gun. In our study, one in four abusers confessed to perpetrate complicated domestic violence, which is similar to an earlier US study.[27] In our attempt to study the correlates leading to a maladaptive behavior like domestic violence a sample of men who perpetrated violence rarely would perhaps not do. However, an assessment of the frequency of domestic violence found that for around half of these families, physical and verbal abuse by men on their wives occurred at least 1–2 times a week, and for another 30%, it was 2–3 times a month.

Alcohol and domestic violence

The percentage of adults using alcohol in an urban slum in India is far greater than the national prevalence.[28] Changing social norms, urbanization, increased availability, high-intensity mass marketing, and coupled with the poor level of awareness contributed to increased alcohol use among the urban slum dwellers.[28]

Research has established that alcohol consumption plays an important part in IPV.[29] Keeping this in mind our study findings of 94% alcohol consumers within a subset of domestic perpetrators residing in a slum urban area is on expected lines. Twenty-three percent of those fulfilled the criteria harmful use, whereas 77% were diagnosed as dependent on alcohol. Some controversy regarding the role of alcohol in domestic violence has arisen. Various groups are of the opinion that calling an abuser an alcoholic takes the responsibility away from the perpetrator and perhaps gives him an excuse to abuse further. However, alcohol dependence is a neurobiological disorder and the effects it has on behavior cannot be denied. In our study, a statistically significant association was found not just between the presence and absence alcohol dependence and the severity of domestic violence (P = 0.01), but also more the severity of the dependence (as assessed by the AUDIT score) the more were the number of complicated domestic abusers. This difference is also found to be statistically significant (P = 0.04).

These findings are similar to an earlier study which concluded that alcohol use may contribute to the occurrence and/or severity of domestic violence.[30]

Adverse childhood consequences and domestic violence

Intergenerational transmission of domestic violence has been a nature of scientific scrutiny for a few decades now. It has been hypothesized, examined, and proved that children who witness their father beating their mother grow up to be perpetrators of domestic violence on their spouses. One of the largest studies followed 543 children over a period of 20 years to examine the independent effects of parenting and exposure to domestic violence between parents on the risk of violence to and from an adult partner. This prospective cohort study found that the children who witnessed violence between their parents had a 2.34 times increased odd of physically abusing their partners as compared to the ones who did not witness violence between their parents even after controlling for other risk factors. An odds ratio of 1.20 was also found between the association of physical or sexual abuse as a child and perpetrating domestic violence.[31] In agreement with the above, the present study also found a statistically significant association between the severity of domestic violence and the severity of physical abuse witnessed between the parents (P = 0.010). Also, our study found that the more harshly the physical punishment on the child (in terms of simple and complicated punishment), the harsher was the domestic violence on the wife. This association was also statistically significant (P = 0.015).

Another significant association was found with separation of parents (P = 0.007). About 60% of the perpetrators whose parents had separated (due to death or divorce) in their childhood committed complicated abuse as opposed to only 16% among those who did not report parental separation. In agreement with our findings, it has shown that separation due to parental marital disharmony accounts for violent behavior in sons.[32]

Among all the associations between the additional ACEs and severity of domestic violence, our study brought out a significant association between deprivation of basic needs as a child and severity of domestic violence (P = 0.006). One explanation of this could be that economic hardship has been linked to violence within families, being a victim of or witnessing such violence in childhood could have led to an increased severity of domestic violence as an adult.[33] Another more direct explanation could be that sustained economic hardship leads to poorer physical, psychological, and cognitive functioning including being cynically hostile.[34]

A study examining the role of “stress sensitization” on 34,653 adult males found that high-level childhood adversity was associated with an 8.8% increased risk of perpetrating IPV compared to a 2.3% increased risk among men with low-level adversity.[3] Though a statistically significant association between the dose of ACE and severity of domestic violence was not found in our study, an upward trend of percentage of complicated domestic abusers was observed as the dose of adversity increased along with a downward trending percentage of simple domestic abusers [Figure 3].

Domestic abuse, alcohol, psychiatric comorbidities, adverse childhood experiences

Heavy drinking and recurrent depression has long been considered predictors of IPV.[35] We divided the perpetrators in our study group into four categories depending on the presence or absence of alcohol dependence and the presence or absence of anxiety/depression and ran an association with each of the childhood adversities. We found that for almost all of them the proportion of individuals who perpetrated violence without alcohol dependence or anxiety/depression was the least. We found that the proportion of abusers who were suffering from alcohol dependence along with anxiety/depression was very high. Certain adversities' namely verbal punishment, physical punishment, bullying, community violence, caretaker of sibling, and household work was significantly associated with male perpetrators of domestic violence. Moreover, when the associations with the dosing of ACE were analyzed, we found that there was a steep increase in the proportion of abusers who were alcohol dependent and diagnosed with anxiety/depression as the dose of the adversities increases from low to medium to high (P = 0.010 and 0.001, respectively).

Poor relationship at home, especially maltreatment in childhood, is associated with later depression.[36] The relationship between alcohol dependence and other psychiatric comorbidities is complex. First, some alcoholic patients may develop a major depressive disorder or anxiety in their life without any relationship with their alcoholism (accidental co-occurrence). Second, the vast majority of alcoholic patients have experienced depressive and anxiety symptoms, especially during withdrawal. Finally, some alcoholics may present a severe depressive episode that require a specific treatment, the alcoholism being explained by the depressive disorder.[37] Based on previous research, we propose four possible explanations for our findings:

ACEs could be a predictor of alcohol dependence, psychiatric morbidities and domestic violence independently

ACEs could be a predisposing factor for psychiatric morbidities which could lead to substance abuse and dependence which in turn could perpetuate domestic violence

ACEs could predispose to alcohol dependence which can lead to psychiatric comorbidities which in turn increase violent behavior

ACEs could predispose to alcohol dependence which independently increases the chances of psychiatric comorbidities and domestic violence.

Limitations

The major limitation of the study was the small sample size and cross-sectional nature of the study.

A largescale prospective study following a cohort from childhood till a few years after marriage would have been ideal for determining causal association.

CONCLUSION

This community-based descriptive study in an urban slum revealed that the male perpetrators of domestic violence experienced an extremely high number and multiple types of ACEs. The proportions of alcohol dependence, anxiety, and depression were also found to be higher than that of the general population. There was a strong association between the presence and severity of alcohol dependence and the more severe complicated type of domestic abuse. A few of ACEs, namely witnessing parental abuse, suffering harsh extreme physical punishment, were also associated strongly associated with the complicated type of domestic abuse. The more the number of ACEs suffered by the abusers, the more the chances of them having alcohol dependence and comorbid anxiety or depression. Rather than only focusing on managing the consequences of domestic violence on its victims, our study makes a strong case for paying attention to factors in male perpetrators, which could make them more likely to abuse their spouses. Community-based strategies to prevent possible adversity in children and mental health interventions to curb and restrict alcohol-related problems and other psychiatric morbidities are needed to break the intergeneration cycle of domestic violence.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.