[acb-hsp] Domestic Violence Among Male & Female Victims

[J.Rayl]

Domestic Violence among Male and Female Patients Seeking Emergency Medical Services

by Mary Beth Phelan , L. Kevin Hamberger , Clare E. Guse , Shauna Edwards

It has been well established that significant health risk exists for victims, perpetrators

and witnesses of domestic violence (Edleson, 2000; Koss & Heslet, 1992). Clinical

and research efforts largely have been focused on the identification and care of

victims of domestic violence, primarily women (Eisenstat, 1999; Hamberger, Ambuel,

Marbella, & Donze, 1998; Johnson & Elliott, 1997; Kyriacou et al., 1999).

Although controversial, the idea that women commit acts of intimate partner violence

against their male partners has been studied in various ways since Straus, Celles,

and Steinmetz published the first national study of partner violence prevalence (1980).

In fact, in a recent meta-analysis of over 80 studies, Archer (2000) summarized the

research by reporting that women were slightly more likely than men to report using

physical aggression against their intimate partners. This trend is echoed in recent

clinical research that has also begun investigating men's and women's use of intimate

partner violence (Cantos, Neidig, & O'Leary, 1993, 1994; Hamberger & Guse, 2002;

Vivian & LanghinrichsenRohling, 1994). Other research also has begun to investigate

men who have been injured in domestic disputes and who sought emergency medical care

(Mechem, Shofer, Reinhard, Hornig, & Datner, 1999), and the perpetration histories

of men who self-report as victims (Muelleman & Burgess, 1998). Still other research

has begun to investigate the prevalence of male partner violence perpetrators in

health care settings (Oriel & Fleming, 1998). In fact, similar to results observed

in national survey studies, research with clinical samples that has investigated

prevalence of bi-directional partner violence has found that violence perpetrated

exclusively by a single partner is the most uncommon pattern. About 80% of men and

women in clinical samples with partner violence report a bi-directional pattern (Cantos,

Neidig, & O'Leary, 1993, 1994; Cascardi, Langhinrichsen, & Vivian, 1992; Langhinrichsen-Rohling,

Neidig, & Thorn, 1995).

A superficial analysis of research on men's and women's use of intimate partner violence

would suggest that the violence is mutual, and that both men and women who report

experiencing violence from their partners are equally victimized, and should be treated

as such. However, other research with clinical samples is beginning to demonstrate

important gender differences with respect to contextual issues and outcome of the

violence. For example, in the area of motivation for committing partner violence,

it appears that men more frequently intend abusive behaviors to control or frighten

(Barnett, Lee, & Thelen, 1997), and dominate or punish a partner (Hamberger, Lohr,

Bonge, & Tolin, 1997), while women more frequently report assaulting their partners

to protect themselves, escape their perpetrator, or in retaliation for prior violence

(Barnett et al.; Hamberger et al.).

The physical impact of partner violence on men and women appears to differ as well.

In particular, women are more likely than men to be injured by an assault from an

intimate partner (Cantos et al., 1993, 1994; Cascardi et al., 1992; Cascardi & Vivian,

1995; Langhinrichsen-Rohling et al., 1995). Further, women appear to be more severely

injured than men, and to seek medical care more often for injuries resulting from

domestic assault (Cantos et al., 1993, 1994; Cascardi et al.; Langhinrichsen-Rohling

et al.). This may be the result of gender differences in physical size and strength,

socialization to use violence, frequency and severity of injuries, and/or the use

of verbal or emotional abuse and dominance and isolation tactics (Coben, Forjuoh,

& Gondolf, 1999; Ganley, 1989; Pagelow, 1984). Additionally, women who are victims

in domestic disputes are far more likely to be subjected to more severe violence

(Hamberger & Guse, 2002) and execute more requests for law enforcement intervention

than men during these encounters (Barnett, Lee, & Thelen, 1997; Hamberger & Guse,

2000).

Another impact variable that has begun to receive study is fearfulness. If the ultimate

impact of battering and abuse is domination and control, then one important mechanism

by which such control is accomplished may be through fear induction in the victim

(Hamberger & Guse, 2002). Hence, analyzing the degree and impact of fear in men and

women who experience intimate partner violence is an important area of study. Langhinrichsen-Rohling

and coworkers (1995) studied the impact of fear using a single item to which respondents

provided a Likert-type rating of fearfulness of their partner. Women scored significantly

higher than men on this measure of fear. Jacobson and colleagues (1994) studied the

interactions of men and women in violent relationships within a controlled laboratory

setting. The researchers found that women were significantly more fearful during

intense arguments than men, and that during these arguments, men were not fearful

at all. Barnett and Thelen (1995) compared women in shelter programs with a demographically

matched sample of men in abuse abatement programs and found that men were significantly

more likely than women to report frightening their partners as an emotional outcome

of physical aggression. More recently, Hamberger and Guse (2002) reported that, among

men and women who used violence toward their intimate partners, women reported significantly

higher levels of fear of their partners' physical aggression than men. In addition,

compared to women, men reported being amused and likely to laugh at their female

partners' initiated violence than vice versa. Hence, it appears that in relationships

where there is violence, women are more fearful, generally, of their partners than

are men. In addition, women are more fearful of their partners' violence and aggression

than are men. Moreover, men actually appear to report feeling amused by at least

some of their partners' initiated violence, whereas women do not.

Taken as a whole, the above review suggests that although women, both generally and

in clinical samples, report perpetrating violence against their male intimate partners

at rates similar to men, there are important gender differences in the motivation

for using such violence, as well as physical and emotional impact of the violence

itself. In particular, it appears that, within clinical samples of men and women,

women are more likely than men to report using partner violence for purposes of self-defense

and retaliation, whereas men are more likely to report using violence to dominate

and control their partners. In addition, women report more injuries than men that

result from partner violence. Further, compared to men, women are more likely to

seek health care services for abuse-related injuries than are men. Finally, women

report higher levels of fear in the context of intimate partner violence than do

men.

These data serve as an important reminder that it is not sufficient to merely catalogue

and consider rates and severity of partner violence when trying to understand gender

differences in the commission of acts of partner violence. Rather, as asserted by

Vivian and Langhinrichsen-Rohling (1994) and Vivian and Malone (1997), it appears

that to fully understand partner violence, it is important to assess gender differences

in the context and impact of intimate partner violence, such as injury, fear, and

motivation or function of violence.

With significant differences in outcomes and contextual issues identified between

men and women involved in violent relationships, men and women may also likely have

differences in treatment and referral needs. This point, as well as those made by

Vivian and Malone (1997), is best illustrated by comparing two clinical studies evaluating

men injured in domestic disputes. The first, by Mechem and colleagues (1999), surveyed

all male patients presenting to an urban emergency department (ED) and found that

12.6% of these men had experienced violence perpetrated by a female partner during

the year preceding the ED visit. The screening survey was brief and did not include

questions that explored important contextual issues surrounding the violence. The

authors concluded that further research evaluating violence perpetrated by women

against men should be undertaken, and that efforts to identify men who are experiencing

physical violence within relationships should be increased. The authors acknowledged

that not exploring the contextual issues surrounding the violent events was a limitation

of the study. Nevertheless, at least one potential implication of the findings of

Mechem and coworkers is that many men who attend EDs are victims of partner violence,

therefore leading to the conclusion that men should be screened for their victimization

experiences. However, without closer examination of the context and impact of violence,

it is premature to conclude that the men identified by Mecham and coworkers (1999)

are partner violence victims in need of victim advocacy services.

A separate study of men seeking ED services took some of these contextual issues

into consideration, and obtained important information that qualifies men's experiences

with partner violence. Muelleman and Burgess (1998) examined prior domestic violence

arrests of men identified as victims of violence within a relationship. Using a case-controlled

analysis, the authors noted that 51 % of the case patients who were identified by

the ICD9 code of "adult maltreatment syndrome" had prior arrests for domestic violence

perpetration, versus 20% of the control patients (p = .009). They determined that

men who present with injuries inflicted during a domestic dispute have a high rate

of prior arrest for partner violence perpetration. These authors also conclude that

a more in-depth look at the male patient injured in a domestic dispute is necessary,

as this patient may not be a primary victim, but rather a perpetrator injured as

the result of his female partner's attempts at self-defense. Hence, treatment referrals

for many such men would be to abuse abatement treatment rather than victim advocacy.

In order to further delineate the issues surrounding men and women who disclose involvement

in violent relationships in the clinical setting, the present study aimed to survey

injured patients attending an ED about their experiences with incurring and perpetrating

physical partner violence, as well as key contextual issues surrounding the violence,

such as the psychological and behavioral impact of the violent events experienced

from the intimate partner. Emotional impact included feelings of anger, fear, amusement,

insult, and feelings of intimidation by the partner. Behavioral impact included using

force back, calling police, escaping, laughing, threatening violence, and acquiescing

to partner's demands. In addition, respondents were queried about injury prevalence,

frequency, and severity. Finally, respondents provided information about emotional

abuse and partner behaviors considered dominating and isolating.

On the basis of the above review, the following hypotheses were tested about men

and women involved in heterosexual, violent relationships during an evaluation of

a cohort of injured ED patients:

1. Respondents of both genders would report that men are more likely than women to

initiate violence.

2. Women would report being injured more frequently than men and incurring more severe

injuries than men.

3. Women would report higher levels of fear than men in the context of a partner-initiated

assault.

4. Women would report being physically intimidated by the size of their partner at

a higher rate than men.

5. Women would report higher rates than men of experiencing partner behaviors intended

to dominate and isolate.

METHOD

Study Design

This was a cross-sectional survey of a convenience sample of injured male and female

patients presenting to a Level I trauma center. Study subjects were administered

a survey that inquired about interpersonal relationship violence by a trained research

assistant.

Study Setting and Study Participants

The study was conducted at an adult, Midwestern, urban, Level I trauma center that

evaluates approximately 42,000 emergency patients per year. During designated study

times, English-speaking male and female patients over the age of 18, and who were

seeking services for a traumatic injury, were eligible for study inclusion. Patients

who were hemodynamically unstable, head injured, intoxicated, victims of motor vehicle

crashes or unable to provide informed consent were excluded. Stable victims of major

trauma (e.g., stab wounds) were included, following initial evaluation, stabilization,

and permission from the treating physician. Because the intent of the study was to

determine gender differences in patients experiencing domestic violence in the clinical

setting, patients considered to be at high risk for domestic violence were sought.

Injured patients were chosen because of the correlation between this variable and

acute domestic violence involvement for women (Hartzell, Botek, & Goldberg, 1996;

Kyraicou et al., 1999; Le, Dierks, Ueeck, Homer, & Potter, 2001; Perciaccante, Ochs,

& Dodson, 1999; Zachary, Mulvihill, Buton, & Goldfrank, 2001). Men with any type

of injury seen in the ED also have been noted to have had prior arrests for domestic

violence (DV) perpetration (Muelleman & Burgess, 1998). Data were collected during

two separate study periods, June to August 2000, and June to August 2001. Initially,

the study was designed to assess men's experiences with partner violence. Subsequently,

we hypothesized that a gender comparison would provide valuable information. Thus,

women were added to the study in June 2001.

Survey Content and Administration

After obtaining informed consent from eligible patients, the assistants administered

a survey, in the form of a 30-minute, structured interview. The interview schedule

consisted of three main parts. The first part, adapted from Hamberger and Guse (2002),

sought to obtain information pertaining to participant demographics, including race,

age, education level, and occupational status. In addition, respondents answered

questions about the respondent's own use of force in relationships and episodes of

adult intimate victimization, as well as the emotional and behavioral impact of the

partner's initiated violence in a current adult, heterosexual relationship. Specific

emotional impact variables included feelings of anger, fear, insult, and amusement.

Specific behavioral impact items included use of force back, call police, threaten

violence, escape, and acquiesce to partner's demands. Impact items were responded

to in a Likert-type format. For emotional impact, ratings ranged from I, "do not

experience that emotion," to 5, "experience that emotion very strongly." Behavioral

impact items were rated from 1, "I never do that in response to my partner's initiated

violence," to 5, "I always do that in response to my partner's initiated violence."

In addition, respondents were asked to describe if they felt intimidated by their

partner's physical size. Further, respondents answered questions related to past

year frequency of violence victimization, frequency and severity of injury, as well

as whether they had ever sustained a DV injury.

Conflict Tactics Scale. Violence victimization and perpetration were assessed using

a modified version of the Conflict Tactics Scale (CTS), which was administered as

part of the structured interview. The CTS (Straus, 1979), is one of the most widely

used measures of partner violence. Its psychometric characteristics are well documented

(Straus, 1990). A major strength of this tool is its ability to record physical aggression

perpetrated or experienced on a scale of increasing severity. Increasing severity

of tactics has been shown to correlate with increasing injury severity (Coben et

al., 1999). The CTS has been criticized, however, for not taking into account the

contextual issues of the violent event. For the present study, we developed a survey

that combined the strengths of the CTS, with focused contextual questions, as noted

above. In addition, questions about violence initiation were asked in two different

ways. First, respondents were asked, generally, whether they and their partners had

ever initiated physical violence in the relationship. Second, following questions

about the overall frequency of physical violence in the relationship, respondents

were asked to report the percentage of incidents initiated by themselves and their

partners.

Psychological Maltreatment of Women Inventory. The Psychological Maltreatment of

Women Inventory short form (PMWI, Tolman, 1999) is a 14-item instrument that was

modified for use with both women and men to detect perceived psychological abuse

experienced by the patient within a heterosexual adult relationship. The PMWI was

further modified in response format from a 5-point Likert scale to a yes/no format

to facilitate the interview process. As with the CTS, the PMWI was integrated into

the structured interview. The PMWI consists of two, factorially validated subscales.

One subscale measures verbal abuse. The other subscale measures behaviors related

to domination/isolation. The latter subscale was used in the present study as a proxy

for function of DV. That is, gender differences on the domination/isolation subscale

was used to determine sex differences in experiencing partner behaviors that served

to dominate and/or isolate the respondent.

Procedure

Medical students were trained as research assistants by the investigators to conduct

the structured interview with patients in private. The research assistants worked

random shifts (day (7 am-3 pm), pm (3 pm-11 pm), and night shifts (11 pm-7 am)).

They wore a pager that was activated by ED staff whenever a patient with major trauma

arrived in the department. The research assistant reviewed the patient's record to

determine whether any of the exclusion criteria outlined above were present. If the

patient appeared to be eligible for the study, the assistant approached the patient,

ascertained privacy, and made a final determination of eligibility. Eligible patients

were given a detailed verbal description of the study. If the patient agreed to participate,

they read and signed an informed consent form and participated in the structured

interview, which lasted for about 30 minutes. If the patient was accompanied by family

members or others, efforts were made to excuse them for the duration of the study.

If that was not possible, the patient was not recruited for the study. Informed consent

forms and interview schedules were stored separately from the patient's medical record,

and their treating professionals were not given information about responses to the

interview. Following interview completion, if the respondent had reported relationship

violence, they were given information about appropriate community referrals for domestic

violence intervention. Support services from the hospital Social Services Department

were offered to all patients who requested these, and to any patient who experienced

emotional upset or trauma as a result of discussing their experiences.

Data Analysis

Tests of univariate associations between demographic and other background data, with

measures of violence severity, frequency, emotional impact, and injury frequency

and severity, were conducted using chi-square, Fisher's exact test, or exact Kruskal-Wallis

tests as appropriate for unordered and ordered categorical data. All analyses were

conducted with either Stata 7.0 or StatXact 2.11 (Cytel Software Corporation, 1992;

StataCorp, 2001).

RESULTS

One hundred fifty-seven patients were approached for participation in the study,

with 129 (90 men and 39 women) agreeing to participate (82.2%). Of these participants,

94 (65 men and 29 women) were in heterosexual relationships. Of those individuals

currently in relationships, 34 were experiencing violence within the relationship

(23 men and 11 women). These individuals comprised the study group.

Demographic characteristics of study participants are summarized in Table 1. The

women in the study group were primarily African-Americans (82%). Nine percent of

the women were Caucasian, and another 9% were Hispanic. The demographic makeup of

the men was African American (52%) and Caucasian (43%). With respect to relationship

status, 13% of men and 18% of women were married, 61% of men and 27% of women were

cohabiting, and 17% of men and 36% of women were dating. Eight percent of men and

18% of women were separated at the time of the study.

Employment status of the individuals in the study was as follows. Forty-eight percent

of men had full-time employment versus 18% of women. Part-time employment was reported

by 17% of men and 18% of women; 55% of women in the study describe being unemployed

as compared to 13% of men. Seventeen percent of men were disabled, and no women in

the study reported being disabled. Four and 9% of the men and women (respectively)

had no employment data captured.

Data regarding past abuse history was obtained. Fifty-five percent of women and 43%

of men reported witnessing parental fighting. Large differences between men and women

existed for prior history of emotional abuse (26% vs. 100%, p < .001), physical abuse

(26% vs. 91%, p < .001), and sexual abuse (9% vs. 36%, p < .07, all Fisher's exact

test, respectively).

A statistically significant difference in violent tactics experienced by women and

men was also noted. One hundred percent of the women and 29% of men indicated that

their partner utilized one or more of the most severe forms of violence from the

modified CTS. In particular, 30% of women, compared to 5% of men, reported their

partner beat them up. Fifty percent of women, compared to 5% of men, reported their

partner choked or strangled them. Finally, 20% of women and 14% of men indicated

that their partner used a knife or a gun in an attack (p < .008, exact chi-square

test).

Sex differences were assessed for rates of initiation of partner violence in two

ways. First, respondents were asked if their partner had ever struck first in a violent

episode. The group difference was marginally significant and in the predicted direction.

Ninety-one percent (10/11) of the female respondents indicated their partner had

ever struck first, compared to 52% (12/23) of the male respondents (p = .07, exact

chi-square test). The second method for assessing rates of partner violence initiation

asked respondents what percent of the total episodes of physical fights were initiated

by themselves and by their partners. Results of self-reported initiation were significant

(p = .0034, exact Kruskal-Wallis test). Only one woman (9%) responded that she initiated

a violent episode 100% of the time. Further, only 36% of the female respondents reported

initiating violence between 10% and 20%. Fifty-five percent of female respondents

reported not initiating any violent episodes. In contrast, 100% (n = 5) of the men

admitted to initiating a violent episode, with the range being between 50% and 100%

of the incidents. In addition, similar findings were observed for respondents reporting

on the percent of time their partners initiated violence. Specifically, while one

female respondent (9%) reported that her partner never initiated violence, 91 % of

the female respondents reported that their partners initiated violence at least 80%

of the time. In contrast, 60% of the male respondents reported that their partners

initiated violence 50% of the time, and 40% indicated that their partners initiated

violence between zero and 40% of the time (p = .0034, exact Kruskal-Wallis test).

These data are summarized in Table 2.

Sex differences were also assessed in behavioral responses to partner's initiated

violence. Compared to men, women were more likely to report using force back and

calling police. Regarding using force back, only 30% of the women stated that they

never or almost never did so, compared to 74% of the men. Further, while 40% of the

women reported always or almost always using force back when their partner initiated

violence, none of the men reported doing so (p = .003, exact Kruskal-Wallis test).

Women were also more likely than men to call police in response to their partners'

use of violence. Specifically, whereas 95% of the men never or almost never called

police, 50% of the female respondents reported always or almost always doing so (p

= .001, exact KruskalWallis test).

Injury and frequency of injury as the result of domestic assault were recorded and

are summarized in Table 3. One hundred percent of the female respondents versus 39%

of the male respondents reported being injured in a domestic assault at some point

in their life (p < .001, Fisher's exact test). Thirty percent of the male respondents

reported that they had been injured in the past year by a female partner, while 100%

of the women in the study indicated that they had been injured by a male partner

within this same time frame (p = .0002, exact Kruskal-Wallis test). Of those respondents

injured in a domestic dispute in the last year, women reported a higher frequency

of injury than men. When injuries experienced in the past year are categorized into

four mutually exclusive categories by the most severe level experienced (range 0

= none to 4 = severe), higher proportions of women appear at every level above "none,"

compared to men (p = .0009, exact Kruskal-Wallis test). Twenty-six percent of male

and 73% of female respondents sustained their current injury in a domestic dispute.

Women were significantly more likely to have sustained their current ED injury due

to a domestic assault (p = .02, Fisher's exact test).

Women reported being intimidated by their partner's size (36%), while none of the

men reported being intimidated by their partner's size. Women were also more likely

to report feeling fearful when their partner initiated violence. Seventy percent

of women responded that they were very strongly afraid during partner-initiated violence,

while only one man reported experiencing this degree of fear (p = .0001, exact Kruskal-Wallis

test). The majority of men (85%) reported not being afraid at all when their female

partner initiates violence. See Table 4 for a summary.

Results from the modified PMWI showed that both men and women experienced verbal

and emotional abuse, with some items on the inventory being near equal (Table 5).

Women, however, did report experiencing more overall emotional abuse when the items

were summed. Eighty-two percent of women reported suffering six or more of the verbal/emotional

abuse items versus 22% of men. Women were far more likely than men to experience

domination/isolation techniques, with 64% of women reporting experiencing six or

more of these behaviors versus 22% of men (p = .03, Fisher's exact test). See Table

6 for a summary.

DISCUSSION

The present study investigated five hypotheses related to men's and women's expression

and experience of intimate partner violence. In particular, the study assessed the

relative frequency with which men and women initiate intimate partner violence, inflict

injuries, experience fear and intimidation, and are subjected to dominating/isolating

behaviors of their partners. The degree to which each of the hypotheses was supported,

and their implications, are discussed subsequently.

Hypothesis 1

Hypothesis 1, that both male and female respondents would report that men initiate

partner violence at higher rates than women, was supported. Both male and female

respondents disclosed that they initiated violence within an intimate relationship;

however, significant differences were observed along gender lines, supporting our

hypothesis that both men and women would identify that men initiated partner violence

at higher rates than women. Overall, a trend was observed suggesting that men initiate

violence at higher rates than women. In addition, when evaluating percentage of total

incidents initiated by women or men, the data further suggest that women initiate

violence at far lower rates than men. In particular, both men and women were in agreement

that, on a percentage basis, men were more likely than women to initiate violence.

The women in our study reported that over 80% of the time their male partner initiated

violence during a dispute, with 55% of the women stating that men initiated the violent

episodes 100% of the time. Men selfreported initiation rates closer to 50%. But the

lower initiation rate notwithstanding, like women, men identified themselves, as

a group, as initiating violence at higher rates than women. Further, no men reported

female partner initiation rates above 50%.

The fact that our study also detected that women were more likely than men to use

force back or call police when their partner initiates violence during a dispute

further supports the likelihood that much of the violence experienced by men in a

domestic dispute is the result of their partner's attempts at self-defense and retaliation,

rather than an overt attempt to control a partner. A similar conclusion has been

drawn from another study (Hamberger & Guse, 2002).

In fact, the present results are quite similar to those reported by Hamberger and

Guse (2002). In particular, both the present study and Hamberger and Guse found that

although women involved in violent relationships readily report using violence, when

asked to describe their own and their partners' violence initiation in terms of relative

frequency, or percentages, they report initiating violence at fairly low levels,

and describe their male partners as initiating violence at high levels. Male respondents

also report initiating violence at higher rates than females, but the pattern is

different. Specifically, men tend to report the violence initiation as more "50-50."

However, even with the more mutualappearing report of men, no men reported their

partners as initiating violence more than half of the time.

Nevertheless, it is difficult to explain the apparent sex differences in percentage

estimates. It could be a defensive denial related to social desirability response

bias. Research does support this perspective (Arias & Beach, 1987; Dutton & Hemphill,

1992). However, research on social desirability and reports of perpetration does

not predict the sex differences reported in the present paper and by Hamberger and

Guse (2002). Another possible explanation for this sex difference is that women and

men may view episodes of fighting differently. Specifically, women may strike a partner

at some point after he has stopped his violence, but consider her use of force as

retaliation and part of an ongoing violent interaction. The man, in contrast, may

view her later use of force as initiating a new fight sequence.

Yet another possibility for this difference is that couples were not studied. Studying

partners from the same violent relationship may be a way to analyze gender differences

with respect to reporting violence initiation. More work is needed to clarify sex

differences in reports of partner violence perpetration, and particularly why women

tend to view the violence as more one-sided, whereas men tend to view it as more

mutual.

The present study also found that women were more likely to experience severe forms

of physical violence from their intimate partners than men. In particular, women

experienced higher rates of being beaten up and choked or strangled, and a trend

toward having a knife or gun used on them, than men. These findings are consistent

with those of Langhinrichsen-Rohling and coworkers (1995), as well as Cantos and

colleagues (1994). In particular, Cantos and coworkers found that husbands were more

likely to use such acts as choking or strangling and beating their partner up. Women

were more likely than husbands to kick, bite or punch, threaten with a knife or gun,

and use a knife or gun. Further, Morse (1995), in evaluating a national cohort sample,

found that men used more severe forms of partner violence than women. The issue is

not fully resolved, however. Barnett and colleagues (1997) did not find consistent

gender differences in using various forms of violence. Hence, the caution of Vivian

and Langhinrichsen-Rohling (1994) applies. Topographic characteristics of violence,

that is, frequency, form, and severity, may be less important for ascertaining gender

differences in partner violence than other factors, such as impact, context, and

function.

Hypothesis 2

Hypothesis 2, that women would report higher rates and severity of injury than men

from DV was supported. Specifically, women reported higher rates of both lifetime

and past year injury than men. In particular, 100% of the women reported having been

injured from a domestic assault at some time in their history and in the past year.

In contrast, 39% of men reported being injured at some time in their life, and 30%

reported a DV injury from their partner in the past year. Even though the rates of

injury for male respondents are quite high, the data clearly support the notion that

women are more likely to be injured from a domestic assault by their partner than

vice versa. If pain and injury are key components of fear induction and establishment

of control in a relationship (Hamberger & Guse, 2002), then the present results lend

support to the idea that in relationships where partner violence is bi-directional,

it is nevertheless not mutual (Vivian & Langhinrichsen-Rohling, 1994).

The current study also supports other data linking injury frequency with increasingly

severe tactics on the CTS (Coben et al., 1999). The women in our study reported experiencing

higher levels of violence perpetrated by their male partners. Women experienced three

severe violence tactics (multiple blows, choked or strangled, and partner used a

knife or gun) more frequently than men. This corroborates prior research that demonstrated

that men, in general, use higher levels of force within violent relationships than

do women (Coben et al.; Morse, 1995) and cause more injury (Cantos et al., 1993,

1994; Langhinrichsen-Rohling et al., 1995).

When injury severity was examined, higher numbers of women reported more injuries

than men in each severity category. The results of the present study are similar

to a number of studies that have found that, compared to men, women who sustain DV-related

injuries also sustain more severe injuries (Cantos et al., 1993, 1994; LanghinrichsenRohling

et al., 1995). The latter studies defined seriousness as having sought medical attention.

In the present study, injury severity was defined by specific injury markers, for

example, bruise, laceration, broken bone, etc. Hence, it appears that men and women

differ in DV injury severity even when injury is measured in different ways.

Results of the present study related to injury severity differ from those reported

by Hamberger and Guse (2002) and Cascardi and Vivian (1995). In these prior studies,

both women and men reported roughly equivalent rates of seriousness of injury. Hamberger

and Guse investigated men and women involved in community abuse intervention services,

and Cascardi and Vivian studied couples seeking marital therapy. Neither group was

selected for health care seeking. Further, Hamberger and Guse used a fairly crude

marker of injury, whereas the present study used an injury index similar to that

used in other DV injury research (Hamberger, Saunders, & Hovey, 1992).

Interestingly, though this study evaluated systematic samples of injured patients

presenting to an ED, 73% of the women in the study group reported that they had sustained

their current injury in a domestic dispute. Only 26% of the men sustained their injuries

in domestic disputes. Despite the small sample size, this was statistically significant.

This finding is congruent with other injury related data that demonstrates gender

differences in mechanism for the same injury. Hartzell and coworkers (1996) noted

that one-third of orbital blow-out fractures (fractures of the orbital wall) in women

were sustained as the result of a domestic assault. No men in their study group with

orbital blow-out fractures incurred this injury in a domestic assault. Clearly, more

work is needed to determine the relationship between gender and injury severity.

Areas of such continued research include determining the most appropriate injury

measurement methods, as well as the types of patients for which injury is most likely

to differentiate the sexes.

Hypotheses 3 and 4

Hypotheses 3 and 4, that compared to men, women would report higher levels of fear

toward their partners' initiated violence and experience more intimidation at their

partners' physical size, were supported. Regarding fearfulness toward partners' initiated

violence, 70% of the women reported being very frightened of their partners' physical

violence whereas 85% of the men reported experiencing no fear when their female partners

initiated violence. Indeed, only one man reported feeling very fearful when his partner

initiated violence. These results are highly consistent with previous research that

has investigated fearfulness as an emotional impact of partner violence (Barnett

et al., 1997; Hamberger & Guse, 2002; Jacobson et al., 1994). Indeed, across studies,

sex differences in fearfulness of partner's violence are typically large. One possible

explanation for this consistent, large sex difference is that it is socially unacceptable

for men to acknowledge their fear, and that men may be denying their fearfulness

out of social desirability. To date, no study has attempted to control for social

desirability in examining gender-based fearfulness of an intimate partner's initiated

violence. Hence, this is an area of study that remains to be conducted. Nevertheless,

prior research on the relationship between social desirability and self-reports of

violence victimization may be instructive in this area. Research in this area has

consistently found that reports of victimization, for males and females, are not

associated with socially desirable response bias (Arias & Beach, 1987). Hence, in

reporting levels of victimization, males appear to be similar to females in that

such reports may not be overstated or minimized in systematic ways that vary by sex.

Similar results were found for feelings of intimidation regarding a partner's physical

size. Over one-third of the women reported feeling intimidated by their partner's

physical size, whereas none of the men reported such intimidation. These results

appear to be similar to those reported by Langhinrichsen-Rohling and coworkers (1995)

that found that women reported greater fearfulness than men of their intimate, domestically

violent partners. Present results are also consistent with the work of Pagelow (1984)

that found male partners in violent, intimate relationships to be physically larger

and stronger than females, and thus more able to use force as a mechanism to induce

fear and control. It may be, then, that differential physical size, together with

the greater tendency of men to actually initiate physical violence and inflict injury,

combine to increase the salience of fear and intimidation for females, much more

so than for males. More detailed information could be obtained by adding questions

regarding physical fitness, physical ability, or the partner's perceived physical

ability to control.

Hypothesis 5

Hypothesis 5, that women would report higher rates than men of experiencing partner

behaviors that function to dominate and isolate, was well supported. Specifically,

while 30% of male respondents reported either no or one form of domination/isolation

from their female partners in the past year, only 9% of the female respondents reported

experiencing no or one form from their male partners. In contrast, 22% of male respondents

reported experiencing six or seven domination/isolation tactics from their partners

in the past year. However, 63% of the female respondents reported experiencing at

least six domination/isolation tactics from their male partners in the same time

frame. These data suggest that, while some men do experience efforts by their partners

to dominate/isolate them, such experiences are much more prevalent among women involved

in violent intimate relationships. Although the present study did not directly assess

respondent motivation for committing acts of physical partner aggression, these data

do provide some insight into gender differences in the experience of the function

and impact of the partner's aggression upon the respondent. However, these findings

lend support to those of Barnett and coworkers (1997), who found that men reported

using violence to show dominance (show who was the boss) and teach her a lesson.

Given that physical violence is part of a more general pattern of domination and

control, it makes sense that women who experience more initiated violence, more injury,

fear, and intimidation would also experience other behaviors from their partners

as dominating and isolating, compared to men. That is, women experience more pervasive

domination and control from their partners than do men, even in relationships where

the violence is bi-directional. This finding is consistent with that of Cascardi

and Vivian ( 1995), who studied couples with bidirectional violence and reported

that wives consistently reported husbands to engage in significantly more psychological

coercion and abuse. In addition, women studied by Cascardi and Vivian viewed their

own violence as more defensive than did the men.

Implications

These findings have several implications for understanding partner violence, generally,

and for applying such understanding to developing and implementing health care systembased

interventions. First, in terms of general knowledge of men's and women's use of partner

violence, the present findings add to a growing body of literature, primarily emerging

from study of clinical samples, that as a group, men and women experience intimate

partner violence differently. In particular, compared to women, men, as a group,

tend to initiate violence more frequently, inflict more injury, and use more severe

forms of violence than women. In contrast, compared to men, women, as a group, report

more fearfulness and intimidation, and experience more dominating and controlling

behaviors from their partners. Because the present study, as well as the literature

based on clinical samples, generally, is limited only to those who seek help for

partner violence, this conclusion is not generalizable to the general population.

However, the consistency of findings across clinical samples as diverse as those

seeking marital therapy, abuse-related services, and emergency medical care, does

add an element of convergent validity to our understanding of partner violence across

clinical samples and populations. Further, as pointed out by Hamberger and Guse (2002),

research methodology developed to investigate gender differences in forms, contexts,

and impacts of intimate partner violence in clinical samples can be adapted to general

population studies to address more global, theoretical questions.

Another implication of the present findings relates to methodology for studying partner

violence in clinical settings. Given the wealth of data from a growing number of

studies that have investigated gender differences in context and impact of partner

violence, the standard has been set for continued research in this area. Specifically,

it will not suffice to conduct studies that merely catalogue gender-based prevalence

rates of partner violence (e.g., Mecham et al., 1999). Rather, it will be necessary

to document the differential physical and emotional impact of partner violence, including

forms of emotional control and coercion, as well as motivations and associated behavior

patterns related to both perpetrating and experiencing partner violence for men and

women. For the field to continue to progress through both continued data gathering

and reduced polemical debate, it will be necessary for researchers to utilize methodologies

that provide the greatest opportunity to clarify both the dimensions and limits of

sex differences in the commission and experience of intimate partner violence, in

all of its forms.

Results of the present study also have implications for clinical practice. The history

of screening for DV in medical practices has been, up to the present, one of screening

for female victims, and conducting appropriate interventions and referrals (Ambuel,

Hamberger, & Lahti, 1997). The present study found that women were more highly exposed

to severe forms of violence than men, and also experienced more severe injury. These

findings suggest that it is feasible to consider that the number and severity of

injuries to women may decline if interventions aimed at reducing violent behavior

in men are initiated early in the identification process. There have been increasing

calls to screen men for DV, as well (Oriel & Fleming, 1998). Although men surveyed

in medical settings do report being victimized (Gin, Rucker, Frayne, Cygan, & Hubbell,

1991; Mecham et al., 1999), data from the present and other studies (Barnett et al.,

1997; Cascardi et al., 1992; Hamberger & Guse, 2002; Muelleman & Burgess, 1998) clearly

show that, as a group, men involved in violent relationships, even when their partners

assault them, appear to be predominant aggressors (Hamberger & Guse). Therefore,

health care-based screening initiatives that are designed for identifying men involved

in partner violence will need to develop different screening, assessment, and community

referral components that are unique to the male experience. Specific examples could

include screening men directly for DV perpetration. Alternatively, screening of men

could begin with questions about victimization, with positive reports followed-up

with further assessment of context and impact of partner violence to determine predominant

victim or perpetrator status. It remains for further research to determine the most

optimal ways to screen men to determine both victimization and perpetration experiences.

It should be pointed out that data from the present study do not support the idea

that all men involved in DV are predominant perpetrators. Rather, a small, but palpable,

percentage of such men from the present study, as well as other studies in this area

(e.g., Cantos et al., 1993, 1994; Langhinrichsen-Rohling et al., 1995; Vivian & Langhinrichsen-Rohling,

1994), appear to be predominant partner violence victims. Hence, careful, individual

evaluation and assessment will be necessary to determine the most appropriate treatment

options for men who report experiencing partner violence (Hamberger & Potente, 1994).

It is also important to note, that in order to ensure that perpetrator identification

is beneficial to families and to the community, appropriate treatment programs must

be in place.

Another implication of the present findings, together with the emerging body of findings

related to gender-specific contexts and impact of partner violence, relates to training

health care providers to screen and assess men (and some women) for perpetration.

This prospect raises important questions, as health care providers frequently have

difficulties asking patients about victimization experiences due to discomfort with

the topic (Minsky, Pape, & Hamberger, 2000; Parsons, Zaccaro, Wells, & Stovall, 1995;

Sugg & Inui, 1992). It is reasonable to expect, therefore, that training health care

providers to screen male patients for perpetration will require considerable effort

to address provider concerns for safety, as well as provider attitudes toward perpetrators,

in addition to provision of requisite skills for screening, assessment, and referral.

Limitations

The present study has a number of limitations that require caution in drawing firm

conclusions from the findings and call for continued research in this area. First,

the overall sample size of the present study is rather small, with some cells having

as few as five respondents. Small sample sizes can be problematic for several reasons.

First is the question of statistical power. Typically, with smaller sample sizes,

group differences must be considerably larger to detect significant differences.

Failure to find significant differences with small samples leads to a question as

to whether the failure is due to no actual group differences or due to insufficient

power to detect real differences. In the present study, numerous differences were

observed, all in the predicted direction. Hence, although it will be important to

continue replication of this type of research with larger samples, the findings reported

in the present study suggest that sample size was sufficient to demonstrate gender

differences in partner violence impact and context.

A second limitation of the present study is restricted sample range. Small samples

may not represent the broad spectrum of patients who attend the ED for medical services,

thus limiting generalizability. With the present study, increasing sample size would

not have affected sample range, as we intentionally sampled only those patients who

attended the ED with a traumatic injury. Hence, we did not attempt to recruit a sample

representative of the entire ED patient population, only those who were injured from

a source other than car crashes. It is well known from prior research with ED samples

that many, if not most, patients who acknowledge DV in their lives are not, at the

time of the survey, seeking services for traumatic injury (Abbott, Johnson, Koziol-McLain,

& Lowenstein, 1995; Dearwater et al., 1998). However, Zachary, Mulvihill, Buton,

and Goldfrank (2001) concluded that trauma was a risk factor for acute DV in women.

Results from the present study may not generalize to patients seeking emergency medical

services. Therefore, it will be necessary to conduct research on men's and women's

experiences with DV with this portion of the emergency medicine patient population,

as well.

Results of the present study are also limited by large amounts of missing data related

to gender differences in relative initiation rates of partner violence. Although

the observed results were in the predicted direction, it is not known if the same

pattern of results would have been observed had the complete data set been available.

Concern about data integrity due to missing data on relative frequency of violence

initiation is mitigated somewhat in that the observed findings were nearly identical

to those reported by Hamberger and Guse (2002) for men and women involved in community-based

DV intervention programs. As with the present study, Hamberger and Guse reported

that women reported violence initiation patterns as highly weighted toward men as

initiating the vast majority of violent incidents, whereas men reported more of a

50-50 pattern of violence initiation between themselves and their partners. Further,

as with the present study, Hamberger and Guse also found that both males and females

reported that men initiated violence at higher rates than females. Clearly, however,

more work needs to be done with larger samples from different settings to further

understand gender differences in partner violence.

It is possible that some patients, perhaps those at great risk of experiencing DV,

were not interviewed. These individuals include those who were highly intoxicated,

those patients whose family members/friends would not leave the patient during an

interview, and/or non-English-speaking patients. The reasons for not including such

patients (lack of reliable data, lack of confidentiality, and inability to communicate)

are understandable. However, while data exist that link the potential increased vulnerability

of these patients for involvement in violent relationships, the impact or prevalence

of such patients in the clinical setting is not fully understood. Future study designs

should strive to include all potential study subjects, including those with the aforementioned

barriers.

SUMMARY AND CONCLUSIONS

The present study sought to investigate gender differences in the expression and

experience of intimate partner violence in a sample of patients seeking emergency

treatment for traumatic injury. This research is motivated in part to help clarify

possible gender differences that will have important implications for development

of policy and procedure to guide health care providers in developing appropriate,

gender-sensitive approaches to screen, assess, and refer for relevant, community-based

intervention. The study found that, beyond a superficial analysis that both men and

women report assaulting and initiating assaults upon their heterosexual, intimate

partners, important gender differences in the expression and experience of partner

violence exist. In particular, women appear to use less severe violence and initiate

fewer violent episodes than men. In addition, women are more likely than men to experience

physical injury from partner violence, and to be fearful and intimidated by their

partners' violence and physical size. Further, compared to men, women report experiencing

more domination and control tactics from their partners. Although individual exceptions

were noted, taken as a whole, the data from the present study suggest that, as a

group, women who seek emergency medical treatment for injury are more likely to be

victims of partner violence. In contrast, men who seek emergency medical care for

injury who are involved in DV are more likely to be predominant perpetrators, even

if they experience violence from their partners. These data suggest that establishing

that violence and injury during domestic disputes are frequently bi-directional necessitates

fully exploring the contextual issues surrounding these events. Without such elucidation,

these conclusions are misleading at best. At worst, men and women involved in violent

relationships will miss important opportunities for successful identification, intervention,

and treatment.

-1-

Questia, a part of Gale, Cengage Learning. www.questia.com

Publication Information:

Article Title: Domestic Violence among Male and Female Patients Seeking Emergency

Medical Services. Contributors: Mary Beth Phelan - author, L. Kevin Hamberger - author,

Clare E. Guse - author, Shauna Edwards - author. Journal Title: Violence and Victims.

Volume: 20. Issue: 2. Publication Year: 2005. Page Number: 187+. © 2005 Springer

Publishing Company. Provided by ProQuest LLC. All Rights Reserved.

Next Page

Jessie Rayl
EM: thedogmom63 at frontier.com
PH:304.671.9780
www.facebook.com/eaglewings10

"But they that wait upon the LORD shall renew their strength; they shall
mount up with wings as eagles. They shall run, and not be weary"--Isaiah 40.31
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.acb.org/pipermail/acb-hsp/attachments/20110725/7d093587/attachment-0001.html>